Incoming stuff, voltage reference..

[robrenz]

I posted the current reversal setup alm mentioned but I since realized I can do offset compensated ohms on the 8846A also. I checked with Fluke and my procedure is valid. I intend to do a post on the technique soon.

Video of the offset compensated ohms technique and making the adaptor.  Thermal EMF effects are clearly shown on this video. https://www.eevblog.com/forum/projects-designs-and-technical-stuff/offset-compensated-ohms-adaptor/msg121362/#msg121362

[Joe Geller]

>>The MV106 is spec'd to < 5uV stability [ I get it practically to 1uV or less under controlled environmental conditions] and resolution to <= 100nV

We have several MV106's too, including one with the built in null meter.  They are very nice units, but definitely not accurate or stable enough to study a reference at the ppm level.  The MV106 spec's in the manual published by Krohn Hite are: 5ppm (50 uV) 1hour 10ppm (100 uV) 8 hours and 25ppm (250 uV) 1 year + 5ppm/C.  1 ppm (one millionth of) 10 V is 10 uV.  5 uV is 0.5 ppm for a 10 V scale.  To see a change of 1 uV on a 10 V scale requires 7 1/2 digit minimum.  100 nV takes a DMM like the 3458A, and even then only short term (a standalone 3458A is spec'd at 8 ppm (80 uV) / year std, 4 ppm / year opt 002).  Below 1 ppm, and certainly by 0.1 ppm, test setups use very special connections (e.g. tellurium copper lugs).  To do 10 V metrology work at the 100 nV generally requires a 10V Josephson junction standards, as found at NIST and Fluke.  By contrast, a hp 3456A can resolve to 10 uV (1 ppm) on a 10 V scale.

In mfgr specs, the scale for a 10 V reference:
1,000 uV, 100 ppm, 0.01% (4 1/2 digits)
100 uV, 10 ppm, 0.001% (5 1/2 digits)
10 uV, 1 ppm, 0.0001% (6.5 digit, 3456A, 34401A, 34410A resolution limit)
1 uV, 0.1 ppm, 0.00001% (Agilent 3458A (for less than the annual calibration period), null meter with a suitably stable reference e.g. the Fluke 732B with historical calibration data going back several years)

More than a few engineer and hobby electronics users are set up to look at the SVR boards at the ppm level, and none have reported any measurable sensitivity to barometric pressure.  We often run a board overnight during calibration, through all kinds of weather and as compared to the 732B transfer standard, using the Fluke 732B with a 3456A recording the 845AR output, the values stay within 1 ppm, all errors considered at a set temperature (in the micro-environmental chamber).  Some of our users periodically check their SVR to a company standard or a local calibration lab, and have reported values within a few ppm of 10.000 00V absolute, some better (our transfer spec is +/- 5 ppm delivered). 

I think the problem might be a combination of units, test setup, and it could be a very broken board too.

[Joe Geller]

I should make another comment about the SVR board and its purpose.  The purpose of the SVR program Since before 2004, is the low cost transfer of an absolute 10 V reference point to 10 ppm (+/- 5 ppm absolute).  The SVR board is not a voltage standard, it is a voltage transfer reference.  For example, a single resistor in a Fluke voltage standard costs far more than each assembled and calibrated SVR board.

The idea is simple.  Many of us buy very high end used equipment.  Some pieces of relatively old gear, with their well aged voltage references, can be remarkably stable over time (where other parts, such as trimmers, electrolytics, and switches are not failing).  The question coming in the door (often pre-stressed by delivery incidents, such as dented or accordioned boxes) is: How close is it?

Most of us live within driving range of a commercial calibration laboratory.  But, do you really want to take that $49 or $12 or $300 used box in and pay $200 to $400 for a cal lab to look at it?  Most say No way!

With the SVR program, anyone from a home lab (e.g. eBay surplus unit) to a light commercial operation (often a one man consulting engineer) can get a relatively quick assessment of their calibration point to about 10 ppm absolute (+/- 5 ppm).  And, you keep the board.

We set the boards to within +/- 1 ppm.  However, taking into account shipping and other possible errors, we rate the boards at +/- 5 ppm (+/- 5 ppm / c).  It is not a rating gimmick, rather this is what we feel we can do with this technology in this price range.  As I mentioned, to further help users, we now calibrate at any requested temperature using our micro-environmental chamber ( http://www.gellerlabs.com/SVR_TempcoTest.htm ), and then measure (takes hours) the actual tempco for each board over a +/- 1c span.

So, can I calibrate a hp 3456A, Agilent 34410A, Agilent 34401A 6 1/2 digit DMM using your SVR board?  The formal cal answer is NO.  However, for most small labs to know they are within +/- 0.0005% of 10 V absolute is plenty good enough.  So, if you set your hp 3456A to that point with the fine trimmer, what harm is done?

So, can I calibrate my used 3458A (8 1/2 digit) DMM with a 10 V SVR board?  NO.  That said, if you measure one of our 10 V SVR boards and are off by 200 uV (20 ppm), you know you have a problem with absolute calibration!  This has happened on more than one occasion. 

[Circuitous]

Circuitous: 34410A 10.000 05 V, same offer. 

Joe, I'll send my SVR to you for re-calibration.  I certainly have no complaints, I think it's a great unit!
Regarding stability, I found it provided a stable output with the input as low as 10.945 V.
http://corgitronics.com/2012/03/22/voltage-references/

Joe, I'll PM you about getting one of the new SVR-T units.

[saturation]

Thanks Joe, for a very erudite analysis and refresher on the MV106 specs.

Of concern here is the stability of the SVR, in this case compared to the MV106.  If the MV106 is not good enough as a comparison, then clearly its inferior to the SVR, and that can easily be demonstrated, QED:  compare outputs of both units over time and compare it to a another reference, a calibrated DMM.

The MV106 reference is a 6V zener IN829? that has the ~ tempco of 5ppm/^oC as the AD587.  Its uses a chopper amp with some gain and a non-KV resistance network to scale 10.000 00V, 1.000 000 V and 10.000 00mV, and a maximum resolution of 10nV.  Accuracy and stability wises, it is at least on paper <= to the SVR, as you describe.

The dividers of the MV106 are all tied to the performance of the IN829 and despite what can go wrong in a discrete design, dial down to 1.0 uV on the 10mV scale and it will hold this reading for at least a month, to ~ +/-2ppm.

You can dial output in 100nV increments, the limit of the 3456a resolution in the 10mV scale, and it will hold for hour.

All the above regardless of weather.

OTAH, tie my SVR to a quality KV divider, even suitably buffered, and you'll be lucky to get 1uV out of it, that's of course out of its spec, 1uV being 0.1ppm. 

Getting it checked by you and getting another would certainly shed more light on this weather or malfunction issue. 

I suspect key to the MV106 performance is that the now well aged zener is fed by a very well regulated constant current source in addition to the chopper amp.  The SVR is at the mercy of line and load regulation which is considerable at 100uV/V, which can explain a lot of its variation.  By itself, the IN829 is far inferior to the AD587, so it would be interesting to build the stability circuits of the MV106 around the AD587.

Barometric effects are known, albeit it doesn't happen uniformly and varies in magnitude, from none to some.  Whether this is my phenomena or something else related to weather [ such as affecting the power feeding the SVR, or changing humidity] remains to be seen.  Suffice to say, the SVR reads best when the weather is best.

http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=852292&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D852292

Rather than calculate a correction coefficient, which is possible once enough data is obtained, I simply make a critical measurement only when the conditions are identical to the original calibration, and saves the trouble.

On the subject of physical issues affecting voltage references, I've never heard of changes with physical position, as suggested in volt-nuts:

http://www.febo.com/pipermail/volt-nuts/2010-September/000465.html

Its the only report I know of this effect, and Dr. Stellmach is reputable.  So, you never know what new things can occur.

>>The MV106 is spec'd to < 5uV stability [ I get it practically to 1uV or less under controlled environmental conditions] and resolution to <= 100nV

We have several MV106's too, including one with the built in null meter.  ...

I think the problem might be a combination of units, test setup, and it could be a very broken board too.

[saturation]

I think this is the bottom line, despite my criticism.  You've an excellent service and I recommend it to others.   To get the 10V traceability you offer, at the cost, is really a boon to all.  Keeping the package small and light is key, imagine shipping a 731 style reference or even the MV106 and the costs far exceed the costs of your services with the board.

I should make another comment about the SVR board and its purpose.  The purpose of the SVR program Since before 2004, is the low cost transfer of an absolute 10 V reference point to 10 ppm (+/- 5 ppm absolute). 

[bingo600]

I can only approve of the Geller SVR's

I just got my HP 3457A back from Cal , and the SVR-T (5 month old) is still "spot on"

Posted pics here
https://www.eevblog.com/forum/testgear/geller-labs-svr-and-svr-t/msg294884/#msg294884

/Bingo