fluke 343a as transfer standard

[muvideo]

Hello everybody,
I'm thinking to buy a used fluke 343a.
The seller offered to "calibrare" it.
I wrote "calibrate" because he will connect the 343 to his 8846 recently
calibrated and move 343's pots to read the correct value on his multimeter.
Now I'm thinking to use this a a starting point for calibrating dcV of my 7061 and 7075.
The 343 output will be loaded, mainly in the 100V and 1000V ranges, by the
multimeters, but it will be loaded the same from the owner's instrument.
The procedure will be not a high precision one, but what accuracy can I expect
from this procedure?
The 343 will be delivered trough post but the travel will be short.

Thank you.
Fabio.

edit:
don't know if this is the right place to make this question, if there is a better section
of the forum let me know.

[amspire]

If you are not sure how well calibrated his Fluke 8846, ask him not to change any settings on the calibrator. Ask him instead to just measure a few key values and to give you the measurements. Also ask him to measure the room temperature when he does it.

The reason is that the calibrator is very stable, and it may actually be more accurately calibrated then the Fluke 8846. Lets say they agreed to 0.001%. You would not go far wrong just trusting the calibrator.

Calibrating the 343a does take some skill, and a seller will probably not be bothered to do anything more then a quick job, that might even make it worse.

Again you may find that your 7061 and 7075 are close enough, If so, don't touch the calibration on them. Just record the differences.

It is best not to start changing calibration unless either something is way out, or you have your Fluke 343a calibrated professionally at a calibration lab with traceable standards.

Richard.

[saturation]

Yes, the real question is more what shape is it in?  The 8846 is less accurate as this calibrator short term, and long term they are nearly equal but TAR would be 1:1, so the adjustment has a higher risk of being off, but better than nothing.

If its in good working condition it would make an excellent calibrator to its rated accuracy.  Its similar to the model restored here, that you've viewed too:

https://www.eevblog.com/forum/product-reviews-photos-and-discussion/fluke-341-dc-voltage-calibrator-restoration/msg97919/#msg97919

[muvideo]

My situation is :
- the 7061 is not calibrated at all.
It had probably some EEPROM problems, I'm waiting another programmer to check it, but
now it has a new eeprom so no cal.
- the 7075 had many problems, I changed reed contacts, it has a bad pot on 10V dc calibration,
also I changed the chopper leds. So it's calibration isn't very trustable.

343 situation is:
343 owner isn't a professional seller, he works in an electronics company, and they
used it for some projects some years ago.  They bought it used, and from last pro calibration,
they adjusted it, don't know why, so the original calibration is "compromised" anyway.
Their 8846 was calibrated by a pro service in 2012.

To resume his 8846 is recently calibrated in a professional laboratory,
my instruments aren't calibrated at all, the 343 lost it's original calibration.
I should at least find a way to minimize the errors on this procedure.
He is too far to go in person to his lab .

Fabio.

[amspire]

I haven't looked at the calibrating procedure for the 343a, but I assume it is a Kelvin-Varley divider, and there is probably adjustments for every resistor in the first 3 or 4 decades. In can be a big job re-adjusting the resistors all to match again.

With any luck, all the idiots who fiddled with the calibration didn't touch the resistor calibration just the voltage source. If that is the case, ask this guy just to calibrate the 1000V, 100V and 10V ranges. Then after calibrating, ask him to write down the measured values.  Don't ask him to touch anything else.

When you get it, it probably has every switch as a 0-10 switch. this means you can check the divider by comparing the 1 output on a switch with the 10 output on the next switch.

Richard.

Edit: it is not a kelvin-varley divider - it is a Beta String series resistor chain, but there are 6 adjustments per decade for the main decades.

[muvideo]

Ok, thank you, I'll ask him to not touch it anymore, write down the values he reads on the multimerter
in multiple voltages and describe the calibration procedure they have done in the past.

Fabio.

[saturation]

This is a better approach as if the seller doesn't do it right, it could could out worse.

Ok, thank you, I'll ask him to not touch it anymore, write down the values he reads on the multimerter
in multiple voltages and describe the calibration procedure they have done in the past.

Fabio.

[casinada]

Did you end up purchasing the Fluke 343A ?

[muvideo]

Yes, later I found that it was surprisingly damaged but still workin ok,
the owner, trying to repair the unit, made some mistakes, fortunately
no precision or rare part was damaged.
I answered you in the other thread about the transistors.

Fabio.