Author Topic: T.C. measurements on precision resistors  (Read 396343 times)

ducus and 1 Guest are viewing this topic.

Offline janaf

  • Frequent Contributor
  • **
  • Posts: 339
  • Country: se
Re: T.C. measurements on precision resistors
« Reply #225 on: February 09, 2015, 11:44:54 pm »
Well most of us don't have a MS in the toolbox. I do have a helium detector of the thermal type but doubt it can detect the low levels required.

As He has 1/5 th the density of air, one gram is ballpark a gallon / 5 liters......
my2C
Jan
 

Offline ManateeMafia

  • Frequent Contributor
  • **
  • Posts: 730
  • Country: us
Re: T.C. measurements on precision resistors
« Reply #226 on: February 10, 2015, 12:13:21 am »
Is it possible to open and reuse the body of a hermetically sealed relay? I have seen pics of old bathtub style caps disassembled but I don't know if it is worth the effort.
 

Offline babysitter

  • Frequent Contributor
  • **
  • Posts: 893
  • Country: de
  • pushing silicon at work
Re: T.C. measurements on precision resistors
« Reply #227 on: February 10, 2015, 06:25:16 am »
I thougth more on a tin can case and using ceramic feed through cap's
which could be useable to get the signals in / out of the box.
These cap's are also available with copper / tin legs.

Otto Schubert boxes - thats where my homebrew voltage reference lives in. With a ugly wire hole, I must confess.
You can get feed-thru caps with actual holes in it - so you can pull your inner wiring to the outside. But this forum told me that those develop cracks and don't stay airtight [citation needed]

Quote
I thought of having a thin metal membrane showing inner pressure. As long as there is over-pressure the metal membrane would bulge out and you'd be safe from in-leakage and . A product someone?


http://www.servometer.com
I am aware of those, saw them on a fair once. Word is that you find a similar something in classic liquid-filled thermostats.
I'm not a feature, I'm a bug! ARC DG3HDA
 

Offline macfly

  • Contributor
  • Posts: 37
  • Country: de
Re: T.C. measurements on precision resistors
« Reply #228 on: February 10, 2015, 07:52:42 pm »
Hi volt-nuts,

thanks for your suggestions.

Hmmm, helium, argon .... please don't forget that I wanted to use medical grade silicon oil as main-filling, not gas. The dry-nitrogen (I mentioned before) is only to swamp out the remaining oxygen.

@ babysitter: why do this kind of compartments crack ?

Regards,

macfly
« Last Edit: February 10, 2015, 07:57:51 pm by macfly »
Genius is one percent inspiration, ninety-nine percent perspiration (Thomas Alva Edison 1903)
 

Offline AndreasTopic starter

  • Super Contributor
  • ***
  • Posts: 3221
  • Country: de
Re: T.C. measurements on precision resistors
« Reply #229 on: February 10, 2015, 08:15:03 pm »
Hello,

its time to go again "on topic"

The "warm" cycle with fixed setpoints
25 deg C
47 deg C
25 deg C
minimum 3-4 hours each.

shows around 0.8 ppm deviation.
But it seems also that the time did not be sufficient to reach final value for the first 25 deg set point.
There is 0.2 K difference to the final value.
This would give togeter with around 0.25ppm/K for the resistor already 0.05 ppm.

the cold cycle (with fan from cooler box active)
shows no hysteresis.

But all in all time constants are a horror.

With best regards

Andreas


 

Offline AndreasTopic starter

  • Super Contributor
  • ***
  • Posts: 3221
  • Country: de
Re: T.C. measurements on precision resistors
« Reply #230 on: February 12, 2015, 10:17:49 pm »
Hello,

since I will still need some days for another test fixture I thought it was a good idea to test the reverse polarity again in a different way.

Instead of reversing the VREF voltage I put the resistor from face up to face down (FD) between the measurement clips.

So against measurement of 05.02.2015 in the measurement of 10.02.2015 only the resistor is sitting face down.

Surprise!! virtually no hysteresis above the noise level.

coefficients of LMS curve 10.02.2015

A 0 =  2.27445998893372E-0002
A 1 = -2.42945611854776E-0001  -> -0.24 ppm/K slope @ 25 deg C
A 2 = -7.22887593273244E-0003
A 3 = -3.87699048469777E-0005

maximum deviation:  8.57934015336459E-0001

On 11.02.2015 I reversed the VREF without any mechanical change.
So it is the same as the measurement of 06.02.2015 only with resistor face down.

coefficients of LMS curve 11.02.2015

A 0 = -2.71122470347014E-0001
A 1 = -2.45528403800187E-0001 -> -0.25 ppm/K slope @ 25 deg C
A 2 = -7.28122515035776E-0003
A 3 =  2.14549190916762E-0005

maximum deviation: 2.09427235197069E+0000


So the direction of the current is changeing the hysteresis from small to large.
And also the mechanical orientation of the resistor (face up / face down) has a certain influence (perhaps thermally).

Ok the resistor is not symmetrically. The leads are on the back side of the ceramic substrate and welded to the front side with the resistive element at the top of the resistor.

With best regards

Andreas
 

Offline Dr. Frank

  • Super Contributor
  • ***
  • Posts: 2377
  • Country: de
Re: T.C. measurements on precision resistors
« Reply #231 on: February 13, 2015, 08:22:22 am »
Hi Andreas,
That's very strange..
Is this resistor  identical to the Z201, you've sent me, 1k also?
Then I'll do the reversal test on my 3458A also.

Although i can't understand that behaviour.
Frank
« Last Edit: February 13, 2015, 08:23:56 am by Dr. Frank »
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 37661
  • Country: au
    • EEVblog
Re: T.C. measurements on precision resistors
« Reply #232 on: February 13, 2015, 09:26:57 am »
FYI I'm getting a pretty awesome 10K metrology grade transfer standard soon from Wekomm
2ppm tolerance
Long term stability better then 1ppm / year
Temperature stability better than 0.3 ppm / °C
And they are conservative specs, and they are working on producing 0.001ppm tolerance parts.
 

Offline babysitter

  • Frequent Contributor
  • **
  • Posts: 893
  • Country: de
  • pushing silicon at work
Re: T.C. measurements on precision resistors
« Reply #233 on: February 13, 2015, 09:54:34 am »
Hi Andreas,
That's very strange..
Is this resistor  identical to the Z201, you've sent me, 1k also?
Then I'll do the reversal test on my 3458A also.

Although i can't understand that behaviour.
Frank

Do you have x-ray inspection at work or can you take it to a dentist to x-ray it? This might give some (pun intended) insight if it is not available in the datasheet.
I'm not a feature, I'm a bug! ARC DG3HDA
 

Offline Dr. Frank

  • Super Contributor
  • ***
  • Posts: 2377
  • Country: de
Re: T.C. measurements on precision resistors
« Reply #234 on: February 13, 2015, 09:59:17 am »
Hi Dave,
This one, obviously:
http://www.vishaypg.com/foil-resistors/case-studies/study/wekomm_1/

Well, that's quite feasible for a good reference resistor, at a limited / fixed temperature range.

But not good for the described big temperature range, as the Vishay foil resistors show hysteresis.
And 0.001ppm stability or uncertainty?
Nope, never... This German engineering GmbH would need a Klitzing Hall standard.
Even the best secondary standard, the ESI SR 104 (which  practically has no hysteresis) does not approach that level..
And third, the SI Ohm is not better 'mise en pratique', than 0.2ppm .
That will probably change in 2018..

But you need now additional metrology grade , pornographic ppm instruments for testing..maybe a nice 3458A, or so.
Then,welcome at the volt-nuts!

Frank
« Last Edit: February 13, 2015, 10:02:22 am by Dr. Frank »
 

Offline Dr. Frank

  • Super Contributor
  • ***
  • Posts: 2377
  • Country: de
Re: T.C. measurements on precision resistors
« Reply #235 on: February 13, 2015, 10:24:43 am »
Hi Andreas,
That's very strange..
Is this resistor  identical to the Z201, you've sent me, 1k also?
Then I'll do the reversal test on my 3458A also.

Although i can't understand that behaviour.
Frank

Do you have x-ray inspection at work or can you take it to a dentist to x-ray it? This might give some (pun intended) insight if it is not available in the datasheet.

Me?
Nope.

There exist many Pictures of the ceramic /metal foil chip, with attached leads.. That always looks symmetrical, so no reason for asymmetric thermal behavior.

Frank
 

Offline JohnnyBerg

  • Frequent Contributor
  • **
  • Posts: 474
  • Country: de
Re: T.C. measurements on precision resistors
« Reply #236 on: February 13, 2015, 10:42:56 am »
Long thread to read. Very nice though  :-+

Would it be possible to get all the results in some kind of sheet or selection guide?
It would make it easy to choose a right resistor for the right job :)

Or, make a comparison against a poor mans 0.1% resistor like the RT0805BRD071KL from Yageo?
What do you get extra when paying $10 for a resitor, instead of $0,10?
 

Offline Dr. Frank

  • Super Contributor
  • ***
  • Posts: 2377
  • Country: de
Re: T.C. measurements on precision resistors
« Reply #237 on: February 13, 2015, 11:31:26 am »
Long thread to read. Very nice though  :-+

Would it be possible to get all the results in some kind of sheet or selection guide?
It would make it easy to choose a right resistor for the right job :)

Or, make a comparison against a poor mans 0.1% resistor like the RT0805BRD071KL from Yageo?
What do you get extra when paying $10 for a resitor, instead of $0,10?

Are you kiddin'?
<sigh>

Well, you get:
T.C. as low as < 1ppm/K (thin film  25ppm/K)
Stability as low as 10ppm/yr. Down to 2ppm/6years (t.f. about 100...1000ppm/yr.)
Trimming (stable) down to 10ppm( T.F. 0.1% but not very stable due to technology. Heat and environmental influence easily or even deterioate initial trimming on the same order of magnitude)

Frank
« Last Edit: February 13, 2015, 04:15:55 pm by Dr. Frank »
 

Offline JohnnyBerg

  • Frequent Contributor
  • **
  • Posts: 474
  • Country: de
Re: T.C. measurements on precision resistors
« Reply #238 on: February 13, 2015, 11:56:44 am »
T.C. as low as < 1ppm/K (thin film  25ppm/K)

I think Yageo specifies 20ppm/K

Thanks for the comparison. Do you have any idea about the spread of spec. between DUT's?
(Specialy when they sit next to each other on the tape.)

When the resistors are thermo coupled, and used as resistors in a circuit, absolute drift and accuracy is not as important as the spread between devices.
 

Offline janaf

  • Frequent Contributor
  • **
  • Posts: 339
  • Country: se
Re: T.C. measurements on precision resistors
« Reply #239 on: February 13, 2015, 11:59:24 am »
Great work! Surely enough for a paper!?!

 
my2C
Jan
 

Offline janaf

  • Frequent Contributor
  • **
  • Posts: 339
  • Country: se
Re: T.C. measurements on precision resistors
« Reply #240 on: February 13, 2015, 12:07:04 pm »
Did you consider any SMTs?

Susumu RG and URG are avaliable to 2ppm/C, with URG series being supposedly a higher quality, better aging. Both have some kind of secret inorganic coating. Datasheet: "Unmatched Reliability and Excellent Stability at different environmental conditions".

my2C
Jan
 

Offline janaf

  • Frequent Contributor
  • **
  • Posts: 339
  • Country: se
Re: T.C. measurements on precision resistors
« Reply #241 on: February 13, 2015, 12:13:14 pm »
One thought;

Thermal cycling is sometimes recommended for removing mechanical stress on sensitive components. Do you have any thought on that based on the measurements?
my2C
Jan
 

Offline Dr. Frank

  • Super Contributor
  • ***
  • Posts: 2377
  • Country: de
Re: T.C. measurements on precision resistors
« Reply #242 on: February 13, 2015, 12:26:03 pm »
T.C. as low as < 1ppm/K (thin film  25ppm/K)

I think Yageo specifies 20ppm/K

Thanks for the comparison. Do you have any idea about the spread of spec. between DUT's?
(Specialy when they sit next to each other on the tape.)

When the resistors are thermo coupled, and used as resistors in a circuit, absolute drift and accuracy is not as important as the spread between devices.

Well,
Thin film SMD resistors, like this 0805 , usually were produced from one ceramic board.
This is completely sputtered, and annealed, and then cut vertically and horizontally, to separate these hundreds of single chips from one ceramic board.
After laser trimming , attaching the metal contacts and protecting the resistance film by tight lacquer, the resistors were put into T&R.
Therefore, all resistors from a single board may have identical /similar T.C., but trimming and sorting into T&R from maybe different batches does NOT guarantee at all that you also get similar resistors by your order.. These will probably differ greatly, instead.

This is another reason, why you need the PWW or BMF technology, 'cause they already have smallest T.C. values and variations intrinsically by the technology implemented.

Frank
« Last Edit: February 13, 2015, 12:43:19 pm by Dr. Frank »
 

Offline quarks

  • Frequent Contributor
  • **
  • Posts: 874
  • Country: de
Re: T.C. measurements on precision resistors
« Reply #243 on: February 13, 2015, 02:00:01 pm »
This one, obviously:
http://www.vishaypg.com/foil-resistors/case-studies/study/wekomm_1/

That is almost identical to what I planned for a DIY project. They even have used the same binding posts (really very nice Model 2758, see att.) as I had on my BOM.

But I wonder why they use the VHA518-7 instead of VHA518-11Z (4-terminal), which I had on my BOM, before I bought a SR104.
 

Offline ManateeMafia

  • Frequent Contributor
  • **
  • Posts: 730
  • Country: us
Re: T.C. measurements on precision resistors
« Reply #244 on: February 13, 2015, 03:09:52 pm »
Maybe the internal layout is similar to the Fluke 742A. Attached is a picture of the internals of my 742A-10K. It had a loose binding post and I was able to fix it before sending it to calibration.

Fluke used what appears to be four 39.992K resistors with a mica card resistor for trim. The lighting wasn't the best but you should be able to make out the five resistors.

I think they use Rose Bopla cases too. This part seems to match the dimensions http://www.digikey.com/product-detail/en/07051100%2050/902-1220-ND/2499610
 

Offline quarks

  • Frequent Contributor
  • **
  • Posts: 874
  • Country: de
Re: T.C. measurements on precision resistors
« Reply #245 on: February 13, 2015, 03:20:26 pm »
Maybe the internal layout is similar to the Fluke 742A. Attached is a picture of the internals of my 742A-10K.

Thank you very much!!!
I have been searching for this inside look for years.  :-+
« Last Edit: February 13, 2015, 03:23:24 pm by quarks »
 

Offline Dr. Frank

  • Super Contributor
  • ***
  • Posts: 2377
  • Country: de
Re: T.C. measurements on precision resistors
« Reply #246 on: February 13, 2015, 04:25:30 pm »
Big Thanks for showing this photo.
Very crude construction, but effective.
These four 39.99k will be matched pairwise for T.C. cancellation, to less than 0.5ppm/K, usually.
A big pity, that they did not implement a thermometer with R(T) curve.

Frank
 

Online Vgkid

  • Super Contributor
  • ***
  • Posts: 2710
  • Country: us
Re: T.C. measurements on precision resistors
« Reply #247 on: February 13, 2015, 04:45:45 pm »
Thanks for posting the internals of the 742.
If you own any North Hills Electronics gear, message me. L&N Fan
 

Offline Dr. Frank

  • Super Contributor
  • ***
  • Posts: 2377
  • Country: de
Re: T.C. measurements on precision resistors
« Reply #248 on: February 13, 2015, 08:59:50 pm »
T :bullshit:
FYI I'm getting a pretty awesome 10K metrology grade transfer standard soon from Wekomm
2ppm tolerance
Long term stability better then 1ppm / year
Temperature stability better than 0.3 ppm / °C
And they are conservative specs, and they are working on producing 0.001ppm tolerance parts.

For the tolerance, you really meant 0.001% [10ppm], not "0.001ppm", right?

I don't think there is any way to even *measure* 0.001ppm-- well, maybe with a QHR and cryogenic comparator, but I think the leads connecting to the QHR inside the liquid helium Dewar will prevent you from getting to 0.001ppm [because the leads are not superconductors].

So, a "typo" right?

Definitely something < 1ppm. Maybe one zero too much..
Comparison between two QHRs has been done -as you describe correctly- by SQUID or so, to 1^-18 uncertainty.
Problem, like JJ array output, is the transfer from quantum to analogue world, which is limited at about 1^-9, due to noise @ 300K and thermo voltages of about 1..10nV.

Therefore, I can't await 2018, ie the new kilogram, the new Volt and Ohm, and so on..

Frank
« Last Edit: February 13, 2015, 09:15:14 pm by Dr. Frank »
 

Offline AndreasTopic starter

  • Super Contributor
  • ***
  • Posts: 3221
  • Country: de
Re: T.C. measurements on precision resistors
« Reply #249 on: February 13, 2015, 09:13:50 pm »
Can you show us a picture of what is "face up" and what is "face down" and which side is closest to your temperature measuring device in each case?

Hello Ken,

the arrangement is shown on page 15 in post
https://www.eevblog.com/forum/projects/t-c-measurements-on-precision-resistors/msg599876/#msg599876

picture:


So face up means that you can read the resistor value and the tolerance from top side.
Face down in this context is when you read the brand (VISHAY), date code (B0940-) and resistor type (Z201T).

The reference temperature sensor (NTC2) is the one lying left of the resistor.
NTC1 is right and NTC_Legs is between the legs.
The NTCs are not calibrated so they have up to 0.3 deg C difference against each other.
For the T.C. measurement I will need only temperature differences and not the absolute value.

When you look at the "naked" Z-Foil resistors (VAR) for "high end audio" applications you will see
that it is impossible that they behave symmetrically.

http://www.vishaypg.com/docs/63140/var.pdf

Is this resistor  identical to the Z201, you've sent me, 1k also?
Then I'll do the reversal test on my 3458A also.

Hello Frank,

you got the Z201#2 which has the same date code (B0940-) as the Z201#3 which I have
used for the last measurements. Both are 1K resistors.
And thanks again for testing.
If you get different results then we have a problem to find the root cause.
One could be different measurement current. Around 2.5 mA in my case and around 1 mA in your case.
So I have factor 6.25 more self heating of the resistor.

There exist many Pictures of the ceramic /metal foil chip, with attached leads.. That always looks symmetrical, so no reason for asymmetric thermal behavior.
They are symmetrical if you mount them in upright position.
But from front and back side you should have different thermal resistances to the surface.
The resistive element (metal foil) is on the front side then comes the aluminium substate then the wires on the back side.
I guess that the epoxy resin is thicker on the wire side than on the metal foil side.
The problem is to know if vishay does really do consistent printing or if they dont mind printing the values on the "wrong side".
http://www.vishaypg.com/docs/63187/zseries.pdf

Or, make a comparison against a poor mans 0.1% resistor like the RT0805BRD071KL from Yageo?
What do you get extra when paying $10 for a resitor, instead of $0,10?

Hello Johnny,

I already made a comparison to a "good" (more in the 1$ range) thin film resistor (RC55Y type) with 15ppm/K spec.
See page 1 of the thread. The question is how much effort is done for long term stability.

Did you consider any SMTs?
Susumu RG and URG are avaliable to 2ppm/C, with URG series being supposedly a higher quality, better aging. Both have some kind of secret inorganic coating. Datasheet: "Unmatched Reliability and Excellent Stability at different environmental conditions".

I think I have already mentioned it in the thread: Definitively no.
The reason is simple: SMTs cannot be measured alone. (At least it makes no sense).

I will in each case get the humidity coefficient and the thermal expansion coefficient of the epoxy material or ceramic substrate where the SMTs are mounted.
It is simply useless to specify T.C. coefficients of SMTs (below 20ppm/K) without exact definition of the mounting method.
With through hole devices I can decouple the resistor from PCB (expansion) by the mounting method.

I think you can reach the 2ppm/K only on ceramic substrates with same thermal expansion than the resistor.
But never on a standard epoxy PCB which shrinks and swells with humidity.

Thermal cycling is sometimes recommended for removing mechanical stress on sensitive components. Do you have any thought on that based on the measurements?
Heating removes humidity from PCB. So the following thermal cycle will get lower hysteresis.
Thats an old trick to let voltage references with plastic package look better in the datasheet.
But it will not help you in real life applications. (except when you heat up the whole device).

With best regards

Andreas
« Last Edit: February 13, 2015, 09:49:39 pm by Andreas »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf