fluke 83 main asic

[diegoperez]

Hi everybody,

After some troubleshooting (whith help of the service manual) of a fluke 83 which measures properly in some ranges mVDC, ohms... but does not in others like VAC or diode,  I came to the the conclusions that there is a faulty main fluke asic, so called u4.

Is it possible to find a replacement for this part somewhere? what would be advisable to do in this situation?

Thanks for the attention, any advice will be appreciated

[armandas]

From my experience trying to source an ASIC for Fluke 27, Fluke will not sell you these parts. eBay/Aliexpress is probably your best bet.

[retiredcaps]

After some troubleshooting (whith help of the service manual) of a fluke 83 which measures properly in some ranges mVDC, ohms... but does not in others like VAC or diode,

I might have some suggestions based on your test results above.

1) When you measure an AA cell, do you get roughly 1.5V DC?
2) When measuring a 1k ohm resistor, do you get 1k ohm reading?
3) When you measure a diode, what do you see on the lcd screen?
4) When you measure mains AC, what do you see on the lcd screen?
5) Was this meter working before?
6) Are you the original owner of the Fluke 83?
7) Did the 83 experience an "oops" moment recently?  That is, high energy?

[diegoperez]

After some troubleshooting (whith help of the service manual) of a fluke 83 which measures properly in some ranges mVDC, ohms... but does not in others like VAC or diode,

I might have some suggestions based on your test results above.

1) When you measure an AA cell, do you get roughly 1.5V DC?
2) When measuring a 1k ohm resistor, do you get 1k ohm reading?
3) When you measure a diode, what do you see on the lcd screen?
4) When you measure mains AC, what do you see on the lcd screen?
5) Was this meter working before?
6) Are you the original owner of the Fluke 83?
7) Did the 83 experience an "oops" moment recently?  That is, high energy?

1) No, VDC has 352V approx which does not goes to zero when shorting probes. However, when measurinc a AA cell it does adds the voltage and displays 353~354V approx.
2) Yes, Ohms range works fine.
3) OL always, however probes do output 3.75V in Open Loop and source  0.54mA when shorted.
4) When in  VAC range it oscillates between OL and a value  with every display update. Measuring mains voltage displays 50VAC steady (should be 230V here in Spain)
5) 6) Don't know, I bought it on ebay with the purpose explained here  https://www.eevblog.com/forum/repair/advice-on-gaining-troubleshooting-skills/msg733107/#msg733107
 and it looks like the eevblog curse has struke me too  (I bought an old HP PSU for parts and it only had a blown fuse, and know this meter looks like it is unrepairable, too easy or too hard )
7) I didn't (as explained above) but I don't see any traces of smoke on the PCB or the components. All pretty clean.

As you may notice I have no experience on troubleshooting so all of your recommendations will be very welcomed.

PS, I did notice that Vdd measures 3.2V and not 3.0V as indicated on the service manual (p. 49) http://assets.fluke.com/manuals/83_85_87smeng0500.pdf however I also notice that R24 is 324K instead of 210K therefore it might be a  red herring, and the different voltage rail is due to a different board revision I guess??

[retiredcaps]

Let me digest your readings.  I also have to head out for a while so I probably won't have any suggestions until later tonight?

In the meantime, can you post clear focused pictures of the 83 pcb (top and bottom side with shield and lcd removed)?

Can you also link your auction so we can see the original description and see if the seller has any experience in selling test equipment?

[retiredcaps]

Just one quick test.  Can you measure the input impedance of the Fluke 83 when it is set to DCV mode?  It should be around 11.11M ohm when measured with another meter.

[krivx]

I have a parts Fluke 87 if you are interested. I don't know for sure if the main IC works...

http://imgur.com/a/JGNQC

[ModemHead]

The 3.2V Vdd measurement is OK.

The trouble-shooting questions are important, but before getting too deep into it, clean the PCB with strong (>90%) isopropyl alcohol first.  DMMs are very sensitive to PCB contamination.  It can cause all manner of strange symptoms.

You'll have to remove the LCD assembly, so be very careful not to break the little tabs and clips in the process.  The first few disassembly steps in this procedure may help.

Look for any foreign matter on the board, like the stuff in between the ASIC pins shown in the first photo below.  If you see that, use a toothpick or something to drag it out.  Then give the board a bath in IPA, using a brush to work in between components and pins.  Allow to dry.

If you see an improvement in the meter's function afterwards but it still isn't quite right, sometimes it helps to repeat the cleaning procedure.

[diegoperez]

Let me digest your readings.  I also have to head out for a while so I probably won't have any suggestions until later tonight?

In the meantime, can you post clear focused pictures of the 83 pcb (top and bottom side with shield and lcd removed)?

Can you also link your auction so we can see the original description and see if the seller has any experience in selling test equipment?

Here are three pictures of the DMM as it came.

And this was the ebay auction http://www.ebay.com/itm/252065578180

Just one quick test.  Can you measure the input impedance of the Fluke 83 when it is set to DCV mode?  It should be around 11.11M ohm when measured with another meter.

It measures Open Circuit in the tree ranges ACV DCV and DCmV

The trouble-shooting questions are important, but before getting too deep into it, clean the PCB with strong (>90%) isopropyl alcohol first.  DMMs are very sensitive to PCB contamination.  It can cause all manner of strange symptoms.

You'll have to remove the LCD assembly, so be very careful not to break the little tabs and clips in the process.  The first few disassembly steps in this procedure may help.

Look for any foreign matter on the board, like the stuff in between the ASIC pins shown in the first photo below.  If you see that, use a toothpick or something to drag it out.  Then give the board a bath in IPA, using a brush to work in between components and pins.  Allow to dry.

If you see an improvement in the meter's function afterwards but it still isn't quite right, sometimes it helps to repeat the cleaning procedure.

It looks like someone has opened it before. Two of the  little tabs that you mention were already broken  . The rotary switch looks pretty greasy but unfortunatlly at the moment I don't have IPA to give a thoughtful bath. I tried to clean the contacts rubbing them carefully with an ear cotton bud to no avail. Actually, now the Ohms range doesn´t work either and gives wrong measures even when probes are open.

[retiredcaps]

Just one quick test.  Can you measure the input impedance of the Fluke 83 when it is set to DCV mode?  It should be around 11.11M ohm when measured with another meter.

It measures Open Circuit in the tree ranges ACV DCV and DCmV

Just to confirm.

1) The Fluke 83 was set to DCV and the meter was turned on (9V battery inserted and working inside the Fluke 83).
2) What was the other meter used to measure the input impedance of the Fluke 83?

While you have meter open and waiting to get some IPA, let's measure the input protection components to make sure they are okay.  It is something I regularly do with used/abused meters since they might have had an "oops" moment subjected by the previous owner.  It doesn't take very long and is easy to do.

3) The red MOVs (circled in red) should measure 0L (resistance reading).
4) The PTC (circiled in green) should measure around 1.1k ohm.
5) The fusible resistor (circled in blue) should measure around 1k ohm.
6) The spark gap (circled in yellow) should measure measure 0L (resistance reading).

3,4,5 and 6 can all be measured "in circuit".

[retiredcaps]

The rotary switch looks pretty greasy but unfortunatlly at the moment I don't have IPA to give a thoughtful bath. I tried to clean the contacts rubbing them carefully with an ear cotton bud to no avail.

You should also take the black rotor switch on the bottom of the pcb off to check.

http://mrmodemhead.com/blog/fluke-83-dmm-repair/

[ModemHead]

The wafer switch has lubricant applied at the factory, as long as the contacts are not damaged it's probably OK.

Those time-delay fuses are useless and somewhat dangerous in a DMM.  They should be replaced with proper fast-blow KTK/KLK-15 and BBS-1 types (if you get the meter working, of course.)

[retiredcaps]

Those time-delay fuses are useless and somewhat dangerous in a DMM.  They should be replaced with proper fast-blow KTK/KLK-15 and BBS-1 types (if you get the meter working, of course.)

I noticed that as well, but I initially found conflicting information about the SC-10. 

One site said it was "fast acting", but Bussmann's official site states time-delay of 12 seconds (minimum) at 200% of rated content.  Obviously, I would trust the datasheet posted on Bussmann's  website.

http://www.cooperindustries.com/content/dam/public/bussmann/Electrical/Resources/product-datasheets-b/Bus_Ele_DS_1024_SC.pdf

I suspect the fuses inside the 83 are purchased from Home Depot in the USA.

[ModemHead]

Forgot to mention: that PCB is very nice and clean for an old instrument.

And: We may need to check for a broken jack assembly.

[retiredcaps]

And: We may need to check for a broken jack assembly.

For OP, see

http://mrmodemhead.com/blog/fluke-80-series-jack-assembly-rebuild/

[diegoperez]

Just to confirm.

1) The Fluke 83 was set to DCV and the meter was turned on (9V battery inserted and working inside the Fluke 83).
2) What was the other meter used to measure the input impedance of the Fluke 83?

While you have meter open and waiting to get some IPA, let's measure the input protection components to make sure they are okay.  It is something I regularly do with used/abused meters since they might have had an "oops" moment subjected by the previous owner.  It doesn't take very long and is easy to do.

3) The red MOVs (circled in red) should measure 0L (resistance reading).
4) The PTC (circiled in green) should measure around 1.1k ohm.
5) The fusible resistor (circled in blue) should measure around 1k ohm.
6) The spark gap (circled in yellow) should measure measure 0L (resistance reading).

3,4,5 and 6 can all be measured "in circuit".

1) Yes it was.
2) A fluke 77

3)  MOVs -> OL   (ok)
4) PTC -> 1582 Ohms 1.5k is the Service Manual reported value (ok)
5) fusible resistor -> OL    we found a possible culprit!
6) spark gap -> OL (ok)

Forgot to mention: that PCB is very nice and clean for an old instrument.

And: We may need to check for a broken jack assembly.

I measured continuity between the tip of the probe and the trace in the PCB that is connected to the Volts jack (and Irepeted the process for the common probe).  It measured continuity even when wiggling the jack. So, no problem found in the jack assembly.

I guess the way to proceed now is to replace the damaged part. I would try to test bodging a 1k resistor  in parallel with the fuse resistor. Will these be a good test before buying the fusible resistor?

The description  found in the service manual is RES,MF,1k,+-1%,100PPM,FLMPRF,FUSIBLE  I don't know where would be a good place to buy it?

[retiredcaps]

5) fusible resistor -> OL    we found a possible culprit!

Yes, that is a problem.

I guess the way to proceed now is to replace the damaged part. I would try to test bodging a 1k resistor  in parallel with the fuse resistor. Will these be a good test before buying the fusible resistor?

The description  found in the service manual is RES,MF,1k,+-1%,100PPM,FLMPRF,FUSIBLE  I don't know where would be a good place to buy it?

For test purposes, you can use any 1k ohm resistor.

For procurement, see

http://mrmodemhead.com/blog/fluke-87-fusible-resistor/

[diegoperez]

For test purposes, you can use any 1k ohm resistor.

For procurement, see

http://mrmodemhead.com/blog/fluke-87-fusible-resistor/

After replacing the fusible resistor Ohms range worked fine again. However, DCV DCmV and ACV still do the same things e.g.

--  VAC range it oscillates between OL and a value  with every display update when probes in Open circuit. Measuring mains voltage displays 146.4V steady (before it was approx 40V and should be 230V here in Spain).
-- VDC displays 352V approx with probes open and does not goes to zero when shorting probes. However, when measuring a voltage with probes it does adds the voltage and displays 352 + 2/3 measured_voltage (approx).
-- mVDC displays -313mV  approx (and sometimes OL) with probes open and does goes to zero when shorting probes. It also measures spot on (the reference was another fluke 77)
-- Diode Testing OL always, however probes do output 3.81V in Open Loop and source  0.92mA when shorted.
-- Ohms and Current ranges work fine.

Now the imput impedance are the following

VAC OL
VDC 18.75 MOhm with one polarity and 5.6MOhm with polarity reversed 

My guess know is the thin film resistor or the main asic might be problem. All sugestions will be appreciated and thanks for helping!

PS, It is amazing the quality of the content ModemHead has in his website. I really appreciate the effort and generosity of this work.

[ModemHead]

Ok, it gets more difficult from here.  Some red herrings are bound to be followed.  That said, here's some thoughts on what you describe:

Statistically speaking,  the resistor array is very unlikely to spontaneously fail but of course it is susceptible to physical damage and overloads.

I find it odd that R1 may have failed open during your troubleshooting?  It takes energy to make it fuse open and the ohms range cannot work without it.

Non-zero residual readings like "352" or "313" generally indicate current leakage into the high-impedance parts of the circuit (between the voltage divider and the ASIC inputs).  #1 cause of this is PCB/ASIC contamination.  #2 is, unfortunately, damaged ASIC inputs.  In this case, it could also mean that some open circuit fault is simply not draining away accumulated charge.

Grounding the bottom end of the divider through rotary switch pins 1 and 2 is common to the ACV, DCV, and Diode Test ranges.  Something to check out, since that sort of lines up with what works and doesn't work?

It may be helpful to go through all the DC volts ranges manually (including the separate mV range) and check the input impedances.  These are very predictable, and your readings indicate something is wrong there.  Leakage currents could be affecting your Fluke 77 readings, leading to the polarity differences.

Thanks for the kind comments regarding my website, I'm glad you find it useful.  I wish I had more time to spend on it these days.

[retiredcaps]

Statistically speaking,  the resistor array is very unlikely to spontaneously fail but of course it is susceptible to physical damage and overloads.

Agreed.  The only resistor array failure I have seen myself is on a cheap 830 like meter.  The "array" is a 1M ohm network made up of multiple resistors in series.  The last resistor was a 100 ohm resistor and it failed open.

You can easily test the resistor network on the 83 to verify that all the values are correct.

I find it odd that R1 may have failed open during your troubleshooting?  It takes energy to make it fuse open and the ohms range cannot work without it.

+1.  Opening your meter for photos and cleaning should not have caused the fusible resistor to fail.

[retiredcaps]

And this was the ebay auction http://www.ebay.com/itm/252065578180

That seller sells a lot of amplifier equipment and tubes.  I don't have any knowledge with tubes, but I do see posts in forums that the voltage associated with tubes in radios/amplifiers can be very high.  Perhaps the 83 was used in this type of environment and the previous owner had an "oops" moment. That would certainly cause the fusible resistor to go open circuit.