Author Topic: Ultra Precision Reference LTZ1000  (Read 1333305 times)

0 Members and 5 Guests are viewing this topic.

Offline Dr. Frank

  • Super Contributor
  • ***
  • Posts: 2377
  • Country: de
Re: Ultra Precision Reference LTZ1000
« Reply #1325 on: February 15, 2016, 11:07:22 am »

What kind of ballpark figure is a set of hand made resistors, like yours, from Edwin going to set me back? Or is it "if you have to ask, it's too expensive for you!"  :-DD

Any stock wirewounds that could be available?

As you are located in UK, you might order most easily from Rhopoint Components.
They offer PWWs, called 'econistor' from Prime Technology / General Resistance.
I had some trouble with the 120 Ohm type, they had too high a T.C.
I received 5 new ones, directly from G.R, not stellar, but inside spec. limits.

All other values were also 'ok', especially the 12k/1k resistors, to realize an LTZ1000 - reference with a numerical T.C. of about 0.05ppm/K.

Rhopoint also offer BMF resistors from AE, 'FLCY', which are quite good.

Btw.: Did anybody calculate the influence of the 200k "T.C. compensation resistor" in the LT circuit?
I wonder, whether this can be chosen to trim to near zero T.C., with given overall T.C. of the five precision resistors.
« Last Edit: February 15, 2016, 03:33:42 pm by Dr. Frank »
 

Offline d-smes

  • Regular Contributor
  • *
  • Posts: 101
  • Country: us
Re: Ultra Precision Reference LTZ1000
« Reply #1326 on: February 15, 2016, 12:37:18 pm »
Btw.: Did anybody calculate the influence of the 200k "T.C. compensation resistor" in the LT circuit?
I wonder, whether this can be chosen to trim to near zero T.C., with given overall T.C. of the five precision resistors.
I was hoping someone would bring this up.  In the LT datasheet, the positive reference circuit feeds the 400k compensation resistor with heater voltage plus the Vf of the 1N4148 which has a strong negative tempco.  But in the negative reference circuit, there is no diode Vf, but instead has the ~7.2V reference voltage in series with the heater voltage feeding the 400k.   Is there a way of analyzing the optimum value for the tempco compensation resistor or must this be empirically derived?   I suspect the latter...
 

Offline lars

  • Regular Contributor
  • *
  • Posts: 132
  • Country: se
Re: Ultra Precision Reference LTZ1000
« Reply #1327 on: February 16, 2016, 06:59:04 pm »
What about humidity sensitivity for wire wound resistors like UPW50 and new Econistors? I haven´t tested UPW50 but many old 8E16. I have seen up to 2ppm/%RH on 8E16. That is 60ppm for a 30%RH variation, 100ppm for 50%RH and 160ppm for 80%RH. For me that is worse than temperature coefficients up to 5ppm/C that also is easier to compensate. On the 8E16 and also on metal foils like RN55 and BMF like S102 and VSMP it seems that higher values, like 10k and 100kohm, is worse with around 1ppm/%RH and 100 and 1kohm sometimes is close to zero sensitivity but variation might be large. As time constants is weeks or maybe months humidity sensitivity is difficult to test.

For the LTZ1000 this means that if you are unlucky you can have say 1ppm seasonal variations that is tricky to check. Seasonal variations may also make your drift assumptions to be completely wrong.

If you are lucky with your wire wounds the 1-12.5-70kohm humidity sensitivities to LTZ sensitivities cancel a little but you never know.

Lars
 

Offline branadic

  • Super Contributor
  • ***
  • Posts: 2378
  • Country: de
  • Sounds like noise
Re: Ultra Precision Reference LTZ1000
« Reply #1328 on: February 16, 2016, 07:28:22 pm »
Quote
I suspect that a larger than the theoretical 0.05ppm/K TempCo was discovered when the LTZ1000 was used, and the ~400K resistor was a "hack" to fix the symptoms instead of fixing the real problem.  Honestly, you should leave this 400K resistor out of your build, and use proper mechanical design that addresses the thermal issues instead of hacking on your circuit until it works

Well, interestingly you wont find this 400k resistor in the circuit diagram of the LTZ1000 (none A) in the paper "The Ultra-Zener ... is it a portable replacement for the Weston Cell?" (Reveived 15 December 1989).
Leads to ask at what time this resistor was put into the circuit diagram.
Computers exist to solve problems that we wouldn't have without them. AI exists to answer questions, we wouldn't ask without it.
 
The following users thanked this post: cellularmitosis

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 14072
  • Country: de
Re: Ultra Precision Reference LTZ1000
« Reply #1329 on: February 16, 2016, 08:21:44 pm »
Many analog circuits run at a temperature well above room temperature - could be intentional or just a nice side effect. This keeps down relative humidity - a 10 K temperature rise already reduces humidity to about half. So variations in humidity are not that large anymore (e.g. only half as well) - at least for instruments that run continuously.

The reduced humidity also helps to keep leakage currents down - surface leakage get worse at high humidity levels (e.g. above about 50% for glass).

With the 400 K resistor for temperature coefficient compensation, I am a little suspicious, as the the voltage at the heater varies with the square root of the heater power which is about linear in temperature. So the effect of the resistor would be larger at a low temperature set-point. Also the thermal design (e.g. long vs. short leads) would have an influence.
 

Offline alanambrose

  • Frequent Contributor
  • **
  • Posts: 377
  • Country: gb
Re: Ultra Precision Reference LTZ1000
« Reply #1330 on: February 17, 2016, 09:27:20 am »
As a couple of data points re humidity - when I ran some regressions on MM's 3458a data a while back (I still need to do some more) I didn't get any obvious coefficients re pressure or humidity - drift was mostly a function of time and internal temperature. Another data point is the Fluke 732 analysis (I appreciate not an LTZ1000, but indicative of precision refs in general) - they hinted at some 'seasonal variations' (which I took to mean RH) with their fitting equation. Fluke didn't elaborate further. This all suggests to me a relatively minor role for humidity and pressure in built refs. Lars - btw how did you calculate your RH coefficients / do your RH experiment?

A.

p.s. I was hoping to get my drift experiment underway before I left for vacation. However, I discovered a lot of noise on my 'low noise' power supply :) It will have to wait until I'm back now.
“A foolish consistency is the hobgoblin of little minds"
 

Offline Andreas

  • Super Contributor
  • ***
  • Posts: 3221
  • Country: de
Re: Ultra Precision Reference LTZ1000
« Reply #1331 on: February 17, 2016, 07:48:44 pm »
Hello,

What about humidity sensitivity for wire wound resistors like UPW50 and new Econistors?

BMF resistors in epoxy packages can vary as much as 25ppm from 0% to 100%, and they can do this in 30 minutes or less!  PWW resistors on epoxy bobbins will also be affected by humidity if there is inadequate spacing between the winding and the bobbin, and also if the welds to the resistance wire are not good [letting the wire move while the epoxy is flexing due to the humidity and/or temperature changes].

in my T.C. measurements I blame the "ageing drift" which I measure over 3 days (up to 5-6 ppm) to the fact that my peltier box traps some humidity over this time.
I have also put a dry-pack into the box and the drift changed direction after this.
In principle I have no significant difference between Z201 and PWW resistors. (both up to 6 ppm within 3 days depending on air humidity).
Both have a epoxy housing which swells with humidity (more or less depending on actual resin) and no one can tell me that this has no influence on the wire.
You simply cannot fully isolate the wire (= strain gauge) mechanically from the epoxy.

I think it has its reason why precision voltage dividers use mica cards for their wire wound resistors.

With best regards

Andreas


 

Offline Andreas

  • Super Contributor
  • ***
  • Posts: 3221
  • Country: de
Re: Ultra Precision Reference LTZ1000
« Reply #1332 on: February 17, 2016, 07:51:40 pm »
Well, interestingly you wont find this 400k resistor in the circuit diagram of the LTZ1000 (none A) in the paper "The Ultra-Zener ... is it a portable replacement for the Weston Cell?" (Reveived 15 December 1989).
Leads to ask at what time this resistor was put into the circuit diagram.

Hello,

interesting, they use a FET for the zener current like my cirquit.
But they have forgotten the pull down resistor at the gate of the FET.
Without this the LT1013 may get instable...

With best regards

Andreas
 

Offline Galaxyrise

  • Frequent Contributor
  • **
  • Posts: 531
  • Country: us
Re: Ultra Precision Reference LTZ1000
« Reply #1333 on: February 17, 2016, 11:22:44 pm »
68k collector resistors, too; closer to 100uA than 70k.
I am but an egg
 

Offline lars

  • Regular Contributor
  • *
  • Posts: 132
  • Country: se
Re: Ultra Precision Reference LTZ1000
« Reply #1334 on: February 18, 2016, 11:57:14 am »
Lars - btw how did you calculate your RH coefficients / do your RH experiment?


About every second month I spend a couple of hours to measure a couple of hundred resistors (and many DIY 10V refs) against my HP3456A. I have done this for almost 20 years.

Lately I have also started to use ice-cream boxes and large bags of silica gel prepared for different humidity’s. Right now I have a VREF10 from Voltage standard under test.
See also https://www.eevblog.com/forum/metrology/calibratory-d-105-dc-precision-voltage-reference-standard/msg803701/#msg803701
for an example of my use of the box that revealed a 20ppm change due to humidity on the 2ppm specified D105 ref.

For the long-term tests with the HP3456 I use regression in Excel to get temperature, humidity and aging. I enclose a couple of screen shoots from Excel with temperature compensation but not aging and humidity.

The first chart is 100ohm, 1k, 10k and 100kohm wire wound 8E16. The 1Mohm is ordinary 1% metal films I happened to have 20 years ago, if I remember correct I selected four 220kohm + some smaller to get to 1Mohm with low tempco. (As a side note I have seen up to 1000ppm/year drift on some new 0.1% metal films from Reichelt).

From 1997-2001 the values are relative to the 3456 and a small extra humidity sensitivity about 0.1-0.2ppm/C can be seen due to the 3456. From 2001 I got a GR1440-10k that is the reference and the 100ohm, 1k, 100k and 1M is related though a home built resistance decade. I enclose a picture of one of many home-built decades I have including my parallel and series-parallel adapters.

I also enclose a chart showing 100ohm, 1k, 2 pcs 10k and 100kohm VHP resistors I have had six years. Remember that the 3456 has 1ppm resolution and to get from the 10k GR1440 to for example the 1kohm I have to measure the GR, the VHP and series-parallell 10x1k+ series 10x1k+one of the 1k. So the ripple is not so bad, sometimes it also can be seen that I have some wrong values. Nowadays I have an automated routine that takes 20 samples with 10PLC but disturbances can still occur. Earlier I averaged a few samples with 100PLC in my head.

The last chart is two each of 100ohm, 1k and 10k of VSMP 1206SMD soldered on two small FR4 board.

Lars
 
The following users thanked this post: cellularmitosis

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4543
  • Country: ua
    • xDevs.com
Re: Ultra Precision Reference LTZ1000
« Reply #1335 on: February 18, 2016, 01:12:44 pm »
This is what I call dedication. Appreciate efforts and sharing results.
YouTube | Metrology IRC Chat room | Let's share T&M documentation? Upload! No upload limits for firmwares, photos, files.
 

Offline blackdog

  • Frequent Contributor
  • **
  • Posts: 736
  • Country: nl
  • Please stop pushing bullshit...
Re: Ultra Precision Reference LTZ1000
« Reply #1336 on: February 18, 2016, 01:31:54 pm »
Hi,


If you want to know how to measure realy low noise references, look @ this webpage:
http://www.edn-europe.com/en/measuring-2-nv/-8730-hz-noise-and-120-db-supply-rejection-in-linear-regulators-the-quest-for-quiet-part-2.html?cmp_id=7&news_id=10008095&vID=1328&page=0

Happy reading  ;)

Kind regarts,
Blackdog
Necessity is not an established fact, but an interpretation.
 

Offline fmaimon

  • Supporter
  • ****
  • Posts: 165
  • Country: br
 

Offline Andreas

  • Super Contributor
  • ***
  • Posts: 3221
  • Country: de
Re: Ultra Precision Reference LTZ1000
« Reply #1338 on: February 20, 2016, 10:10:41 am »
Hello,

after 2 weeks without temperature cycling the popcorn noise on my LTZ#5 has restored.  :(

see reduced noise after temperature cycling
https://www.eevblog.com/forum/metrology/ultra-precision-reference-ltz1000/msg856965/#msg856965

and previous behaviour
https://www.eevblog.com/forum/metrology/ultra-precision-reference-ltz1000/msg833804/#msg833804

With best regards

Andreas
 

Offline Andreas

  • Super Contributor
  • ***
  • Posts: 3221
  • Country: de
Re: Ultra Precision Reference LTZ1000
« Reply #1339 on: February 20, 2016, 10:44:11 pm »
Hello,

further results on
LTZ#4 (LT1013) and
LTZ#6 (LTC2057)

First the operating points compared to the LTZ#3 and LTZ#5 values

and then the PCB twisting test (to check the influence of mechanical stress to the PCB).
I am bending the PCB on different directions and record the zener voltage of the LTZ
as difference to another zener in 100 mV range of a meter.

Result is around 4 uV change (or 0.5ppm) when bending the PCB
with no significant difference dependant of operational amplifier.
-> a good value (I will not try to bend the PCB within metal housing).

With best regards

Andreas
 

Offline Andreas

  • Super Contributor
  • ***
  • Posts: 3221
  • Country: de
Re: Ultra Precision Reference LTZ1000
« Reply #1340 on: February 21, 2016, 09:22:07 pm »
Hello,

today I improved thermal management:
a additional PCB slot between voltage regulator and the LTZ1000
zener to keep the waste heat (battery voltage dependant) somewhat away from the LTZ.

Battery voltage was sweeped between 14 and 18.5V (typical operating range around 14.5 to 17.4V)

Without slot on LTZ#3 the voltage output change was around 5uV / 4.5V (0.15ppm/V) as regression curve.
With slot the change is reduced to 2.9uV / 4.5V (0.09ppm/V).

On LTZ#5 I have 2.1uV/4.5V (0.06 ppm/V) with the slot.
Unfortunately I missed to measure the "before" data.

With best regards

Andreas
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4543
  • Country: ua
    • xDevs.com
Re: Ultra Precision Reference LTZ1000
« Reply #1341 on: February 22, 2016, 04:09:09 am »
How about tempco of your design? Any plans on those tests?
I'm getting some interesting results during last 2 weeks on my three proto LTZ modules. Going to make summary analysis after all data aligned.
YouTube | Metrology IRC Chat room | Let's share T&M documentation? Upload! No upload limits for firmwares, photos, files.
 

Offline Andreas

  • Super Contributor
  • ***
  • Posts: 3221
  • Country: de
Re: Ultra Precision Reference LTZ1000
« Reply #1342 on: February 22, 2016, 06:28:54 am »
Hello Illya,

of course I have planned the T.C. measurements.
But those will be the last tests after all shieldings are finally installed.
on #5 this will take some time. Since I probably have to exchange the zener anyway (popcorn noise)
I want first do a test with short legs and the influence on tilting and local heating.

And with the results of local heating I suspect that the T.C. will appear different according to ramp speed.

Another question is if there is a dependency of T.C. of zener voltage like on LM399.

with best regards

Andreas


 

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 14072
  • Country: de
Re: Ultra Precision Reference LTZ1000
« Reply #1343 on: February 22, 2016, 04:27:29 pm »
The correlation of TC and zener voltage will be likely similar to the LM399 or other similar refs. It's a more or less fixed feature of silicon zener diodes.

There is some popcorn noise, but not that much. The reference may still be useful. The popcorn noise also looks a little strange, showing a sharp jump and a slow relaxation after that in this case even sufficient filtering might be a sufficient countermeasure. This somewhat makes me think the noise could even originate from the temperature stabilization part (e.g. the sensing transistor). For classical popcorn noise I would more expect something like sharp steps up and down at more or less random times, thus essentially jumping between a high and low value.  This would a way nastier case.

It could also be a candidate to test higher currents for the zener diode and the transistors (70 K resistor). The two 70 K resistors in the originally circuit do not have to be the same - the same value is just convenient. 
 

Offline Andreas

  • Super Contributor
  • ***
  • Posts: 3221
  • Country: de
Re: Ultra Precision Reference LTZ1000
« Reply #1344 on: February 22, 2016, 07:48:50 pm »
The popcorn noise also looks a little strange, showing a sharp jump and a slow relaxation after that in this case even sufficient filtering might be a sufficient countermeasure.

Hello,

for me it does not look strange.
Do not forget that there is a high pass with a time constant of 3 sec within the amplifier.
The simulation shows the same behaviour for a 1 sec rectangular jump.

With best regards

Andreas
 

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 14072
  • Country: de
Re: Ultra Precision Reference LTZ1000
« Reply #1345 on: February 22, 2016, 08:29:22 pm »
The filter in the amplifier explains the shape. So nothing unusual there.

Still jumps of something like 1 µV or 1.5 µV are not that bad, that is within the noise specs of the LTZ1000. It may be that other people just can resolve these jumps and see it as 1/f noise. So ref. 5 might not be the best one, but still OK.
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4543
  • Country: ua
    • xDevs.com
Re: Ultra Precision Reference LTZ1000
« Reply #1346 on: February 25, 2016, 08:12:39 pm »
I was running three of my prototype boards for some TCR measurements and testing few ideas how to improve it. I think few tricks could work for my boards.
Still one more thing to check, and then I'll try to post all data and my take on it.

Setup jig:



Final reference will go into own box, such as this:






YouTube | Metrology IRC Chat room | Let's share T&M documentation? Upload! No upload limits for firmwares, photos, files.
 

Offline alanambrose

  • Frequent Contributor
  • **
  • Posts: 377
  • Country: gb
Re: Ultra Precision Reference LTZ1000
« Reply #1347 on: February 25, 2016, 10:29:38 pm »
>>> Still jumps of something like 1 µV or 1.5 µV are not that bad, that is within the noise specs of the LTZ1000. It may be that other people just can resolve these jumps and see it as 1/f noise.

Just reading my new (to me) copy of Henry Ott - he points out that the noise is gaussian (both johnson and 1/f) with unbounded tails, so (in principle at least) any jump is possible. Some kind of noise distribution would be a nice way to compare refs.

A.
“A foolish consistency is the hobgoblin of little minds"
 

Offline doktor pyta

  • Frequent Contributor
  • **
  • Posts: 488
  • Country: pl
Re: Ultra Precision Reference LTZ1000
« Reply #1348 on: February 26, 2016, 12:12:42 am »
Here is my version :)
The red cup covering LTZ1000 is SP163 (http://www.tme.eu/gb/katalog/?art=SP163#)
The board fits into Fischer TUF-series encolsure.
Photos below.

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4543
  • Country: ua
    • xDevs.com
Re: Ultra Precision Reference LTZ1000
« Reply #1349 on: February 26, 2016, 12:49:20 am »


Here is my version :)
Photos below.

Beautiful board, welcome! Output is direct 7V or scaled, based on second LT1013 presence?

Looking forward for your tests ;).
YouTube | Metrology IRC Chat room | Let's share T&M documentation? Upload! No upload limits for firmwares, photos, files.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf