Author Topic: Kelvin Varley Divider [and Precision Voltage Source]  (Read 79500 times)

0 Members and 1 Guest are viewing this topic.

Offline fmaimonTopic starter

  • Supporter
  • ****
  • Posts: 165
  • Country: br
Kelvin Varley Divider [and Precision Voltage Source]
« on: November 04, 2011, 09:35:37 pm »
I'm thinking on doing a Kelvin Varley Divider, but I'm having some trouble finding some information. Is there any advantage of doing each stage with a lower resistance, as the one in Conrad Hoffman's Mini Metrology Lab or using resistances of about the same order of magnitude like the IET Labs KVD-700 (check out page 10)? Maybe lower output impedance? Is that a problem, as it depends more on the input resistors? Doing it with with a smaller set of values should be easier to match them.

One more question. In Conrad Hoffman's KVD, he choose to solder the resistors directly to the terminals, instead of using a PCB. Is that because of thermal EMF?

Thank you,
Felipe Maimon

[edit] Corrected the links
[edit2] Changed the thread title to better match the discussion
« Last Edit: November 08, 2011, 11:38:17 pm by fmaimon »
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider
« Reply #1 on: November 04, 2011, 11:31:58 pm »
 The IET KVD-700 is actually the old Fluke 720a divider, and the manual with schematic is available. It is well worth reading.

http://128.238.9.201/~kurt/manuals/manuals/Fluke/FLUKE%20720A%20Instruction.pdf

The advantages of using higher resistances is that it reduces the errors from switch contact resistance.  The divider can compensate for switch contact resistance, but not for the variation in switch contact resistance.  So if the variation is 10mOhms and you want to have accuracies in the order of 1 part in 10-7 (like the Fluke), then you want the primary decade to have a total resistance of at least 100K, which is exactly what Fluke do with its eleven 10K resistors in the first decade.

The problem with higher resistances is that you have to measure the output voltage with something, and the higher the resistance, the more minute test currents will affect the output voltage.

The output resistance of a Kelvin-Varley divide varies with the switch positions, but say it can be as high as 100Kohms, then measuring the output with a 10Mohm multimeter will cause an error of 1% which destroys your 0.00001% accuracy. So you have to decide how you are going to use the divider, what accuracy you need and design accordingly.

The rules for building a Kelvin-Varley divider are:

  • All resistors in a decade have to match closely - you do not need resistors in one decade to match another.  So if you get a pile or 10ppm resistors, you can choose the closest set for the first decade, the next closest set for the next decade and so on.

  • Each lower decade must have resistors of a value greater then one fifth the higher decade resistors. So if your first decade uses 10K resistors, the second decade must use over 2K resistors. There is nothing at all stopping you using 10K resistors for each decade, so you can buy a big pile of 10K resistors, and select them for the most accurate divider.

  • Each decade needs a parallel resistor across the whole decade divider so that the decades total resistance exactly equals twice the resistor value in the higher decade.  So if the first decade uses 10K resistors, then the second decade will have to have a parallel resistor across the whole decade adjusted so that the total resistance is exactly 20K.

The advantage about using lower resistors in the lower decade is it helps lowering the output resistance as much as possible.

So if a 10Mohm multimeter causes an error of 1%, how do you use it?  Many high end meters have input resistances much more then 1G for input voltages between +- 20V.  Or you can put a fet input low offset amplifier on the output in a inity gain voltage follower configuration and make sure the FET input currents are low enough not to cause a significant  error.  Or you can use the divider in a traditional way where you use it with a galvanometer on the output, and you adjust either the divider or the voltage that you are measuring so that the galvanometer shows zero current is flowing.
 
The following users thanked this post: jorgemef, kees0_15, Svgeesus

Offline fmaimonTopic starter

  • Supporter
  • ****
  • Posts: 165
  • Country: br
Re: Kelvin Varley Divider
« Reply #2 on: November 05, 2011, 12:18:55 am »
The IET KVD-700 is actually the old Fluke 720a divider, and the manual with schematic is available. It is well worth reading.

http://128.238.9.201/~kurt/manuals/manuals/Fluke/FLUKE%20720A%20Instruction.pdf

Thank you for the link. Looks like it's very good. I'll read more thoroughly later.

Quote from: amspire
The advantages of using higher resistances is that it reduces the errors from switch contact resistance.  The divider can compensate for switch contact resistance, but not for the variation in switch contact resistance.  So if the variation is 10mOhms and you want to have accuracies in the order of 1 part in 10-7 (like the Fluke), then you want the primary decade to have a total resistance of at least 100K, which is exactly what Fluke do with its eleven 10K resistors in the first decade.

The problem with higher resistances is that you have to measure the output voltage with something, and the higher the resistance, the more minute test currents will affect the output voltage.

The output resistance of a Kelvin-Varley divide varies with the switch positions, but say it can be as high as 100Kohms, then measuring the output with a 10Mohm multimeter will cause an error of 1% which destroys your 0.00001% accuracy. So you have to decide how you are going to use the divider, what accuracy you need and design accordingly.

So the best way to use the KVD is by buffering it's output with a chopper stabilized amp (OPA735 looks a good candidate), if you intend to read or use its output for anything other than a wheatstone bridge, right?


Quote from: amspire
The rules for building a Kelvin-Varley divider are:

  • All resistors in a decade have to match closely - you do not need resistors in one decade to match another.  So if you get a pile or 10ppm resistors, you can choose the closest set for the first decade, the next closest set for the next decade and so on.[/l][/l][/l][/l]
That's pretty much what I was thinking on doing.


Quote from: amspire
  • Each lower decade must have resistors of a value greater then one fifth the higher decade resistors. So if your first decade uses 10K resistors, the second decade must use over 2K resistors. There is nothing at all stopping you using 10K resistors for each decade, so you can buy a big pile of 10K resistors, and select them for the most accurate divider.
I didn't thought of that rule, but it makes perfectly sense. 10x the resistors of the second decade has to be greater than 2x the resistor of the first, so you can shunt/trim the second decade to a lower value.

Quote from: amspire
  • Each decade needs a parallel resistor across the whole decade divider so that the decades total resistance exactly equals twice the resistor value in the higher decade.  So if the first decade uses 10K resistors, then the second decade will have to have a parallel resistor across the whole decade adjusted so that the total resistance is exactly 20K.

Thats pretty much in sync with the previous rule.


Quote from: amspire
The advantage about using lower resistors in the lower decade is it helps lowering the output resistance as much as possible.

So if a 10Mohm multimeter causes an error of 1%, how do you use it?  Many high end meters have input resistances much more then 1G for input voltages between +- 20V.  Or you can put a fet input low offset amplifier on the output in a inity gain voltage follower configuration and make sure the FET input currents are low enough not to cause a significant  error.  Or you can use the divider in a traditional way where you use it with a galvanometer on the output, and you adjust either the divider or the voltage that you are measuring so that the galvanometer shows zero current is flowing.

Lol. Just noticed I didn't read you first response properly, as you've pretty much answered my first question in this post.  ;)

Thank you,
Felipe
[/list]
 

Offline Conrad Hoffman

  • Super Contributor
  • ***
  • Posts: 1928
  • Country: us
    • The Messy Basement
Re: Kelvin Varley Divider
« Reply #3 on: November 05, 2011, 02:27:34 am »
Hi Felipe,

Everything I did was driven by cost and ease of construction. If I had specified good rotary switches the cost would be so high that nobody would build the project. The same thing for getting very stable wire wound or metal foil resistors. Most KVDs are built point to point on the switch terminals, probably to keep leakage at a minimum. If you use a PCB, clean it very well. I didn't want to shunt the decades by much, thus the traditional resistor values for each decade. You can certainly build a lower impedance divider if you want. KVDs are only useful if no current flows in the pickoff lead. Either use a gig input impedance bench meter, a null detector and second voltage source (the traditional way) or the chopper stabilized buffer.

There are some clever alternatives these days. You can set up a bank of simple dividers and sum the results with a chopper stabilized amp. Done that way you can even get away with using analog switches and maintain accuracy.

Building a KVD is really a learning exercise. If you just want a KVD to use, it's easier to pick up a used Julie Research, ESI or Fluke. The Fluke has the advantage of being trimmable, as as said above, the manual is worth the read.

If you just want to adjust or confirm the ranges of a DVM, go for the Hamon divider, as it's far easier to build.

Best,
Conrad
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider
« Reply #4 on: November 05, 2011, 02:37:21 am »
The OPA735 is pretty good, but Linear Technology have IC's that are an order of magnitude better:

http://www.linear.com/products/zero_drift_amplifiers

One of the great advantages of a Kelvin-Varley divider is that it is totally passive. If the Fluke 720A is connected to 1100V, you can get any voltage out from 0V to 1100V. If I were adding an output amplifier, I would have it so it could be switched out of circuit, so you are not limited to the voltage ranges that the op-amp can cope with.

If you are after accuracy, you do have to invest in good resistors - the more powerful the better as they will heat up less. The last time I was balancing a bridge-type circuit built on 0.1% 25ppm SMD resistors, just the 10V I had across a 10K 10:1 divider  was enough power to see the resistors drift  due to heating.  In a 10:1 divider one resistor gets 9 times the power of the other resistor, so one resistor heats more then the other.  Because of the drift, best accuracy I could get was probably 0.005% even though I had the resolution to adjust within 0.000001%.  I would love to have some of the unbelievable Vishay metal foil resistors, but their prices really hurt.

Richard
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 11790
  • Country: us
Re: Kelvin Varley Divider
« Reply #5 on: November 05, 2011, 03:24:42 am »
instead of using a PCB
Actually, PCBs are a rather new-fangled invention. This is how proper circuits are supposed to be constructed:


 
The following users thanked this post: RAPo

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider
« Reply #6 on: November 05, 2011, 03:34:36 am »
I can't think how long it has been since I last saw a pair of Selenium rectifiers.

The wax dipped paper capacitors that always failed. The carbon composition resistors available in 10% and 20%.

That is great.
« Last Edit: November 05, 2011, 04:03:06 am by amspire »
 

Offline fmaimonTopic starter

  • Supporter
  • ****
  • Posts: 165
  • Country: br
Re: Kelvin Varley Divider
« Reply #7 on: November 05, 2011, 09:39:14 pm »
Everything I did was driven by cost and ease of construction.

And I thank you for that. Your Mini Metrology Lab is what sparked me on starting this.

Quote from: Conrad Hoffman
If I had specified good rotary switches the cost would be so high that nobody would build the project. The same thing for getting very stable wire wound or metal foil resistors. Most KVDs are built point to point on the switch terminals, probably to keep leakage at a minimum. If you use a PCB, clean it very well. I didn't want to shunt the decades by much, thus the traditional resistor values for each decade. You can certainly build a lower impedance divider if you want. KVDs are only useful if no current flows in the pickoff lead. Either use a gig input impedance bench meter, a null detector and second voltage source (the traditional way) or the chopper stabilized buffer.

I really liked the idea of using headers to switch each decade and I intend on doing just that.

On doing the PCB, although I have soldering experience, I think I would heat each resistance even less with it than soldering directly to the pin headers, so I expect to maintain the resistor values during soldering. My only fear was the thermal EMF because of the copper in the PCB.

As I have a Fluke 8840A, I'll initially use it to read the KVD output. It's spec'd at higher than 10G, so it's enough. I just hope that it's still near its accuracy number, as I bought it from eBay and still havn't sent it for calibration.

Quote from: Conrad Hoffman
There are some clever alternatives these days. You can set up a bank of simple dividers and sum the results with a chopper stabilized amp. Done that way you can even get away with using analog switches and maintain accuracy.

Can you please elaborate a bit further? Do you mean using a summing amplifier, where you set the summing ratios of each divider is set by the resistors in the amplifier? How can you mantain accuracy when using analog switches (I believe you mean the 74HC405X or similar series) when the switch resistance can be as high as 200R and highly non-linear?

Quote from: Conrad Hoffman
Building a KVD is really a learning exercise. If you just want a KVD to use, it's easier to pick up a used Julie Research, ESI or Fluke. The Fluke has the advantage of being trimmable, as as said above, the manual is worth the read.

Unfortunately, even used, they are pricey, and with shipping (when the seller ships!) and taxes to Brasil, it's way beyond my league. :(

Quote from: Conrad Hoffman
If you just want to adjust or confirm the ranges of a DVM, go for the Hamon divider, as it's far easier to build.

Initially I just want to have a KVD, but I wan't to use it as a precision voltage source by using a voltage reference and a chopper buffer, something like the one Dave just got.

Thank you,
Felipe Maimon
 

Offline fmaimonTopic starter

  • Supporter
  • ****
  • Posts: 165
  • Country: br
Re: Kelvin Varley Divider
« Reply #8 on: November 05, 2011, 09:40:20 pm »
instead of using a PCB
Actually, PCBs are a rather new-fangled invention. This is how proper circuits are supposed to be constructed:

Lol. That's just what I want to avoid! :)
 

Offline fmaimonTopic starter

  • Supporter
  • ****
  • Posts: 165
  • Country: br
Re: Kelvin Varley Divider
« Reply #9 on: November 05, 2011, 10:17:57 pm »
The OPA735 is pretty good, but Linear Technology have IC's that are an order of magnitude better:

http://www.linear.com/products/zero_drift_amplifiers

What, other than offset voltage, drift and voltage gain, do you usually watch for when looking for a chopper stabilized amp?

Quote from: amspire
One of the great advantages of a Kelvin-Varley divider is that it is totally passive. If the Fluke 720A is connected to 1100V, you can get any voltage out from 0V to 1100V. If I were adding an output amplifier, I would have it so it could be switched out of circuit, so you are not limited to the voltage ranges that the op-amp can cope with.

I'll add an amplifier, later, because I want to make a precision voltage source. But I'll make it detachable, so I can still use the KVD alone.

Quote from: amspire
If you are after accuracy, you do have to invest in good resistors - the more powerful the better as they will heat up less. The last time I was balancing a bridge-type circuit built on 0.1% 25ppm SMD resistors, just the 10V I had across a 10K 10:1 divider  was enough power to see the resistors drift  due to heating.  In a 10:1 divider one resistor gets 9 times the power of the other resistor, so one resistor heats more then the other.  Because of the drift, best accuracy I could get was probably 0.005% even though I had the resolution to adjust within 0.000001%.  I would love to have some of the unbelievable Vishay metal foil resistors, but their prices really hurt.

I'm looking for accuracy, but not that much. I'll use through hole resistors, as these have better thermal dissipation. Did you try making the traces and pads wider, so it acts as a heat sink?
It's already expensive buying low ppm resistors (~$.6-$1 each), but those metal foil are crazy!  :o

Thank you,
Felipe Maimon
 
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider
« Reply #10 on: November 06, 2011, 12:05:04 am »
The OPA735 is pretty good, but Linear Technology have IC's that are an order of magnitude better:

http://www.linear.com/products/zero_drift_amplifiers

What, other than offset voltage, drift and voltage gain, do you usually watch for when looking for a chopper stabilized amp?


Many of the chopper stabilized opamps have a low frequency response, but that is fine for a DC reference.  So you are looking at offset, drift and input current.

Quote
Quote from: amspire
If you are after accuracy, you do have to invest in good resistors - the more powerful the better as they will heat up less. The last time I was balancing a bridge-type circuit built on 0.1% 25ppm SMD resistors, just the 10V I had across a 10K 10:1 divider  was enough power to see the resistors drift  due to heating.  In a 10:1 divider one resistor gets 9 times the power of the other resistor, so one resistor heats more then the other.  Because of the drift, best accuracy I could get was probably 0.005% even though I had the resolution to adjust within 0.000001%.  I would love to have some of the unbelievable Vishay metal foil resistors, but their prices really hurt.

I'm looking for accuracy, but not that much. I'll use through hole resistors, as these have better thermal dissipation. Did you try making the traces and pads wider, so it acts as a heat sink?
It's already expensive buying low ppm resistors (~$.6-$1 each), but those metal foil are crazy!  :o


The stand alone resistors will have a much lower temperature rise then SMD's.  Wider traces may have an effect but probably doesn't matter. Work really hard not to have any mechanical stress on the resistors, so bend the legs to match the mounting exactly before you solder.  Stress on a component is one of the factors that cn result in a slow long term drift.

What you can also do is in the first decade, use 2 or 3 resistors in series to make up each divider resistor.  It gives you more combinations to end up with 10 exactly matched divider resistors, spreads the thermal load, and gives you a higher input voltage level that the divider can take. So if you got 3k3 resistors, the first decade can be 3 in series to make up one divider resistor and all the other decades use one of the 3k3.  If you want to save money, you can go to cheaper resistors after the 3rd decade.  What I have found is that the more you can match resistors rather then have adjustment pots, the better the final long term stability. If you do have adjustment pots, it is better if it only has to adjust over +/- 0.01% rather then +/- 1%. I haven't looked at the fluke, but I would think its adjustment on the first stage resistors is probably something like +/- 0.0005%

Richard
 

Offline Conrad Hoffman

  • Super Contributor
  • ***
  • Posts: 1928
  • Country: us
    • The Messy Basement
Re: Kelvin Varley Divider
« Reply #11 on: November 06, 2011, 02:33:55 am »
Yes, pots are trouble! When Julie Research did their KVDs, they matched the resistors as best they could (which was very good) but then soldered a short length of resistance wire to one end of each resistor to get the final value. You'll also find that regular metal films will change value when you solder them, so use long leads, heat clips and work fast. Read the old article a good number of times because I think it mentions all sorts of things to watch for. BTW, it was 15 years ago, so I don't remember as much as you might think.

See if you can download the manual for an Analogic 8200 (same as Data Precision 8200) voltage source. That will show you how to use analog switches, but basically if you're feeding a high impedance opamp from a divider, having 200 ohms in series with the input doesn't matter, and whatever pickoff point is connected swamps out the leakage of the other ones. It's all about ratio of impedances. Then, the decade amps are all summed by another amp, using different value summing resistors so each decade has the correct weight in the final answer. The 8200 doesn't actually use decades, but octal, since it's no problem for a processor to display it.

Conrad
 

Offline fmaimonTopic starter

  • Supporter
  • ****
  • Posts: 165
  • Country: br
Re: Kelvin Varley Divider
« Reply #12 on: November 06, 2011, 02:01:17 pm »
Many of the chopper stabilized opamps have a low frequency response, but that is fine for a DC reference.  So you are looking at offset, drift and input current.

That's pretty much what I thought. Thank you.

Quote from: amspire
The stand alone resistors will have a much lower temperature rise then SMD's.  Wider traces may have an effect but probably doesn't matter. Work really hard not to have any mechanical stress on the resistors, so bend the legs to match the mounting exactly before you solder.  Stress on a component is one of the factors that cn result in a slow long term drift.

What you can also do is in the first decade, use 2 or 3 resistors in series to make up each divider resistor.  It gives you more combinations to end up with 10 exactly matched divider resistors, spreads the thermal load, and gives you a higher input voltage level that the divider can take. So if you got 3k3 resistors, the first decade can be 3 in series to make up one divider resistor and all the other decades use one of the 3k3.  If you want to save money, you can go to cheaper resistors after the 3rd decade.  What I have found is that the more you can match resistors rather then have adjustment pots, the better the final long term stability. If you do have adjustment pots, it is better if it only has to adjust over +/- 0.01% rather then +/- 1%. I haven't looked at the fluke, but I would think its adjustment on the first stage resistors is probably something like +/- 0.0005%

Using multiple resistors is a pretty good idea. That should also help matching sets, as you can combine them anyway you like.

Note taken about the pots. I'll try to match the resistor as best I can and make the adjustments the pots make as small as possible.

The fluke looks like it has lots of pots! Check the attached schematic. :o

Yes, pots are trouble! When Julie Research did their KVDs, they matched the resistors as best they could (which was very good) but then soldered a short length of resistance wire to one end of each resistor to get the final value. You'll also find that regular metal films will change value when you solder them, so use long leads, heat clips and work fast. Read the old article a good number of times because I think it mentions all sorts of things to watch for. BTW, it was 15 years ago, so I don't remember as much as you might think.

I'm reading the article a lot. :)


Quote from: Conrad Hoffman
See if you can download the manual for an Analogic 8200 (same as Data Precision 8200) voltage source. That will show you how to use analog switches, but basically if you're feeding a high impedance opamp from a divider, having 200 ohms in series with the input doesn't matter, and whatever pickoff point is connected swamps out the leakage of the other ones. It's all about ratio of impedances. Then, the decade amps are all summed by another amp, using different value summing resistors so each decade has the correct weight in the final answer. The 8200 doesn't actually use decades, but octal, since it's no problem for a processor to display it.

That's one thing google can't find... Or maybe my googlefu is not that good... :(
The only manual I found is an users manual describing the functions. It didn't have a schematic.

I think I got the idea. One amplifier in a follower configuration (gain of 1) for each decade and another for summing everything. But that brings the problem of matching ratios of resistors in the summing amplifier. Maybe the Hamon idea can be used in here...

Can you explain better how the octal thing works? As it's easier to find analog switches in multiples of 2, it should be easier to construct.

Thank you,
Felipe Maimon

[edit] added the comment about the schematic.
« Last Edit: November 06, 2011, 05:34:09 pm by fmaimon »
 

Offline fmaimonTopic starter

  • Supporter
  • ****
  • Posts: 165
  • Country: br
Re: Kelvin Varley Divider
« Reply #13 on: November 07, 2011, 12:11:35 am »
I think I may found out how the octal thingy works!

Instead of 10 resistors for each decade, use 8 (octal, remember :) ). Then you just have to work in a octal base. For example, the ratio of 0.3125 is 0o24 in octal (I've put the 0o in front of any octal numbers). Then the first "octal decade" (is there a name for this?) is set at the second tap and the second "octal decade" is set on the fourth tap.

Now it's possible to use an easy to find 8 channel analog multiplexer, like the 4051 or similar, but more decades are needed to match the same resolution. Doing the math, you need (log 10 / log 8 ) = about 1.1073 more decades. And now there is another problem of the summing amplifier needing resistors in ratios of multiples of 8, but some modifications of the hamon divider can do it.

[edit] fixed an incorrect smiley
« Last Edit: November 07, 2011, 01:20:41 am by fmaimon »
 

Offline fmaimonTopic starter

  • Supporter
  • ****
  • Posts: 165
  • Country: br
Re: Kelvin Varley Divider
« Reply #14 on: November 07, 2011, 01:33:19 am »
Just had an idea. Instead of lots of different resistors for each deacade in the summing amplifiers, use a ladder in similar fashion to the r-2r ladder. If my math is correct, it should look like the schematic below. Now I just need 3 different resistor values. Vout is followed by a buffer, of course.



Lol. This thing is looking more like a very big and complex DAC :D
« Last Edit: November 07, 2011, 01:37:32 am by fmaimon »
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider
« Reply #15 on: November 07, 2011, 01:44:07 am »
The summing method means each decade can be built independently, and matching the relative contribution of each decade is done by adjusting one summing resistor for each decade.

The Hamon divider method can be used for ratios that are the square of numbers - ie 22:1, 32:1, 42:1, etc.

That means you get ratios of 4:1, 9:1. 16:1, 20:1  or voltage dividers that divide by 5, by 10, by 17, by 21.

It is hard to see how you would get a divide by 8 ratio.

A better technique that has been used in lots of older instruments is arrange for each decade to go from 0 to 10 (instead of 9).

Then calibration consists of two major steps. Step One is is you get all the resistors within each decade matched to each other.

Step 2 is you adjust each decade, starting from the second most significant so that "10" is exactly equal to "1" on the range above.

If you are using electronic switches, then to match the ranges, you can switch between the "1" on one range and the 10 on the next lower range at rate of 133Hz (or any frequency that is not a mains harmonic). You will get a square wave out when they do not match, and when you adjust the summing resistor for a match, the amplitude of the squarewave will go to zero.  If you built an sensitive AC amplifier, along with a tuned 133Hz bandpass filter to eliminate everything but the 133Hz Ac, and put your multimeter on the output, then you can match the decades with extreme precision without needing anything expensive or precise.

So calibration comes down to matching resistors within a decade, and zeroing an AC value. These are two steps that can be done simply and very accurately without expensive equipment.

In practice, any time you are chasing precision, it is always the detail that you have to be obsessive about.  You have to consider all the factors you normally do not have to consider in other designs. If you are using analogue switches you have to look at the switch resistance, the switch-to-switch variations of resistance, the temperature coefficient of the switch resistance, the OFF leakage current through the switch, and the leakage current from the supply rails to both sides of the switch.  The currents will probably rise exponentially with temperature, so you have to think of the maximum operating temperature.

The beauty of the totally passive resistive-based dividers is that as long as you start with great resistors and great switches, there is nothing else to go wrong. After a few years, it stabilizes and remain incredibly accurate. Put it in the cupboard, drag it out in 20 years and it will still work perfectly.

Part of the genius of the Kelvin-Varley divider is that the whole divider can be build from the built from one single batch of resistors, all of the same value. If all the resistors have matching temperature coefficients, then errors due to ambient temperature changes cancel out. Hopefully, they will age the same too. In the summing method, you have a very wide range of summing resistors, all that have to stay accurate. Is the 1Mohm resistor temperature characteristics going to match the 1Kohm characteristics?  Yes if you were Fluke of 50 years ago, and you make all the resistors from wire of the precisely same composition, and yes if you are HP or Fluke today, and you can make a laser trimmed thick film resistor network, where all the resistors are made from exactly the same deposited film. Probably not if you are a hobbyist getting whatever parts you can find.

Richard
« Last Edit: November 07, 2011, 01:57:52 am by amspire »
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider
« Reply #16 on: November 07, 2011, 03:00:49 am »
Just had an idea. Instead of lots of different resistors for each deacade in the summing amplifiers, use a ladder in similar fashion to the r-2r ladder. If my math is correct, it should look like the schematic below. Now I just need 3 different resistor values. Vout is followed by a buffer, of course.



Lol. This thing is looking more like a very big and complex DAC :D

I think your maths is a bit off with this one. Top marks for getting the values to work successfully as a tree with each stage having the same impedance as the stage below it. That bit is perfect. Very elegant.

The problem is that each stage contributes 7/8 of the previous stage, and that is not much use at all.  There is no reason not to use the R-2R instead. Far more use with far less resistors.

So if the top stage is 1 Volt out, the second stage contributes 7/8V out. The third stage 49/64 volts out. The 4th 73/83 volts out, and so on. It is not that in theory it couldn't work in a very strange way. In practice, it would need about 103 stages to get one part in a million resolution - 206 resistors - with a massive amount of redundancy. All of which have to be very precisely matched. It is just not efficient, or convenient in terms of the weird increments.

Now the R-2R is great for ICs, but for discreets, it is a problem.  To get a resolution of 10-6, you need 20 stages which means 40 resistors.  All 40 resistors have to match precisely.

In theory, each resistor down the chain requires less accuracy then the one above it, but if you want to calibrate it, it is hard without adjustment to one resistor affecting the previous calibrations you have already done.

Someone may have developed a practical technique for calibrating R-2R networks, and I would love to learn of one.

Richard
« Last Edit: November 07, 2011, 03:21:30 am by amspire »
 

Offline Conrad Hoffman

  • Super Contributor
  • ***
  • Posts: 1928
  • Country: us
    • The Messy Basement
Re: Kelvin Varley Divider
« Reply #17 on: November 07, 2011, 03:22:16 am »
Analogic seems rather possessive of their schematics, and they don't appear on-line anywhere. I'll give you a verbal description of the heart of the thing, the 20 bit DAC. A buffered LM299 reference feeds a divider consisting of 7 resistors, giving 7 voltages, 8 if you include ground. The pickoff points are low value trimmers so the top six can be tweaked slightly.

The voltages all connect to 7 4051 switches connected in parallel. The output of each switch is buffered with a simple follower. Thus, by addressing the control lines of each 4051, you can select one of 8 voltages on 7 different identical buffered outputs.

The output of each follower feeds a summing amp, with each one being summed through a different value resistor, giving you the fine to coarse steps. The beauty of this is that one master divider chain provides the voltages for all the "digits" of precision, rather than needing a separate divider for each digit. I use the term digit loosely since this is octal, not the usual decades we're used to.

This scheme probably makes sense only if you're controlling things with a uP, but it gives 1 ppm resolution and is quite stable. A full calibration does require a good Fluke or Julie KVD, and the calibration numbers are not "friendly". Full scale is something like 1048575. IMO, the same technique could be done with ten voltages and manual control.

No doubt that was clear as mud!

Best,
Conrad
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider
« Reply #18 on: November 07, 2011, 04:08:54 am »
Conrad,

That all makes sense. Every Op-amp follower connects to a 4051 one of 8 multiplexer switch. So if you have a 7V reference, each opamp's output can be 0V, 1V, 2V, 3V, 4V, 5V, 6V or 7V depending on the 4051 binary input.  If you are using opamps with a 1uV offset, then they will have negligible contribution to the accuracy. Also in this configuration, the resistance and leakage of the 4051 will have negligible effect, as long as the 0-7V divider chain is made up of fairly low impedance resistors. Perhaps a few hundred ohms per resistor.

So lets say the MSO (Most Signifigant Octal - I made that up  :) ) opamp has a 1K resistor to the summing amp.

The next would have an 8K.

Then 64K, 512K, 4.096M, 32.768M and 262.144M for the LSO opamp output.   I am sure I have plenty of those 262.144 Megohm resistors somewhere........

May as well make it a 21 bit DAC - you already have the switches for it.

I like it!

Richard
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider
« Reply #19 on: November 07, 2011, 04:30:09 am »
It wouldn't be hard to combine the Analogic idea with the idea I mentioned about about making each decade go from 0 to 10.

So you use a 4051 and half a 4052 to get a 12 way multiplexer  - of which 11 switches will be used.

Have a divider that does 0, 1, 2, 3, 4, 5, 6, 7, 8, ,9 and 10V.

If the MSD op-amp has a 1K resistor to the summing amp, the others have 10K, 100K, 1M, 10M and 100M for the LSD.

So to calibrate, you first calibrate the divider, which is a matter of using a bridge-type circuit to make each resistor an identical value.

Then in the software you switch between 1v output from the MSD op-amp, and 1V output from the second opamp. You adjust the 10K resistor until there is zero AC out.

Then you do the same comparing the 0.1V from the second op-amp to the 0.1V of the third op-amp.

So the total calibration is:

1. Calibrating the 10V voltage reference.
2. Calibrating 10 divider resistors to be identical in value.
3. Ajusting  5 resistors for zero AC output.

Done!  You have a 6 digit voltage reference.

Whereas the Analogic design used binary for the voltages, this plan would use binary coded decimal. So you could use binary coded thumb wheel switches to dial the voltage manually, with a calibration using the last two multiplexer switches to do "1" and "10" for the calibration.

Richard.
« Last Edit: November 07, 2011, 04:34:32 am by amspire »
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider
« Reply #20 on: November 07, 2011, 02:16:30 pm »
I have added a concept diagram based in Conrad's idea but organized in decades rather then octal/binary.

The is a cost - 12 x 4051's needed rather then 7, but the gains are that you can control it via thumbwheel decade switches, and it is much easier to calibrate.

I have included 2 decades - 4 more have to be added to make a 6 digit voltage source. Each one will have the resistance from the opamp increasing by 10 from the previous decade.

The 10V resistor chain is powered by a 10V reference. Since I have a positive reference, the final op-amp will put out 0 to -10V. In a final design, there would probably be a switchable polarity, or at least a positive polarity.

I haven't included the wiring to the switches, or the extra wiring for my easy-calibrate idea.

Richard.
 

Offline Conrad Hoffman

  • Super Contributor
  • ***
  • Posts: 1928
  • Country: us
    • The Messy Basement
Re: Kelvin Varley Divider
« Reply #21 on: November 07, 2011, 05:21:45 pm »
By golly I think you've got it! Could be wrong, but I think you can go quite a lot higher on the values for the divider. Since the switches are buffered there isn't any significant load on it, and because something is always connected the leakage resistance of the switches gets swamped out. Analogic/Data Precision did pick off the voltages through low value trimmers, and that probably compensated for any slight loading errors. It also meant you needed a KVD to calibrate the thing. The fixed design just requires identical resistors, something doable with next to no special equipment. I know they have patents on all or some of the design, but I don't know what's still in force.
 

Offline Bored@Work

  • Super Contributor
  • ***
  • Posts: 3932
  • Country: 00
Re: Kelvin Varley Divider
« Reply #22 on: November 07, 2011, 06:03:10 pm »
It wouldn't be hard to combine the Analogic idea with the idea I mentioned about about making each decade go from 0 to 10.

But you could as well stick with octal, or more precisely, septenary (the base-7 system) and apply the same trick. This would allow you to stay with common analog switches. Make each stage go from 0 to 6, and 7 being the same value as the 1 of the next stage. I would, however, recommend a microcontroller to control the switches and do septenary math. And you have to give up the idea of nice decadal voltage steps.
I delete PMs unread. If you have something to say, say it in public.
For all else: Profile->[Modify Profile]Buddies/Ignore List->Edit Ignore List
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider
« Reply #23 on: November 07, 2011, 07:17:05 pm »
I believe the Analogic was pure binary. The only octal thing about it was probably that each 4051 did 3 bit of resolution.

If you are going to have a micro managing all the voltages settings, to can stick to pure binary. However since you will want to increment in steps of 1uV or 10 uV, to end up with very odd numbers for the high end ranges. Also you could loose the simple summing calibration method I planned.

Not a problem if you happen to have a 7 1/2 digit multimeter or a Fluke 720a for calibration, but if you don't, calibrating a binary system will be rather difficult.

But there are a lot of options, which is why I called it a "concept diagram".

If we can work out a good design, then for each decade, or octal, a module could be put on a small PCB, that we could then source very cheaply. Then anyone could use it to build a voltage source with a many decades or binary bits that they want. Just keep adding modules and summing them.

There are details to be looked at. My low offset opamps have input currents high enough to cause errors. So more effort needed there.

Richard

« Last Edit: November 07, 2011, 10:17:42 pm by amspire »
 

Offline Conrad Hoffman

  • Super Contributor
  • ***
  • Posts: 1928
  • Country: us
    • The Messy Basement
Re: Kelvin Varley Divider
« Reply #24 on: November 07, 2011, 09:50:41 pm »
I think easy calibration is the way to go for the average bear. What I don't know is the magnitude of the error sources that "come out in the wash" when the Analogic unit is trimmed. That trimming might have compensated for switch resistance, leakage and who knows what else. The unit did step in 10 uV steps on the 10V range. IMO, one of the nice things about the way they did it was being able to go a bit over 1 or 10V; It's very common to need that feature. I like the idea of modules for each decade. They could have Samtec or similar pin headers so you'd plug them together, creating a block. Doubt it makes much sense to go beyond 1 ppm, as the resistor and reference requirements get way out of hand, not to mention thermal issues. Analogic did make one horrible mistake. They used quadrature switches on the front panel for each decade, that didn't hold up well. The old units can be really unpredictable when setting the voltage, though whatever they read out on the display is what you get. Sometimes you turn it up and it goes down, sometimes down and it goes up. Contact cleaner only helps for a year or two and then becomes ineffective.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider
« Reply #25 on: November 07, 2011, 10:51:00 pm »
I think easy calibration is the way to go for the average bear. What I don't know is the magnitude of the error sources that "come out in the wash" when the Analogic unit is trimmed. That trimming might have compensated for switch resistance, leakage and who knows what else. The unit did step in 10 uV steps on the 10V range. IMO, one of the nice things about the way they did it was being able to go a bit over 1 or 10V; It's very common to need that feature. I like the idea of modules for each decade. They could have Samtec or similar pin headers so you'd plug them together, creating a block.

Ideas for the construction definitely would help.

I would put the summing resistors on the main board as the design of the summing junction is the critical part of the design.  When you are summing currents from a  260Mohm resistor, then I think  shielding surrounding the main summing resistors would be a good ides - just the adjustment pots would be outside the shield.

So you would end up then with all the decade modules being absolutely identical and unless a offset zero pot is added, the modules would have no adjustment.

The way the modules are connected is critical. If there is 1mA flowing through the main reference divider chain, then a total leakage of 1pA causes a 1ppm error. It will be almost impossible to get 1pA absolute leakage. The 4051's are a big concern, and it may be necessary if you are chasing the 1ppm to add extra voltage followers on each of the divider outputs.  It also means that the divider must be calibrated with all the actual loads, so we will need to come up with a method of calibrating all the stages of the divider that does not involve matching resistors.  I think it can be done with a combination of a precision 1:1 resistive divider and a 10:1 resistive divider - ie the elements of a Hamon divider.

But also issues of PCB leakage currents and thermal voltages come into it. It may be best to pit right angle pins on the modules and solder them to the main board.

Quote
Doubt it makes much sense to go beyond 1 ppm in terms , as the resistor and reference requirements get way out of hand, not to mention thermal issues. Analogic did make one horrible mistake. They used quadrature switches on the front panel for each decade, that didn't hold up well. The old units can be really unpredictable when setting the voltage, though whatever they read out on the display is what you get. Sometimes you turn it up and it goes down, sometimes down and it goes up. Contact cleaner only helps for a year or two and then becomes ineffective.


In terms of the number of decades, you are right that trying to go for more that 1ppm would be a bit too much. However, if we find opamps that give a real 1uV offset or less - whether it is through chopper auto zero, or through zero calibration, then it may be possible to have useful resolution down to 1uV or less with a low voltage output. you can achieve something similar by having multiple ranges on the voltage source, but every time you add something like a range switch, it is another 1ppm voltage divider you have to consider.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider
« Reply #26 on: November 08, 2011, 05:15:13 am »
Here is another approach to throw into the mix.

It is the BCD tree used by Fluke in its Programmable Voltage Source like the 4210A. These were actually programmable supplies with voltages up to +/-65 V and currents to +/-1A.



Its merits are the big choice of switch IC's for each decade, 5 resistors per decade, all resistors in the range of 25K to 200K and only 5 different precision values are required.

Also the ranges overlap - each decade can go up to 15 and so 10 on a lower range can be calibrated from 1 on a higher range.

Problems are the same as a R-2R tree - calibration changes in the lower ranges affect the higher range calibration, so the calibration cycle may require going around the calibration loop until everything settles down to the calibration spec.

Fluke were only after 0.01% accuracy in voltage sources, so they limited the calibration problem by limiting the number of calibration pots to about 5.  For 1ppm though, calibration may get too difficult.

Richard
« Last Edit: November 08, 2011, 05:24:46 am by amspire »
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider
« Reply #27 on: November 08, 2011, 06:21:28 am »
I stumbled across these switches that look like they might be OK for a Kelvin-Varley divider. 2 pole 11 positions (which I prefer to 10 position).

Ceramic wafer, and it looks like silver plated.

$4 each with free delivery. Ebay Item 300562748033



 

Offline fmaimonTopic starter

  • Supporter
  • ****
  • Posts: 165
  • Country: br
Re: Kelvin Varley Divider
« Reply #28 on: November 08, 2011, 11:36:05 pm »
In practice, any time you are chasing precision, it is always the detail that you have to be obsessive about.  You have to consider all the factors you normally do not have to consider in other designs. If you are using analogue switches you have to look at the switch resistance, the switch-to-switch variations of resistance, the temperature coefficient of the switch resistance, the OFF leakage current through the switch, and the leakage current from the supply rails to both sides of the switch.  The currents will probably rise exponentially with temperature, so you have to think of the maximum operating temperature.

That's why I'm here asking all these questions. I don't have much experience in precision stuff. If I didn't thank you and Conrad yet, let me thank you both for your insight and knowledge.

The beauty of the totally passive resistive-based dividers is that as long as you start with great resistors and great switches, there is nothing else to go wrong. After a few years, it stabilizes and remain incredibly accurate. Put it in the cupboard, drag it out in 20 years and it will still work perfectly.

Quote from: amspire
Part of the genius of the Kelvin-Varley divider is that the whole divider can be build from the built from one single batch of resistors, all of the same value. If all the resistors have matching temperature coefficients, then errors due to ambient temperature changes cancel out. Hopefully, they will age the same too. In the summing method, you have a very wide range of summing resistors, all that have to stay accurate. Is the 1Mohm resistor temperature characteristics going to match the 1Kohm characteristics?  Yes if you were Fluke of 50 years ago, and you make all the resistors from wire of the precisely same composition, and yes if you are HP or Fluke today, and you can make a laser trimmed thick film resistor network, where all the resistors are made from exactly the same deposited film. Probably not if you are a hobbyist getting whatever parts you can find.

Be sure I'll build one pure KVD because of exactly that. Actually this message just made me realize that we went a little off topic and started discussing a precision voltage source. I'll change the thread title to reflect that.

I think your maths is a bit off with this one. Top marks for getting the values to work successfully as a tree with each stage having the same impedance as the stage below it. That bit is perfect. Very elegant.

When I first design this 8R-56R circuit, it seemed that it would work, but I didn't made all the calculations. After your post, I've applied the thevenin theorem to the circuit and saw it really didn't work... Thank you for pointing out.

No doubt that was clear as mud!

Actually that was very clear! Thank you for sharing.

If the MSD op-amp has a 1K resistor to the summing amp, the others have 10K, 100K, 1M, 10M and 100M for the LSD.

Liked it! Nice design, but isn't the 100M resistor a bit high? Isn't a little bit better to use a 0.1 hamon divider with it´s output a little higher (about 0,1%) followed by a 10M resistor?

I think easy calibration is the way to go for the average bear. What I don't know is the magnitude of the error sources that "come out in the wash" when the Analogic unit is trimmed. That trimming might have compensated for switch resistance, leakage and who knows what else. The unit did step in 10 uV steps on the 10V range. IMO, one of the nice things about the way they did it was being able to go a bit over 1 or 10V; It's very common to need that feature. I like the idea of modules for each decade.

Easy calibration is everything I hope for!
On going a bit over 10V, Richard's design can do that too. If you select the 10V and  the 1V outputs, you get 11V. With 6 decades, you can go to 11.1111V.

Good ideas with the modularisation of the design. I'll keep it in mind.

I stumbled across these switches that look like they might be OK for a Kelvin-Varley divider. 2 pole 11 positions (which I prefer to 10 position).

Nice find! But why 11 positions are better? Don't you only need 10 positions for a KVD?

Thank you,
Felipe
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider
« Reply #29 on: November 09, 2011, 12:36:45 am »

If the MSD op-amp has a 1K resistor to the summing amp, the others have 10K, 100K, 1M, 10M and 100M for the LSD.

Liked it! Nice design, but isn't the 100M resistor a bit high? Isn't a little bit better to use a 0.1 hamon divider with it´s output a little higher (about 0,1%) followed by a 10M resistor?


100M is a bit high, and using a 10:1 divider or a 100:1 is a great solution. Accuracy is not important - you can calibrate that out. Stability is important. One thing you can do is to use the divider networks that have great temperature matching, so you get good stability.  Since it is a one-off build, it is perfectly reasonable to manually select resistor combinations until it is very close to correct, then use a pot to adjust for the last 1 part in 100 of accuracy. Also the high resistances are for low order decades, so an an accuracy of 1% is adequate. You do not need to go overboard. It is the first 4 decades that you have to be obsessive about.

Quote
I think easy calibration is the way to go for the average bear. What I don't know is the magnitude of the error sources that "come out in the wash" when the Analogic unit is trimmed. That trimming might have compensated for switch resistance, leakage and who knows what else. The unit did step in 10 uV steps on the 10V range. IMO, one of the nice things about the way they did it was being able to go a bit over 1 or 10V; It's very common to need that feature. I like the idea of modules for each decade.

Easy calibration is everything I hope for!
On going a bit over 10V, Richard's design can do that too. If you select the 10V and  the 1V outputs, you get 11V. With 6 decades, you can go to 11.1111V.

Good ideas with the modularisation of the design. I'll keep it in mind.


I wasn't sure if my explanation for the easy calibration method was clear, but if you followed it, then great!.

Quote
I stumbled across these switches that look like they might be OK for a Kelvin-Varley divider. 2 pole 11 positions (which I prefer to 10 position).

Nice find! But why 11 positions are better? Don't you only need 10 positions for a KVD?

Thank you,
Felipe

To calibrate the resistors in a KVD stage, you need to be able to break the connection between at least one leg of the chain and the rest of the KVD. Going all the way with the Fluke 720a system with the whole calibration for the first 3 stages managed by switches is the ultimate, but they can afford to buy great switches  - and great switches can easily cost hundreds of dollars.  Having an 11th position that breaks the chain free of the next lower divider, combined with a calibrate procedure that perhaps using plugable wire links rather then Fluke's switches,  is a workable solution for a home builder. There would be other ways to do it I am sure, and some people will just hate having that 11th position.

There still need to be another switch that can disconnect all the resistor in parallel with the dividers, or maybe not with guarding during calibration? Exactly how how to make it all work will need some thinking.

Calibrating the KVD is all about keeping the box sealed and untouched as much as possible. If there was a removable panel that gave access to the adjustment pots, and the pins that you could link for calibration, that would be ideal.

You can safely match resistors to 0.01% before assembly, so like the Fluke, it is best if the lower 4 ranges have no calibration at all. That means they really only need 10 position switches. Just have to find good ones at a decent price. The supply of affordable good quality rotary switches is not what it used to be.

On the first decade, an 11th position is essential, as it means you can you can have a "1.1" input like the Fluke 720A. This allows you to do plus-minus variations around the 10.00000 mark. But you also need to be able to break the connection to the second divider, so does that mean we need a 12 way? Very possibly.

It is very annoying if you can only go to 9.99999 and no higher.

Richard.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #30 on: November 09, 2011, 01:07:15 am »
I have ordered 10 of those switches, and I will test them when they arrive from Hong Kong.

If you order them in lots of 100, the price comes down to $1.75 each. (ebay item 290571818531)

Looks like they have good stocks, so I am hoping they are good switches.

Richard.
 

Offline Flavour Flave

  • Regular Contributor
  • *
  • !
  • Posts: 76
  • Country: 00
  • Never knowingly unoffensive.
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #31 on: November 11, 2011, 06:35:20 pm »
Linear techs Kelvin Varley Voltage divider  8)



You can get the schematic here:

http://www.edn.com/article/459869-Video_Design_Idea_Build_your_own_laboratory_precision_voltage_reference.php
« Last Edit: November 12, 2011, 02:36:18 am by Flavour Flave »
 
The following users thanked this post: RoadDog

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #32 on: November 11, 2011, 11:48:52 pm »
There are some good ideas in that video.

Unfortunately, those dividers are a problem.

http://www.ietlabs.com/voltage-divider/esi-dp-1211-dp1311-voltage-dividers.html

He used a 4 decade divider. Price for 3 decade concentric dividers on the IET Labs website is $3940 to $4895, and they don't mention the 4 decade version at all. I think the 4 decade Tegam coaxial resistors are available from IET, but you have to contact them for details.

They are impressive though:

5ppm/C temp coefficient, 0.3ppm/mW power coefficient and short term switch repeatability of 1mOhm.

Richard
 

Offline Flavour Flave

  • Regular Contributor
  • *
  • !
  • Posts: 76
  • Country: 00
  • Never knowingly unoffensive.
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #33 on: November 12, 2011, 12:43:55 am »
This would be way cheaper "How to make custom resistors"

http://www.all-electric.com/schematic/res_trim.htm

This could be a cheap way to make a divider?
Using a file, nail polish and cheap carbon resistors oh and metal film aswell.
There is good tip for tuning filters on that page as well.
« Last Edit: November 12, 2011, 01:14:14 am by Flavour Flave »
 

Offline codeboy2k

  • Super Contributor
  • ***
  • Posts: 1836
  • Country: ca
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #34 on: November 12, 2011, 04:10:40 am »
This would be way cheaper "How to make custom resistors"

http://www.all-electric.com/schematic/res_trim.htm

This could be a cheap way to make a divider?
Using a file, nail polish and cheap carbon resistors oh and metal film aswell.
There is good tip for tuning filters on that page as well.

Ouch.  That webpage hurts my eyes.  1999 called, they want their web back. :)
 
The following users thanked this post: RoadDog

Offline Conrad Hoffman

  • Super Contributor
  • ***
  • Posts: 1928
  • Country: us
    • The Messy Basement
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #35 on: November 12, 2011, 03:20:47 pm »
Carbon won't get you where you want to go. It's not hard to make precision resistors, at least for low frequencies. Get yourself a roll of manganin wire and wind them on mica squares, just like they did years ago. If you want to rig up a simple spot welder, you can get "800 series" wire. It can't be soldered (easily) but has very low TC, better than 5 ppm/C. I think many people used to trim the value (trade secret revealed here) by abrading the side of the wire using an electric eraser- the rotary drafting kind with an ink eraser tip. If you look inside things that use card wound resistors, you'll see this frequently.

When Jim Williams wrote on KVDs (not sure if app notes or his books) he talked about getting Julie Research KVDs for something like a few dollars at a surplus market. They were tagged "bad" because someone tried to measure them with an ohm meter, not knowing that you can't pull current from the pickoff point. I bought a used KVD directly from Julie. It was the prototype of the 307, and proudly wears serial number 1.
 

Offline Flavour Flave

  • Regular Contributor
  • *
  • !
  • Posts: 76
  • Country: 00
  • Never knowingly unoffensive.
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #36 on: November 12, 2011, 04:02:40 pm »
Carbon won't get you where you want to go.


Why is that. Is it to do with temperature?

I thought that I could knock up a Wheatstone bridge and file the carbon resistors to the same value or use metal film, but they are bit harder to file. Well the aim is get them to the same value, not an exact absolute value, to use in the KVD for a precise voltage reference.

Also that mica seems expensive(show me a source please) and wouldn't I need a lot of Manganin wire to do resistors for KV divider ?(for 10k 2k etc..). And a lot of winding.

Thanks
Richard
« Last Edit: November 12, 2011, 04:16:47 pm by Flavour Flave »
 

alm

  • Guest
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #37 on: November 12, 2011, 05:45:01 pm »
Filing won't do anything about tempco or drift. I think there's a table in the Art of Electronics that indicates several percent change in value after soldering, load cycling and other kinds of stress. Don't remember if this was carbon film or carbon comp. The important spec for precision equipment is its drift (short term, long term) and tempco, not so much the actual value. This is why precision components are sometimes aged by manufacturers. You can cal out any change in value (eg. put a label with 1.010 ohm on it), but you can't compensate for variations, apart from indicating that the instrument has +/-2% tolerance.

I think there might be a job for you in China somewhere where you can use some blue paint to turn carbon film resistors into metal film, although I doubt they actually bother tweaking them to 1%.
« Last Edit: November 12, 2011, 06:47:30 pm by alm »
 

Offline Flavour Flave

  • Regular Contributor
  • *
  • !
  • Posts: 76
  • Country: 00
  • Never knowingly unoffensive.
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #38 on: November 12, 2011, 06:05:25 pm »
Filing won't do anything about tempco or drift. I think there's a table in the Art of Electronics that indicates several percent change in value after soldering, load cycling and other kinds of stress. Don't remember if this was carbon film or carbon comp. The important spec for precision equipment is it's drift (short term, long term) and tempco, not so much the actual value. This is why precision components are sometimes aged by manufacturers. You can cal out any change in value (eg. put a label with 1.010 ohm on it), but you can't compensate for variations, apart from indicating that the instrument has +/-2% tolerance.

I think there might be a job for you in China somewhere where you can use some blue paint to turn carbon film resistors into metal film, although I doubt they actually bother tweaking them to 1%.

Well that's knocked it on the head for carbon film resistors then. But on that webpage I linked to about filing resistors it also said you can do it to metal film reisistors. In the Art of Electronics it says that with metal film you can get stability of 0.1% under normal conditions. Well, howabout if you file them roughly near  to each others value, then solder them to those switches like the ones you just bought( well if they are any good) and then remeasure them  and trim them with file. And then pot them with thermally conductive potting compound. Wouldn't that be abnormal condition for them and you might get better than 0.1% ?

Using this KVD with the Linear Tech circuit it will only have 10mA (I think?) going through it if that helps with stability.
Well I'm learning a lot about the humble resistor.  ;D
« Last Edit: November 12, 2011, 06:44:20 pm by Flavour Flave »
 

alm

  • Guest
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #39 on: November 12, 2011, 06:55:12 pm »
Filing removes the lacquer and might make them more susceptible to oxidation, but feel free to try it and find out. It will take a few years to get results about long term stability, though, and you need a stable reference (eg. DMM) that will drift little over that time span.

10mA through 1k is 0.1W dissipation, which for 0.25W resistors will likely result in some heating, since it's fairly close to the max. It would also drop 10V over the resistor, so it's likely a few orders of magnitude off for a buffered voltage source.

The precision wire wound resistors often used for these applications are often physically large compared to their power rating to reduce heating.
 

Offline Flavour Flave

  • Regular Contributor
  • *
  • !
  • Posts: 76
  • Country: 00
  • Never knowingly unoffensive.
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #40 on: November 12, 2011, 07:31:18 pm »
Filing removes the lacquer and might make them more susceptible to oxidation, but feel free to try it and find out. It will take a few years to get results about long term stability, though, and you need a stable reference (eg. DMM) that will drift little over that time span.

10mA through 1k is 0.1W dissipation, which for 0.25W resistors will likely result in some heating, since it's fairly close to the max. It would also drop 10V over the resistor, so it's likely a few orders of magnitude off for a buffered voltage source.

The precision wire wound resistors often used for these applications are often physically large compared to their power rating to reduce heating.

Nail varnish would stop oxidation(maybe?)
Anyway  wirewounds is probably the way to go then. But the precision ones seem very expensive. Well making your own might be an option as Conrad has written.

I just thought that I could just buy one stage at a time. So 0.01% wire wounds and  10ppm/C are the ones to use?

I think there might be a job for you in China somewhere where you can use some blue paint to turn carbon film resistors into metal film, although I doubt they actually bother tweaking them to 1%.

One Hung Low Precision Resistors Inc  ;)
« Last Edit: November 12, 2011, 07:33:00 pm by Flavour Flave »
 

Offline fmaimonTopic starter

  • Supporter
  • ****
  • Posts: 165
  • Country: br
Re: Kelvin Varley Divider
« Reply #41 on: November 12, 2011, 09:22:48 pm »
Step 2 is you adjust each decade, starting from the second most significant so that "10" is exactly equal to "1" on the range above.

If you are using electronic switches, then to match the ranges, you can switch between the "1" on one range and the 10 on the next lower range at rate of 133Hz (or any frequency that is not a mains harmonic). You will get a square wave out when they do not match, and when you adjust the summing resistor for a match, the amplitude of the squarewave will go to zero.  If you built an sensitive AC amplifier, along with a tuned 133Hz bandpass filter to eliminate everything but the 133Hz Ac, and put your multimeter on the output, then you can match the decades with extreme precision without needing anything expensive or precise.

So calibration comes down to matching resistors within a decade, and zeroing an AC value. These are two steps that can be done simply and very accurately without expensive equipment.

Rereading this, I just can't understand how to calibrate the decades. If you sense the output of the summing amplifier, you will get exaclty 0.2 x the AC voltage, not 0! If you remove the summing amplifier and sense the output of the pots, you will get 0 allways! Am I missing something?

One more thing, the first 2 decades need a transistor (something like Art of Electronics figure 4.21) in it's output to source enough current, as the opamps can't source the 10 mA (first decade) or 1 mA (second decade) needed. As the idea is doing as a module, it is needed in every decade.

And instead of using 2 4051, it may be better to use a single 4067. BTW, wht are you connecting the 1V and 0V taps to the X3, X6 and X7 of the second 4051?

Thank you,
Felipe Maimon
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider
« Reply #42 on: November 12, 2011, 11:18:23 pm »
Step 2 is you adjust each decade, starting from the second most significant so that "10" is exactly equal to "1" on the range above.

If you are using electronic switches, then to match the ranges, you can switch between the "1" on one range and the 10 on the next lower range at rate of 133Hz (or any frequency that is not a mains harmonic). You will get a square wave out when they do not match, and when you adjust the summing resistor for a match, the amplitude of the squarewave will go to zero.  If you built an sensitive AC amplifier, along with a tuned 133Hz bandpass filter to eliminate everything but the 133Hz Ac, and put your multimeter on the output, then you can match the decades with extreme precision without needing anything expensive or precise.

So calibration comes down to matching resistors within a decade, and zeroing an AC value. These are two steps that can be done simply and very accurately without expensive equipment.

Rereading this, I just can't understand how to calibrate the decades. If you sense the output of the summing amplifier, you will get exaclty 0.2 x the AC voltage, not 0! If you remove the summing amplifier and sense the output of the pots, you will get 0 allways! Am I missing something?

In this case, we have electronic range switching, and I proposed that lower decades could go to "10".

Now each decade has to fist be calibrated so it is linear to the final spece. All that is left is to calibrate each lower decade to the higher decade.

So you have your precision 10V DC +/- 0.001% as the reference voltage. Switch between these two settings at, say, 133 Hz:

Decade1 = "1" - all other decades = 0
Decade2= "10" - all other decades = 0

When calibrated, both these two voltages will be exactly equal and you will DC out of the summing amplifier. If they are not equal, you will get a 133Hz squarewave out. When you adjust the calibration resistor of the second decade so that the 133Hz squarewave disappears, then the second decade is calibrated. The nice thing is that no precision equipment is needed to do this adjustment, just a simple audio preamp (probably with a low pass filter to remove switching transients) connected to the AC volts on any multimeter.

Quote

One more thing, the first 2 decades need a transistor (something like Art of Electronics figure 4.21) in it's output to source enough current, as the opamps can't source the 10 mA (first decade) or 1 mA (second decade) needed. As the idea is doing as a module, it is needed in every decade.

And instead of using 2 4051, it may be better to use a single 4067. BTW, wht are you connecting the 1V and 0V taps to the X3, X6 and X7 of the second 4051?

Thank you,
Felipe Maimon

Of course you choose the first decade summing resistor so that the opamp can handle the current. You can easily get opamps that can do 10mA or more, but if your chosen opamp needs to work below 2mA for best accuracy, you use a summing resistor large enough so that the current is below 2mA. This will make all the lower resistors proportionally larger. Or  as you correctly suggested, you just add a transistor or two to the opamp output on the first decade, and then the 10mA is then easy. Keeps the heat out of the opamp, and that is a good thing.

You can use the 4067 - I was thinking of 2 4051's just because it might make calibrating within a decade slightly easier - I had an idea.  But as I also said, precision voltage sources are all in the details, and it may be the 40XX just have too much leakage for great precision. you have to do the calculations and see if it works or not. Once the numbers are right, you have a design.

Richard.
« Last Edit: November 12, 2011, 11:27:05 pm by amspire »
 

Offline fmaimonTopic starter

  • Supporter
  • ****
  • Posts: 165
  • Country: br
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #43 on: November 17, 2011, 10:37:34 pm »
I'm about to buy the resistors for the KVD at digikey and thinking on trying to optimize the cost/benefit ratio, instead of buying 250 10k 1% 50 ppm resistors @ $20, I was thinking on buying 50 10k 0.1% 25 ppm resistors @ $24. They are both 0.5W or better. The problem is that my probability skills are non existant and I don't know the minimun number of 0.1% resistors that I need to get 2 sets of resistors that match better than 40 ppm. Can anyone help me with this? :)

Or can use the idea of using 3 3k3 resistors, each of 0.25W, but it will probably make my head hurt when matching.

I'll do the 10k-10k-4k-1k-1k-1k KVD as the fluke 720 and the iet kvd-700 as I'll need less values to make it all work. The shunts will have the same specifications of the previous decade and for the 4k and 1K, I'll buy 100 of the 50 ppm.

Thank you,
Felipe Maimon
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #44 on: November 17, 2011, 11:05:05 pm »
If you are going to all the trouble of making a KVD, it is worth trying to source some lower temp coefficient parts for the first decade at least. Perhaps the first two decades.

You really notice the drift of 25ppm parts with 10V or more applied. The trouble is if you are using a KVD and you see the voltage drifting a little, it is easy to loose confidence in the accuracy. If you see it sitting rock solid at the set voltage, then you have confidence.  It is really worth investing that extra bit of effort and make something really good.

The negative is that if you buy a small number of 10ppm or better resistors, you definitely cannot select a matching set as well - you will need a trimmer pot on each one.

Now, back to your question on matching,  40ppm is 0.004%.  You need one superb set for the first decade. The second can be a little worse. The third worse again.  So it comes down to how many resistors do you need to get 11 resistors within 0.004%. The chances are good with 100 resistors. With 150, you probably can do it easily.  With 50 0.1%, you may be able to find 11 within 0.01%. 

I wouldn't be overly concerned about using all 10K resistors in the KVD which means you just buy lots of the one value.  When the 720a was designed, they didn't have the pico-amp fet input opamps that are available now to make a output buffer amplifier. So to minimize the output resistance, they used lower values in the lower decades.  They were often relying on very sensitive mechanical galvanometers that can't show lower currents then a few microamps.

You cannot put a 10Mohm multimeter across a KVD output - it draws too much current and will change the output voltage, so to use it, you either need a very high impedance input, or you use the KVD to compare with a second voltage, and you are nulling the current between the two.

Richard
 

Offline fmaimonTopic starter

  • Supporter
  • ****
  • Posts: 165
  • Country: br
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #45 on: November 18, 2011, 12:00:33 am »
If you are going to all the trouble of making a KVD, it is worth trying to source some lower temp coefficient parts for the first decade at least. Perhaps the first two decades.

You really notice the drift of 25ppm parts with 10V or more applied. The trouble is if you are using a KVD and you see the voltage drifting a little, it is easy to loose confidence in the accuracy. If you see it sitting rock solid at the set voltage, then you have confidence.  It is really worth investing that extra bit of effort and make something really good.

That's why I'm thinking on using 25ppm instead of 50ppm. With 10k resistors and 10V, you get about 100uW dissipation in each one and that's about 0.015C temp rise (150 C/W, according to the resistors datasheets). At 50ppm/C it is less than 1 ppm of error if my math is correct. The lower thermal resistance is why I choose 0.5W resistors instead of 0.25W or 0.125W.

This 10ppm resistor looks nice as, according to the datasheet, its thermal resistance is only 110C/W, but it's expensive, as a bag of 250 is nothing less than $165!  :o
 
Quote from: amspire
The negative is that if you buy a small number of 10ppm or better resistors, you definitely cannot select a matching set as well - you will need a trimmer pot on each one.

Now, back to your question on matching,  40ppm is 0.004%.  You need one superb set for the first decade. The second can be a little worse. The third worse again.  So it comes down to how many resistors do you need to get 11 resistors within 0.004%. The chances are good with 100 resistors. With 150, you probably can do it easily.  With 50 0.1%, you may be able to find 11 within 0.01%. 

These numbers are for 1% resistors, right? I did check Dave's 1% resistor data (400 resistors) and it had 4 sets of 40 ppm. So I assumed (ASS-U-ME?) that 0.1% needed 10 times less resistors to get about the same numbers of sets.

Quote from: amspire
I wouldn't be overly concerned about using all 10K resistors in the KVD which means you just buy lots of the one value.  When the 720a was designed, they didn't have the pico-amp fet input opamps that are available now to make a output buffer amplifier. So to minimize the output resistance, they used lower values in the lower decades.  They were often relying on very sensitive mechanical galvanometers that can't show lower currents then a few microamps.

That's a good idea. I'll just need 10K and 25K (24.3K + 1K pot) resistors.

Quote from: amspire
You cannot put a 10Mohm multimeter across a KVD output - it draws too much current and will change the output voltage, so to use it, you either need a very high impedance input, or you use the KVD to compare with a second voltage, and you are nulling the current between the two.

I won't do that. The output will only go to my Fluke 8840A (>10Gohm) or a LT1050 or similar...
 

Offline Conrad Hoffman

  • Super Contributor
  • ***
  • Posts: 1928
  • Country: us
    • The Messy Basement
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #46 on: November 18, 2011, 01:59:47 am »
Based on experience with supposedly good expensive wire wound resistors, and with the bags of 200 plain metal films I'd get from Digikey for $9 (a while back), I'm going to suggest that you go through the exercise with the cheap metal films first. You're going to learn things about matching and TC that will serve you well if you try to do it with better parts. You can even compare TC by immersing resistors in warmed mineral oil. One thing I learned was that the resistors in a bag didn't necessarily have the same TC. The other was that soldering was very risky business and I always needed to have a few extra matched resistors to fine tune a decade after it was all assembled. This was true for the wire wounds as well, and I did use heat sinks. Trying to do a 1 ppm first decade is remarkably difficult and there's a lot to be said for trimmers, if you choose wisely. The Fluke uses trimmers, as does the Analogic source. Loebe Julie didn't, nor did ESI, at least on the one I've got. General Radio didn't either, but theirs was only 4 decades, though surprisingly good decades if one measures one in good shape. Of course you could also look for their precision ratio transformer that was good to better than 1 ppm and considered a primary divider by NIST. It was a close relative of the one used in their capacitance bridge.
 

HLA-27b

  • Guest
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #47 on: November 18, 2011, 03:41:53 am »
I was pondering the idea of constructing a KVD on the cheap a while ago. My line of thinking was, instead of purchasing expensive resistors, to buy a spool of low tcr wire like evanohm and wind it around a glass rod in order to obtain a whole decade. Then I could tap that at exactly determined intervals to obtain decade divisions.

The best way to tap leads seemed to be to weld them instead of soldering them. Perhaps by discharging a cap through the junction. But I have never tried this so I don't know if it would work in high precision resistors.

Another point I considered was the winding method. Normal helical wining would make the resistor inductive. There are low inoctance windings like bifilar and Ayrton - Perry windings but these are hard to tap precisely because the exact point may not be exposed. So my idea is to wind the wire around two glass rods simultaneously in a figure of eight fashion which would cancel out inductance and at the same time expose the wire enough for easy tapping.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #48 on: November 18, 2011, 03:45:48 am »
Quote from: amspire
The negative is that if you buy a small number of 10ppm or better resistors, you definitely cannot select a matching set as well - you will need a trimmer pot on each one.

Now, back to your question on matching,  40ppm is 0.004%.  You need one superb set for the first decade. The second can be a little worse. The third worse again.  So it comes down to how many resistors do you need to get 11 resistors within 0.004%. The chances are good with 100 resistors. With 150, you probably can do it easily.  With 50 0.1%, you may be able to find 11 within 0.01%. 

These numbers are for 1% resistors, right? I did check Dave's 1% resistor data (400 resistors) and it had 4 sets of 40 ppm. So I assumed (ASS-U-ME?) that 0.1% needed 10 times less resistors to get about the same numbers of sets.

No that is for 0.1%. It is likely that the 0.1% resistors will have a wider spread within the 0.1% limits then Dave had with his 1% limits. But it is one of those things that you will not know till you get the resistors. You could find 90% are within 0.01% if you are lucky.

But you are probably right. I was probably being far too conservative.
Quote
Quote from: amspire
I wouldn't be overly concerned about using all 10K resistors in the KVD which means you just buy lots of the one value.  When the 720a was designed, they didn't have the pico-amp fet input opamps that are available now to make a output buffer amplifier. So to minimize the output resistance, they used lower values in the lower decades.  They were often relying on very sensitive mechanical galvanometers that can't show lower currents then a few microamps.
That's a good idea. I'll just need 10K and 25K (24.3K + 1K pot) resistors.

A 1K pot is way to big. You will regret it.  If you have resistors that match between decades to at least 0.05%, you don't want the pot doing much more then that. I would have the 24.3K (25ppm) and a 820 ohm (1% 100ppm) in series and put a 10K pot across the 820 ohm resistor. I prefer this circuit as only a fraction of the current is going through the pot wiper.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #49 on: November 18, 2011, 04:02:05 am »
I was pondering the idea of constructing a KVD on the cheap a while ago. My line of thinking was, instead of purchasing expensive resistors, to buy a spool of low tcr wire like evanohm and wind it around a glass rod in order to obtain a whole decade. Then I could tap that at exactly determined intervals to obtain decade divisions.

The best way to tap leads seemed to be to weld them instead of soldering them. Perhaps by discharging a cap through the junction. But I have never tried this so I don't know if it would work in high precision resistors.

Another point I considered was the winding method. Normal helical wining would make the resistor inductive. There are low inoctance windings like bifilar and Ayrton - Perry windings but these are hard to tap precisely because the exact point may not be exposed. So my idea is to wind the wire around two glass rods simultaneously in a figure of eight fashion which would cancel out inductance and at the same time expose the wire enough for easy tapping.

You could try that, but it is not at all easy.  Lets say you had 10 meters of wire per section - 100 meters overall. to get a match of 0.001%, you have to tap with a 0.1mm accuracy. If you use a finer wire so it is 100 meters per section, then you are talking about 1.1km of wire for the whole network. All that 1.1 km has to be wound perfectly with a minimum stress on the wire (stresses of resistance wire is one of the biggest causes of long term drift in a wirewound resistor). And that is just one divider!

In a KVD, the tap has to carry half the current, so the tap has to be done in a way that will not degrade with corrosion of the wire.

Richard
 

HLA-27b

  • Guest
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #50 on: November 18, 2011, 04:45:56 am »
0,1mm is easy to achieve with steady hand, with aditional gentle filing it should be possible to improve precision by another order of magnitue. As you said strain would be a major concern. The initial strain from the wining process could be eliminated by annealing the assembly after winding.

On the other hand since glass has much lower tce than the metal wire, we could expect the wire to slacken when temperature is hot and to tighten (and get strained) when cold. If the wire is wound in figure of eight fashion there would be enough slack to prevent this from happening in most cases.

To prevent corrosion either immersion in oil or enclosing the resistors in inert gas should do the trick. It seems like a big deal but after the trouble undergone to achieve 1 ppm it should seem trivial.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #51 on: November 18, 2011, 06:11:46 am »
0,1mm is easy to achieve with steady hand, with aditional gentle filing it should be possible to improve precision by another order of magnitue. As you said strain would be a major concern. The initial strain from the wining process could be eliminated by annealing the assembly after winding.

You cannot really anneal it. The wires are coated in varnish, so the heat would destroy the varnish. When making the Thomas Standard one Ohm resistors, they anneal the wire at something like 500 degC from memory. Not sure if it is possible to do that with a large value resistor on a former.
Quote

On the other hand since glass has much lower tce than the metal wire, we could expect the wire to slacken when temperature is hot and to tighten (and get strained) when cold. If the wire is wound in figure of eight fashion there would be enough slack to prevent this from happening in most cases.
To get it right is not easy. The former and the wire probably has different expansion coefficients, but also the glass has a greater thermal mass. So the wire can be hotter then the glass or cooler. It is hard to stop the wire being stressed at some point.

I remember reading about the Hewlett Packard refinement of the Rosa Standard resistor (the secondary standard resistors in the 40mm diameter cans with the two bent conductors out the top). I think the Rosa resistor uses a thin-walled brass tube as the former, and HP used something like a thin thickness polyester tube as the former that gets immersed in the oil. The idea is the polyester is not strong enough to stress the wire.

Quote

To prevent corrosion either immersion in oil or enclosing the resistors in inert gas should do the trick. It seems like a big deal but after the trouble undergone to achieve 1 ppm it should seem trivial.
« Last Edit: November 19, 2011, 12:38:19 am by amspire »
 

HLA-27b

  • Guest
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #52 on: November 18, 2011, 07:58:42 am »
You cannot really anneal it. The wires are coated in varnish, so the heat would destroy the varnish. When making the Thomson Standard one Ohm resistors, they anneal the wire at something like 500 degC from memory. Not sure if it is possible to do that with a large value resistor on a former.

Uncoated wire perhaps. Strain should not be a problem if wound with a little slack. As long as windings do not short it should work.
I should seriously try to get my gands on some low tempco wire then...
 

Offline fmaimonTopic starter

  • Supporter
  • ****
  • Posts: 165
  • Country: br
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #53 on: November 18, 2011, 12:48:00 pm »
The other was that soldering was very risky business and I always needed to have a few extra matched resistors to fine tune a decade after it was all assembled. This was true for the wire wounds as well, and I did use heat sinks.

I'm thinking on soldering the resistors with their bodies immersed in deionized water, with only the tips of the leads outside the water. This may help preventing the heat changing the value of the resistors.

A 1K pot is way to big. You will regret it.  If you have resistors that match between decades to at least 0.05%, you don't want the pot doing much more then that. I would have the 24.3K (25ppm) and a 820 ohm (1% 100ppm) in series and put a 10K pot across the 820 ohm resistor. I prefer this circuit as only a fraction of the current is going through the pot wiper.

Ok, I'll do that.

Thank you,
Felipe Maimon
 

Offline Conrad Hoffman

  • Super Contributor
  • ***
  • Posts: 1928
  • Country: us
    • The Messy Basement
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #54 on: November 18, 2011, 03:41:28 pm »
Having built a couple high precision resistance standards, I can tell you that something as soft as a spray coating of Krylon will completely destroy the stability of the standard. The ones I built used a bite of wire, held at one end with a piece of thread, and loosely wound on a brass tube. I think I threaded the OD of the tube so the wire would have a nice predictable place to sit. IMO, you could do a great job with the KVD using wire, but for one problem. To get any sort of higher values using wire, you have to go to a really small gage. That still means a huge number of turns and working with the stuff is no picnic. Don't over estimate the inductance cancellation of the various winding schemes, as they work less well than you might think. Flat windings on mica cards are easiest to build. I've used other things as well, from pieces of PC board, to metal, to plastic. I saw a new (to me) form in diyaudio.com that I want to try. It was essentially two small star shaped pieces, with a stand off in the middle. The wire was wound top to bottom around the outside, with a phase shift between the legs of the top, vs the bottom. This gave a winding very similar to an RF inductor, where the wires are never close and parallel, but cross each other at an angle. It looked like it would give far less inductance than the various old fashioned ways you find in the electrical measurement texts like Stout, Laws, Farmer, Harris, Terman et al. FWIW, I don't think the high temp bake is as critical as for under 1 ppm standards like the Thomas resistors. Also, for whatever reason, people seem to use manganin for low resistance standards and Evanohm for higher ones. There are some really good "800 series" wires from Kanthal (formerly H.P. Reid in Palm Coast, Florida, USA) that have both low TC and low thermal emf to copper. So who's ready to machine up some tellurium copper binding posts?
 

Offline nukie

  • Frequent Contributor
  • **
  • Posts: 799
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #55 on: November 18, 2011, 11:27:44 pm »
I don't mind turning a couple but where do I find that raw material?
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #56 on: November 18, 2011, 11:52:38 pm »
There are some really good "800 series" wires from Kanthal (formerly H.P. Reid in Palm Coast, Florida, USA) that have both low TC and low thermal emf to copper. So who's ready to machine up some tellurium copper binding posts?

I can't seem to find any mention of 800 series resistance wire. The best wire from Kanthal current products seems to be +/- 15ppm/C coefficient wire. 

http://www.kanthal.com/products/materials-in-wire-and-strip-form/wire/resistance-heating-wire-and-resistance-wire/list-of-alloys/

Richard
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #57 on: November 19, 2011, 12:17:02 am »
There are some really good "800 series" wires from Kanthal (formerly H.P. Reid in Palm Coast, Florida, USA) that have both low TC and low thermal emf to copper. So who's ready to machine up some tellurium copper binding posts?

I can't seem to find any mention of 800 series resistance wire. The best wire from Kanthal current products seems to be +/- 15ppm/C coefficient wire. 

http://www.kanthal.com/products/materials-in-wire-and-strip-form/wire/resistance-heating-wire-and-resistance-wire/list-of-alloys/

I have found  Jelliff Alloy 800 that could be like the "Series 800" wire.  It is rated at 5ppm over a -55 to 150 C range. They don't specify the coefficient near room temperature. It has good solderability and low EMF with copper.

http://jelliff.thomasnet.com/item/resistance-wire/jelliff-alloy-800-electrical-resistance-alloy/w08000056?
 

Offline Conrad Hoffman

  • Super Contributor
  • ***
  • Posts: 1928
  • Country: us
    • The Messy Basement
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #58 on: November 19, 2011, 06:38:07 am »
Most of what I know is old and obsolete! I think the old name was Rediohm 800. That Jelliff looks like the stuff. Usually it has to be spot welded because soft solder won't wet it. There's also a Kanthal wire down the list a bit that's available in 5 and 10 ppm/C- the Nikrothal LX. These are more modern and higher performance than manganin, but manganin has more history on its behavior since its been studied so much. I know the Julie KVDs used tellurium copper binding posts and you can get the copper rod from speciallty copper/brass suppliers, but no idea the cost. Then we have to talk about the low thermal emf solder used by Leeds & Northrup. It was loaded with cadmimum, so might be a big no no today. I believe they put a dab of paint (green?) on any joints that used it in their equipment and standards.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #59 on: November 19, 2011, 07:20:21 am »
I found out a bit about making resistors.

After winding, you want to age them - cycling them between room temperature and a maximum temperature. So if you are making resistors or buying, it would make a lot of sense to set up a rig to cycle current on and off for a few days before testing the resistors. Cycling between room temperature and 100 deg C sounds reasonable.

If you have bare resistance wire, you never want to let the wire get about 100deg C or it will oxidize and you need to pickle it to remove the oxide.

With enamel coated wire, you have to keep the maximum temperature below 140 degrees so you do not damage the varnish.

One of the main problems with soldering is the high temperature coefficient of tin - 4600ppm/degC. That is why it is best to spot weld wire rather then solder.  However, for a 10K resistor, all you need is for the solder resistance to be less than 10mOhms and you are fine. Amazing the small things that make a difference when you are working to 1ppm.

Another thing that could be tried is if you have resistors - wirewound or metal film - and they have a slight negative coefficient, you could wind some extra turns of copper wire on the resistor and put it in series. Copper has a positive coefficient of something like 4000ppm.  So if you measure a 10K at different temperatures and decide it has a -10ppm coefficient near room temperatures, you could add 25 ohms of thin copper wire in series and you would then get a zero temperature coefficient. It would probably be easier to use lower resistances for the divider like 3k3 and put 3 in series for the first decade. That way, there will be less copper wire needed.

But don't tell anyone - this idea is patented in 1976 by a Japanese guy. Since when was combining different temp coefficients together a new idea - what kind of idiots are in the patent offices?

Richard
« Last Edit: November 19, 2011, 07:43:01 am by amspire »
 

Offline robrenz

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
  • Real Machinist, Wannabe EE
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #60 on: November 19, 2011, 12:01:37 pm »
tellurium copper rod   http://www.onlinemetals.com/merchant.cfm?pid=15266&step=4&showunits=inches&id=1112&top_cat=0

The tellurium is for machinabilty not electrical properties.  Pure copper is nasty to machine.

Offline nukie

  • Frequent Contributor
  • **
  • Posts: 799
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #61 on: November 20, 2011, 01:48:28 am »

The tellurium is for machinabilty not electrical properties.  Pure copper is nasty to machine.

Tellurium by itself might not be good for electrical conductor but it's Tellurium Copper which makes it ideal for low thermal emf connection.

http://www.tequipment.net/Pomona3770.asp

Very nice it makes copper softer and easier to machine. I have machined 101 oxygen free copper with carbide and coolant. Need to add a bit of patient and slow feed rate.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #62 on: November 20, 2011, 02:14:39 am »
A thing I have never understood.

Why is gold plated tellurium copper alloy better for EMF then gold plated copper or gold plated brass terminals? I cannot see how the metal under the plating makes any difference.

Now if you are talking about unplated terminals , then it does make sense to me.

Richard
 

Offline robrenz

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
  • Real Machinist, Wannabe EE
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #63 on: November 20, 2011, 02:40:10 am »
A thing I have never understood.

Why is gold plated tellurium copper alloy better for EMF then gold plated copper or gold plated brass terminals? I cannot see how the metal under the plating makes any difference.

Now if you are talking about unplated terminals , then it does make sense to me.

Richard

  Doesnt the base metal and the plating form a another thermal emf junction.  that would explain why they are different even though they all have gold plating. Any two disimilar metals in contact whether its just physical contact or wetted metals(solder) or plating are a thermocouple arent they?  My understanding is all these need a temperature gradient to generate a voltage. There is no thermal EMF if there is no temperature gradient.

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #64 on: November 20, 2011, 03:02:36 am »
As far as I can see, if the connector is gold plated, and you are not touching the underlying base metals at all, then any gold to base metal EMF voltages are irrelevant.  All that matters is the test lead to gold EMF voltage.

The only thing I can think of is if you are working at a high current, then there can be local heating at the contact points with the leads.  Perhaps the gold to base metal EMF voltage differentials caused by this heat are enough to cause a voltage gradient on the surface of the gold plating. The must be a reason like that since gold plated brass does appear to be a no-no for low EMF work.

Richard


 

Offline robrenz

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
  • Real Machinist, Wannabe EE
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #65 on: November 20, 2011, 03:30:19 am »
As far as I can see, if the connector is gold plated, and you are not touching the underlying base metals at all, then any gold to base metal EMF voltages are irrelevant.  All that matters is the test lead to gold EMF voltage.

Richard

I am not an authority, I am just thinking out loud.  IMO you cant pick and choose what different metal to metal joints you are going to account for. Every single connection in a circuit that is a dissimilar metal counts as a themocouple. and the connection of a plating to the base metal counts just as much as a copper wire crimped to the gold plating.


As far as I kinow the EMF of a thermocouple is not generated at the junction of the two metals. It is generated in the thermal gradient of each wire.  So how does this apply when you have few microns of plating?

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #66 on: November 20, 2011, 04:58:43 am »
As far as I can see, if the connector is gold plated, and you are not touching the underlying base metals at all, then any gold to base metal EMF voltages are irrelevant.  All that matters is the test lead to gold EMF voltage.

Richard

I am not an authority, I am just thinking out loud.  IMO you cant pick and choose what different metal to metal joints you are going to account for. Every single connection in a circuit that is a dissimilar metal counts as a themocouple. and the connection of a plating to the base metal counts just as much as a copper wire crimped to the gold plating.


As far as I kinow the EMF of a thermocouple is not generated at the junction of the two metals. It is generated in the thermal gradient of each wire.  So how does this apply when you have few microns of plating?

Here is what I have found out so far. The best connections are copper to copper  - less then 0.3 uV/C.  The problem is that copper oxidizes easily, and copper to copper oxide has 1000uV/C EMF. Obviously any copper oxide is a disaster. So if you were after the absolute best possible EMF performance, you would use a solid copper wire with a copper post, and you would lightly sand both before using to remove of all traces of copper oxide.

Copper to gold or silver is 0.5uV/C.  So perhaps they flash the copper with enough gold to restrict corrosion, but for EMF reasons, they don't put the usual nickel plating first to seal the copper. This means the copper can still oxidize through the pores in the gold. Tellurium-Copper has a similar EMF properties to pure copper, but has a high corrosion resistance, so you don't get much copper oxide leaching through the pores in the gold plating. Tellurium bonds tightly enough to copper to prevent reaction of the copper with oxygen. I guess if the Gold plating is not very thick, the Tellurium-Copper somehow still dominates in the EMF voltages.

Tin/Lead solder connections to copper is pretty bad at 1 to 5uV/C, but if you have a copper-solder-copper joint, then to you need a temperature differential across the joint to generate an EMF. Definitely could happen at high currents, but probably not at low currents. Cadmium-tin solder is down at 0.3uV/C but is very toxic. Companies like Leeds and Northrup used it a lot, so be warned if you are working with old precision equipment.  10%tin/90% lead solder is sometimes used as a safe compromise.

Richard
 

HLA-27b

  • Guest
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #67 on: November 20, 2011, 10:12:47 am »
Do we know the thermal noise generated by a Manganin - Copper connection ?

I recall that insteard of copper, phosphor bronze was preferred as connection material in a paper about precise measurement of temperature.
If its thermal noise properties are suitable phosphor bronze is decidedly easier to keep free of oxidation.
 

Offline nukie

  • Frequent Contributor
  • **
  • Posts: 799
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #68 on: November 20, 2011, 11:49:01 am »
Hold on, does oxygen free copper 101 turns into copper oxide? The HP34401a input connectors are all made of copper.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #69 on: November 20, 2011, 11:54:32 am »
Hold on, does oxygen free copper 101 turns into copper oxide? The HP34401a input connectors are all made of copper.

No, they are gold plated tellurium-copper alloy connectors. The specifications describe them as "Copper Alloy", but they are tellurium-copper Alloy.
« Last Edit: November 20, 2011, 12:07:51 pm by amspire »
 

Offline robrenz

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
  • Real Machinist, Wannabe EE
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #70 on: November 20, 2011, 03:06:45 pm »
What about the precision multi pulse capacitor discharge spot welders for making the connections to the resistance wire. Would the extremely quick but high temperature be worse for long term stability.

HLA-27b

  • Guest
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #71 on: November 20, 2011, 06:03:03 pm »
What about the precision multi pulse capacitor discharge spot welders for making the connections to the resistance wire. Would the extremely quick but high temperature be worse for long term stability.

We don't really know if it would. On one hand we hope that even if it changes value at least the ratio would stay the same. On the other hand one would anneal the whole thing just to be on the safe side.
 

Offline lowimpedance

  • Super Contributor
  • ***
  • Posts: 1245
  • Country: au
  • Watts in an ohm?
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #72 on: November 24, 2011, 01:15:20 am »
Just thought that it would be interesting to show a few pictures of an old ESI dekapot DP1311 (1OK) Kelvin Varley divider for reference on how a commercial product was built.
note; the 34401 has been set for an input R of 10g ohm.
(Since this is not a real review, and not many would be interested anyway I thought this would be the best place to post, hope no one minds).
cheers
John
The odd multimeter or 2 or 3 or 4...or........can't remember !.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #73 on: November 24, 2011, 02:10:56 am »
I have been contemplating the issue of building a KVD. I am not impressed with the quality of affordable resistors. Some people are luck enough to snag a 6/7 digit KVD at a bargain, but the sellers seem to now know that a good one is worth $1000+

So how else can you get great resistors + switches affordably?

I saw this and bought it for $16.50. Worth half a US family average income back in 1960. See the row of decade switches - could that be a Kelvin-Varley divider?



I have it now, and it is not a KVD, but what it does have is very interesting and as useful as a KVD. I am sure someone in the forum recognizes what the switches are. From the build dates and the last calibration date, I suspect it was used for calibrating avionics in Lockheed Martin Orion P-3C aircraft. I know for sure it was built for Lockheed Martin. Last calibrated  about 2000 and at this time, Australia was upgrading all its Orions to AP-3C's with all new avionics. I think I will do a teardown, before I "re-manufacture" it.

The next idea is that the Fluke Differential Voltmeters on ebay all contain Kelvin-Varley dividers, and you do see some on sale under $100.

Now there are some that have 3 decades + a 0-100 pot and some that have 4 decades + a 0-100 pot.  The 4 decade ones are the way to go as they have a fully calibratable first decade, and I think they use 0.01% resistors in the second decade. So using it as it is, you can get at least 0.001% out of the divider, and knowing Fluke, it may be much better. The first decade is usually 5K resistors in the divider.

The 3 decade ones are all fixed resistors with no calibration. The first decade is 100K resistors in the divider which is very high.  To have less then  0.0001% error from the following voltmeter or amplifier, it would need an input resistance of over 1000 GOhms.

I feel a bit guilty suggesting messing with these great old instruments, but if you turn something that is now hardly used into something useful, well it is not such a bad thing.

Richard
« Last Edit: November 24, 2011, 02:24:36 am by amspire »
 
The following users thanked this post: cellularmitosis

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #74 on: November 24, 2011, 02:46:11 am »
Just thought that it would be interesting to show a few pictures of an old ESI dekapot DP1311 (1OK) Kelvin Varley divider for reference on how a commercial product was built.
note; the 34401 has been set for an input R of 10g ohm.
(Since this is not a real review, and not many would be interested anyway I thought this would be the best place to post, hope no one minds).
cheers
John

Thanks John. I had been wondering what was in them.

I mentioned the price before  - I would love to have a home lab kitted out to the same dollar value as that one divider.
 

Offline lowimpedance

  • Super Contributor
  • ***
  • Posts: 1245
  • Country: au
  • Watts in an ohm?
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #75 on: November 24, 2011, 03:22:29 am »
I mentioned the price before  - I would love to have a home lab kitted out to the same dollar value as that one divider.
[/quote]

Ouch, I had better treat it well  then :)
This unit plus a 1 k version were rescued from dumpster doom at work (some old custom bit of gear that no doubt cost a few arms and legs in its day!!)
By the way there is a Fluke 720a collecting dust in the instrument store at work, if anyone is interested in some internal pictures let me know.
John
The odd multimeter or 2 or 3 or 4...or........can't remember !.
 

Online BravoV

  • Super Contributor
  • ***
  • Posts: 7547
  • Country: 00
  • +++ ATH1
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #76 on: November 24, 2011, 03:33:49 am »
By the way there is a Fluke 720a collecting dust in the instrument store at work, if anyone is interested in some internal pictures let me know.

John, yes please, really love to see how is it look inside.

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #77 on: November 24, 2011, 03:35:01 am »
This unit plus a 1 k version were rescued from dumpster doom at work (some old custom bit of gear that no doubt cost a few arms and legs in its day!!)

EIT only mention the price of the 3 decade version, but that one is probably worth well over $6000 to buy one new - possibly creeping up towards $10K.  A pretty good find in a dumpster.
« Last Edit: November 24, 2011, 03:37:04 am by amspire »
 

Offline lowimpedance

  • Super Contributor
  • ***
  • Posts: 1245
  • Country: au
  • Watts in an ohm?
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #78 on: November 24, 2011, 11:19:30 pm »


EIT only mention the price of the 3 decade version, but that one is probably worth well over $6000 to buy one new - possibly creeping up towards $10K.  A pretty good find in a dumpster.
[/quote]
 
Well that is rather scary for what is in it (of course precision costs and a little, and a bit of profiteering on top too!! ::).)


By the way there is a Fluke 720a collecting dust in the instrument store at work, if anyone is interested in some internal pictures let me know.

John, yes please, really love to see how is it look inside.

Ok as requested attached is a collection of pic's of the Fluke 720a with covers off.
The first thing that hits you are the trailer load of trimmers (a good number also hidden from view in the pic's.)
The green block at the rear is a 'can of precision resistors in oil!.
The green switches are special low thermal types (Leeds and Northrup)
The reference used in the 'lash up test' is a ltz1000 around 12 years young'
Again the 34401 is set for an input r of >10g ohm.
I hope that satisfies your curiosity
cheers
John
The odd multimeter or 2 or 3 or 4...or........can't remember !.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #79 on: November 24, 2011, 11:49:44 pm »
Quote
EIT only mention the price of the 3 decade version, but that one is probably worth well over $6000 to buy one new - possibly creeping up towards $10K.  A pretty good find in a dumpster.

Well that is rather scary for what is in it (of course precision costs and a little, and a bit of profiteering on top too!! ::).)
Not necessarily profiteering. It looks like the resistors are more or less hand made. They are probably made, aged and then someone has to adjust the value of every resistor until the whole set of resistors match. With things like this, it is easy to match the resistors, construct everything, a night or two in the thermal chamber to age, test again, and then find it is a little out. Then you pull it apart again, adjust, assemble, another night in the thermal chamber,  and check again. This is repeated till it is right.

It looks like there are no adjustments in it, so everything has to be accurate.

I don't think they can be made very quickly at all, and at that price, they will be made to custom order.
Quote

The green block at the rear is a 'can of precision resistors in oil!.

The can of resistors for the first decade is the huge silver looking box taking up most of the rear with the "CAUTION" sticker on. It didn't look green to me, but the colour may not show up in the photo. It should contain 12 sets of 3 resistors, so 36 large Fluke precision wirewound resistors in total. I do see large green switches.

Thanks for the photos. I haven't seen inside a 720a before. So that is all you need to make a 1 part in 10-7 divider!  :(

Richard
« Last Edit: November 24, 2011, 11:54:59 pm by amspire »
 

Offline lowimpedance

  • Super Contributor
  • ***
  • Posts: 1245
  • Country: au
  • Watts in an ohm?
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #80 on: November 25, 2011, 12:11:43 am »
Richard
 When looking at it from that perspective of hand made it can easily mount up in price.
Did a search and found the price for the DP1311 to be mid $5000 , must say I'm still a little surprised.
May be it has something to do with being around this sort of gear in my work that I dont give it much thought as to the costs.
(I dont do the purchasing of this type of gear! , more just general test equipment).
Sorry about the color rendition in the photo, but the box is green.
The manual says if something goes wrong inside the box, send it back ( read betwwen the lines; you are screwed!).

John
The odd multimeter or 2 or 3 or 4...or........can't remember !.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #81 on: November 25, 2011, 12:26:56 am »
The manual says if something goes wrong inside the box, send it back ( read betwwen the lines; you are screwed!).

I think the current price for the IET KVD-700 (the same as the Fluke 720a) has a "Starting Price" of $28,895, and that box could be worth at least $20,000 of that.

Richard
 

Online BravoV

  • Super Contributor
  • ***
  • Posts: 7547
  • Country: 00
  • +++ ATH1
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #82 on: November 25, 2011, 03:16:20 am »
Ok as requested attached is a collection of pic's of the Fluke 720a with covers off.
The first thing that hits you are the trailer load of trimmers (a good number also hidden from view in the pic's.)
The green block at the rear is a 'can of precision resistors in oil!.
The green switches are special low thermal types (Leeds and Northrup)
The reference used in the 'lash up test' is a ltz1000 around 12 years young'
Again the 34401 is set for an input r of >10g ohm.
I hope that satisfies your curiosity
cheers
John

Thanks John, great pics.

About those truck load of trimmers, just curious, I assume those golden trimmers are not like your average trimmer out there right ?

It didn't look green to me,

May be all your gadgets/instruments, monitor included are maintained and fully calibrated properly, you just forget to calibrate your eyes Richard.  ;D  ...j/k ...

Btw, it looks green here.

Offline lowimpedance

  • Super Contributor
  • ***
  • Posts: 1245
  • Country: au
  • Watts in an ohm?
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #83 on: November 25, 2011, 03:39:08 am »
Ok as requested attached is a collection of pic's of the Fluke 720a with covers off.
The first thing that hits you are the trailer load of trimmers (a good number also hidden from view in the pic's.)
The green block at the rear is a 'can of precision resistors in oil!.
The green switches are special low thermal types (Leeds and Northrup)
The reference used in the 'lash up test' is a ltz1000 around 12 years young'
Again the 34401 is set for an input r of >10g ohm.
I hope that satisfies your curiosity
cheers
John

Thanks John, great pics.

About those truck load of trimmers, just curious, I assume those golden trimmers are not like your average trimmer out there right ?

It didn't look green to me,

May be all your gadgets/instruments, monitor included are maintained and fully calibrated properly, you just forget to calibrate your eyes Richard.  ;D  ...j/k ...

Btw, it looks green here.
Actually they are just single turn trimmers all 25ohm ! but I suspect a little bit better quality than some :)
Also there are a truck load of multiturn trimmers under the PCB too (access from the front, shown with the 'flap' up).
The odd multimeter or 2 or 3 or 4...or........can't remember !.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #84 on: November 25, 2011, 03:44:37 am »
It didn't look green to me,
May be all your gadgets/instruments, monitor included are maintained and fully calibrated properly, you just forget to calibrate your eyes Richard.  ;D  ...j/k ...

Btw, it looks green here.

I thought that was a board sitting on top of the box, but now I have done the eye recalibration, I see it is the top of the resistor box.
 

HLA-27b

  • Guest
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #85 on: November 25, 2011, 08:37:15 am »
Thank you very much for the pictures John,

One thing that I could not see from the pictures - What is the winding type of the resistors? Is it normal helical winding or is it bifilar?
 

Offline lowimpedance

  • Super Contributor
  • ***
  • Posts: 1245
  • Country: au
  • Watts in an ohm?
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #86 on: November 25, 2011, 10:21:20 am »
Thank you very much for the pictures John,

One thing that I could not see from the pictures - What is the winding type of the resistors? Is it normal helical winding or is it bifilar?
Hang tight until Monday and I will check when I get to work.
John
The odd multimeter or 2 or 3 or 4...or........can't remember !.
 

Offline Conrad Hoffman

  • Super Contributor
  • ***
  • Posts: 1928
  • Country: us
    • The Messy Basement
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #87 on: November 25, 2011, 04:51:08 pm »
Cool, thanks for the photos; I'd never seen what's under her skirt either. IMO once you have a procedure down for building the resistors, the rest is just hard work. Remember that the values on the upper decades are high, so it involves a huge amount of very fine and difficult to work with wire. Creating a bight (I misspelled that some time back) is a PITA. Julie wound their resistors on plastic bobbins with separators. So many turns one way, then move to the next section and wind the other way. Loose is the rule to avoid stress. Then they added a short length of resistance wire externally on the switch terminal to trim to final value. Sensible solution, IMO, and they're said to remain stable for many years. The Fluke trimmers look just like the trashy ones I find in old hi-fi amps for bias. They may be wire wound. The low value resistors are card wound and very likely are a simple unidirectional winding. All the Fluke resistors I've seen are. For a while the Fluke KVDs could be had cheap, while everybody was going out of business in the USA, but the test equipment market seems to solidified at a high price now.
 

Offline lowimpedance

  • Super Contributor
  • ***
  • Posts: 1245
  • Country: au
  • Watts in an ohm?
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #88 on: November 27, 2011, 11:35:23 pm »
Just had a look at the ESI dekapot and the Fluke 720a and the card resistors are as wound as Conrad suggested. The others ...well cannot see inside the can!.
The single turn pots that ideed appear somewhat cheep on the pictures are wire wound and were made by CTS.
John
The odd multimeter or 2 or 3 or 4...or........can't remember !.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #89 on: November 28, 2011, 12:40:13 am »
Just had a look at the ESI dekapot and the Fluke 720a and the card resistors are as wound as Conrad suggested. The others ...well cannot see inside the can!.
The single turn pots that ideed appear somewhat cheep on the pictures are wire wound and were made by CTS.
John
They will be totally adequate. In the critical first decade, the way they are connected, they are adjusting the 10K divider stages by +/- 0.2 ohms. So they pick a Select-On_test resistor to get each divider section in the first decade matching to within 10ppm, and the pot only has to be stable to 1% to get a final 0.1ppm accuracy.

The pots may look cheap, but Fluke have probably made sure that the same type of resistance wire is used in both the pots and the divider resistors, so all the temperature drifts will match. They probably do a better job in the 720a then a modern 10 turn trimpot would do.

Richard
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #90 on: November 28, 2011, 02:19:18 am »
I am starting to pull the elements together for a KVM, so I think I will set a target.

I want to make a 6 digit KVM with a budget (excluding case) of $60. I could spend more money on better resistors, but I just don't want to. I thought about winding resistors, but the wire is not cheap at all, I don't know where to get good formers from, and when you are talking resistors in the 1000's of ohms, there is a lot of winding to be done with very fine wire. So scrap that.

Here is the plan:

Switches: $4 each  (ebay item 300562748033). I will review these switches in my next post.

Resistors: 3k3 3W 1% 50ppm metal film. $10 per 100 resistors. A total of 103 resistors will be needed, but I can use cheap  1% 1/w watt on the lower ranges. To calibrate, I will use 100 10 ohm 1% resistors.

Knobs: 6mm shaft cheap comfortable knobs with a printed mini-dvd disk stuck/screwed to the bottom with the 0 - 9 printed on it.

Terminals: Cheap gold plated brass 4mm terminals.  Thermal voltages are only a problem when temperatures differ, so I will make sure they don't.

Adjustment Pots: I will first test the resistors, then work out how much adjusting I will allow. Bulk packs of Chinese multiturn pots are easy to get.

The first decade will use three 3k3 resistors in series per divider. The other decades will use a single 3k3 resistor.

I will age all the resistors first. This will be done by connecting all the resistors in parallel and cycling between 80V and 0V for a day or two. To do the cycling, I will use a mains timer with alternate setting tabs up or down.

I will then work out the temperature coefficient of the resistors, and assuming it is a bit negative, I will wind some extra copper wire on the resistor to cancel out the temperature coefficient. This will be done by putting a bunch of resistors between two pieces of metal with just the leads sticking out, add some thermocouple probes, and heat to a bit over 100 deg in an oven. When it comes out, I will wrap in some insulation, and let the temperature stabilize. Then as the resistors slowly cool, I will measure the resistance of each. I only need the first decade done really well.  The each lower decade only has to be 1/10th as good as the decade above it. The copper wire has a positive coefficient of about 4000ppm/C, so if a 3k3 has a -20ppm coefficient near room temperature, I would need to add 33 ohms of copper wire in series with the resistor to get a zero coefficient. 40 gauge copper wire is 1 ohm per foot, so that would be 33 feet or 10M wound around the resistor. The 3W resistors are large, so there is plenty or room to add the copper wire.

Errors between resistors in a divider set are cancelled out by adding selected 10 ohm resistors in series. If the resistors turn out to be stable in the short term, I will aim for better then a  100ppm match in the first decade plus use a pot to adjust down to 1ppm. Realistically, if I can get 10ppm reliably from the KVM, I have done well.

Richard

« Last Edit: November 28, 2011, 03:03:25 am by amspire »
 

Offline lowimpedance

  • Super Contributor
  • ***
  • Posts: 1245
  • Country: au
  • Watts in an ohm?
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #91 on: November 28, 2011, 02:39:27 am »
Richard
 Lood forward to see the result when your done. The old 'sperry', couldnt bring yourself to use the switches from that (or are they not suitable?).
I like the idea of the mini dvd disk dial to get the numbers.

Just to note in the 720a oil filled resistor can, each section in the divider is made up of 4 resistors of 2.474k in series then externally in series with those is a combo of
 factory selected R + (8.45k+5k trimmer//100R), getting the temp co. right here must have been real fun. This divider uses the low thermal (green) switch.

cheers
John
The odd multimeter or 2 or 3 or 4...or........can't remember !.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #92 on: November 28, 2011, 02:51:56 am »
I received my 1 way 2 wafer switches.

https://www.eevblog.com/forum/index.php?action=dlattach;topic=5442.0;attach=16882

The ceramic wafers are 25mm diameter, and the shaft is a 6mm metric shaft.

It looks like the contacts are silver plated brass.

So are they good enough?

Quality: Contacts are very firmly riveted onto the ceramic wafer and do not move. Alignment is poor, so that some contacts are not centered properly. The contacts seem to work though. The batch I received had some tarnishing on the silver. Initial contact resistance was high - sometimes over 0.1 ohms, but after turning the switch back and forward 10 times, most contacts were 12 milliohms. I pulled some switches apart and detarnished the contacts and resistance reduces down to 9 milliohms on the good contacts. Some contacts can be bad, so get extra switches and pick the best one for the first decade. I think they are good enough to get started.  I compared it with some 50 year old super quality switches - the 50 year old ones had a contact resistance of 0.002 ohms with a variation of 0.0001 milliohms. To at anything like that today costs hundreds of dollars per switch.

Construction: Can be taken apart. Good quality ceramic wafers which are the  best for low leakage. Ceramic wafers typically have resistances between contacts of over 900GOhms. Heavy duty beryllium copper spring for the detents is not going to wear out or break.

Feel: Nice positive detents. Satisfied.

Thermal EMF voltages. In a practical test with soldered copper wires, and heating up one contact with my finger, it looks like there will be less then 0.3uV thermal voltage in total, so I am not going to worry there. If nothing is heating the pins significantly, it will not be measurable. The first decade resistors will not be mounted on the switch, and they are the most likely resistors to produce heat.

This switch is about $4 in one of quantities, and you can get much cheaper if you buy in quantity. I can get as low as about $1.60 if I buy hundreds. If you try and buy something similar at Digikey, Mouser, etc, you will probably pay 5 to 10 times the price.

So I will give them a go.  If I later get my hands on better switches, i can replace the higher decades.

Richard
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #93 on: November 28, 2011, 02:59:51 am »
Richard
 Lood forward to see the result when your done. The old 'sperry', couldnt bring yourself to use the switches from that (or are they not suitable?).
I like the idea of the mini dvd disk dial to get the numbers.

Just to note in the 720a oil filled resistor can, each section in the divider is made up of 4 resistors of 2.474k in series then externally in series with those is a combo of
 factory selected R + (8.45k+5k trimmer//100R), getting the temp co. right here must have been real fun. This divider uses the low thermal (green) switch.

cheers
John

I will do a seperate post on that Sperry today, because it is is very interesting.

I bet is would be really hard finding anyone who could make a switch today like that green 720a switch. Very consistant low resistance, high voltage as it has up to 1.1kv on it. Very low current leakage. If someone does make an equivalent, the cost would be incredible.
 

alm

  • Guest
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #94 on: November 28, 2011, 03:01:24 am »
Metal film should have a positive tempco if I remember correctly. Apart from this, sounds like a well thought-out project. Will make sure to follow your progress. Thermal EMF will be an issue if you go down to the uV/nV levels. Copper to gold is about 0.5 uV/K, so with 10mV across the divider, the resolution is 10nV, requiring the temperature of the two binding posts to be within less than 0.01 or so K. If you stick to reasonable voltages, it becomes less critical. With a resolution of 1uV, you only need to be within 1 K or so. Not unreasonable, but not in the 'you need a blowtorch to see any difference' territory either.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #95 on: November 28, 2011, 03:29:40 am »
Metal film should have a positive tempco if I remember correctly. Apart from this, sounds like a well thought-out project. Will make sure to follow your progress. Thermal EMF will be an issue if you go down to the uV/nV levels. Copper to gold is about 0.5 uV/K, so with 10mV across the divider, the resolution is 10nV, requiring the temperature of the two binding posts to be within less than 0.01 or so K. If you stick to reasonable voltages, it becomes less critical. With a resolution of 1uV, you only need to be within 1 K or so. Not unreasonable, but not in the 'you need a blowtorch to see any difference' territory either.

The thermal voltages are only an issue if something is generating a temperature differential. My testing showed that with the switch carrying no current,there is just no measureable thermal emf (with a 0.1uV resolution meter).

I will not mount the first decade resistors on the switch, so the only place where a significant heat can be generated to the switches is in the second decade resistors. These will be large 3W resistors and even if I apply 1000 volts to the KVM, their heat will be 3mW per resistor which could possibly cause a rise of one degree in the wire to the switch. With 10V applied to the KVM, the power will be 3uW per resistor which is nothing. So if I have a problem, it will only be at extreme voltages, and at 1000v, I can afford to have emf's up to 1mV without any change in accuracy. I think I would only trust the switches up to about 250V, and I do not have any supplies that go above 60Volts.

I will probably need to somehow shield the heat from the first decade resistors from the rest of the system, but I wont go as far as putting them in a can of oil.

As far as the terminals are concerned, I can put a foam insulator over them inside, so they are not getting any significant radiated heat. Thermal EMF's generated through handling the front panel terminals with fingers will change as they cool again, so if the emf's are relevant, I will wait till the drift stops. I am just not going to spend $20 per terminal to get the low EMF copper alloy ones.

Richard.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #96 on: November 28, 2011, 03:37:40 am »
Metal film should have a positive tempco if I remember correctly
The 1/4W one I have tested here have a negative tempco, but I will just have to wait and see what the 3W ones do when I get them.
 

Online BravoV

  • Super Contributor
  • ***
  • Posts: 7547
  • Country: 00
  • +++ ATH1
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #97 on: November 28, 2011, 03:41:16 am »
Richard, looking forward to see your build in progress in the new thread, lots of pics please !  :)

Offline fmaimonTopic starter

  • Supporter
  • ****
  • Posts: 165
  • Country: br
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #98 on: November 28, 2011, 12:06:27 pm »
Richard, where did you find those 3W 50 ppm resistors? Digikey nor Mouser seem to have them...
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #99 on: November 28, 2011, 12:42:17 pm »
Richard, where did you find those 3W 50 ppm resistors? Digikey nor Mouser seem to have them...

Only the best for me.

Brand is totally unknown, so I will have to see if they are good enough to equal the world famous "One Hung Low" resistors.  :)

http://www.uxcell.com/wire-leads-33k-ohm-metal-film-resistors-100-pcs-p-161865.html

I will test them first, and if after aging, I think I can get enough short term stability, I will press on. I ordered them today, so in a week or two, I should know.

I did look at Digikey and Mouser, but it seems as soon as you get above 1/4 watt, the resistor prices skyrocket.

I just have a feeling that if I can't afford to buy the Vishay 1ppm foil resistors, then I will get better results from nice big chunky resistors of lesser quality then expensive brand name 25ppm resistors in a small package. I think it is the duty of the hobbyist to try and do things incredibly cheaply that normally are done by spending big bucks. Well, it is fun anyway.

Now between the decades made of 3k3 resistors, you need to add a 8k2 resistor + some extra calibrating parts ( a pot + small resistors) that goes across the next decade. So I ordered 8k2 3W resistors as well.

Between the  first decade with 10K (three 3k3 resistors) per division and the second stage with 3k3 per division, I will need a resistor across the second decade of 50K. I haven't decided what I will use for it yet.

Richard.
« Last Edit: November 28, 2011, 12:43:58 pm by amspire »
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #100 on: November 28, 2011, 01:59:04 pm »
This is an interesting source of wirewound resistors:

http://www.surplussales.com/resistors/WireWound/WW5k-799k.html
 
The following users thanked this post: Overspeed

HLA-27b

  • Guest
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #101 on: November 28, 2011, 02:23:55 pm »
Just had a look at the ESI dekapot and the Fluke 720a and the card resistors are as wound as Conrad suggested. The others ...well cannot see inside the can!.
The single turn pots that ideed appear somewhat cheep on the pictures are wire wound and were made by CTS.
John

Many thanks John, you helped a lot. This taught me that there is no point in overengineering beyound the point that Fluke et. al. did. There is no need for example to be concerned with things like interwinding capacitance and high voltage behavior.
 

Offline fmaimonTopic starter

  • Supporter
  • ****
  • Posts: 165
  • Country: br
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #102 on: November 28, 2011, 02:54:19 pm »
Only the best for me.

Brand is totally unknown, so I will have to see if they are good enough to equal the world famous "One Hung Low" resistors.  :)

http://www.uxcell.com/wire-leads-33k-ohm-metal-film-resistors-100-pcs-p-161865.html

I saw them on ebay (150707751358), but they don't ship to Brazil... :(
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #103 on: November 28, 2011, 03:13:22 pm »

I saw them on ebay (150707751358), but they don't ship to Brazil... :(

Well let's see if they are good enough. If you need them, something can be worked out.

Richard.
 

Offline fmaimonTopic starter

  • Supporter
  • ****
  • Posts: 165
  • Country: br
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #104 on: November 28, 2011, 03:41:19 pm »
I've sent them an email just to be sure. Maybe just nobody has asked them...
 

Offline lowimpedance

  • Super Contributor
  • ***
  • Posts: 1245
  • Country: au
  • Watts in an ohm?
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #105 on: November 29, 2011, 01:59:04 am »


I will do a seperate post on that Sperry today, because it is is very interesting.

I bet is would be really hard finding anyone who could make a switch today like that green 720a switch. Very consistant low resistance, high voltage as it has up to 1.1kv on it. Very low current leakage. If someone does make an equivalent, the cost would be incredible.
[/quote]

One last post regarding the special switch used in the 720a. There are some like the two in the Fluke, made by Leeds and Northrup, at work. I have scanned the info sheet for them and attached here for those interested in the details of such a switch.
John
The odd multimeter or 2 or 3 or 4...or........can't remember !.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #106 on: November 29, 2011, 02:07:25 am »
What a brilliant switch! 1500V rating. 1 milliohm contact resistance.  1012 ohms insulation resistance after 2 months in 100% humidity. Amazing.

Thanks John.
 

Offline Conrad Hoffman

  • Super Contributor
  • ***
  • Posts: 1928
  • Country: us
    • The Messy Basement
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #107 on: November 29, 2011, 02:45:02 am »
That's an impressive switch; never realized how good they were. L&N made some high resolution bridges for testing RTDs and such. They sometimes go cheap because nobody has a clue what to do with them. I disposed of one years ago. In hindsight, it probably had more value from the switches in it, than as a bridge. I don't think it was a KVD, just a big divider network, but I could be wrong. L&N made some bizarre stuff; their big DPDT knife switch for reversing bridge connections comes to mind. I also have a couple of their projection galvos. Incredibly ugly things, sort of like a miniature gray bathtub lying on its back.
 

Offline quarks

  • Frequent Contributor
  • **
  • Posts: 874
  • Country: de
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #108 on: December 07, 2012, 09:48:47 pm »
I know this is quite old, but I wonder if there came out a final KVD and would be very interested to read/see it.
Thx quarks
 

Offline Spikee

  • Frequent Contributor
  • **
  • Posts: 568
  • Country: nl
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #109 on: January 02, 2013, 06:48:35 pm »
I am also interested if there is any progress .
Freelance electronics design service, Small batch assembly, Firmware / WEB / APP development. In Shenzhen China
 

Offline JBeale

  • Frequent Contributor
  • **
  • Posts: 298
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #110 on: January 29, 2013, 07:37:29 pm »
I'm also curious how this project came out, and also the quality of the 3W 3.3k 1% 50ppm/C resistors, which are being sold today at $15/100.

By the way, if anyone else is building their own KVD on the cheap, I also recommend Conrad Hoffman's 1996 article in "Electronics Now"  (plus voltage reference, and null meter). The link in the original post of this thread was on his own web page as separate JPG scans, but the whole series is also assembled at k04bb.com as one convenient PDF.
http://www.conradhoffman.com/mini_metro_lab.html
www.ko4bb.com Mini Metrology Lab.pdf
« Last Edit: January 29, 2013, 07:40:38 pm by JBeale »
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #111 on: January 29, 2013, 10:59:16 pm »
I'm also curious how this project came out, and also the quality of the 3W 3.3k 1% 50ppm/C resistors, which are being sold today at $15/100.


I think I did post about them previously.

The ones I got had a temp co of -45ppm/degC. The match from memory was probably less then 5ppm/C. The thing is this is still enough to cause a significant change (if you have better then a 4 digit meter) in resistance just by applying a significant voltage, such as a few volts.

In a kelvin varley divider, the two of the resistors in a divider chain connected to the switch for the next lower range dissipate less power then the rest so the temperature differential will cause a significant error. There is also the temp co of the resistance of the next lower range to add to this.

I had a plan to correct for the negative temp co with some copper wire would around the resistor.  What I found is that it is not just matching the temp co that is important, but if the copper wire is too thick a gauge, it heats at a much slower rate then the resistor. I would probably need to find something like 0.2mm diameter wire that is not that easy to get, and it would be an absolute pain to work with.

I think the best bet would be to put all the resistors of the upper ranges (except for the trimpots) in mineral oil (easily available from a chemist) and select for matching temp coefficient. This hopefully would help to keep the resistors at a uniform temperature, and also minimize local temp differentials on the resistor body when power is applied.

I would need to do some tests, but I haven't had the time lately. I do have a nice diecast box with a seriously good seal that I could use to mount the resistors in an oil bath.

Richard.

 

Offline PA4TIM

  • Super Contributor
  • ***
  • Posts: 1161
  • Country: nl
  • instruments are like rabbits, they multiply fast
    • PA4TIMs shelter for orphan measurement stuff
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #112 on: January 30, 2013, 11:58:18 am »

The inside of my ESI 722 decavider ( 200 euro)

http://www.pa4tim.nl/wp-content/uploads/2011/05/720openkapot.jpg. Fluke 720A before restoration
http://www.pa4tim.nl/?p=2332. More pictures from my 720A ( got it for free together with a 750,760, 731, 332, 845 , guildline cabinet and some more goodies)

I have restored a GR1608, bought an expensive 1M 0.01% vishay for that. The original WW was dead. That is really thin wire, like a hair. The 1M turned out to be 100 Ohm low so I added a 100 Ohm that nicely zeroed the TC. But after installing and adjusting the rest the range was off. I removed the 100 Ohm and all was perfect. So maybe the old one was also 100 Ohm off .
« Last Edit: January 30, 2013, 12:05:47 pm by PA4TIM »
www.pa4tim.nl my collection measurement gear and experiments Also lots of info about network analyse
www.schneiderelectronicsrepair.nl  repair of test and calibration equipment
https://www.youtube.com/user/pa4tim my youtube channel
 

Offline quarks

  • Frequent Contributor
  • **
  • Posts: 874
  • Country: de
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #113 on: January 30, 2013, 08:06:29 pm »
...ESI 722 decavider ( 200 euro)

...my 720A ( got it for free together with a 750,760, 731, 332, 845 , guildline cabinet and some more goodies)

Hello PA4TIM,
you are indeed a really lucky man :-+
I hope you get even more of these special goodies you can show us!

bye
quarks
 

Offline EyeDontKnow

  • Contributor
  • Posts: 12
  • Country: us
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #114 on: December 09, 2013, 03:12:52 pm »
What would be nice, is if someone made a board for a KVD.
Given the advancements in surface-mount resistors (and lowering cost)....possibly a multi-deck switch that had solderable boards for each deck.....resistors soldered on each deck.

Maybe I'm being too optimistic or hopeful that there might be a demand for us retro old-school metrology geeks.  O0

I have several (new-old-stock) Shalco silver-plated 6 deck open frame rotary switches..........but I would love to see a more miniaturized opportunity.

The resistor accuracy could be up to the builder.
The ultimate being Vishay foil,  VSMP series, .01%, 0.2ppm types (or better). If bought in quantity.....the cost comes down.

What is really cool, is their "bulk metal foil" resistors can be specifically made at any value, to-order.....(I have not tried it yet)


 

Offline PA4TIM

  • Super Contributor
  • ***
  • Posts: 1161
  • Country: nl
  • instruments are like rabbits, they multiply fast
    • PA4TIMs shelter for orphan measurement stuff
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #115 on: December 09, 2013, 03:43:32 pm »
The Fluke and ESI can do 1000V. I needed such a resistor for a GR bridge and asked around. A company in the stated still makes them but 1 resistor of 1M would cost over 100 dollar. I then used a precision vishay through hole resistor with a very low tempco. That costed me 35 euro. You need a lot those for a KV.

SMD would be a lot cheaper but that limmits the use to 50V or 100V or so.
The KV's from Fluke and ESI have low EMF contacts and switches and everything is isolated very well to make sure leakage current stays low (10xE14 Ohm if I recall it right, and it was a hell of a job to get it within that specs again)
www.pa4tim.nl my collection measurement gear and experiments Also lots of info about network analyse
www.schneiderelectronicsrepair.nl  repair of test and calibration equipment
https://www.youtube.com/user/pa4tim my youtube channel
 

Offline EyeDontKnow

  • Contributor
  • Posts: 12
  • Country: us
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #116 on: December 09, 2013, 05:46:44 pm »
Yes, my ESI kelvin dividers still do a great job.....over 20 years old ?
Clean and re-lube the contacts.

Often, old boat-anchor test stuff suddenly comes close to original spec just by cleaning the contacts (and replacing power supply caps).

I repaired a JRL voltage divider that had an open 15k resistor inside the oil-bath housing......no way to get to it. I used a Vishay RNC90y resistor as a replacement across the dead area...300v/0.6watt....that's the limit now.
 

Offline babysitter

  • Frequent Contributor
  • **
  • Posts: 893
  • Country: de
  • pushing silicon at work
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #117 on: June 22, 2014, 08:04:24 pm »
Some Kelvin Varley pron showed up during repairing the cheapest KVD available on eBay when i felt I need a little brother of my RV722 O0. There was a device on ebay USA employing a Electro-Measurement Inc. (later ESI) DV411, together with a power supply. Everything besides the Divider was removed, the remaining problem was the third decade switch was not latching to the digit positions :scared:. So i removed the three screws and opened the second can (my coworker from Taiwan was reminded of landmines  >:D)

Overview picture

Above you see the switch without the latching mechanism on top,there is a inner and a outer ceramic wafer, the inner one is turning with the outer drive shaft and has the moving contacts. The outer part is fixed.

Under it the resistors that make up the rest of the KVD network. The second decade here is realized by a potentiometer with two fixed R in parallel, in the first can with 2 decade switches i would simply expect that it repeats.


The Potentiometer without its cap. Although sporting 270° range, it is really turning 360°. Asked the gurus at volt-nuts, they confirmed this strange behaviour is okay.  :-//


The broken latching mechanism short after repair-the cogweel on the drive shaft is intended to be locked in positions by a ball-on spring. The cogwheel was loose, but to my surprise only soldered to the drive shaft.  :wtf: Employing a agressive flux, that is what I reworked after repositioning the wheel on the shaft.

After putting everything back together in its housing it has shown to be a sucessful repair. :-+
It divides on demand, Ratio is good for 4 decades and the 5th depending on operators skills :)
Cons: US screws! The handwheels had to kiss TORX 7 to align to the shafts, not the real US imperators imbus.


« Last Edit: June 22, 2014, 08:18:45 pm by babysitter »
I'm not a feature, I'm a bug! ARC DG3HDA
 

Offline Flump

  • Frequent Contributor
  • **
  • Posts: 520
  • Country: gb
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #118 on: June 25, 2014, 07:40:36 pm »
nice welding job there babysitter  :-+

is it fully operational now and what do you plan do with it ?
I had some old decade boxes that had similar resistors inside.

can we get a pic of it all together ?
 

Offline babysitter

  • Frequent Contributor
  • **
  • Posts: 893
  • Country: de
  • pushing silicon at work
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #119 on: June 25, 2014, 07:50:38 pm »
lets call it soft welding, maybe ? :)

I am not going to open the can with the first two decades out of respect, and the wiring between the cans is not interesting.

It is working on first sight, 4 decades fine on 34401A, 5th user skill dependent.

As explained I removed the integrated power supply and the second potentiometer (Helipot 10 turn) from the big box. As I have 2 and a half labs (way too small home lab, my hobby basement about a km away and my workplace) this is my mobile ratio box. (The 7 decade part is a pita to move without a car and you just dont haul such things around.)

Teaching our trainee, simulating a NTC-resistor series circuit as temperature simulator are the things it is scheduled for in the next weeks.

I'm not a feature, I'm a bug! ARC DG3HDA
 

Offline cellularmitosis

  • Supporter
  • ****
  • Posts: 1111
  • Country: us
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #120 on: April 20, 2018, 03:20:31 pm »
Time to bring this thread back from the dead!   :horse:

Here's a question: do all of the switches need to be of equal quality?  I haven't found a definitive answer on this, but my intuition says that yes, they do.  Variation in contact resistance, or leakage should be just as detrimental at any decade, since they are all in series.

What sayest the forum?
LTZs: KX FX MX CX PX Frank A9 QX
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 14072
  • Country: de
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #121 on: April 20, 2018, 04:50:18 pm »
The less significant digit switches can get away with lower quality switches. There are parallel resistors and thus less current flows through the lower switches and less influence of the lower decades contacts.

This is even more so if extra shunts are used to keep the resistors in a smaller range: So not strictly the KV circuit, but from a certain point on the same value resistors can be used in the lower decades with an extra resistor in parallel to each decade. The first 2 decades are usually still higher impedance (classical KV) to keep the load reasonable.
 
The following users thanked this post: cellularmitosis

Offline cellularmitosis

  • Supporter
  • ****
  • Posts: 1111
  • Country: us
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #122 on: April 20, 2018, 06:00:32 pm »
Thanks Kleinstein, that makes sense.  Some current is diverted by the parallel resistance (blue), and some by the shunt in each decade (green), so not all of the current makes it to the final decade switch (red).
LTZs: KX FX MX CX PX Frank A9 QX
 

Offline Vgkid

  • Super Contributor
  • ***
  • Posts: 2710
  • Country: us
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #123 on: April 20, 2018, 06:50:23 pm »
From the teardowns I have seen , a few milliohms are acceptable. 2-5 is the average for a good high quality switch.  As long as they are consistent , you can even parallel them up , though that price it may be cheaper to buy a used unit. Though I do want to test those Chinese ceramic rotary switches some day...
If you own any North Hills Electronics gear, message me. L&N Fan
 

Offline lars

  • Regular Contributor
  • *
  • Posts: 132
  • Country: se
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #124 on: April 20, 2018, 07:02:37 pm »
What about thermal EMFs, could that be something to consider even for the less significant digit switches?
Lars
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 14072
  • Country: de
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #125 on: April 20, 2018, 07:10:58 pm »
Thermal EMF for the switches would add up for all stages, at least most of it (the average of the 2 switches of 1 decade). So low thermal EMF is still important for the lower stages.
 
The following users thanked this post: cellularmitosis

Offline klimm

  • Contributor
  • Posts: 34
  • Country: ro
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #126 on: January 03, 2021, 05:31:51 pm »
amspire has a concept diagram in post#20 on the first page and I understand it is just a concept.
Specifying the LTC1050 was meant as a generic chopper/zero drift Op amp  in that schematic,  but  not to be used  with a 10V reference at input, right?
It was inviting at first to use the LTC1050 as I have them, but I read in the DS that  the max ratings are±9V  and surely I would not use the chopper at max ratings. Characteristics are given only for ±5V in the DS and I suppose this should be the nominal ±Vs to be used at.
 
What would you recommend to use as followers and summing amp  of the above mentioned concept for the following; 
 AD588 Vref (±3mV(or even 0.002% if I get it calibrated))   outputs max  10mA) followed by a decade of 10x10k or 10x100k (hesitate which to use)  mechanical switched 3 decades  to get a simple DC check unit for 3 1/2 to 4 1/2 DMMs.
I already built the reference and baked it for about 260 hours but  now of course I need  more voltages.
« Last Edit: January 04, 2021, 09:01:13 am by klimm »
 

Offline Conrad Hoffman

  • Super Contributor
  • ***
  • Posts: 1928
  • Country: us
    • The Messy Basement
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #127 on: January 03, 2021, 08:06:25 pm »
You might take a look at how Analogic/Data Precision did their voltage reference- http://w.ko4bb.com/getsimple/index.php?id=manuals&dir=06_Misc_Test_Equipment/Data_Precision

IMO, once get into summing buffered dividers, their method makes a lot of sense. I think they used an LM399, but the parts are not very exotic to get excellent performance.
 
The following users thanked this post: klimm

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 14072
  • Country: de
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #128 on: January 03, 2021, 08:18:25 pm »
For the voltage followers there are several options, depending on the impedance (resistors values) and precision needed. The version with separate buffer for the decades may get away with lower grade buffers for the lower digits. 
The LTC1050 should be OK with a 10 V range (e.g. +12 V and maybe -0.6 V supply)

A few reasonable options are:
OPA202  (non AZ, relatively low bias and drift)
OPA140, OPA191  FET based OP with low offset (rel. expensive)

ICL7650,LTC1050  (AZ OP, some needs external capacitors, up to some 15 V)
LTC2057  (AZ OP up to some 30 V, but bias)
MCP6V51 (cheap AZ OP up to 45 V, but bias  and current noise higher than LTC2057  (useful up to some 30 K, but hardly with 100K)
            - the DS number on current noise is too good :bullshit:)
LTC2055, max4238, LMP2011, LTC2050,...   AZ ops with low bias, but only 5 V - these OPs would need some additional circuit for bootstrapping the supply. The classical way is driving the supply from the output, with some additional low pass filtering. The faster way uses a second amplifier from the input (can be faster and may be a little easier).

For high drive the BS supply version would need an additional in loop buffer for the current. For better precision the additional in loop buffer can also be good for a the other AZ OPs.

Usually a good concept avoids to many critical parts / buffers.
 
The following users thanked this post: klimm, SigurdR

Offline klimm

  • Contributor
  • Posts: 34
  • Country: ro
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #129 on: January 04, 2021, 02:33:25 pm »
Thank you Conrad,
before asking I already read the whole topic at least twice and  was looking at 8200 and took notes. I will use some info from there but I don't want to use analog switches or Muxes as I would dive in, for me, unknown territory.  I would use some  rotary switches I have,  as they are metrological grade and work in decades not in heptal or octal system.

Kleinstein,
trying to be short, I think I was not very clear in my description. When talking about 3 decades I meant actually that after the one and only  resistor decade of say10X10K,  follow 3 rotary switches and 3 buffers(1 following each switch)  that deliver their signals to a summing amp. Going down the ranges from volts to tenths and hundredths of volt,  I would need at least same low offset and drift op amps  as for volts range, am I  right?



 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 14072
  • Country: de
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #130 on: January 04, 2021, 02:45:03 pm »
In the circuit as in post 20 (summing amplifier with weights 1:0.1:0.01:..., the lower decades have less stringent requirement on the buffers, as there weight is lower. They still need very low input bias, if a common 1..10 chain is used.  So one may be able to use some good FET types (e.g. OPA145, OPA191) at least for the 3rd and lower decade. The 3rd decade has 1/100 the weight and thus the offset effect 100 times smaller. It would also still be a constant added to the output.

The common decade divider should be low impedance, so 10 x 1 K. With 10x10 K the demands on buffer input bias gets quite high !
 
The following users thanked this post: klimm

Offline klimm

  • Contributor
  • Posts: 34
  • Country: ro
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #131 on: January 04, 2021, 03:06:27 pm »
Thank you for clarifying this.  I was only thinking at the lower voltages not taking weighting into account although I saw the  TL062s in the 8200.
In the D.P. 8200 they use 500 Ohm resistors  resulting a current demand from the Vref of 2mA. As my reference delivers max 10mA I think I am on the safe side with 10x1K and 1mA through the divider  but I can't  analyze  how this would affect if wanting  to hang more buffers to get  more resolution; there would be more current necessary to flow through the divider I suppose as energy won't come from nowhere even if Ib is really low, if using  say 5 Buffers? Sorry for bothering but I am by no means comfortable with  Op Amps

« Last Edit: January 07, 2021, 09:08:48 am by klimm »
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 14072
  • Country: de
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #132 on: January 04, 2021, 05:18:34 pm »
The bias can be quite different between OPs:
The TL062 or similar are likely in the 10 pA range - so really low bias, but also poor DC specs.
With AZ OPs like LTC2057 / MCP6V51 expect a few 100 pA. So with 5 K impedance in the middle this could change things by a  may be 1 µV. Probably OK if not in the absolute volt nut range.
BJT base OPs may be a few nA  - and thus in the 10s of µVs.

One would probably not use the same type of buffer for all cases.
A possible combination may be ICL7650, LTC2057, MCP6V51 or maybe OPA191  for the 1st stage and output.
e.g. OPA191, OPA145 or maybe OPA171 for the 2nd digit.
e.g. OPA171 or TLC277 for the 3rd
e.g. TLC277, TLC272 or even lower grade for the 4th / 5 th
Especially for the lower grades there are many options.
For the later digits low bias is as important as for the first, but offset, noise and drift get less important.
It may help to have single supply capable OPs, so that one would only need a small negative supply, like -0.6 V.
 
The following users thanked this post: MegaVolt, klimm

Offline klimm

  • Contributor
  • Posts: 34
  • Country: ro
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #133 on: January 09, 2021, 12:14:46 pm »
I need to compare some 1K  wirewound resistor (flat wounded on 2mm epoxi-fiberglass PCB in the picture) in order to select them for the divider of 10X1k  .  They are about max 0.1% to each other according to my 4 1/2 DMM but resolution  is not enough. I think of using a Wheatstone bridge.
Now comes my problem;
The resistors are made of really thin wire,  0.1 to 0.2mm only, so very low power.  They were used into a Feussner Kompensator that was working at a current of 100uA. So I suppose that when measuring with the bridge I need to limit the current to  a maximum  of  200uA, 100uA for each branch, in order not to stress the resistors. Yes I would use a morse key to power the bridge for the time of the measurement  but still can't use a voltage like 9V
Any sugestions?
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 14072
  • Country: de
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #134 on: January 09, 2021, 12:37:53 pm »
One can always reduce the current to the bridge with an extra resistor in series to the whole bridge. So maybe start with some 1.5 V and than have some 5-10 K in series to limit the current. Another option would be to use a transformer with a low frequency AC and measure the bridge with AC. AC amplifiers are usually easier than DC and 1 K is low enough to work well with some 100 Hz. The transformer could also provide 1 side of the bridge.

Just looking for equal resistors one can have 3 fixed resistors and just compare the bridge reading for each each resistor.

The picture already looks like a good resistor decade or KV divider. So maybe do a little reverse engineering to see if it could be used largely as is.
 

Offline klimm

  • Contributor
  • Posts: 34
  • Country: ro
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #135 on: January 09, 2021, 12:49:24 pm »
Kleinstein,
in the picture you see a Feussner Kompensator.

I was thinking of a series resistor but I thought that I would see a lower voltage at the bridge input as the supply sees the resistor and the bridge as a divider. Is this wrong?
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 14072
  • Country: de
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #136 on: January 09, 2021, 01:56:59 pm »
The extra resistor and the bridge would form a divider - that is right. For the bridge the driving signal does not have to be very stable of exactly known, so the extra divider is not a problem. Of cause one would have a lower voltage at the bridge - can't change that is the current needs to be small.

The resistors don't look that small, so something like 1 or 1. V and 1 to 1.5 mA should be OK for the test. This is still only 1 - 2.3 mW, so not yet a real problem with self heating.

That original apparatus looks really nice and could also be used to set fine voltage steps - likely better than the circuit from post #20. The main part is a precision divider, just using more resistors than the KV circuit, but just as good, even with some slight advantages.
 

Offline klimm

  • Contributor
  • Posts: 34
  • Country: ro
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #137 on: January 09, 2021, 02:42:57 pm »
I will try to feed the decade with 10V and see if it heats at 1mA. Supposedly it would hold max 16mA as I can read on a decade box, I just opened and,  made with the same resistors.
this part added:==================
A short computation gives a max power P=0.016A*0.016A*1000R=0.256w. So, you should be right :)

I let here a link about bridge sensitivity that I found really helpful  https://electronics.stackexchange.com/questions/110588/sensitivity-of-wheatstone-bridge
End of added part=================
Yes, the Feussner kompensator is a neat thing. I opened it  to try getting the schematic as I don't have much info.  The lower range of switches  is an auxiliary decade to set the current in the auxiliary circuit.  One could calibrate 200mv and 2V ranges of 0.1 accuracy DMMs  in small steps, provided a Weston cell is present for calibration but  I don't have one. I was thinking of making a divider buffered by an OP07 on my Vref10V as the accuracy of the device is about 0.01 so far I know,   This one has some defective resistors I was said. Measuring the 1k decade I found one of them interrupted.  I am not determined to break it apart but can do some measurements to see how close resistors  are to one another and asses if I could use them.
Of course, best would be to be able to repair and use the thing as is.
« Last Edit: January 09, 2021, 06:28:30 pm by klimm »
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 14072
  • Country: de
Re: Kelvin Varley Divider [and Precision Voltage Source]
« Reply #138 on: January 09, 2021, 05:12:53 pm »
A broken resistor is a very real possibility.  On could very well be lucky and have only one broken resistor. Once broken the instrument my no be used very much.
A repair would be very well possible, if only a few bad resistors.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf