DMM cemetery .. #1 Fluke 87

[Giorg]

I've received as a 'gift' a lot of broken multimeter  and I'm trying to get them back from the dead ..

The first one is a Fluke 87:

Symptoms:

mV --> OL
V --> around 0.030 V with crossed leads. I can measure correctly a 9v battery
  --> jumping around 64K. Crossed leads -> still 64k
diode --> OL

test made:
dial test (rotary switch test - holding range button while powering )-> OK
input fusible resistor -> around 1k
spark gap -> open
2 mov -> Open
thermistor -> 8.5 K (should be 1,5k but schematic is not 100% matching. swapped with a 1.5 resistor, same behavior)
voltage reference zenner -> around 1.2 V
mA and A works fine.

all power supply lines indicated in service manual are within specs.

I cannot spot any damaged components on the board itself.
I've looked at the schematics, very well made but I'm not able to spot a component that could trigger this kind of behavior.
Is there a way to bypass or force a reading directly to the main uC pins?
If anyone as some kind of advice, it would be a great opportunity to learn something new and to put a great instrument into my drawer!

[Giorg]

more pics

[retiredcaps]

I suggest you start a new thread for each meter.  I'm willing to make suggestions, but I and others will quickly get confused what you are testing and reporting with 3 meters in one thread.

I suggest starting with the Fluke as I have several suggestions and there is lots of documented solved cases on

http://mrmodemhead.com/

and full service manual for the Fluke 87 on Fluke's website.

[Giorg]

You are right, I've modified this thread for the Fluke only.

I've looked on your website but I cannot find any information specifically related to my issue 

[ModemHead]

Is it possible that some has attempted to replace the rotary switch?  It doesn't look quite right.  Especially the contact sticking out.

[Giorg]

it appears to be just a floating contact, it has no connection with the pcb or anything else. In other pictures found online is just missing. Also solder joints do not look like they have been reworked.
pin to pin connection is consistent with schematic.

[Celectronicamx]

Maybe this can help?

https://elektrotanya.com/fluke_87_service_manual.pdf/download.html

[ModemHead]

it appears to be just a floating contact, it has no connection with the pcb or anything else. In other pictures found online is just missing. Also solder joints do not look like they have been reworked.
pin to pin connection is consistent with schematic.

Okay, I don't recall ever seeing an actual contact installed on pin 7 on any 80-series meter that I have.  There's usually no contact or rivet there.  Having that thing flapping in the breeze doesn't seem like a great idea. 

Okay I'll offer my usual disclaimer, which is I don't know what's wrong with your DMM, but I can throw some stuff out there to ponder on.

The thermistor is a problem, but doesn't sound like the main cause of the malfunctions.  If it's cold resistance goes too high, it will limit the amount of test current that can be provided in ohms mode, and throw ohms readings off on the low ranges, otherwise not much effect unless completely open.

That you can measure a 9V battery correctly bodes well I think.  Not zeroing out in DCV mode with the input shorted is usually some leakage current caused by contamination and can often be solved by one (or more) IPA baths and some action with a brush.  Resist the urge to turn the calibration pots until all other repairs are finished.

The biggest problem seems to be with mV, ohms, and diode-test.  These functions share the same voltage sense input on the processor U4P93 (AP1).  The route to this input depends on rotary switch pins 11 & 12 being closed, and R2, which is the blue 909Meg resistor beside the fusible resistor.  Verify everything along this path.  Refer to the manual pages 5-22 and 5-24 to see where I'm getting this.

[retiredcaps]

Since modemhead spotted that floating contact, a quick easy test is the rotary switch test from the service manual section 3-16 (attached).

[retiredcaps]

More rotary and code switching and IPA by modemhead ...

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

[Giorg]

Thank you retiredcaps but I've already done it ( "dial test" in my first post) and results are perfect.

I've also already done the general cleaning with IPA..

[retiredcaps]

Thank you retiredcaps but I've already done it ( "dial test" in my first post) and results are perfect.

Oops, sorry didn't see that you edited the first post.

[Giorg]

it appears to be just a floating contact, it has no connection with the pcb or anything else. In other pictures found online is just missing. Also solder joints do not look like they have been reworked.
pin to pin connection is consistent with schematic.

Okay, I don't recall ever seeing an actual contact installed on pin 7 on any 80-series meter that I have.  There's usually no contact or rivet there.  Having that thing flapping in the breeze doesn't seem like a great idea. 

Okay I'll offer my usual disclaimer, which is I don't know what's wrong with your DMM, but I can throw some stuff out there to ponder on.

The thermistor is a problem, but doesn't sound like the main cause of the malfunctions.  If it's cold resistance goes too high, it will limit the amount of test current that can be provided in ohms mode, and throw ohms readings off on the low ranges, otherwise not much effect unless completely open.

That you can measure a 9V battery correctly bodes well I think.  Not zeroing out in DCV mode with the input shorted is usually some leakage current caused by contamination and can often be solved by one (or more) IPA baths and some action with a brush.  Resist the urge to turn the calibration pots until all other repairs are finished.

The biggest problem seems to be with mV, ohms, and diode-test.  These functions share the same voltage sense input on the processor U4P93 (AP1).  The route to this input depends on rotary switch pins 11 & 12 being closed, and R2, which is the blue 909Meg resistor beside the fusible resistor.  Verify everything along this path.  Refer to the manual pages 5-22 and 5-24 to see where I'm getting this.

I've desoldered ant tested all the components on "ohm path " and they are all ok.

I've measured -2.8V across COM and V input terminal. I've lifted U4 pin 93 and it's like at -3V from COM.
Is it normal ? If it's a sensing input ..is it supposed to be at a negative potential ? I'm starting to worry that U4 might be bad

[ModemHead]

I've measured -2.8V across COM and V input terminal. I've lifted U4 pin 93 and it's like at -3V from COM.
Is it normal ? If it's a sensing input ..is it supposed to be at a negative potential ? I'm starting to worry that U4 might be bad

It would of course be normal to have a positive voltage across the jacks in Ohms or DT mode, but this test current comes through the fusible resistor/thermistor path.

So what you're seeing does not seem normal at all.  The ADC input should not source (or sink, in this case) enough current to develop that much voltage.  It should be very high impedance.  Lifting the pin is exactly what I would have done (as a last resort), and pretty much narrows it down to U4, unfortunately. 

[retiredcaps]

I have a 87 with a dead U4 as well. It won't boot at all.

Time to start a new thread for your other 2 meters.