Fluke 8200A read problem.

Fluke 8200A read problem.
Posted by neo on 23 Sep, 2017 13:41
I have a Fluke 8200A which reads ohm's and volts perfectly and accurately. It turns on all the nixies work and there are no problems except for one small annoyance.

When i press one of the ohm's ranges a light turns on and i hear a faint click, for simplicity i will call this a "range latch", the ohm's range has no problems with range latching however of the three ranges for volts, DC, AC, MVDC; all ranges have iffy range latches.

That is to say sometimes it works, often enough that i could verify that it will read DC volts very accurately, but most of the time there is no light or click.

By deduction i assume this means it is relay related but, and this is where the point of this thread comes in, why is it all three ranges?
How would i fix it if it is a relay? I mean more specific than just soldering a new one in, for example identifying which one it is and a viable replacement.
To be honest i am somewhat intimidated by my lack of success on other things as of yet and i am slightly intimidated by the inner workings of this thing without a manual.

So in closing, does anybody have a manual or at the very least a schematic for this thing? I'd be happy with one drawn on the back of a napkin so long as it's legible.
I know i can buy one but i would prefer to save that for a last resort, please and thank you.

#1 Reply
Posted by Cubdriver on 23 Sep, 2017 21:08
neo -

If I'm understanding this, when you change the meter's measurement mode, the resistance works all the time, but the other modes (DC, AC or mVDC) don't always activate properly, and when they don't the associated indicator light does NOT come on, correct? Am I also correct in concluding that if the indicator comes on that the associated mode works correctly too?

In short, if the light comes on, the mode works, but the light (other than for resistance) doesn't always come on?

-Pat

#2 Reply
Posted by Cubdriver on 23 Sep, 2017 22:34
The signals traipse through the schematic like drunken sailors, but to begin with I'd suggest looking at the Ohms mode selector switch - as shown in the schematic below, there is a ground signal that goes through the series of mode switches, as traced by the pink line. As the various modes are selected, the associated 'control' line is pulled low (control lines in green). If I understand your description of the issue correctly, it seems that the problems are in the modes that lie beyond the 'Ohms' mode switch, so I'd begin by looking at the contacts at the top left of S4 on the schematic.

Good luck!!

-Pat

#3 Reply
Posted by neo on 24 Sep, 2017 02:52
Quote from: Cubdriver on 23 Sep, 2017 21:08
neo -

If I'm understanding this, when you change the meter's measurement mode, the resistance works all the time, but the other modes (DC, AC or mVDC) don't always activate properly, and when they don't the associated indicator light does NOT come on, correct? Am I also correct in concluding that if the indicator comes on that the associated mode works correctly too?

In short, if the light comes on, the mode works, but the light (other than for resistance) doesn't always come on?

-Pat

Exactly. Thank you for the schematic, i'll look at it and update here.

#4 Reply
Posted by neo on 24 Sep, 2017 03:04
Small question , how do i get at the switches? Short of complete disassembly that is.

#5 Reply
Posted by neo on 24 Sep, 2017 03:31
To answer my own question, no it is not. Completely disassembly IS required to get to switches, now that i have their board loose what should i do now?

#6 Reply
Posted by Cubdriver on 24 Sep, 2017 03:57
I'd shoot some contact cleaner into the switches and exercise 'em (ohms especially). If the meter was used in a test setup and left in 'Ohms' forever, the contacts on that switch may have oxidized; perhaps repair would be as easy as cleaning them.

If you have access to the bottom of the board, trace the circuit and try to see which sets of contacts are the ones in question, and check their resistance with the switches in each position. I'm guessing that they will be one of the pairs at the rear of the switch body (furthest from the button), but cannot be 100% certain of this.

-Pat

#7 Reply
Posted by neo on 24 Sep, 2017 04:39
Quote from: Cubdriver on 24 Sep, 2017 03:57
I'd shoot some contact cleaner into the switches and exercise 'em (ohms especially). If the meter was used in a test setup and left in 'Ohms' forever, the contacts on that switch may have oxidized; perhaps repair would be as easy as cleaning them.

If you have access to the bottom of the board, trace the circuit and try to see which sets of contacts are the ones in question, and check their resistance with the switches in each position. I'm guessing that they will be one of the pairs at the rear of the switch body (furthest from the button), but cannot be 100% certain of this.

-Pat

I don't have any contact cleaner as of yet, unfortunately.

http://www.ebay.com/itm/11-Oz-Premium-Electronic-Contact-Cleaner-Spray-The-Best-Quick-Drying-Specialist-/263212323991?epid=1325135137&hash=item3d48ad5497:g:N~0AAOSwsm1ZviWe

Would that work or is it "deoxit or nothing" when it comes to this?

Anyways, i will try to sort out which pin is which. I notice that the AC button does not seem to want to stay depressed.

#8 Reply
Posted by neo on 24 Sep, 2017 05:03
The ac button will stay depressed if pushed hard enough.

#9 Reply
Posted by Cubdriver on 24 Sep, 2017 05:08
I think the CRC stuff is geared more towards electrical contacts in vehicles, but it's worth a shot. I doubt it would hurt anything. Contact cleaner was much more readily available back in the days of mechanical TV tuners, but those days are long gone. The Caig stuff seems to be the gold standard these days, but I believe the company that makes WD-40 also has a contact cleaner out, and I seem to recall hearing good things about it. FWIW.

The AC button not latching is likely an issue with gummed up lube on the moveable bar that latches and releases the buttons. The contact cleaner will probably help that, too, by cleaning away the sticky, dried out grease.

-Pat

#10 Reply
Posted by neo on 24 Sep, 2017 05:22
The wd-40 contact cleaner isn't all that more expensive so i will go with that, thank you for mentioning it.

#11 Reply
Posted by Cubdriver on 24 Sep, 2017 05:46
-Pat

#12 Reply
Posted by neo on 24 Sep, 2017 06:39
Mapping switches now, also someone has been in this before i had suspicions that were confirmed by a nixie that did not match the others. Same nixie just a few years off, the glass is all that is different. It doesn't appear they did anything else and did well soldering it in, but still something to keep in mind.

#13 Reply
Posted by neo on 24 Sep, 2017 09:11
OK this is the way the continuity on the switches worked out
30 connects directly to M $\$\Omega\$$ 3 and 4, K $\$\Omega\$$ 9 and 11, $\$\Omega\$$ 9 and 10, VAC 9 and 4, MV 1 and 4

M Toggled, 1 and 2; Not toggled, 5 and 6.
K Toggled, 2, 4 and 7; Not toggled = nothing.
O Toggled, 2, 4 and 8; Not toggled 3 and 5
VAC Toggled, 7 and 2; Not toggled, 11 and 6.
M-V Toggled, 2, 3, 8 and 10; Not toggled 64 ohms between 1 and 3, 3 and 5 connected, 6 connected to 1

DC switch is weird, not toggled 3 and 5 have a direct path to ground but 4, 6, 8, 10 and twelve all have a 64 ohm resistor to ground. 4 and 6 are connected. toggled 3 and 1 have a direct path to ground and the only change in the other pins is that 4 is no longer connected.

01 02
03 04
05 06
07 08
09 10
11 12
the orientation for the switches, nixies upside down and facing me

I checked it with two multimeters, three times each so i am relatively sure that what i listed is correct to my unit.

#14 Reply
Posted by neo on 25 Sep, 2017 05:25
I have now also checked each pin of both MV and VDC and i cannot find any that connect to what is labeled as their corresponding control pin, 26 for MV and 25 for VDC.

#15 Reply
Posted by Cubdriver on 25 Sep, 2017 07:45
I'll try to pop the cover off of mine tomorrow after work and see if I can take some readings for comparison.

-Pat

#16 Reply
Posted by neo on 27 Sep, 2017 00:55
OK so as to provide an abridged version of my work on it.

Didn't work, sprayed contact cleaner on switches and then it didn't work because boards were loose.

Fixed the loose board and now works, VAC up to at least 10V, VDC up to at least 230V.
OHM's works up to a little above 100k before it loses its mind.

NOTE, the only reason i stopped voltage tests there was that it was the highest voltage i could safely generate.

Switches felt fine before reassembling it, but a few hours after the contact cleaner they are gummy and hard to push again.

Took front off and contact cleanered again and it feels fine now.

Ohms works above 100k now, however anything above 100K swings the reading alot.

300K reads as 285K-336K
1M reads as 900K-1087M
and 10M swings the widest between 8M-12M

#17 Reply
Posted by neo on 28 Sep, 2017 21:38
All in all i think it is currently working well enough to call it good, unless someone has something to say about the $\$\Omega\$$ reading swinging.

EDIT:
See below.

#18 Reply
Posted by neo on 29 Sep, 2017 00:21
This thing is going to drive me up a wall.

The contact cleaner once applied makes the switches easy to work, however a few hours later they are an unuseable gummy mess. As if that wasn't enough it only selectively works, AC works in testing but not later for example.

EDIT:
These switches are seriously screwy i think, if the contact cleaner is fresh it works great if not it gets screwy.
For example the ranges don't all work even though they all did before, or the buttons become inoperable.

#19 Reply
Posted by Cubdriver on 29 Sep, 2017 01:31
Crap. I've been pulled away to help a friend with painting every night, so I haven't been able to open mine. From the sound of it, it's a switching issue, and somehow the contact cleaner is incompatible with the grease in the switches. What did you wind up using, the WD-40 stuff?

Have you thoroughly flushed them with it, and did crud come out, or did you just give them a little squirt?

I just grabbed mine; I'm pulling the bottom cover now. (Have to poke a hole in the August 1987 cal sticker to do so. )

-Pat

#20 Reply
Posted by neo on 29 Sep, 2017 02:36
I sprayed it, worked the switches, sprayed a little more and then worked the switches a little more. Yes, the WD-40 contact cleaner is what i am using and i have yet to see crud but i also have not investigated the switches recently as that requires full dismantle.

Shame about the cal sticker, thanks a lot for your help though.

Just out of curiosity what year is yours? Mine was assembled august 1971, at least thats what the ink stamp on the chassis says.

#21 Reply
Posted by neo on 29 Sep, 2017 02:53
I think i may be able to fix the switches by bathing them in contact cleaner and working them until my fingers don't work. Maybe, i think it might be getting better because,at least currently, everything works including the switches. Time will tell.

#22 Reply
Posted by bitseeker on 29 Sep, 2017 04:46
It sounds like you're starting to scratch the surface, as it were, on the contacts in the switches. When the cleaner evaporates, the goo settles and causes problems. If the cleaner can't dissolve the muck and nothing will come out of the switches, it may not improve without disassembly or replacement of the switches. Perhaps, as Pat implied, a different cleaner may work better to dissolve the old lubricant.

#23 Reply
Posted by Cubdriver on 29 Sep, 2017 06:35
This beast will certainly not win any 'ease-of-serviceability' awards, at least not for the mode and range switches. It seems that they start with those and then build the rest of the meter around them.

Here is the chassis with the outer and inner top and bottom covers removed (four screws and one screw each, respectively), and the front panel (four additional screws after top and bottom outer covers are removed) disconnected and taken off. I left the boards that were hooked to the input jacks (four of them) connected and removed them as a unit. The black and white leads visible at front left are the rear analog input connections; they too connect to the front panel jacks but as the cable is fixed to the chassis, I disconnected them to remove the panel:

Boards and front panel:

Left to right: AC Converter, input buffer, ohms converter, and millivolt converter PCBs.

Here is the display/interconnect board assembly removed from the meter chassis (four screws in the bulkhead, and two more at the rear adjacent to the edge connectors, along with two wires to the power switch:
Front oblique:

Rear oblique:

Note that the display board is directly above the switches, supported on standoffs that also hold the bulkhead with the card guides on it. To make matters more difficult, they do not appear to be easily separated - in the 'front' photo, note the long, low grey connector visible just behind the two rightmost nixies and the small daughter board with the trim pot on it. Its opposite end can be seen in the 'rear' view just below the card guide bulkhead, beneath the guides second and third from the left. This is the ribbon cable that ties the display board to the interconnect board. On my meter, at least, it is about 4-5" long, and those are IDC connectors that are soldered to each of the boards. The display can be removed, but it will remain attached and flopping around. I suppose for easier access to the switches, the standoff screws could be removed and the now loose display assembly moved over the edge connectors and laid on its back, then affixed to the two rear mounting points with small two hole brackets and the screws that previously held it to the chassis...

Closeups of the switches causing neo grief are below - the mode switches that are giving him trouble are soldered to PC boards both above and below. They solder to the interconnect board on the bottom, then have a small board with the mode indicators soldered to the top. NOT easy to clean, alas. :

I'll see if I can further disassemble things tomorrow, and get some more photos up. Other photos of the meter are at: https://pmanning.smugmug.com/Electronics/Fluke-8200A-Digital-Voltmeter

-Pat

#24 Reply
Posted by Cubdriver on 29 Sep, 2017 06:41
Oh, and neo, mine is a baby compared to yours. I haven't found a date stamp anywhere on the chassis, but the component date codes range from the late 70s to ~~mid 1980, so I'd assume that it was made late in 1980.~~ I'm curious - does yours use LEDs for the mode indicator lights, or incandescents, or...?

-Pat

Wait - belay that previous date estimate - a closer look at the ohms converter board reveals trim pots with 1981 production dates (light blue ones on the right), (and a big honkin' 6.99M precision resistor with a mid '79 date - Fluke must have bought things in big lots!), so I now guess mid 1981 for a build date?