General Question

[tomato]

Threads about voltage references seem to dominate this forum.  As a general question, why is everyone interested in this?

[Terry01]

Maybe because it's in the metrology part of the forum?

[tomato]

Sorry, I'm not asking why the posts are in this sub-forum.  I'm asking why so many people are interested in voltage references.

[TiN]

It is a good engineering exercise and involves topic from basics like circuit design, PCB layout, power design and component selection to deep rabbit holes in statistics, physics and analysis methods.
And above all, our world is analog by nature, so to convert real world properties into digital information A/D and D/A converters are using references, and all that are surround us everywhere today.

Also reference modules design is deceiving simple looking, so some of people here got into it, with just simple 10 component designs and before they realized, they ended up with five 8.5digit meters and house-worth of old equipment at home. 

[David Hess]

Voltage references can be a difficult and underappreciated problem.  This is especially the case with high resolution sampling converters and precision applications.

[GEOelectronics]

Very respectfully, is the a forum somewhere for slightly advanced but definitely amateur electronics and metrology enthusiasts interested in improving and seeing other's techniques but haven't/can't or don't need quite Volt-Nut standards?

George Dowell

[classicTEK]

You are in it!!!

It is my experience that we have some VERY talented engineers from the likes of FLUKE, TEK, HP and many others that hang out right here and provide unparalleled support in all forms.  Not for current products you understand, but to share the knowledge with those that care to learn from their decades of experience in real world metrology applications.

All are welcome here, otherwise gee.........who would let me in!!!

:  )

Cheers,

David

[GEOelectronics]

Maybe because it's in the metrology part of the forum?

[zhtoor]

But frankly I would be embarrassed to post my projects here as being too "not 3458A" worthy".

you should'nt be!

according to Dmitri Mendeleev: Science begins with measurement.
and HP 3458A is not the only way to measure things.

best regards.

-zia

[tggzzz]

... and before they realized, they ended up with five 8.5digit meters and house-worth of old equipment at home. 

Please remove that smiley; it really isn't remotely amusing. Pertinent perhaps, but not funny.

[tggzzz]

Maybe because it's in the metrology part of the forum?

Respectfully, metrololy is the science of measurement. My hobby is metrology and I don't do A to D work at all. This sub forum seems to be focused on one very small aspect of the science of measurement.
Perhaps I too was lured ere by the title and not the content necessarily, although it does hold some interest to me. But frankly I would be embarrassed to post my projects here as being too "not 3458A" worthy".

It is focussed on that solely because people post about it. It is in your power to change that

[rhb]

It's volt oriented because there are several people who are very prolific working on that and the associated precision resistors.

It's actually a rather small group of people.  I've not seen any significant activity related to other electrical or physical measurements other than temperature.

Mostly I think the availability of voltage reference devices is the cause.

I'm hoping to get my shop sufficiently functional that I can restart work on a gauge block interferometer and some other physical metrology projects.  I have bought every book on metrology I have come across, so it is now approaching 6 ft of books.  Lots of redundancy in the electrical measurement field, but also lots of stuff on optics, pressure, temperature strain, etc.  I've got the pieces for a monochromatic light box and hope to make some optical flats.  My biggest obstacle is I have too much stuff both that I have collected and what my Dad collected.

I've spent far more time on physical metrology than I have or intend to ever spend on voltage references.  My sole voltage reference effort is seeing if I can reduce the error due to aging of a reference. I'm far more interested in measurement of arbitrary angles  of rotation to fractional arc seconds and sub-micron distances.  Powerful MCUs provide a lot of opportunities for calibrating out errors in physical construction.

As an example, consider the problem of measuring deviation from a straight line or a plane surface to 0.0001" or better.  The tenth is difficult, but less than that is fiendishly difficult.  You can buy low temperature coefficient resistors, but low temperature coefficient materials are expensive and fragile. And such things can't be used in a small temperature controlled box.

In general I'm far more interested in metrology projects that are not in 3458A territory.  My primary concern is self calibration of my equipment with a minimum number of external calibrations.  That equipment is not limited to electronics.

[zhtoor]

As an example, consider the problem of measuring deviation from a straight line or a plane surface to 0.0001" or better.

hello Reg,

imagine people making glass vials with a "slight" curvature to put in a machinist's level to be able to measure deviation in microns over feet of lengths!
and the best use we pakistani's have for abundantly available (and cheap ca. $5 / sq foot) granite is "low tempco" kitchen tops 

best regards.

-zia

[blackdog]

Hi tomato,

Welcome! You've asked the right question... 

This is your first step towards acquiring a great deal of general and complex knowledge, as TiN has already explained.

You can learn here that 1mm copper wire also has a resistance, it is not "0" Ohm, and it also has a thermal mass and al kind of other characteristics that can be important...

And if you had an 8.5 digid meter like a 3458A and you gave it a "Normal Calibration" at KeySight,
after more than 1 day you are no longer sure that your 10V measurement is within 10PPM and that for about 800$
Fortunately, these measuring instruments are much better, when they are in good condition.

Furthermore, it is 'normal' human behaviour to seek for certainty.

So, my opinion is that if you want to have a lot of and deeper knowledge of electronics and materials, this metrology forum is a nice place to acquire this knowledge.

Happy learning en kind regards,
Bram

[GEOelectronics]

"According to Dmitri Mendeleev: Science begins with measurement."

My thanks to Dmitri, he laid out the Periodic Table of the Elements

[RandallMcRee]

So, the question is--why voltage references?

I've asked myself that same question.  I think there are a few reasons...

a) a voltage reference is the granite surface plate for electrical measurements. It's basic, you need one (or three).
b) there are a *lot* of options each manufacturer has ten or so, and then there are discrete versions
c) there are endless variations on how to handle temperature, noise, and noise is multi-dimensional--pink noise, white noise, emi...

I'm sure there are more.

Randall

[Andreas]

A voltage reference is the "hello world" of metrology

[GEOelectronics]

"A voltage reference is the "hello world" of metrology "

A voltage reference is the "hello world" of ELECTRICAL metrology

Perhaps this is understood, since this is EEV blog. What does EEV stand for?
Thanks
George Dowell

[rhb]

A straight line and a vertical reference are the "hello, world" of metrology.  Proper flat surfaces didn't exist until Joseph Whitworth showed everyone how to do it in the 1830's. He also constructed the first instrument capable of measuring to the millionth of an inch.  The Egyptians had basic metrology, but it took thousands of years to refine it to what we have now.  Almost all of that was done in the last 200 years.

Metrology is not my hobby.  It's a hill I have to climb to get to where I want to go.  I do find it interesting.  In large part because of how interconnected everything is when you get involved in precision measurements.

[Tony_G]

I find it fantastic to read the posts from the 'volt nuts' (and I use that term with the greatest respect) - I think you can learn a lot just listening in on the discussions.

That doesn't answer the why it's so popular, it's just my observation that these posts are valuable to everyone.

TonyG

[GEOelectronics]

Quote from: GEOelectronics on Today at 03:03:28 am>
"A voltage reference is the "hello world" of metrology "

A voltage reference is the "hello world" of ELECTRICAL metrology

Perhaps this is understood, since this is EEV blog. What does EEV stand for?
Thanks


They say the 34470A and the 34458 are the only HP/Keysight products to use the LTZ1000 and have ACAL feature.
 Now I'm happy and will go on learning about this amazing DMM and the new world it has opened up to me. It was purchased because of a need for something that can read 1 G Ohm directly.

George Dowell

[Conrad Hoffman]

The forum is a sub-forum of Electronics, so that's what you tend to see here. I spent much of my previous job building interferometers of various types and doing "machining" that many would say is impossible. IMO, it ties back into electronics if you want to build things like calculable capacitors! Alas, my home shop isn't up to those standards. The Richardson Grating Lab is right here in Rochester- http://www.gratinglab.com/Home.aspx

[GEOelectronics]

Cool Conrad I understand now. No problem.

[zhtoor]

I spent much of my previous job building interferometers of various types and doing "machining" that many would say is impossible.

hello Sir,

would that be precision-lapping by any chance?

forever grateful to you.

-zia

[tomato]

Thanks for all the responses.  It's interesting to hear what makes everyone "tick."

[GEOelectronics]

Thanks for starting the thread Tomato.

(joke removed by author)

George Dowell

[Conrad Hoffman]

I spent much of my previous job building interferometers of various types and doing "machining" that many would say is impossible.

hello Sir,

would that be precision-lapping by any chance?

forever grateful to you.

-zia

Honing of round parts actually. We built piezoelectric motors that required a fit within 10-15 millionths of an inch of a certain target. With a well tuned Sunnen machine and the right air gauging, it can be done quickly and in reasonable quantities. The optics for the big Fabry-Perot interferometers were purchased from the UK and would pair up for lambda/200. I think telescope people get all excited if their secondaries get to lambda/20.

[GEOelectronics]

Tomato, Frank's site explains one of the main reasons to have a Voltage reference in an electronics shop very well:

http://www.frankshospitalworkshop.com/electronics/diy-voltage_reference.html


George Dowell

[kj7e]

Threads about voltage references seem to dominate this forum.  As a general question, why is everyone interested in this?

In order to measure anything, you have to start with a known reference.  I have always found "the volt" very interesting, the history and why 1 volt is... 1 volt.  Who and why did they decide this specific amount of EMF will forever forth be known as 1 volt.  I asked this question to a Ham Radio friend of mine with I was just getting in to radios and electronics as a kid.  A few dozen questions later he told me I needed a Standard Cell (which lead to a few dozen more questions).  Dr. Frank had a nice post here regarding the history of the Volt;
https://www.eevblog.com/forum/metrology/1000-my-(hi)story-of-the-weston-cell-of-the-volt-and-of-being-a-volt-nuts/

Its amazing to me, how precise we can get, but at the same time it still seems there is so much uncertainty and how difficult it is to achieve any greater precision and or accuracy.  The best voltage reference (not counting a JJA as that cant be practicality owned)  to date was developed 33 years ago and its not been improved upon since.

[hwj-d]

Threads about voltage references seem to dominate this forum.  As a general question, why is everyone interested in this?

Take one or more of those answers:
First: yes
Second: don't know, what others ("everyone") are thinking
Third: ask a locksmith, why he is interested in his vernier calipers
Fourth: we are all going to need "splitting ball hairs"[sic! mjlorton(yt)]
Fifth: build one, to find correct answers

HTH 

[tomato]

I have bought every book on metrology I have come across, so it is now approaching 6 ft of books. 

Can we assume you meant to write 6 m of books?

[tomato]

A voltage reference is the "hello world" of metrology

Yes, that makes sense - it's cheap and easy to get your feet wet.

[JS]

Because you need to know the 8th digit,
to calibrate your 7 digits instruments,
to calibrate your 6 digits instruments,
to calibrate your 5 digits instruments, 
to calibrate your 4 digits instruments,
to calibrate your 3 digits instruments, 
to meassure a voltage at ±10% of the PS of a TV and say the it should be working but still doesen't   

JS

[GEOelectronics]

Because you need to know the 8th digit,
to calibrate your 7 digits instruments,
to calibrate your 6 digits instruments,
to calibrate your 5 digits instruments, 
to calibrate your 4 digits instruments,
to calibrate your 3 digits instruments, 
to meassure a voltage at ±10% of the PS of a TV and say the it should be working but still doesen't   

JS

I came to this sub group of metrology to learn new (to me) and better (than I have been using for 50 years) techniques of electronics measurements, expecting discussions about bridges, instrument amplifiers, filters etc. etc. etc.
Your explanation sums up the majority of the preferred subjects nicely.
I go away with the feeling that there are three things in Metrology, HP3458 and LT1001 and $75 resistors.
Nothing else matters or is allowed.

My article about a 0.03% resistor box was met with "Put some better resistors in there".
My hack to a China LM399 board's response was "why bother".
My question asking was there a mid level metrology board somewhere got no response at all, however there is one somewhere, or will be soon
Respectfully, but not impressed,
George Dowell

[kj7e]

George, I read your posts you reference.  I did not have the same takeaway you mention.

Even though most all posts here are in English, this is a very international forum and context is not always translated well.  I have seen some responses that come off harsh but really are not meant to be (lost in translation).  Also, particularly with the metrology group, its always about how can something be done better.  So yes, anything posted here will be critiqued and criticized in the effort to improve whatever the desired outcome is. 

[Vgkid]

@ George Dowell , an I have a link to your decade box thread.

[GEOelectronics]

Thanks L+N, there are several other one page articles in the works concerning the L&N "Boxes" of various model #s. Some have milliohm resistors that can be used as references and others have quite a few wire-wound resistors that look particularly hack-able.

Now that I can measure milliOhms for the first time, with 2 instruments, one a vintage HP-analog and the other a modern digital, I've been having a ball measuring things  that were unimportant in my business and hobby before Not that I get paid to or need to, it's just a hobby now.

Those will be published on eHam along with a homage to William Edward Ayrton


Geo

[TiN]

I go away with the feeling that there are three things in Metrology, HP3458 and LT1001 and $75 resistors.
Nothing else matters or is allowed.

I can excuse your statement, due to fresh member effect  . Proper phrase should be "three things in metrology, Fluke 792A, Fluke 734A and ESI SR104". Truckload of rest M$ gear is expected as default. 

Jokes aside on serious note, I don't quite see that much members here to bloat on having 3458A(/8508A/5730A/XXXXA/whatnot else) , like you suggesting to grant the "must have 3458 or you are not worthy" attitude . It is probably just biased because people here (in metrology section of the EE forum) are biased and sucked into this hobby, so on average more of us have expensive meters, than usual electronics engineers, who don't need such meters for daily work.

If you looking for feedback on low-cost projects, that is fine too, if you don't take too everything that people post too personal. After all it's only this much once can do with low-cost part, that is designed and manufactured to meet it's spec (usually, take that VPG , but it's bit much to ask for much better. Some go to great length doing sorting/matching/baking/cooking/soaking low cost part, be it resistor or voltage reference chip, but in the end it would be financially, time-wise and project wise CHEAPER just to spend those $5-75-250 on higher spec part that designed as such and not to worry. Now some of high-priced parts have issues and limitations too, so entering into realm of testing those limits and how to live with them is bit another matter, as it may help us to understand how to make things better. It is what some of nuts here about, taking something best commercial market can offer, and attempting futile attempts trying to go step above. 

Remember, lot of us here doing metrology toys and builds at home as a hobby, just for own amusement in free time, and often that extra hour of free time spent on soldering, measuring and documenting (taking photos and making post like this take time too!) is more expensive than even 75$ resistor. So think about return factor, what you are after, it is always a choice somewhere.

So yes, anything posted here will be critiqued and criticized in the effort to improve whatever the desired outcome is.

This! Also we already have General electronics talk, Test equipment, Repair and Projects sections. I see no reason why need to turn Metrology section into mix of these, posting general questions.
Back when I joined EEVBlog forums, all volt-nuts/ohm-nuts/time-nuts just hang out in T&M section and repair section, making all meaty threads that might be of interest for *nuts, but not so for general EE folk sink into abyss quick.
As refining Dave created dedicated subsection so bees are happy and separate from wasps and hornets now. So this supposed to be high-SNR area from the start, and members expected to have some level of understanding of things discussed.

[GEOelectronics]

Back when I joined EEVBlog forums, all volt-nuts/ohm-nuts/time-nuts just hang out in T&M section and repair section,

Finally the answer to my very first questions, thank you.
By the way tin, are you an administrator/moderator/owner here?

Geo

[TiN]

No, I'm mere humble contributor 

[Vgkid]

Thanks L+N, there are several other one page articles in the works concerning the L&N "Boxes" of various model #s. Some have milliohm resistors that can be used as references and others have quite a few wire-wound resistors that look particularly hack-able.

Now that I can measure milliOhms for the first time, with 2 instruments, one a vintage HP-analog and the other a modern digital, I've been having a ball measuring things  that were unimportant in my business and hobby before Not that I get paid to or need to, it's just a hobby now.

Those will be published on eHam along with a homage to William Edward Ayrton


Geo

That's right. I remember seeing your post on the "Ayrton Shunt does Double Duty" on the volt nuts mailing list. I have a relative of that box , the 4385
https://www.eevblog.com/forum/metrology/ln-4385-teardown/
along with a 4232 Wheatstone Bridge
https://www.eevblog.com/forum/metrology/ln-4232-wheatstone-bridge/
A 4393 URS, I plan on buying another one once I get a real job , and gutting both of them...
https://www.eevblog.com/forum/metrology/ln-4393-universal-ratio-set/
and a 4271 linear amplifier , that doesn't seem to work correctly. It will be stable for several days , then go drifty. Unfortunately it is unstable.
When it comes to milliohms , that is also one thing that greatly interested me as well.

[cellularmitosis]

I came to this sub group of metrology to learn new (to me) and better (than I have been using for 50 years) techniques of electronics measurements, expecting discussions about bridges, instrument amplifiers, filters etc. etc. etc.

I would love to see posts about this, or youtube videos!  This would be fantastic.

My article about a 0.03% resistor box was met with "Put some better resistors in there".
My hack to a China LM399 board's response was "why bother".
My question asking was there a mid level metrology board somewhere got no response at all, however there is one somewhere, or will be soon
Respectfully, but not impressed,
George Dowell

It can be easy to get discouraged on the internet, because the voices of detractors can be so loud, especially relative to the silent majority who are interested in such posts and eagerly awaiting more.

I was enjoying your posts so far, I hope you keep chugging along! 

[rhb]

I came to this sub group of metrology to learn new (to me) and better (than I have been using for 50 years) techniques of electronics measurements, expecting discussions about bridges, instrument amplifiers, filters etc. etc. etc.
Your explanation sums up the majority of the preferred subjects nicely.
I go away with the feeling that there are three things in Metrology, HP3458 and LT1001 and $75 resistors.
Nothing else matters or is allowed.

My article about a 0.03% resistor box was met with "Put some better resistors in there".
My hack to a China LM399 board's response was "why bother".
My question asking was there a mid level metrology board somewhere got no response at all, however there is one somewhere, or will be soon
Respectfully, but not impressed,
George Dowell

I made a partial list of my ongoing projects this morning:

DIY RF power sensor

OSHW diode noise source

Reference aging study using <$3 AD584s from eBay

VNWA 3E S parameter test fixture

Arc second electronic level

Multiplexed sampling system

TEC environmental test chamber

Gauge block interferometer

Monochromatic light box

All of these are fundamentally metrology projects, but except for the AD584 project, none of them are voltage reference related. And there are interdependencies among them.  I need the multiplexer to do the AD584 project.  But none of these involve a high end DMM , voltage reference or resistor. I have lots of equipment and measuring tools.  I'm trying to acquire the ability to maintain and calibrate them.

Some people in the forum are very narrowly focused.  Many others are much more broadly focused.

[kj7e]

I came to this sub group of metrology to learn new (to me) and better (than I have been using for 50 years) techniques of electronics measurements, expecting discussions about bridges, instrument amplifiers, filters etc. etc. etc.

I would love to see posts about this, or youtube videos!  This would be fantastic.

I was enjoying your posts so far, I hope you keep chugging along! 

Jason,
http://www.qsl.net/k0ff/

Lots of cool stuff there.

[GEOelectronics]

That's right. I remember seeing your post on the "Ayrton Shunt does Double Duty" on the volt nuts mailing list. I have a relative of that box , the 4385
https://www.eevblog.com/forum/metrology/ln-4385-teardown/
along with a 4232 Wheatstone Bridge

[/quote]

Thanks for those links. So far in the collection and have been torn down are  2 different  "Volt Boxes" "Shunt Boxes" and the "Ayrton Shunt Box".
Cat 7592 Volt Box
Cat 7593 Volt Box
Cat 4385  Shunt Box
Cat 4385-S Shunt Box
Cat 2664 Ayrton Shunt Box.

Were there others?
Also is there an archive holding L&N manuals/schematics/ads etc?
I saw your teardown of the LN ratio bridge, thanks for that it answered all my questions about that device.

Perhaps we should pick this up in PRODUCTS sub category now that I have found it (not what I thought was in there).

Geo

[GEOelectronics]

George, I read your posts you reference.  I did not have the same takeaway you mention.

KJ7E- I mean no criticism of your recent resistor standard project posts, I envy them.

I'm building one with a $5.00 eBay resistor. And another with a $10 USD eBay resistor.
Please comment

.

By the way can anyone identify the connectors that came on these? They are 0150" OD.

http://www.qsl.net/k/k0ff/VOLT_NUTS/DIY_0.01%25_and_0.005%25_Resistance_Standards/PicFile/

[kj7e]

No worries here George, I just meant I don't think people where being overly critical of your posts.

If you want to have some fun, test those parts for TC.  Its really eye opening.

[GEOelectronics]

Yes TC is on top of the bucket list.

Geo

[GEOelectronics]

The best link is :

http://www.qsl.net/k/k0ff//VOLT_NUTS/

Geo

[JohnnyMalaria]

Tomato, Frank's site explains one of the main reasons to have a Voltage reference in an electronics shop very well:

http://www.frankshospitalworkshop.com/electronics/diy-voltage_reference.html


George Dowell

That's interesting. I'll open the back of a couple of mine

But I have what I hope isn't a dumb question: if it's so straightforward to construct a reference from a few inexpensive components, why don't DMMs have internal references? (Maybe the top-notch ones do?)

[GEOelectronics]

1) Tomato, maybe we should move this to the PRODUCTS Forum. I was told that's where this sort of ting belongs.

2) I agree with your suggestion and yes top DMMs do, I looked long and hard for one that self-calibrated to a known good internal standard, and found the HP-34470A, it one-button autocals in 15 seconds. Manual calibration of old DMMs takes a long time, requires tons of calibration gear, charts etc.(like my HP-3478A- totally beyond my skill-set).
The Keysight 34470A was a good investment for me.

Geo

[JohnnyMalaria]

So how do you calibrate the constant current source in a DMM required for resistance measurement?

In my case, I just want to adjust two hand-held DMMs so that they read as close to each other as their precision allows. Right now there's a couple of percent between them for V, I and R.

(Sorry if this is getting off-topic - I can ask this in a new thread if necessary).

[GEOelectronics]

I don't have that knowledge but would suggest asking at PRODUCTS. It is really a test equipment and measurement sub category despite the name.

I'll be watching there too, as I am interested in general.

Geo

[rhb]

Most cheap DMMs have a single trim pot that sets the voltage and everything else is derived from that.  I have s slew of the cheap (now free with any purchase) Harbor Freight DMMs.  All but the more expensive model have a trim pot in the upper right.  There's a footprint to add one on the one that lacks it.   They are 3.5 digit DMMs good to ~0.1% if adjusted.

The best thing to do is find the service manual for the DMMs you have. But often they will mark on the PCB what the trim pots adjust if there is more than one.