Author Topic: Fluke 8840A stuck in measurement  (Read 7601 times)

0 Members and 1 Guest are viewing this topic.

Online ulixTopic starter

  • Regular Contributor
  • *
  • Posts: 118
  • Country: de
Fluke 8840A stuck in measurement
« on: July 23, 2014, 05:43:03 pm »
Hello Guys,

I bought last week a Fluke 8840A-AF on ebay. The Problem is that the meter is stuck in a measurement. I can't change anything by switching into A, V or Ohms as well as making the self test. When I opened the lid, I discovered a inflated cap, it's a 6800mF cap (C601), the first for the 7805. But this didn't change the problem, I'm at this point I can't find a error.
Can anyone give some tipps?
cheers
« Last Edit: July 23, 2014, 05:45:07 pm by ulix »
 

Online ulixTopic starter

  • Regular Contributor
  • *
  • Posts: 118
  • Country: de
Re: Fluke 8840A stuck in measurement
« Reply #1 on: July 24, 2014, 09:25:01 pm »
is there no one which has any idea?
cheers
 

Offline xwarp

  • Frequent Contributor
  • **
  • Posts: 367
  • Country: us
Re: Fluke 8840A stuck in measurement
« Reply #2 on: July 24, 2014, 10:02:21 pm »
So what does the meter do upon first power up?

Also, make sure the switch for rear inputs is OUT.
 

Online ulixTopic starter

  • Regular Contributor
  • *
  • Posts: 118
  • Country: de
Re: Fluke 8840A stuck in measurement
« Reply #3 on: July 25, 2014, 12:25:08 pm »
Hi,
here is the start up.



Rear side:
cheers
« Last Edit: July 25, 2014, 12:35:05 pm by ulix »
 

Offline ManateeMafia

  • Frequent Contributor
  • **
  • Posts: 731
  • Country: us
Re: Fluke 8840A stuck in measurement
« Reply #4 on: July 25, 2014, 02:39:02 pm »
I have seen something similar to this with a 3458a on the AC range. It was a bad op-amp. It wasn't stuck it just had a constant output. I am not familiar with this model but I hope you can trace through the ranging and A/D sections.
 

Offline xwarp

  • Frequent Contributor
  • **
  • Posts: 367
  • Country: us
Re: Fluke 8840A stuck in measurement
« Reply #5 on: July 26, 2014, 02:27:01 am »
Hold the "SRQ" button down and see if it runs through the self tests. If it faults, you will know it.

Looks to me, and this is just a guess, that the calibration is bad.
 

Online ulixTopic starter

  • Regular Contributor
  • *
  • Posts: 118
  • Country: de
Re: Fluke 8840A stuck in measurement
« Reply #6 on: July 26, 2014, 09:18:51 am »
Self test runs, and each test fails!!! From 1 to 21 eache with ERROR :-//
 

Offline xwarp

  • Frequent Contributor
  • **
  • Posts: 367
  • Country: us
Re: Fluke 8840A stuck in measurement
« Reply #7 on: July 26, 2014, 01:27:12 pm »
Do you have the service manual?

If you have it, starting on page 138 of the pdf, it talks about the calibration of the A/D converter and the precision voltage reference.

It appears that you may be having an issue with one of the two, and probably the reference being that ALL tests fail.

« Last Edit: July 26, 2014, 01:36:30 pm by xwarp »
 

Online ulixTopic starter

  • Regular Contributor
  • *
  • Posts: 118
  • Country: de
Re: Fluke 8840A stuck in measurement
« Reply #8 on: July 27, 2014, 09:29:29 am »
I have a printed Manual. Which page do you mean, 6-33?
 

Offline xwarp

  • Frequent Contributor
  • **
  • Posts: 367
  • Country: us
Re: Fluke 8840A stuck in measurement
« Reply #9 on: July 27, 2014, 03:29:45 pm »
Sorry,

page 6-29.

I must have been looking at the service manual for the 8842.
 

Online ulixTopic starter

  • Regular Contributor
  • *
  • Posts: 118
  • Country: de
Re: Fluke 8840A stuck in measurement
« Reply #10 on: January 13, 2025, 06:49:31 pm »
Hi Everyone,

sorry for digging up this thread. At the time I didn't get the problem solved and the instrument still sits here. Hope someone can lend a helping hand to solve the problem.

At the time I remember I had found something odd but did not pursue this further at an OPAMP. Currently I'm trying to solve the problem and found some thing odd located at the prescision voltage reference. At the Dual Opamp U702 (LF412). The first image summarizes the mearuemnts on the U702 pins.

I'm wondering, shouldn't I measure foltages on Pins 2,3,5 and 6, I find it odd that I cannot meaure anything with my multimeter.

Second image is a section of the mentioned schematic.

Any help is much appreciated. thanks
« Last Edit: January 13, 2025, 06:51:11 pm by ulix »
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8397
  • Country: us
Re: Fluke 8840A stuck in measurement
« Reply #11 on: January 13, 2025, 06:59:34 pm »
First question--do you have the manual for the 8840A/AF or the 8840A?

Second--can you take off the covers, completely remove the AC and GPIB boards and then take a clear photo of the main PCB?  I have several 8840A but not an A/AF, so if I follow along and post voltages at test points and so forth it would be handy to have the photo and correct manual to reference.

Third--does shorting the inputs make any difference in the numbers displayed on the display?

Last, I don't remember the specifics of the #1 vs #2 commons on these, but if you have the negative lead of your DMM on something that is connected to the #1 common then you should measure about 0V on those pins because two of them are connected directly to #1 and the others are the opposite pin of an op amp.  If the op amp is operating within its normal range its basic function is to make those two input leads equal.  If your +7V and -7V test points are correct--or close to it--then your problem is not in this area.
« Last Edit: January 13, 2025, 07:12:52 pm by bdunham7 »
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Online ulixTopic starter

  • Regular Contributor
  • *
  • Posts: 118
  • Country: de
Re: Fluke 8840A stuck in measurement
« Reply #12 on: January 13, 2025, 07:38:14 pm »
Hi bdunham7,

thanks a lot for your reply.

1) Yes, I have the manual for the 8840A.
2) Please see attatchments. If more are needed please let me know. (Not sure if there is a difference between the 8840A and 8840A/AF).

For measureing the voltages on U702, I had disconnected the GPIO board, the AC Board was still connected. Can I power the device without the AC board savely?
thanks
 

Online ulixTopic starter

  • Regular Contributor
  • *
  • Posts: 118
  • Country: de
Re: Fluke 8840A stuck in measurement
« Reply #13 on: January 13, 2025, 07:40:19 pm »
Place holder images.
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8397
  • Country: us
Re: Fluke 8840A stuck in measurement
« Reply #14 on: January 13, 2025, 08:37:12 pm »
For measureing the voltages on U702, I had disconnected the GPIO board, the AC Board was still connected. Can I power the device without the AC board savely?

I was asking whether you had the 8840A or the 8840A/AF manual as they are different instruments to a degree.  AFAIK the 8840A/AF manual isn't available so everyone makes do with what they have, but if you or someone has an actual 8840A/AF manual that would be even better.

Yes, you can and should power it up with the AC board completely removed.

If you could, back up a step and perform some basic checks just so I know we're on the same page.  Check the power supply voltages, TP601-607 (refer to the PSU schematic page on the manual) and post the results.  Also, after you power it up select ACV and AC current ranges and see if you get an Error 30 or if something else happens.

I have an old 8840A I can disassemble and poke at for you, but it is in storage waiting for me to get to it--it's on the "someday list".  It will be a few days before I will retrieve it but when I do I can do some comparisons for you as you probe your way through the circuits.

Also, what other equipment (DMMs especially) do you have available?   
« Last Edit: January 13, 2025, 08:39:03 pm by bdunham7 »
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Online ulixTopic starter

  • Regular Contributor
  • *
  • Posts: 118
  • Country: de
Re: Fluke 8840A stuck in measurement
« Reply #15 on: January 13, 2025, 09:28:02 pm »
As I understand it, I have the 8840A manual, not the 8840A/AF. However, I’m not entirely certain, as the manual does not specify which model it pertains to (I couldn’t find a reference to the model, and the front page is missing).

1) Test Points (TP601 to TP607):
These test points are not labeled on my board. However, I found test points that reference voltages matching those specified for TP601–TP607. Based on this, I’m using these points for testing. Please see the attached image for the measured voltages.

2) Shorting the Inputs:
Does this make any difference? No, it doesn’t appear to have any effect.

3) Current Ranges:
Pressing the mA DC (second-to-last white button at the top) twice results in no number being displayed. (Only blinking S at the top of the display). Pressing mA AC (the last white button at the top), the display shows 30 (error).

Available equipment:
- Digital oscilloscopes (100 MHz) and analog
- Measurement was done with a Gossen Metrawatt Metraline DM62. I additionally have stowed away a Metrahit World, Metrahit Xtra, and a Fluke 87, as well as a Fluke current clamp meter.
- DC power supplies and frequency generator (up to 100s of kHz, I think, stowed away as well)


« Last Edit: January 13, 2025, 09:39:56 pm by ulix »
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8397
  • Country: us
Re: Fluke 8840A stuck in measurement
« Reply #16 on: January 13, 2025, 09:46:45 pm »
OK, one more set of tests if you can before I have to run off and do something.

Set up your other DMM to measure DC milliamps if you can and connect it to the INPUT leads (not SENSE) on the 8840A/AF.  Press the "kΩ-2W" button and then the "2" button and read the mA shown on the other meter.  It should be about 1mA, if it isn't then try working the Front/Rear switch repeatedly and see if you can get a reading.  After that, set your DMM up to measure volts again and set the 8840A/AF to the DCV function and press the "20" button, insuring the the "AUTO" indicatior is off.  Check and post the voltages at TP301-303 as shown on the DC Scaling/Input schematic page.  Also check and post the voltages at TP701-702 on the precision reference schematic you were working on earlier.

A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Online ulixTopic starter

  • Regular Contributor
  • *
  • Posts: 118
  • Country: de
Re: Fluke 8840A stuck in measurement
« Reply #17 on: January 13, 2025, 09:57:35 pm »
Just tried the first step, not getting any current only 3uA reading at Input for "kΩ-2W" and "2". Yet when removing the multimeter, the display changes to "1" and OVER on the 8840A/AF.

Could you let me know what you mean with working the front/rear switch? Do you mean the on/off switch?
« Last Edit: January 13, 2025, 10:05:09 pm by ulix »
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8397
  • Country: us
Re: Fluke 8840A stuck in measurement
« Reply #18 on: January 13, 2025, 10:24:24 pm »
The front/rear switch is the input selection switch near the front input jacks.  It should be in the out position.  Sometimes the contacts get funky and excercising the switch a few dozen times will clear it.

What did the 8840A/AF read in 2WΩ before you disconnected the other meter?

Regardless of how that turns out, go ahead and continue with the rest of the tests I suggested.  I forgot a step in that--when you set it to the 20VDC range, please short the input jacks as well.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Online ulixTopic starter

  • Regular Contributor
  • *
  • Posts: 118
  • Country: de
Re: Fluke 8840A stuck in measurement
« Reply #19 on: January 15, 2025, 08:47:57 pm »
1) I have tested the front switch previously. There is no change now either.

2) What happens when the input is shorted in 2WΩ mode? Just tried it again with a bridge, it reads 4.3kOhm and "1"+Over when bridge removed.

Please find the voltages of TP301–TP303 and TP701–TP702 listed below.

One odd observation: the capacitor C309 shows a short between its two pins.
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8397
  • Country: us
Re: Fluke 8840A stuck in measurement
« Reply #20 on: January 16, 2025, 02:22:44 am »
1) I have tested the front switch previously. There is no change now either.

2) What happens when the input is shorted in 2WΩ mode? Just tried it again with a bridge, it reads 4.3kOhm and "1"+Over when bridge removed.

Please find the voltages of TP301–TP303 and TP701–TP702 listed below.

One odd observation: the capacitor C309 shows a short between its two pins.

If you haven't already, try setting the switch to rear and using the rear inputs instead of the front.

A high resistance reading when shorted is usually high resistance in the input circuit somewhere.  So check R309 and then measure from R309 all the way to the positive input connector.

Those voltages look good and you can rule out problems with most of the input and reference circuitry, at least as far as causing the major problems and self-test failures.  One more thing--when the 20VDC range  is shorted and you read ~0V at TP 302, what does the meter's display show?

How did you come across the short on C309?  Did you test it out-of-circuit?   A short would cause the final stage of the T&H amp would go to unity gain instead of open loop, but IDK what the overall effect would be.  Maybe the T&H amp is oscillating?  If C309 is truly shorted, you can probably just remove it and put an oscilloscope on it to see what it is doing. 
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Online ulixTopic starter

  • Regular Contributor
  • *
  • Posts: 118
  • Country: de
Re: Fluke 8840A stuck in measurement
« Reply #21 on: January 16, 2025, 08:56:18 pm »
If you haven't already, try setting the switch to rear and using the rear inputs instead of the front.
=> Switching to the rear does indicate other values. Yet these values seem to be "stuck" as well.

A high resistance reading when shorted is usually high resistance in the input circuit somewhere.  So check R309 and then measure from R309 all the way to the positive input connector.
=> I'm intimidated by the sheer schematic. I have indicated the values measured. Please let me know if this is okay. All values where measured in the circuit, and the fluke powered down + inputs shorted.

One more thing--when the 20VDC range  is shorted and you read ~0V at TP 302, what does the meter's display show?
=> (for the front) -4.3192V (for V-DC); -430.89 mA DC (for mA-DC)
=> (for rear) it shows +0.9999 DC (yes the "V" is missing)

How did you come across the short on C309?  Did you test it out-of-circuit?   A short would cause the final stage of the T&H amp would go to unity gain instead of open loop, but IDK what the overall effect would be.  Maybe the T&H amp is oscillating?  If C309 is truly shorted, you can probably just remove it and put an oscilloscope on it to see what it is doing.
=> Just check this again in the circuit and not able to measure this short again (when powered down and powered up) - it was definitly there when I noticed it.
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8397
  • Country: us
Re: Fluke 8840A stuck in measurement
« Reply #22 on: January 16, 2025, 10:07:48 pm »
=> (for the front) -4.3192V (for V-DC); -430.89 mA DC (for mA-DC)
=> (for rear) it shows +0.9999 DC (yes the "V" is missing)

I'm sorry--I wasn't thinking too clearly.  Your A/AF uses the rear for ratio, not regular inputs.  The display and missing "V" are completely normal.  Your measurements look normal.  Barring something extra-funky, the input circuitry is not the issue, at least not the immediate issue.

We're getting closer to the problem.  Set the meter to the 20VDC range and short the inputs.  If it still reads -4.3V or so, then measure the voltage at TP103 on the A/D converter schematic or if you can't find that, at the R318/C311 junction near U307.  If you see about -4.3V then your problem is to the left with the T&H amplifer; if you see about 0V then your problem is to the right in the A/D converter.  If the former, re-verify that you get about 0V at TP302. 
« Last Edit: January 16, 2025, 10:14:40 pm by bdunham7 »
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Online ulixTopic starter

  • Regular Contributor
  • *
  • Posts: 118
  • Country: de
Re: Fluke 8840A stuck in measurement
« Reply #23 on: January 17, 2025, 08:16:09 pm »
I'm sorry--I wasn't thinking too clearly.  Your A/AF uses the rear for ratio, not regular inputs.  The display and missing "V" are completely normal.  Your measurements look normal.  Barring something extra-funky, the input circuitry is not the issue, at least not the immediate issue.
No worries, I'm extremly appreciative for helping me!

We're getting closer to the problem.  Set the meter to the 20VDC range and short the inputs.  If it still reads -4.3V or so, then measure the voltage at TP103 on the A/D converter schematic or if you can't find that, at the R318/C311 junction near U307.  If you see about -4.3V then your problem is to the left with the T&H amplifer; if you see about 0V then your problem is to the right in the A/D converter.  If the former, re-verify that you get about 0V at TP302.
=> I measured at TP103 it is 0V (to reference GND1) or -1.120V (to reference GND2).

I guess this means that the fault seems to be at the A/D converter?
« Last Edit: January 17, 2025, 10:09:45 pm by ulix »
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8397
  • Country: us
Re: Fluke 8840A stuck in measurement
« Reply #24 on: January 17, 2025, 10:08:29 pm »
=> I measure at TP103 either 0V (to reference GND1) or -1.120V (to reference GND2).

I guess this means that the fault seems to be at the A/D converter?

Yes, I think so.  Additionally, the service manual describes desoldering and lifting one end of R318 (at the junction of R318 and U307) and then shorting TP103 to REF LOW (GND 1), this should result in a near-zero reading on the meter.  You can then also put a test voltage of no more than +/- 2V between TP103 and REF LOW (make sure R318 is disconnected) and the display should be the inverse of this voltage in the 2V range.  I'm pretty sure the result will be the same, but it might be interesting to see if anything happens when you disconnect R318.  I'm not sure there is supposed to be that much of a discrepancy between GND 1 and GND 2, so that might be a clue.  There might be a fault current running somewhere in the system.  I'll probably retrieve my spare meter today or tommorrow morning and I'll check this.  It's been a while since I poked at one of these and I'm not sure how GND 1 and GND 2 are interconnected.  I don't see anything about it in the manual, but if someone knows perhaps they'll chime in.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf