Fluke 77 AC wandering offset

[BurningTantalum]

Hi,
I have a trusty old Fluke 77 'series 1' that has been all around the world with me over the last quarter century. It has always been a bit of a nuisance in humid climes, but recently it started to wander about in the AC volts range. I cleaned the PCB and relubricated the wafer switch which cured much of the problem but was left with an offset of up to 500mV on the AC volts range, and the AC amps range, regardless of whether the probes where connected or not, or inputs shorted or not. When I remove the two lower case screws the offset disappears.
I find that when the case top is placed onto the bottom half without the screws and pressure is applied mainly in the area below the switch the offset climbs back up relative to the applied pressure. With the case removed this effect cannot be replicated regardless of twisting and pressing anywhere on the PCB and/or any components AT ALL. I have tried placing the top case inverted over the bottom case but this has no effect. I have really tried everything that I can think of.
This does not occur AT ALL on any of the other ranges, just AC volts and amps.
Does anyone have ANY ideas or suggestions as to the cause of this odd fault.
Regards,  BurningT

[tautech]

A trick an old EE told me to find PCB problems was in a quiet room to twist, bend, flex PCB and listen for clicks that might indicate a cracked solder joint on SMD components. You might hear something but finding it is not always easy but it indicates what the problem is. 
Then you might have to re-solder most components for success. 

[retiredcaps]

I don't have the original 77, but I have a 75 series II.  The service manual for it says that (page 2-4)

http://transmille.net/ProCal/Procedure%20Library/Fluke%20Digital%20Multimeter%2023%20MKII%20%5B1.10%5D/Technical%20Data/Fluke%2070%20Series%20MKII%20(PN%20896204%20July%201991%20Rev%201,%2009-92)%20Service%20Manual.pdf

"AC CONVERTER

The ratio of R17 and R18 determines ac voltage and current accuracy, and C11 is the averaging capacitor for the half wave ac coverter portion of U1."

If you look at page 4-9, you will see the above 3 components are near the bottom of the multimeter just above F1.  If you look at the backside, you will see they are all connected to U1.  I suggest cleaning the backside of these 3 components and all the pins to U1 using a stiff toothbrush with IPA. 

A tip I learned from modemhead, another member here, is to use a dental pick or very sharp probe to clean the pins in between U1.  A 10x jeweler's loupe helps to make sure there are no contaminents on U1.  I have "fixed" a number of used/abused multimeters that had abnormal offset readings with nothing more than a generous amount of IPA.

If cleaning doesn't work, you might want to resolder these 3 components as well just in case they have a bad solder joint?

PS. I measured all 3 components "in circuit" with my 87V and compared them to the BOM on page 4-8 and here is what I got.

R17 - BOM says 20.5k - I got 20.51k ohm
R18 - BOM says 9.20k - I got 9.199k ohm
C11 - BOM says 2.2uF - I got 5.50uF (no relative mode).

The reading for C11 is high, but I did it in circuit and other components may be affecting it?

[retiredcaps]

Just found the manual for the original 70 series.  R17 , R18 and C11 are mentioned in the AC section (page 3-1) so nothing has really changed in that respect.

http://www.testitnow.co.uk/media/files/test_equipment_manuals/Fluke/Fluke%2070%20Series%20(PN731034%20January%201984%20Rev%201%2005-89)%20%20Service%20Manual.pdf

[BurningTantalum]

Thanks for the replies, and apologies for posting and then going quiet- I have been away from the workshop for two days, and now is Friday evening here in Western Australia so I have opened a beer!
I will have a concerted effort tomorrow and report back on success/observations for anyone interested.

Thanks again!  Tantalum

[BurningTantalum]

Hi again,
  The meter is now back to 99% of what it should be: I had already extensively checked for DJs and no matter what I deformed or pressed or pushed I could not replicate the offset that appeared when the top cover was placed back on and a low pressure exerted below the rotary switch - weird, almost like an electrostatic issue.
I thoroughly cleaned the PCB south of the display with IPA and dried it well for 10 mins with a hairdryer. The fault was still there but not as bad.
I had a sleep, then blasted the area around and under U1 with switch cleaner (!) mainly because it has a violent jet, then blew it dry with filtered compressed air. I then did a standard flood with IPA, bodging it around with a soft artist brush, then dried it for several hours in the fierce Aussie sun inside a glass box.
The problem seems to be solved!
I recall having immense grief with this meter everywhere tropical, and while in the Philippines I had to resort to buying a cheapy from the mall just to get the job done.
Worth noting that there was dust debris around the bottom of the lead sockets - I have got out of the habit of putting the meter back in its grey soft case, and for the last few years have left it in its holster on the bench.

Thanks to all for the info.

Tantalum

[retiredcaps]

I'm glad you resolved the problem.

I might have damaged mine while taking readings.     No one to blame but myself.   

It is winter and super dry due to the heat coming on.  I may have killed the U1 ACLO (pin 28) with ESD?  Mine shows -0L in ACV and AC current mode.  All other functions work fine and are with its accuracy specifications. 

Time to re-read the service manual, look at the schematics and take some readings.

On the upside, I wanted to learn more about how multimeters work and here is a good troubleshooting opportunity for me.  Luckily, I don't have the same emotional attachment to this 75 series II.  I bought it used/abused and it was previously mistreated by its previous owner.

[ModemHead]

... weird, almost like an electrostatic issue. ...

Just a thought, make sure the little spring that grounds the back side shield is still there and is free of oxides.