Fluke 343A repair

[Rax]

I got a 343A this weekend at the swap meet for a song and a dance and ran it, as I always do, for a day, continuously, to trigger any dormant for imminent issues that may it has.

It worked very well for about 24hrs, which readings reasonably close to spec in all ranges and outputs all the way to 1000V - the only issue I could see is some hesitance from the protection circuit (seemed to trigger a bit randomly for the first second or maybe a minute of running, but so little that I didn't give it much thought, especially as it didn't repeat for hours of play, and deemed it just what a unit having sat for years and maybe decades would do) - then, after running unsupervised for a while (frankly, I forgot it on while leaving the house), it seems to have triggered its protection and now it won't successfully switch from STANDBY/RESET to ON (protection would kick in).

I didn't yet get a chance to poke around too much, just did the resistance checks at 4.39 in the manual (p.4-8), and came out with this:

• TP2: 2.7k
• TP3: 8.3k
• TP5: 140k
• TP6: 2k
• TP7: 53k
• -SENSE: 2M

I don't put a lot of currency in these readings (in fact, the last makes no sense to me after looking at the schematic...?), given they were supposed to be done with an analog meter and I'm trying this with a Fluke 189. I assume the latter would interpret some of these very differently, if only given how a digital reading would fluctuate in ways that a needle would simply not see. But I will check the parts that are suspect for #2, #4, $5, and #6 anyway. Next, I'm thinking of doing the STANDBY/RESET voltage readings and hopefully there's a smoking gun somewhere. I can't do anything in ON state, as the protection system kicks in, apparently, every time I try now.

Although there's apparently some coating to the PCB, I may try a recap next, which I render mandatory given the age. I don't have any >=100uF/450V at hand, but they'll be in my next parts order. I do have to decipher why so many legs for the Sprague 68D in there...

Any thoughts and prior experience with this unit are very welcome.

[Rax]

in fact, the last makes no sense to me after looking at the schematic...?

This is because I don't see how R169 could "show up" between +SENSE and -SENSE terminals. I also measured it and it seems to be reasonably close.

I've also checked the auxiliary supply voltages and they're also reasonably close, except the HV:

• TP1: 21.4V
• TP2: 14.996V
• TP3: -14.990
• TP4: 339V
• TP5: 500V

I am tempted to just go ahead and rebuild the power supplies, including both caps and diodes.

[Rax]

Before I dive in too deep into this project, I am wondering if anyone has dealt with the coating applied to the PCB (a polyurethane, apparently - "epocast") - and what were some of the solutions used. Judging by the manual, some of the darkened spots on the PCB may not be due to heating during operation (as I've seen some you tube videos assume), but to replacing parts and the desoldering/rework involved (p.4-5).

The other point I'd be curious regarding the experience of others is whether there's some sustained failure of the rectifiers. If I go by the manual's recommendations to curing some of the issues I'm seeing, they may have to do with 6-8 rectifiers. I have heard reports that many Fluke and Tektronix units from the time have some sustained rectifier reliability issues, but not something I've experienced personally.

[Rax]

I do have to decipher why so many legs for the Sprague 68D in there...

Essentially, from a bit of probing and ("educated") guesses on the geometry of the package, it looks like the middle pin in the line of three is the positive terminal, and the rest of them (a square) are all connected to the case/negative.

[Rax]

The other thing I'm not quite sure of is whether I'm supposed to keep the OUTPUT and SENSE connected during the tests. I kept them disconnected, because I am specifically required to connect to +SENSE, for instance (not "+," or "+OUTPUT/SENSE").

[Rax]

Anyway, this far no smoking gun. I typically pulled one leg of any part I wanted to test. All the diodes I tested (those mentioned before) seem OK. From the caps I wanted to look at, one tested bad (I think it was C23), but all others fine (not great). I went ahead and replaced all of those tested.

TP4 and TP5 are still too low and the "ON" mode still trips the protection.

[Rax]

Well, as one progresses with their troubleshooting in the succession of steps in the SM, one may (fore)see a smoking gun?... In the POWER-ON TEST, at the FAULT ANALYSIS section they mention the HV fuse F2 (which I didn't realize can trigger the protection circuit if going open, but, I guess... duh!). Checking it renders it... [drumroll] open. So there's a smoking gun for yours truly. And humbly. 

I don't have any 1/16A fuses handy - nor they seem easy to source - so I guess I may have to wait for it a little bit before testing if it's the cause of all this hardship. Or, tempest in a bottle, depending on your vantage point.

[wn1fju]

I fixed a couple of these years ago, but I don't remember the details and so I didn't have any useful information to provide.

But I will say that the 343A is only one of two pieces out of the hundreds of test equipment things I have repaired that zapped me real good (the other
was the HV section of a CRT display board).  I had the 343A on and stuck my hand where it shouldn't have gone, not realizing of course that there is
about 1,000 VDC lurking inside the unit.

I don't know if you are planning to calibrate the 343A after you have fixed it as it requires accurate voltage sources, differential voltmeters, null detectors or a mix of those with a real good voltmeter (like an HP 3458A).  I found the cal procedure to be an extreme test of one's patience.

[Rax]

In waiting for the 62.5mA fuse ( ) I've replaced most of the caps I think absolutely needed that. I left the big reservoir ones in there (they measure OK), and some of the better quality Spragues.

Given about a third of the caps tested defective, no wonder the fuse popped... Once I have a fuse at hand and can establish this works again, I may go ahead and finish the recap job including what I skipped this time. Or maybe I change my mind tomorrow and do it then. The fuse order should be delivered on Thu the soonest.

[Rax]

But I will say that the 343A is only one of two pieces out of the hundreds of test equipment things I have repaired that zapped me real good (the other
was the HV section of a CRT display board).  I had the 343A on and stuck my hand where it shouldn't have gone, not realizing of course that there is
about 1,000 VDC lurking inside the unit.

I got my training years ago with exposure to about 550V from the B+ of my monoblock amps... Heck, on thinking about it, I may have done that in early childhood, testing the effect of inserting a bare copper bus into a 220V outlet... I just recall a time/space gap (I suddenly found myself not where I remembered I was just prior... ).

But more seriously, my other unit with this issue is my 332D, which uses a more careful insulation crate around the innards, which is then covered with the actual enclosure. I didn't manage to shock myself with that, but there's always another opportunity....

I don't know if you are planning to calibrate the 343A after you have fixed it as it requires accurate voltage sources, differential voltmeters, null detectors or a mix of those with a real good voltmeter (like an HP 3458A).  I found the cal procedure to be an extreme test of one's patience.

I'll reassess its condition after it's back up and running by using my Prema 6048, and my 5440A. I'll also probably by then have a couple of LTZ1000 kits back from a cal lab I plan to send them to. I got a lot of other useful things around here, such as a Keithley null meter, 732A, etc. I am not convinced the baroque procedures of those times are a must; they didn't have the luxury of 8.5 digit meters... But any attempts to adjust this though are quite a bit down the road and I'll assess things at that time. I have a lot of DCV references/calibrators, this 343A will not very likely even end up as a backup unit.

[Rax]

In undertaking the recapping job, a word of warning is paying attention to the conformal coating. Not a thick or really hard to pierce/melt polyurethane coating - I wore a mask half way through and had the door wide open for ventilation - but it can mess with the desoldering tips and require constant cleaning and retinning. I clogged one tip, which is very frustrating,  given the cost of a Pace tip.

[Rax]

Well, I wasn't aware of this, but these "small current" fuses have not negligible resistance. So my continuity test was likely not a proper test. I got replacement fuses, and both the replacements and the original fuse measure around 80 ohms... So I guess my issue is not the HV fuse having gone open. Live and learn.

So I'm back to the drawing board and head scratching. I may decide to continue my recapping job - my money's still on caps having failed during my "bake out" ("-in...?") initial run - or just take a closer look at the SM and see what my next troubleshooting steps should be.

[Rax]

I decided to start over more slowly and methodically. I've redone the 4-39 tests, and got slightly different results. One thing I did differently this time is to bridge the OUTPUT and SENSE terminals, and that answered one of my prior questions (see below).

• TP2: 2.78k
• TP3: 5.3k
• TP5: >160k
• TP6: 1.98k. Not compliant?
• TP7: 680k
• -SENSE: 57

The last on that list answered my own question: the OUTPUT-SENSE bridging is supposed to be in place for these tests. Hopefully, this helps others that may run into the same dilemma I did.

[Rax]

• TP6: 1.98k. Not compliant?

The SM calls into question CR26, CR27 and lamps DS5, 6, or 7. But I diode-tested the CRs out of circuit and the DSs are lighting when switching voltage range. The resistance from TP6 to +SENSE doesn't change at all either when operating the range switch.

[Rax]

I rechecked the voltages at the test points, and I'm not seeing any significant variations.

Due to the non-compliance of TP4 and TP5, I am becoming suspicious of SINK SUPPLY + CROWBAR DRIVER and maybe the HV POWER SUPPLY. I've diode-checked CR1, 2, 3, 4, 5, and they tested fine.

[Rax]

This unit needed its HV rails electrolytic capacitors replaced. With those replaced, the unit works without a hitch.

Interestingly, in circuit all those caps seemed to measure correctly. I guess the lesson to be learned here is to take those in circuit measurements with a grain of salt. And the SM fault finding procedures as well. But it is important to observe that the common from your meter needs to go to + SENSE, which internally is ground for this instrument.

Before I got those HV caps delivered, I've also replaced all electrolytic capacitors everywhere else. This unit is now ready for very long trouble-free service.

[Rax]

Well, that success was short lived. I had it running for quite a while with no issues, so stable that I started considering adjustments, and just now I turned it on and it seems to output 150V-200V regardless of the "voltage out" setting. Right when this issue started occurring, it's switch to the LIMIT ON mode, now it's not, but the unruly output is still there.

Back to the drawing board.

[bastl_r]

Hi
Did you changed al of the HV-caps?
In my 341, the 450V electrolytic capacitors were all almost at the end and when the 250V electrolytic capacitor C3 died, the calibrator failed.

bastl_r

[daisizhou]

You need to take some pictures and videos to let everyone know about this 343A so that everyone can help you

[Rax]

Hi
Did you changed al of the HV-caps?
In my 341, the 450V electrolytic capacitors were all almost at the end and when the 250V electrolytic capacitor C3 died, the calibrator failed.

bastl_r

I actually replaced all of the electrolytics. The job is relatively easy and having the unit open, I thought why not give it many years of future service.

[Rax]

I've not yet taken any measurements to start diagnosing this, but one thing I'm sure is this is a new, or possibly intermittent issue. Differently put, the power supply work was necessary, and it brought TP4 and TP5 within spec.

But there's more work ahead.

[bastl_r]

Hi
I am also very interested in the progress of the repair as I will soon be getting a 343 which is supposed to work.
If it works already I can also make comparative measurements for you.

Regards

[70bytes]

Hi,
my 343A stopped working after running it for 5 hours. The limit LED went on and the needle wanted to go far below 0V.
After investigating the problem i found that the 63mA fuse has blown.  I replaced the fuse with a new one and turned it on. The limit LED turned off and immediate turned on again. This time the needle wanted to go to the other site over its max limit. I measured 200V+ at the outputs in the 10V range.
The cause of this problem in my unit was a dirty standby/On switch. After cleaning the switch the problem was solved.

I few moths ago i had a problem that it wouldn't tun on. No relay clicking sound. The cause of this problem was diode CR31.

Regards,
Maurice

[Rax]

I measured 200V+ at the outputs in the 10V range.
The cause of this problem in my unit was a dirty standby/On switch. After cleaning the switch the problem was solved.

Regards,
Maurice

Thank you, Maurice. This may be exactly what I have here.

One question I have is how does the cover masking the switch come off, if you recall. There's of course two, one on top, one on the bottom. They're attached with two screws, but they won't come off if I remove the screws. Does this need the handles to also come off, or something like that? The manual seems to indicate they should just come right off.

[70bytes]

For the top one you need to pull op from the middle. It will fold a bit and come loose. It has slots on the sides.
The bottom one i think is the same.

I have a very early one. Serial# 183
I don't know if later units are build the same.

[Rax]

For the top one you need to pull op from the middle. It will fold a bit and come loose. It has slots on the sides.
The bottom one i think is the same.

I have a very early one. Serial# 183
I don't know if later units are build the same.

I wasn't clear enough in my request, sorry. I believe that, in order to access that switch - STANDBY/ON - I need to remove the cover(s) that are further to the front of the unit (see pic below). It inserts in side channels (see arrows) and it won't just come off if the screws are removed. Unless I'm missing something obvious, it looks like the handles have to at least be loosened to allow for the cover to no longer be captive to those two side channels.

[70bytes]

Ok.
I opened mine to check. No problem. Just needed to tilt. See photo.

[Rax]

Ok.
I opened mine to check. No problem. Just needed to tilt. See photo.

Interesting. Mine won't release even with some force applied (as much as I feel I can before I bend it out of shape or break it). It's captive under the side channels, and it's pretty clear to me it's not intended to just come off by tilting. It may be a later redesign or something like that.

Thank you very much for going into your unit and trying this out.

[Rax]

Well, stop press... I pushed my luck and tried harder and it very reluctantly came off. I think there was some possibly conformal residue that was petrified in place holding it in, and it was also probably never removed since coming out of the factory. It looks like it needs to slide in a bit, so I may need to remove that residue to make room on that narrow transversal channel it needs to lock onto.

I just conditioned the STANDBY/ON switch with some Deoxit D5 - being the exact same symptom as in your case, 70bytes - and I'll let it cure for a few minutes and retest... crossing fingers.

[Rax]

The limit LED turned off and immediate turned on again. This time the needle wanted to go to the other site over its max limit. I measured 200V+ at the outputs in the 10V range.
The cause of this problem in my unit was a dirty standby/On switch. After cleaning the switch the problem was solved.

Regards,
Maurice

Maurice - that was just spot on. Brilliant, and thank you. Once I conditioned that switch, the unit seems to be working again with no hitch. I should probably say "we'll see," as it seems these may be a bit cranky until several issues (probably all related to age) are resolved.

[70bytes]

Nice to hear.
I replaced all the electrolytic capacitors. Most of them measured still fine. The oldest component i could find was from week 18 1968.
2 of the big capacitors were bad. 1 was replaced earlier. I read somewhere that the sprague capacitors mostly are bad.

 

[bastl_r]

Hi
Unfortunately I don't look in here every day...
I could have told you about the flap.
There is a 1Ohm resistor behind the lower flap which burns out if the connections are short-circuited at a high output voltage or if the overvoltage arrester of a measuring device responds. Then the protective circuit is activated immediately when the power is switched on. Please don't ask how I know this.

I have now also received my "new" Fluke343. As it looks inside, it is still in its original and very clean condition. So it still has the original capacitors.

Unfortunately, of course, it has a small problem...
The zero points are shifted. However, the deviation matches the range.
7.998mV in the 10V; 79.97mV in the 100V; and ~800mV in the 1000V range so I suspect a common error. The offset then also runs through all stages with the same deviation.
The respective settings seem to fit exactly so that I don't want to adjust anything for the time being, although I'm sure from my 341 that I couldn't adjust that much with the zero adjustment.
Should I start a new thread for this? I think so. Or what do you say?

@Maurice: How did you cleaned the switch contacts?

bastl_r

[coromonadalix]

deoxit should clean correctly  the switches,  and do verify the shafts  extenders  they crack  and become loose

[Rax]

How did you cleaned the switch contacts?

bastl_r

I also used Deoxit (D5), though I'd recommend holding off from using it on any of the output adjustment switches. For those, I'd just use IPA to clean them, and if I'd conclude they need conditioning (mine don't seem to), I'd very parsimoniously apply a bit of Deoxit on the copper contact surfaces only. I'd be very careful not to spill any on the substrate of the switch. But I'd leave them alone if they operate OK.

Another contact conditioner I came to like is MG Chemicals 801B. Very expensive, but seems to work very well. It helped me bring back to life a few DM501As that otherwise have been very reluctant to work reliably.

[bastl_r]

Sometime, it is a good recommendation to read the ****ng manual! 
On page 35 ist the solution for my problem, i think.

[Vertamps]

Hello, i had made these Fluke 343A high voltage capacitor mod boards (a few months ago). Iv intended to use solid copper wire to solder the mod capacitor to the the Fluke PCB, either bend the leads or use cheap  spacers i bought off Amazon. I will solder one in this week to show you it installed and see if i need a minor tweak to the pcb. I did think of adding holes for a few plastic snap in spacers as there is room on the board, standing off is doable a few ways. I can put the Gerber files up on my drop box so you can order your own, or PM and i can email but wait till my install test. I paid $8.16 total for 20 of these boards shipped to me.

[bastl_r]

Hi
I am very interestet on the gerberfiles.
Can you send me a link for Download?

Thank you

[daisizhou]

Hello, i had made these Fluke 343A high voltage capacitor mod boards (a few months ago). Iv intended to use solid copper wire to solder the mod capacitor to the the Fluke PCB, either bend the leads or use cheap  spacers i bought off Amazon. I will solder one in this week to show you it installed and see if i need a minor tweak to the pcb. I did think of adding holes for a few plastic snap in spacers as there is room on the board, standing off is doable a few ways. I can put the Gerber files up on my drop box so you can order your own, or PM and i can email but wait till my install test. I paid $8.16 total for 20 of these boards shipped to me.

You choose to replace the original capacitor with a smaller capacitor.
Although their capacitance and withstand voltage meet the requirements, large capacitors have high power, while small capacitors usually have low power.
In other words, the redundancy performance has become weaker

[Vertamps]

You choose to replace the original capacitor with a smaller capacitor.
Although their capacitance and withstand voltage meet the requirements, large capacitors have high power, while small capacitors usually have low power.
In other words, the redundancy performance has become weaker

Well, i suppose if you wanted to pick out 3 skinny/tall capacitors you could parallel them for lower ESR than the original capacitor and fit them on the pcb size.
But I went with a single United Chemi-Con and assuming it will be ok. With 3 filter caps i would add thermal relief on pads for quicker install. I think my initial design will have
good contact area with the capacitor leads.

[Vertamps]

Hi
I am very interestet on the gerberfiles.
Can you send me a link for Download?

Thank you

Hello, let me install one first so im not sending out files with an issue. Ill try to tonight after work (sorry wait one more day ). Worse case my spacing measurements are off by a mm. Mating hole size board to board are 1.2mm to fit 1mm/18 gauge solid
core copper wire.

[Rax]

You choose to replace the original capacitor with a smaller capacitor.
Although their capacitance and withstand voltage meet the requirements, large capacitors have high power, while small capacitors usually have low power.
In other words, the redundancy performance has become weaker

If you mean smaller physically, things are nowhere near that simple. Manufacturing of electrolytics capacitors has come a very long way since the late '60s, and using 105C 5k+ hrs life capacitors of equal or higher capacitance and voltage from a reputable manufacturer should provide a very high reliability from the unit (in addition to a much smaller size). I am using 12k duty caps for my recapping and am not concerned with their reliability. I also don't think ripple current handling in this application is very demanding.

[Rax]

Hello, let me install one first so im not sending out files with an issue. Ill try to tonight after work (sorry wait one more day ). Worse case my spacing measurements are off by a mm. Mating hole size board to board are 1.2mm to fit 1mm/18 gauge solid
core copper wire.

Assuming you get the spacing right, what do you plan to use for the five leads going to the unit's board? Is that the 18 gauge solid core wire? If yes, I wonder if you've figured a convenient way to attach them to the adapter PCB for soldering. Unless you're using some PCB mount header pin type that snaps onto the board or something similar, I assume keeping attached for soldering five terminals to each of the boards makes the recapping very laborious.

[Vertamps]

Hello, let me install one first so im not sending out files with an issue. Ill try to tonight after work (sorry wait one more day ). Worse case my spacing measurements are off by a mm. Mating hole size board to board are 1.2mm to fit 1mm/18 gauge solid
core copper wire.

Assuming you get the spacing right, what do you plan to use for the five leads going to the unit's board? Is that the 18 gauge solid core wire? If yes, I wonder if you've figured a convenient way to attach them to the adapter PCB for soldering. Unless you're using some PCB mount header pin type that snaps onto the board or something similar, I assume keeping attached for soldering five terminals to each of the boards makes the recapping very laborious.

The solder holes on the fluke are larger to accompany the original caps, if i remember not all of the ground lugs are connected to ground so its universal to keep all the mounting holes filled or not. For a one and done project the extra time to install is fine, i will try 1mm headers and show if it works. After thinking about it I will also make a spacer pcb file with no copper layer, just the drill holes. Could just remove the layer files and reorder but it won't take long, ill add some overlay info. You could stack the pcb standoffs and solder the mounting leads if using solid wire to keep em more straight. Copper is flexible and im sure there's a way to pinch them in or just clip extra lead length after install. Ill really try to reply tonight with an update, maybe quick video. I found some 3mm adhesive trips at work im gonna use as a spacer for now.

[Rax]

The solder holes on the fluke are larger to accompany the original caps, if i remember not all of the ground lugs are connected to ground so its universal to keep all the mounting holes filled or not.

That's right - see pic below - and to me using a bit of insulated wire to connect the ones needing it is fine. What's a bit more messy is cleaning the board after soldering, given the conformal coating which gets charred by the soldering iron. But I feel the work is good and will be long lasting.

[bastl_r]

Well, i suppose if you wanted to pick out 3 skinny/tall capacitors you could parallel them for lower ESR than the original capacitor and fit them on the pcb size.

The ESR will only play a subordinate role here. I therefore see no need to use special electrolytic capacitors. At most some with a view to a longer service life.

bastl_r

[Vertamps]

*New PCB links Below

[Vertamps]

Here i have uploaded pcb mod files of your choosing. Both 1mm and 1.2mm mating holes, a Wide hole cutout spacer that clears all solder pads, one tight fit spacer board with 1.2mm holes.
I am fine with sticky pad spacers i used in trial, however i will tack on each of my boards for my next order. If using JLC PCB just drop the selected zip file, select 10 board qty and your good to go.
Let me know of any issue with the link. -William
https://www.dropbox.com/scl/fo/nz4almtqw9p33j7ymv4mh/ADGSwEl5e5UdivuPPLuvtyQ?rlkey=6nn4h3je2gnniqhheckspqx6j&st=61lgvxhj&dl=0

[bastl_r]

Oh man  "head --> table"

The zero points are shifted. However, the deviation matches the range.
7.998mV in the 10V; 79.97mV in the 100V; and ~800mV in the 1000V range so I suspect a common error. The offset then also runs through all stages with the same deviation.

Wen i was looking for a video disassembling the switches i found the repair of defpom where he repaired a switch shaft.
Later in bed I remembered what else can happen with a broken switch shaft 
And yes the shaft of the 4th digit ist totally destroyed.
Unfortunately, the latching mechanism is firmly screwed to the device housing and suggests a correctly functioning switch, whereby the contacts set to 8 do not move any further...
Defpom has uploaded a print file for a new shaft piece to thingiverse, but unfortunately it doesn't quite work with our printer. But that's not a problem. At least the error has been found.
I will now replace all the shafts because they are all cracked
Now I'll solder the old UA709 and the old Q39 back in and see if I can manage without recalibration. I didn't have to replace the electrolytic capacitors I've replaced so far. They were all still within the specified capacitance range, even though some of them were a bit high-impedance.

Regards

[TheDefpom]

Oh man  "head --> table"

The zero points are shifted. However, the deviation matches the range.
7.998mV in the 10V; 79.97mV in the 100V; and ~800mV in the 1000V range so I suspect a common error. The offset then also runs through all stages with the same deviation.

Wen i was looking for a video disassembling the switches i found the repair of defpom where he repaired a switch shaft.
Later in bed I remembered what else can happen with a broken switch shaft 
And yes the shaft of the 4th digit ist totally destroyed.
Unfortunately, the latching mechanism is firmly screwed to the device housing and suggests a correctly functioning switch, whereby the contacts set to 8 do not move any further...
Defpom has uploaded a print file for a new shaft piece to thingiverse, but unfortunately it doesn't quite work with our printer. But that's not a problem. At least the error has been found.
I will now replace all the shafts because they are all cracked
Now I'll solder the old UA709 and the old Q39 back in and see if I can manage without recalibration. I didn't have to replace the electrolytic capacitors I've replaced so far. They were all still within the specified capacitance range, even though some of them were a bit high-impedance.

Regards

The downloadable file is an STL, which you can drag into Cura for example and have it use your normal print settings and chosen material configuration, so you should be able to print it out like anything else you normally do, I do recommend something like PLA+ rather than just PLA as it will be stronger, unless you have something else that performs well.

[bastl_r]

Hi
We had to adjust the dimensions a little so that the extensions fitted and printed in PLA.
Then you have to make sure that you plug the switches together in the correct position. Otherwise the calibrator won't go to zero either.

I also had to replace Q38 because the zero did not remain stable and drifted away over a longer period of time. I installed a BSV80 instead.
Now I still have to adjust the device completely because the previous owner had probably adjusted deck A and deck B but had not adjusted the ranges on deck C beforehand. 
At X on deck C 99.989mV and 1 on deck B 100.02mV are output...

[bastl_r]

Hello
How much drift of the 15V reference voltage can realistically be expected during a few hours on a Fluke343 calibrator?
The temperature is approx. 25°C +/-1°.
Measurements are made with a K2002.

Greetings bastl_r

[bastl_r]

Renewed question about the drift of the reference voltage of my 343::
Is it normal that I find a difference of about 20µV on the min/max display after about 8 - 9 hours with the K2002?
So 14.999682xV and 14.999665xV.
This means a drift of approx. 1.1ppm. It feels like 20µV is too much, but 1.3ppm is still quite low for such a device.
Or is it?

regards bastl_r

[Rax]

Sorry, bastl_r, I no longer have one at my bench. Hopefully those that do will assess your condition.

[bastl_r]

Thank you for your answer.
I realise that the stability of any calibrator is finite and I am aware of the efforts being made to constantly increase it. However, the 20uV over 8 to 9 hours spontaneously seems a bit much to me.
But I would still be happy if I am wrong.

Greetings bastl_r

[unseenninja]

Thank you for your answer.
I realise that the stability of any calibrator is finite and I am aware of the efforts being made to constantly increase it. However, the 20uV over 8 to 9 hours spontaneously seems a bit much to me.
But I would still be happy if I am wrong.

Greetings bastl_r

I'm wondering if I'm seeing a similar problem which my 343A has developed. When I first received it, the calibration was poor, but the calibrator seemed stable, so I performed the adjustment procedure using my recently adjusted and calibrated Solartron 7061. After that, everything seemed fine and I didn't use the 343A for some time.

Recently, I became the owner of a Datron 1271. As part of checking its performance, I started a log of 10V from the 343A. To my horror, what used to be 10.00000V from the 343A was now more like 10.001V and is falling by a rate of 4ppm every 24 hours. I logged a whole week before I got bored of waiting for the curve to level out.



The specifications for the 343A are very conservative, but even worst case, this is outside of specifications. I have not yet had the time to investigate what the problem could be and have been following this thread with interest in your progress.

[bastl_r]

Thankyou for your response.
Your problem is an other than mine, i think.
Your log was on the outputvoltage.
My measurment was directly on the referencespannung(TP2). The voltage at this point oscillates over long periods of time by around 20uV around the mean value of approx. 14.999670V.
Your Problem, i think, is the chopperamp. Maybe the ageing capacitors in it.
Regards bast_r