Recovery of an old Fluke 8020a with a bad LCD

[lowimpedance]

donbrenrik,
I have been away from the eevblog, and apologize for the delay. I Emailed the information to the address you provided, and will try to attach it here as a zipped .pdf file. The plastic support under the original LCD is too deep and a little too tall to fit the new LCD under the cover glass. I removed the plastic that got in the way with small files and a Dremel tool. There was not much holding a couple of the legs to the support when I was done, but it has worked very well so far. It may be possible to save a bit more plastic on top, by shortening the legs a bit, but that may present some problems too. There is room under the plastic support to route some of the individual wires that replace the ribbon cable. Just be a patient and neat as possible. I do not believe the "Over-Range" indication can be made to work without additional logic circuits, but the digits "blank" on an over-range, so you know if (when) it happens.
Cheers!
jwhitmor

 Handy to have JW. Coincidently I have just finished this week the replacement of another dud LCD on my second unit (using a LUMEX lcd this time), and found my old wiring diagrams.
Yours is MUCH neater!. Just to note the 8020a only used the '1' and the last three digits blanked to indicate an overange condition (see manual page 2.3 parra 2.18).
 Of course my 8020's LCD's were so bad I could not make out the digits let alone any other announciators.
regards
LowZ

[E.Stern]

It's been a while since the last post in this informative thread, and I'm thankful I found it today.

My employer was a longtime user of the 8020A. He's retiring in the coming year and I have his original 8020A, gifted to him by his previous employer already. (see the nameplate on the unit). I'd like to fix it and give it to him again at his final retirement.



Uploaded with ImageShack.us


The LCD is toast, as appears common on these units.

I'll follow what I'm learning so far on the thread, but I have a question. Are you all replacing both the LCD screen and the IC, or just the LCD?

I'm handy enough to attempt the repair on the dead unit, but I don't know enough about electronics to understand some of your lingo. I am comfortable to order a Lumex LCD as spec'd earlier in the forum. That may be all I need, plus some of the soldering expertise from the electronics guys in this building (it's been a decade since I was in school for it, and my job doesn't require me to solder anything-boats are all crimp.)

[PaulAm]

Just ran across this thread, great info!

I have an old D802 with a bad LCD.  I bought an 8020 off of ebay to use for parts, but it turned out that meter works perfectly.  So now I have a way of fixing the original which has some sentimental value.  It was the first piece of pro level test equipment I ever purchased.

[lowimpedance]

@ E.Stern, highly unlikely to need the main IC replaced (its not the common part ICL7106 by the way). Just read all the info already posted and it should be down to just CAREFULLY fitting the LCD mechanically to the plastic mount the original sat on, this requires some material to be machined away for the LCD pins to sit properly, (use a dremel etc if you don't have access to a machine shop!). The rest is just wiring using thin single core wire like wire wrap wire for example. Once the job is done and wired correctly, (a crossed wire is likely at first attempt!!), that area of the meter will never need touching again.
Good luck
 LowZ

[TerraHertz]

Interesting thread! I have an old Fluke 8020A and 8022A, which both still work EXCEPT that the 8022A has a couple of the decimal points on the LCD permanently on. I've tried disassembling it, cleaning the LCD edge, the zebra strip and the PCB, reversing the zebra strip, visually checking for any board shorts, etc, but no luck. Without the circuit I couldn't think of anything more to do, and decided it was probably a fault in the main chip anyway.

So, the schematics are available! Cool. Time to have another go at it.
I don't suppose anyone else has ever had this problem, and fixed it? See pic.

Edit: FWIW, I recall hearing long ago that sunlight was bad for LCDs, so I always avoided leaving these meters in sunlight. This might be why their LCDs are both still perfectly good.
Maybe this doesn't apply to modern LCDs?

Edit2: Dammit. All the amazing 'restoration of old test gear' threads here - cannot unsee. Now I feel guilty every time I look at some of my old gear and realize it could do with cleaning.

[ModemHead]

Interesting thread! I have an old Fluke 8020A and 8022A, which both still work EXCEPT that the 8022A has a couple of the decimal points on the LCD permanently on. I've tried disassembling it, cleaning the LCD edge, the zebra strip and the PCB, reversing the zebra strip, visually checking for any board shorts, etc, but no luck. Without the circuit I couldn't think of anything more to do, and decided it was probably a fault in the main chip anyway.

There's a CD4030 quad XOR gate (U7) that drives the decimal points and the lo-bat annunciator.  The steady-state decimal point logic signals come from the range selector switches and then are XORed with the LCD backplane clock to get the required 50Hz bipolar drive signals for the LCD.

Check that there is 5V or so from U7P7 to U7P14, because Vss (negative supply) for U7 is generated from the ICL7106 and if that failed there may only be about 2.5V or so.

DP1 is U7P13.  DP10 is U7P6.  DP100 is U7P8.  Only one of these signals should be logic high (with respect to U7P7).  If all three are high, then something is wrong with the range selector switches.  If not, then U7 may be bad.

[TerraHertz]

Yay! Fixed. It was the 4030. Quite a strange failure mode - two of the outputs were not XORs of their respective inputs, but a much more complex function of multiple inputs, including some intervals of 'linear' not digital output.  But still roughly square waves as they should be, except at half the frequency of the LCD drive square wave. Some kind of odd internal defect in the chip that managed to produce a div-by-two circuit? Not just blown outputs or inputs as usual with 4000 series logic.

I don't know why I had assumed no schematic would be available for this. Lazy I guess.
Anyway, given the schematic, easy. And I cleaned the case too. Nice to have the old thing back.

[ModemHead]

Excellent.  And nice job on the clean-up. 

[E.Stern]

On disassembly I noticed that the glass over the existing LCD on the 8020A was gripped so tightly by the plastic tabs that it bowed. It puts more pressure on the edges of the LCD via this bowing. Would the point-loading on the edges of the LCD screen be an issue for failure?

Also, with the new LUMEX LCD screen that I'm preparing to order, should I expect the original polarizing glass to fit back in place?

Thanks for the vote of confidence. I have a senior computer tech on hand for the soldering challenge, and I'll post once the job is done.

Cheers from Newfoundland

[lowimpedance]

@TerraHertz - Nice and shiny  , Its nice to see so many of these old timers still chuuging along out there!
 
@E.Stern - If you are referring to the clear acrylic, (perspex), sheet that holds the LCD in place ?! , then when fitting the new LCD (with pins!!) , just make sure this 'perspex sheet' is a good sliding fit and not a tight fit , when modifiying the plastic mount . Most definately does not want to be tight!. Yes its a good idea to retain this window for safety aswell as mechanical protection for the LCD.
good luck
LowZ

[PaulAm]

In a similar vein, I ran across this site:

http://www.spurtikus.de/basteln/repair/fluke_8010a/index.html

where an 8010 was repaired.  It's in German, but google does a pretty fair job of translation.  He has a nice table showing the pins/connections.

[ModemHead]

I recently performed this repair on an 8020A and documented the process.  Photos and write-up are available at http://mrmodemhead.com/blog/fluke-8020a-lcd-replacement/.

[lowimpedance]

I recently performed this repair on an 8020A and documented the process.  Photos and write-up are available at http://mrmodemhead.com/blog/fluke-8020a-lcd-replacement/.

Very nicely documented repairs, not just the 8020a but also the 'others' too. Lots of nice pic's.
I too have an 8062a which suffered the same 'electrolytic' fate as yours. I still have on going prob's. with the elastomeric strip.
 Again , nice web site.

[ModemHead]

Thanks for the inspiration.  And yep, 8060A electrolytic problems seem to be almost guaranteed. Hope you get yours sorted!

[ElektroQuark]

Nice blog ModemHead. I have just bookmarked it.
A suggestion: can you embed the images? Having to load them while reading stops the joy.

[retiredcaps]

Not to get off topic, but I actually prefer the thumbnails.  It gives it a cleaner look and makes for better printing.

[ModemHead]

Nice blog ModemHead. I have just bookmarked it.
A suggestion: can you embed the images? Having to load them while reading stops the joy.

OT: Thanks.  I would hate to de-rail this fine thread with a discussion of my meager blogging skills. Always open to suggestions via PM/email though. Suffice it to say I have not yet discovered a 'perfect' way to present detailed photos and commentary at the same time within the Wordpress framework.

[PA4TIM]

Fixed my 8022 a few years ago. Very easy, only a few euro from Conrad. To bad I could not find a suitable display for a Philips.
http://www.pa4tim.nl/wp-content/uploads/2010/11/DSC_3600.jpg     
http://www.pa4tim.nl/wp-content/uploads/2010/11/DSC_3601.jpg

[PaulAm]

Just finished my 802, that was an easy fix :-)

I used wirewrap wire on LCD pins 1-20, hot glued to the carrier and then soldered to the board contacts so those wires would all fit under the display.  Added stranded wires to the rest of the PCB contacts, mounted the display then soldered the wires to the LCD.  Made for a fairly neat repair.

I'm very happy to get the old soldier back on my bench.