Fluke 8024B...

[knockbill]

I was given a Fluke 8024B today with issues...Physically it's in very nice condition, missing only the bottom push button,,, The display is very clear, all segments are present, but thats the end of the good news...  It does not read in any mode,,, when turned on in ACV, DCV or R  mode there are 2-3 decimal points and the numbers fluctuate in any scale... I googled for info but this seems to be beyond the scope of the trouble shooting guide in the  manual... I did find a thread on removing the boards from the case, which I did to clean the them with alcohol, and check for leaking cap contamination, it was surprising clean inside...
so, is this one worth trying to repair, or should I just keep it for parts for the 8020?
Thanks for any insight...

[retiredcaps]

I know modemhead fixed a 8024B in similar condition to yours. 

1) As you probably know, the caps in these old Flukes leak from the bottom and won't be obvious until you remove the caps.  If you look, you may not spot the leakage.  I think, without checking the manual, there are only two electrolytic caps so it will only take 2 minutes to remove each.

2) If your main IC is socketed, remove the main chip, dunk it in IPA and reseat it after it is thoroughly dry.

3) Modemhead does have an unpublished 8024B blog entry with extensive photos and tips, but due to his situation, he has not yet made it public yet.  However, I can make suggestions based on trying #1 and #2 first.

If this worth repairing?  For me, yes because I learn something new each time I try to repair something even though I may not be doing it myself (like in this case).

[retiredcaps]

4) I forgot that the gang switches in these 30+ year old instruments could be problematic.  Cleaning or repeated push in/out might be another suggestion.

[knockbill]

OK,,,
Replaced the 2 x 22/16V caps with 22/35V, closest I had on hand, they were leaking, so I scrubbed the PCB after I got the old ones out... Removed and scrubbed main chip and socket with alcohol,,, replaced funky, taped up battery power connector with a new one, repaired broken corner(dust shield)  on peak hold switch handle... and polished the display lens w/ toothpaste...

Here's some results against a shop meter I've been using...
     
           FLUKE                          SHOP METER
100k     99.0k                            98.9k
1m        95k and declining           1010k

9V batt  10.8VDC                        9.36VDC  (fluke starts at 1.4 on DC scale

AC TX    17.64VAC                      15.5VAC  (fluke starts at 1.32 on AC scale

Seems we're making  progress, and fluctuation could be dirt/corrosion on R scale...

Thanks again for the info and help... we'll get it!!!

[retiredcaps]

Replaced the 2 x 22/16V caps with 22/35V, closest I had on hand, they were leaking,

Yep, they leak from the bottom bung.  A visual inspection will not be sufficient in this Fluke 8000 series meters.

Here's some results against a shop meter I've been using...
     
           FLUKE                          SHOP METER
100k     99.0k                            98.9k
1m        95k and declining           1010k

9V batt  10.8VDC                        9.36VDC  (fluke starts at 1.4 on DC scale

AC TX    17.64VAC                      15.5VAC  (fluke starts at 1.32 on AC scale

Seems we're making  progress, and fluctuation could be dirt/corrosion on R scale...

A cleaning of the gang switches and another round of IPA cleaning might help.  I know, firsthand, that 2 to 3 IPA baths are necessary/required.  When making this suggestion, I know I get the skeptical expression from people asking for help, but any contamination will affect the readings.  Sometimes the contamination wicks it way up the legs of the through hole components and cannot be easily seen.

I know I have tried to help some and when it comes to suggesting an IPA bath, the thread just ends from the OP.  They don't believe it and think I'm nuts.  I'm sure it is the same right now for some people reading this.

As for the DCV readings being a bit high, I would try another IPA cleaning first and if it still reads high, you can  it using POTS.

[tautech]

A cleaning of the gang switches and another round of IPA cleaning might help.  I know, firsthand, that 2 to 3 IPA baths are necessary/required.  When making this suggestion, I know I get the skeptical expression from people asking for help, but any contamination will affect the readings.  Sometimes the contamination wicks it way up the legs of the through hole components and cannot be easily seen.

I know I have tried to help some and when it comes to suggesting an IPA bath, the thread just ends from the OP.  They don't believe it and think I'm nuts.  I'm sure it is the same right now for some people reading this.

As for the DCV readings being a bit high, I would try another IPA cleaning first and if it still reads high, you can  it using POTS.

+1 to all that ^^^
Had a moment in the last few days to revist this great thread:
https://www.eevblog.com/forum/testgear/old-fluke-multimeters/

Can't remember which page or post but DR Taylor made the statement that one of the downfalls of these DMM's was external contamination and is one of the reasons why most DMM's are rotary dial these days....much easier to keep the muck out.

[krivx]

Replaced the 2 x 22/16V caps with 22/35V, closest I had on hand, they were leaking,

Yep, they leak from the bottom bung.  A visual inspection will not be sufficient in this Fluke 8000 series meters.

Here's some results against a shop meter I've been using...
     
           FLUKE                          SHOP METER
100k     99.0k                            98.9k
1m        95k and declining           1010k

9V batt  10.8VDC                        9.36VDC  (fluke starts at 1.4 on DC scale

AC TX    17.64VAC                      15.5VAC  (fluke starts at 1.32 on AC scale

Seems we're making  progress, and fluctuation could be dirt/corrosion on R scale...

A cleaning of the gang switches and another round of IPA cleaning might help.  I know, firsthand, that 2 to 3 IPA baths are necessary/required.  When making this suggestion, I know I get the skeptical expression from people asking for help, but any contamination will affect the readings.  Sometimes the contamination wicks it way up the legs of the through hole components and cannot be easily seen.

I know I have tried to help some and when it comes to suggesting an IPA bath, the thread just ends from the OP.  They don't believe it and think I'm nuts.  I'm sure it is the same right now for some people reading this.

As for the DCV readings being a bit high, I would try another IPA cleaning first and if it still reads high, you can  it using POTS.

I've posted my experience before with an 8060a (possible in response to one of your posts...) and an IPA bath was completely necessary. I had residual readings on all voltage ranges until I gave the main IC itself a bath to properly clean the underside and legs, even after installing a new socket. After a couple of through cleans it was reading within a count of 0.

[ModemHead]

I made the unpublished 8024B photos available in case they can provide some ideas or inspiration.  I never wrote up any text, but some of the photos have captions.

That particular unit suffered from a nasty solvent spill, as well as leaky capacitors. And the hardest-to-find fault was some current leakage between the conductors on the flex connector between boards.

Unless someone else has twiddled the knobs, it is unlikely the calibration is that far out.  Keep cleaning!

[knockbill]

Studied your pics, and it seems the more I learn the more questions I have,   so here goes,,, Its been said the main IC needs a "bath", but to me that's soap and water in sitting in the tub!!! I'm assuming it means to soak the IC pins in IPA for a while then polish them dry??  Thats what I did for a 2nd time,,, the  pins look shiny, with just the socket marks showing... scrubbed the socket also... Also went over all the component pins/leads with contact cleaner and a toothbrush, then washed off with IPA, and blow dry with a hair dry on low heat... From the pics, I saw the flex connectors were removed with 2 screws and a bar clamping them to 36 contacts at each end,,, I removed them and polished both teh flex and the contacts...   

The switches are a challenge, contact cleaner can be shot thru them, and they can be exercised, but it seems its only dislodging dirt and moving it to other places in the switch,,,I just did the same thing with a SS amp  volume control,,, cleaner and exercise, for days,,, never got it right til I replaced the pot... I opened the old pot and found a broken carbon trace which was the problem... That said, I can only think that switch cleaning is a crap shoot at best, with the only sure fix is disassembly and polishing the contacts, however these push switches look like the springs are held on with a lock ring and then they would push out the back to disassemble,  of course there is no room for that with them mounted to the board...

Right now I'm convinced the PCB is clean, new caps operating correctly, flex is clean, display and connector are clean, but the meter still isn't good enough,,, there is residual voltage present,  probably from muck that moved around in a switch, as they're  really the only parts I can't see in to... Your pics also showed a process of isolating the stray V, and injecting a frequency signal but I'm not quite sure how you did it... And there's always the bad chip or other component cleaning won't fix!!
So, for now I'll reassemble the DMM to keep the pieces together, and keep trying to find new trouble shooting ideas...  Could the whole assembly be put in a ultrasonic cleaning tank? what needs to be removed 1st?  Maybe that would get into the  switches...

Thanks for the continued help and info...

[retiredcaps]

Its been said the main IC needs a "bath", but to me that's soap and water in sitting in the tub!!! I'm assuming it means to soak the IC pins in IPA for a while then polish them dry??

Yes IPA bath means letting the IC soak and being scrubbed with/in IPA.

The switches are a challenge,

Yes, which why I don't have too many of those meters.

there is residual voltage present,

How much?  Before it was 1.4.  What is it now after more cleaning?  If it is lower, then cleaning helped.

Could the whole assembly be put in a ultrasonic cleaning tank? what needs to be removed 1st? 

I know modemhead has an ultrasonic cleaner and has dunked some things in there, but I'm not on my main computer and I can't remember right now.

At minimum, you would have to remove the lcd, fuses, plastic push buttons, 9V battery, battery snap and the electrolytic caps.

[retiredcaps]

Could the whole assembly be put in a ultrasonic cleaning tank? what needs to be removed 1st? 

Wait a minute, we had this same discussion almost a year ago regarding ultrasonic ...

https://www.eevblog.com/forum/testgear/dmm-troubleshooting-help/?all

[ModemHead]

Deja vu...

My ultrasonic cleaner is too small for circuit boards, never tried it for whole-board cleaning.  I suspect the question of damaging something depends on the power of the cleaner.

Your pics also showed a process of isolating the stray V, and injecting a frequency signal but I'm not quite sure how you did it... And there's always the bad chip or other component cleaning won't fix!!

The pics show me grounding the input of the A/D converter to make sure it really can zero out.  Then I injected a stable 100mV DC level straight in to the A/D input, to see that it integrates and measures properly, and it did.  So I concluded that the main chip was OK and the stray current was originating in the front end somewhere.  Eventually I found that the flex ribbon was at fault.  I found that by measuring resistance between each adjacent conductor, and one adjacent pair did not read "OL".

It takes such a tiny amount of current to throw off the A/D converter, it is almost always some sort of contamination somewhere.  But you're right, it could also be a component.  This meter uses CMOS bilateral switches for input configuration, which is why you see U14 missing in the photos.  As I recall, I was trying to rule it out as a source of leakage current.

[knockbill]

Thanks for the info,,, not sure I'm up to removing/replacing multi pin chips,,,

Let it sit overnight, and blew it out with hair dryer again before reassembling it,,,
Ohm readings are pretty close, but the meter won't lock on to the reading, it fluctuates around it...
DCV readings are close to proper,  meter starts at 000
ACV readings are close but fluctuate, meter starts at 0 to 001... 

I "think" the cleaning-drying cycle helped a little, not sure its worth doing again without being  sure the gunk is getting out of the switches, and not just moving inside them... Its clear the switches are the weak point of these meters, just as they are in a lot of amplifiers and preamps...
I've read threads were folks have removed switches of this type from boards, disassembled them to polish, and replaced them,,, I may have the patience to do that, but not the tools or eyesight..
Its frustrating to take on a project like this knowing you don't have the capabilities to do what ever is necessary to finish it, but hoping its something simple,,, which it never is!!!!
Thanks again for the insight...

[retiredcaps]

Ohm readings are pretty close, but the meter won't lock on to the reading, it fluctuates around it...

This could be another issue?  Measure the PTC.  Without looking at the manual, I think it should measure 1.1k ohm.

DCV readings are close to proper,  meter starts at 000

So cleaning did help since initially it was at 1.4V offset.

ACV readings are close but fluctuate, meter starts at 0 to 001... 

I rarely measure ACV and personally I would grab a newer safer meter to measure ACV.  The only time I measure ACV, hands free using clips, on these old Fluke 8000 is to initially test to make sure it works.  After that, I never use ACV.  I find the push button manual arrangement interface too cumbersome to use.  I know some love it, but I prefer the automatic rotary.

[knockbill]

PTC???
 
Seems to have helped DCV, but R scale is a little wonky in some ranges, that's why I think there is still gunk in the switches... but moving it around without knowing its outta there, is a fools errand in my book...

I mentioned ACV as a bench mark for the progress we're making, thought all the scales may all tie in specially in the switching...... I use my rebuilt Triplett 850 for ringing out transformers and figuring Turns Ratios...  As a retired electrician, I have many decent ACV and current measuring meters (Wiggins, Amprobes etc)...
Thanks again for checking in...

[ModemHead]

... but R scale is a little wonky in some ranges, ...

Check the input jacks carefully.  I've seen those tubular jacks that are swaged/soldered to the board get loose.  It may escape detection at first, but you can find it by shorting the inputs with banana patch cord while in ohms mode, and put a little stress on each jack.  The reading will jump around a bit if there's a crack.

[knockbill]

Will do,,, I thought I noticed the R scale fluctuating with the V-R switch and maybe a couple of the range switches,,, thats what made me condemn the switches in the 1st place,,, and probably why Fluke stopped using them,,, I will check what ever is suggested,  if I can...  It would be bnice if i could see/clean them, and cross them off teh list!

What is PTC?

[retiredcaps]

PTC???

Thermistor.  It is RT1 on your board, manual, schematic.

https://en.wikipedia.org/wiki/Thermistor

It should be 1k ohm according to manual.  While likely good, it only takes 1 minute to check.

[ModemHead]

What is PTC?

Positive Temperature Coefficient thermistor.  Labeled RT1 on the schematic.  Right offhand I don't know where it is physically, but in this generation of meters it's probably a small gray rectangular device about the size of a 1/2 watt resistor.

[retiredcaps]

@OP, BTW, have you had a chance to check out the speed of the continuity tester on this model?  Modemhead says in his tests this is the fastest meter that does continuity.  Faster than the modern 87V, 17x series, etc.

[knockbill]

The buzzer responds quickly...

Thanks, I'll find the position of the thermistor in the manual...

[knockbill]

Looks like a lead is broken off RT1,,, I tried to solder it back on but it wouldn't take... separated the lead from RT1  for picture...  Parts list says RT1 1k, +- 40%  25C I'll search for something to  replace it with....
Sad part is its right on the edge of the PCB,,, I probably snapped it off while handling the chassis...
Thanks for mentioning it...

[retiredcaps]

Parts list says RT1 1k, +- 40%  25C I'll search for something to  replace it with....

Any 1k ohm resistor will do for testing purposes.

Does RT1 out of circuit measure 1k ohm?

[knockbill]

It reads 905 out of ckt, but I can't solder a lead back on,,, 1K seems to let the R scale be stable, so now i'll try to find  one...    May be time to get varistors for the other Fluke form last year also!

[retiredcaps]

It reads 905 out of ckt, but I can't solder a lead back on,,, 1K seems to let the R scale be stable, so now i'll try to find  one..

905 ohms should be fine, but if a normal 1k ohm resistor makes ohms reading stable, I guess we found the culprit.  If yes, I will have to remember this troubleshooting tip for future.  As I mentioned in post #2

"If this worth repairing?  For me, yes because I learn something new each time I try to repair something even though I may not be doing it myself (like in this case)."