Fluke 732C mystery

[ap]

The (relatively) new Fluke 732C tripple (10V, 1V, 0.1V) voltage standard does not have any adjustment means for any of their output voltages (at least I could not find anything in the documentation, calibration is limited to verification/measurement, no adjustment hints). So seams set to nominal once, and then left unchanged. That makes some sense, as probably most users dont re-adjust the output voltages anyhow on the older devices (where it was possible through trimmers or resistor switching) to keep track of drifts easier and avoid any impact of potential trimmer instabilities.

However, the deviation from nominal as delivered is not specified either. For curiosity, does anybody have any clue whats typically the case for the three voltages on items shipped?

[TiN]

You mean there is no switches on oven block, when you remove the top metal cover?
Perhaps can share a photo? User's manual clearly shows same design as 732B with FPC connected board and 4 switches to trim 10V output in 0.1, 1.0, 10 and 100 ppm steps.
But it could be just that Fluke didn't update drawings and reused old ones from 732B, if new design actually changed. 1V/100mV have fixed division from 10V so no adjustment for those..

[ap]

I dont have one, so unfortunatelly I cannot open one up 
My question is of pure interest, since this surprised me when reading the 732C manual. Obviously the 732B describes two calibrations, one with and one without adjustment, the 732c manual not.
Could be they just omitted that section of course.

[ManateeMafia]

Not surprised as they also omitted the calibration section/manual on their new meters.

[HighVoltage]

The local CAL lab told me, that Fluke has a strategy now, to keep all of this info to themselfs so that nobody except Fluke will repair and calibrate their metrology gear in the future.

So, no surprise that this info is missing in the new manuals.
 

[Magnificent Bastard]

The local CAL lab told me, that Fluke has a strategy now, to keep all of this info to themselfs so that nobody except Fluke will repair and calibrate their metrology gear in the future.

So, no surprise that this info is missing in the new manuals.

If that's true, it would prevent me from buying ANY of their products.  They can get away with this in the civilian market, but the [at least the US] military demands not only calibration information, but full schematics to the component level.

[TiN]

Military customers may get required documents per special request to vendor.
There is a reason why Mil-version Fluke calibrators even had special index

[ap]

Well, I would say even for commercial users, adjusting a meter (maybe not a reference, where one might accept that it is not adjusted to see drift history) is something one would like to be free to do. Some companies e.g. have their own cal labs. Or if you want to go to a third party cal lab (potentially doing a better and cheaper job), this would no longer be possible. It always takes two to play this game...
Will be interesting to see how the 3458B will handle this.

[CorporateReference]

The 732C has no built in adjustments. The factory laser trims Z402 (R402 on 732B) as a matched set to the reference to get it in the 9.9-10.1V range.  Once the factory assembles and ages the unit for delivery, the assembly engineers have the option to only increase output by adding a resistor on R426 which by default is unpopulated. R426 is the equivalent to the BCD switch string on the 732B which parallels with the 180ohm laser trimmed resistor ( R405 On 732b).  There is typically a dummy  59k resistor at R427 (removed in photo). The 732C is nearly identical to the 732B with only laser trimmed hermetics and the lower 1v 0.1v now buffered via 2x LT2057. The 1v and 0.1V are laser trimmed individually, both voltages are based off the main 10v just like the 732b. 

[Echo88]

Thanks for the insight 
Could you share more photos for example of the hermetic resistor network and the oven structure if possible?

[martinr33]

[ Specified attachment is not available ]I have a few pictures from a current calibrator model 5730A. This unit is at specification 30 minutes after power on (although I am sure that it gets much, much better over sustained power on). The construction is likely to be very similar to the 732C, barring the aluminum thermal block that houses the hybrids.

Note the thick, blank ceramic plates bonded to the rear of some of the devices. This is sililar to the Caddock thermally compensated resistor construction, where a second material helps compensate for mechanical expansion and contraction as temperatures change.


First, the front side of the reference. In the calibrators, there are two "stacked" references to deliver a higher Vref



Then, the rear of the reference board. Here you can see the heater element that covers the back of the ceramic PCB.



In a similar vein, the low voltage dividers on the switch matrix board. These dividers support the lower ranges.



And the rear of the divider.

[Conrad Hoffman]

This make a lot of sense to me. 1) You want history and the best history is with no adjustments. 2) How many cal labs could possible calibrate a 732? Unless you have a JJ unit, the cal won't be as good as the unit requires.

OTOH, I want my meters to read as correctly as possible!

[bdunham7]

Not surprised as they also omitted the calibration section/manual on their new meters.

Which meters have they withheld the calibration information for?

[bdunham7]

Some companies e.g. have their own cal labs. Or if you want to go to a third party cal lab (potentially doing a better and cheaper job), this would no longer be possible. It always takes two to play this game...

Very few have their own primary standards lab. Those that do can probably get the info they need.  In the case of a voltage reference, it doesn't seem wrong that they intend that it not be adjusted during any normal calibration or maintenance. 

[alm]

I guess a good reason to adjust is if you want to do analog averaging of multiple 732Cs. To have all cells contribute equally to the drift of average, you ideally want all cells at approximately the same distance from the average.