Author Topic: Show me the stability of your voltage references  (Read 2946 times)

0 Members and 1 Guest are viewing this topic.

Offline JornTopic starter

  • Contributor
  • Posts: 24
  • Country: dk
Show me the stability of your voltage references
« on: November 12, 2017, 08:46:53 pm »
This topic is dedicated to the discussion of the various methods to document stability of our voltage references.

Each voltage reference deserves a few graphs and statistics so that we get to know them better.

But what is appropriate for sub ppm voltage measurement?  HP3458A?

Maybe. But I think we can do better at much lower cost.  Use more than one reference and measure them back to back with a 6.5 or 7.5 digit DMM.

I currently have 2 LTZ1000 based references. One that has been on 24/7 for over 2 years and one that has been on 24/7  for almost one year. 

Measuring them back to back for almost 2 months and calculating the overlapping allan deviation result in the following "fingerprint" of the combined stability of the 2 references:




Note how the graph spans from 40 milliseconds to almost 2 million seconds.

Actually this graph is combined from two separate measurements sessions:

1) 360,000 samples (NPLC=100, autozero on) with a sampling period of 10 seconds.
2) 92,000 samples (NPLC=1, autozero on) with a sampling period of 0.04 seconds.

as shown here



The graphs are combined at t=100 second.

A lot of other graphs could also be relevant to show:





Unfortunately pressure and relative humidity sensors was not ready until the last 300 hours. :-\

When you have this amount of data available some data aggregation seems obvious.

Let's down sample data from T=10 second to T=1 hour (mean value of every hours of measurements):



and we can go even further. Here the mean value of each day is shown:



And then there are statistics of all sorts:


(disregard pressure and humidity due to missing data)
 
Some are probably useless but looking for sub ppm improvements correlation figures comes true.

And then there is TC measurements but I'm currently working on a setup with two thermal chambers - one for each reference. More on that when ready.

I hope this post will bring more focus on the actual characterization of all your reference builds.

Show me the stability of your voltage references below!
 
The following users thanked this post: zhtoor

Offline Andreas

  • Super Contributor
  • ***
  • Posts: 3351
  • Country: de
Re: Show me the stability of your voltage references
« Reply #1 on: November 12, 2017, 09:40:05 pm »
Hmm,

- what instrument do you use?
- in which range
- are you measuring the noise/offset drift of your instrument + the references
   (zero volts difference) or are you measuring the tempco stability of your instrument (near full scale?)

just for interest: do you use buffered/unbuffered references?
at which temperature setpoint?

with best regards

Andreas
 

Offline JornTopic starter

  • Contributor
  • Posts: 24
  • Country: dk
Re: Show me the stability of your voltage references
« Reply #2 on: November 12, 2017, 11:59:16 pm »
Both references are LTZ1000A based with a 12K5 + 1K = 50 deg setpoint

They are both buffered using LTC2057HV + LT1010. 

I'm measuring the difference in voltages, which is about 24.543 mV, on the 100 mV range of a Keysight 34470A multimeter.
In this range the resolution is 10nV  admitted with a lot of noise on the last digit. 

The drift and TC of the meter is of course included in the measurements.
The temp coefefficient in the 100 mV range is specified as 5 ppm of reading + 5 ppm og range. The  instrument temp spans about 6 degrees in the entire 1k hour measurement period so the total TC error is 6 * (5 ppm * 25 mV + 5 ppm * 100 mv) = 3.75 uV  = 0.52 ppm.  1y accuracy of the meter in this range is 40 ppm of reading + 35 ppm of range = 4.5uV = 0.63 ppm   

This add up to 1.15 ppm. Compare that to a 3458A in a direct measurement in the 10V range with 6 deg temp span having 3 ppm TC error.

I would of cause love to keep the room temperature more stable but that's not possible at the moment.

Regards Jorn
« Last Edit: November 13, 2017, 07:20:40 am by Jorn »
 
The following users thanked this post: zhtoor

Offline MisterDiodes

  • Frequent Contributor
  • **
  • Posts: 457
  • Country: us
Re: Show me the stability of your voltage references
« Reply #3 on: November 14, 2017, 05:42:01 pm »
Just because this is the metrology section...

Out of curiosity, why are you using the ancient LT1010?  Do you really need that much drive current for you application??  That part isn't even spec'd for use below 10Hz and usually adds so much noise we never use them anywhere near an LTZ - or any DC Vref for that matter, at least for a metrology application. 

Usually a discrete large area BJT follower on the output of your '2057 will give you reasonable drive current with very little added noise and very little added cost - see how a 732b Vref is built.

Be careful with back-to-back measuring - make sure you're not kidding yourself into better results when both Vref DUT's are drifting the same direction.  Which LTZ's tend to do.
 

Offline zhtoor

  • Frequent Contributor
  • **
  • !
  • Posts: 337
  • Country: pk
Re: Show me the stability of your voltage references
« Reply #4 on: November 14, 2017, 09:29:54 pm »
hello jorn,

very very nice work. would'nt it be better if you could add at least 2 more LTZ1000's
and then followed up with a couple of LTFLU / SZA263 type references into the mix,
and follow the standard 6-cell protocol by NBS430.

regards and keep up the good work.
 
The following users thanked this post: Jorn

Offline JornTopic starter

  • Contributor
  • Posts: 24
  • Country: dk
Re: Show me the stability of your voltage references
« Reply #5 on: November 14, 2017, 10:11:58 pm »
@MisterDiodes

Just had a couple of LT1010 laying around and besides that, it's within the feedback loop of the LTC2057 so noise should not be an issue (at least i theory ;)). 0.1-10Hz noise measurements confirms that.

When measuring a LTZ1000 with a HP3458A the references could also drift in the same direction....

I'm building another 3 references (waiting since august for LT to deliver LTZ1000) and hope to be able to cross correlate them all. That would help spotting which references that drift at distinct points in time.

Of cause absolute precision is only obtained when comparing one of the references with a known voltage standard from time to time. But that's not within reach for the moment :(

Regards
Jorn
 
The following users thanked this post: MisterDiodes


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf