Products > Test Equipment
Old Fluke Multimeters
drtaylor:
I just started work on my 8060A restoration project. I have the 8060s from the eBay auction and a few of my own to fix up. My lab is all set up for a production line operation for 15 8060s and one 8062. I'm keeping notes on each one, condition, functional, rev of the assembly and PCB. Several were fully functional. A few just need their LCD connections cleaned. A few turned on but have obvious problems. Only 1 was completely DOA. Only 1 had plating on the LCD, but it was the 8062. Regardless if they were fully functional, I am recapping all of them. I am also cleaning the two elastomeric connectors, the carbon one for the LCD, and the silver filled one that connects the SM4 to the MAC. Also cleaning the PCB patterns that contact the elastomerics.
My process is to take them all apart. Remove the uC assembly. Remove all the loose plastic parts. Then using my $5 Pace MTM100 (Auction at my former company...what a great thing to have), I'm removing all the electrolytic caps. Most came out easily with a application of flux then using the hollow sucking head of the Pace rework station. The ones that were corroded gave more problems, but I managed to get them out with not too much damage. I used all the tricks of the trade, solder sucker (Edsyn), solder wick, and so on. I've gotten through 9 so far.
After the caps are out, I cleaned the boards thoroughly with 99.9% IPA and an acid brush. Then soaked the whole PCA in a glass tray with clean IPA. Then another acid brush scrub and blowing out the IPA with a low heat air gun. Then I put the cleaned assembly aside to dry for a few more hours. I've only recapped three so far for testing. They got a similar cleaning process after the new caps were put in.
As far as the replacement caps, I've decided to use Polymer Aluminums. These have the advantage of lower ESR and higher ripple current (not really necessary for the 8060), but best of all, due to the organic polymer, they will not spew electrolyte. These 8060s should last for another 30 years. I only had one value I could not fit physically in the position. So I went with the Nichicon UTT for the 22uF/16V part. All the rest are Nichicon RS, RNS or RNU types.
All these parts are in stock at Mouser. These all fit the physical diameter, height, and lead pitch of the 8060A PCB. They do have much higher leakage current on paper, but in practice and at the voltages used, the effect on battery life should be minimal. I've got three units recapped now and ready for testing. I will publish the list later after I'm sure there is no ill effect from the added leakage current. But I do think that will not matter. I'm actually hoping the RMS noise floor will be improved with the new 7660 caps.
I've contacted an old friend at Fluke to see if I can use a calibrator so I can tweak them in. I can calibrate the DC ranges with what I have, but I do not have an accurate AC source at 10kHz which is required to trim the AC freq flatness. There's a real possibility that Fluke has policies against this, but my friend is actually the marketing manager for the Calibration group. SO I'm crossing my fingers on that. If I'm unable to trim the AC, I will let prospective buyers know.
In a side note, one of my 8060s is really a prototype unit. It has a ceramic MAC chip, the RMS converter in a socket, and it has several kludges. The constant current output uses a LM334 and a couple of bodged resistors. There's also a couple of jumpers. This very likely is the oldest 8060 around as it is clear I was still fooling with it. My IBM version 8060A also is an early version, with two jumpers, but it has worked well for years. Most of the eBay units were mid to late 80s with a board rev of H and an assembly rev of H. Some were labeled Rev H-1. No idea what the difference was. I did not get any of the late production units that subbed the Analog Device TRMS chip on a daughter board. All have the Fluke designed Motorola BiFet TRMS converters.
I'm done with my G-Job for now, so I hope to finish these up in a couple of weeks. I'll report progress soon.
WastelandTek:
a couple pics of your production line would be really neat to see doc
Fungus:
--- Quote from: drtaylor on August 10, 2017, 04:02:35 am ---I just started work on my 8060A restoration project. I have the 8060s from the eBay auction and a few of my own to fix up. My lab is all set up for a production line operation for 15 8060s and one 8062. I'm keeping notes on each one, condition, functional, rev of the assembly and PCB. Several were fully functional.
--- End quote ---
:-+
--- Quote from: drtaylor on August 10, 2017, 04:02:35 am ---A few just need their LCD connections cleaned. A few turned on but have obvious problems. Only 1 was completely DOA. Only 1 had plating on the LCD, but it was the 8062. Regardless if they were fully functional, I am recapping all of them. I am also cleaning the two elastomeric connectors, the carbon one for the LCD, and the silver filled one that connects the SM4 to the MAC. Also cleaning the PCB patterns that contact the elastomerics.
--- End quote ---
I would be good to know how to clean LCD connections, elastomeric connectors, etc. Those things scare me.
I only ever took one important LCD apart and the rubber was stuck to the glass really hard. I worried about ripping tracks off the LCD's glass or something. Is that possible?
It all worked out in the end and the faded digits came back to life after I put it back together. I had no idea what I was doing at the time though and I always wonder how close I was to destroying the device.
Also: Are zebra connectors completely generic? ie. If a connector is rotten can you simply replace it with another one of the right size or do they have internal tracks with spacings that need to be matched up?
Let's see if I can find anything on LCD repairs while I'm waiting for the meter to arrive...
frozenfrogz:
--- Quote from: Fungus on August 10, 2017, 09:44:28 am ---I would be good to know how to clean LCD connections, elastomeric connectors, etc. Those things scare me.
I only ever took one important LCD apart and the rubber was stuck to the glass really hard. I worried about ripping tracks off the LCD's glass or something. Is that possible?
It all worked out in the end and the faded digits came back to life after I put it back together. I had no idea what I was doing at the time though and I always wonder how close I was to destroying the device.
Also: Are zebra connectors completely generic? ie. If a connector is rotten can you simply replace it with another one of the right size or do they have internal tracks with spacings that need to be matched up?
--- End quote ---
AFAIK the rubber zebra stripes are just 1-1 conducting strips, meaning generic replacements of the same width should do the job. They consist of conductive and non-conductive strips in an alternating arrangement (usually black being the conductive component). I fully remove the zebra stripe and clean its both conductive surfaces with rubbing alcohol as well as the PCB contacts (might need so additional treatment because of corrosion) and the glass contact area of the LCD. Never had a problem with that :)
Fungus:
--- Quote from: drtaylor on November 04, 2013, 06:10:29 pm ---I only had peripheral contact with the 8050. It used a discrete analog dual slope converter with a custom controller IC. Getting that circuit working well was a trial for the designer. So the 8050 was still a dumb design, no software, and therefore only slightly related to the 8060. I remember studying its circuitry carefully while designing the 8060. I knew the 8050 input design would never support wideband AC readings, so I pretty much went my own way on the input structure of the 8060.
--- End quote ---
I was thinking about looking for an 8050A to go with my 8060A. It's a teeny bit more accurate and I assumed they were related. Now I'm not so sure. Maybe I should get another 8060A instead. It's the one true meter.
PS: I didn't know the blue IBM meters were rare collectors items. There was a very nice one on eBay last week that went cheap(ish) but I didn't buy it because I wanted a real Fluke, none of that IBM rubbish. :-DD
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version