Fluke 87-V mystery - low impedance input

[Robomeds]

So I have a fluke 87-5 with an interesting problem.  The meter seems to have some sort of low impedance input.  The meter is reading DC voltages low by ~1.6V on a 9V battery.  It also seems to read ~30k ohm in resistance mode with open leads.  Well the 30k ohms seems right to me because when I measure the input impedance it comes out to ~30kohm instead of 10M ohm.  I've opened the thing up and it looks to be in great shape.  No signs of anything odd on the circuit board except for perhaps a wee bit of what appears to be extra flux near some components near the screen.  I'll get back to that in a moment.

None of the typical suspects (movs, PTCs, fusible resistors etc) look bad.  I checked the calibration info and the meter says 1 so I can assume it wasn't miscalibrated.  I was wondering if perhaps the low impedance mode which is supposed to only be on in AC reading is perhaps stuck on.  Perhaps that extra flux was actually heat damage from a over heated transistor.

Anyway, I've had no luck finding a schematic for this thing.  I'm open to any and all thoughts on the mater.  I think it would be fun and quite gratifying to fix the thing. 

Thanks!

[alm]

My guess would be one of the MOVs having failed after an overload. Try desoldering them (one leg would be sufficient) and see if that fixes the problem.

[retiredcaps]

None of the typical suspects (movs, PTCs, fusible resistors etc) look bad.

You should get the following resistance readings.

A good MOV should read "0L" in resistance.  A good PTC should read around 1K ohms.  A good fusible resistor should read around 1K ohms.

Don't forget to give it an IPA bath if there are contaminants.

[retiredcaps]

This thread re 87V and MOVs might also help.

https://www.eevblog.com/forum/testgear/fixing-a-fluke-87v/

[Excavatoree]

I had a similar situation - it turned out to be a tiny burned spot on the circuit board between two traces on the input circuitry.   It was difficult to see, so do a good through inspection.   That's rare, however.  I'd check the MOVs first.

[Robomeds]

Thanks all that's very helpful!

I only had a few minutes to look at things yesterday and won't get back to the project until tomorrow.  However, I do recall reading resistance in the 10s of k ohms on the MOVs.  I couldn't remember if those were in series with the input circuit or in parallel.  Sounds like parallel and almost certainly the problem.  I've been trying to find exact or near exact replacements since in the other thread there was a concern that 10mm discs didn't fit.  Based on the picture in the other thread these appear to be EPCOS parts perhaps a model number 05k575.  Any idea where they or another sub 10mm part could be found in the US?

[Robomeds]

OK, I got a chance to check things out.  Definitely the MOVs.  In the other thread there were comments about 10mm discs not fitting.  However, I can't find any 5 or 7mm MOVs rated for over 500V much less the 575.  So looking at it I think I can fit 10mm parts if they are placed carefully.  2 of the MOVs appear to have plenty of vertical clearance.  One is tight but I think it can be tilted sideways to fit. 

[retiredcaps]

For my own education, I looked at various datasheets including the ones from Epcos.  They make a reference that S10 are available in 5.0 mm lead spacing. 

This leads me to believe they make custom parts for manufacturers like Fluke, but not widely available to the general public? With a 5.0 mm lead spacing, it is possible that the diameter is also less than 10.0 mm?

You can try and bend, fit the bigger MOVs, possibly 12.5 mm in diameter according to some datasheets, into place.

One alternative is to desolder the K575 MOVs from other Flukes so you get the proper size for the tight fit 87V.  Then use the bigger diameter replacements in another Fluke where you have more space?

For example, here are pictures of the Fluke 117 from Dave's review.  As you can see, there is a lot more room for the MOVs.

http://www.flickr.com/photos/eevblog/sets/72157626631973147/

PS. My Fluke 113 uses the same 05, K575 MOVs from Epcos.  Diameter measures less than 10.0 mm.

[Robomeds]

So I have a follow up to this question.  With the change of the MOVs I thought everything was fine.  The meter seems to work well in all ranges but for the lowest resistance ranges.  A 100 ohm resistor reads as 88 ohms.  To verify things I removed all three MOVs to make sure they didn't affect the readings. 

Any idea why I would get what appear to be good readings on higher resistance ranges but I'm off on the low range. 
Thanks!

[retiredcaps]

Any idea why I would get what appear to be good readings on higher resistance ranges but I'm off on the low range.

I bought a "working" 77 III and intially everything checked out.  Two months later, I was measuring a 100 ohm resistor and noticed that the 77 III would not read a stable 100 ohms. 

In fact, the reading of a fluctuating 98 ohm was out of calibration which was a complete surprise to me.  I had never seen an used/abused Fluke out of calibration.  I checked the PTC and it was 110 kohm instead of 1 kohm.  I replaced it temporarily with a 1k ohm 5% carbon resistor and I got a solid 100.1 ohm reading.

[ModemHead]

With the 87V in manual 600 ohm range, it should be able to source 1mA through a short circuit.  If it can't, then the PTC and fusible resistor would definitely be suspect.  As a benchmark, the 87V I have here measures 1.005mA (100 ohm shunt) with an open-circuit voltage of 7.317V.

[Robomeds]

retiredcaps and ModemHead,
Thanks to both of you for the help.  Well the 600 ohm range current is low (86mA).  When I switched to the 6000 ohm range the 100 ohm reading was spot on.  So I'm going to borrow an image here...

I'm assuming the large green resistor at the edge of the board (brown-black-red-silver) is my fusible resistor.  In my meter (as installed) it reads 965 ohms.  I'm assuming the error is just a parallel circuit.  However, my black thermistor (green in the picture, presumably the PTC) reads 4kohm.  Checking a different meter I see 1.2kohm across the PTC.  Looks like the PTC might be the ticket. 

I will let you know.

Thanks!

[retiredcaps]

I'm assuming the large green resistor at the edge of the board (brown-black-red-silver) is my fusible resistor.

Yes.  It is present in pretty much every Fluke multimeter.

In my meter (as installed) it reads 965 ohms.

That is fine.

However, my black thermistor (green in the picture, presumably the PTC) reads 4kohm.

Yes. 4kohm is bad for the PTC.

In my initial response to you, I listed all the readings for what good input protection components should measure.  Whenever I get a new to me used/abused multimeter, I always take 3 minutes to measure all the input protection components especially if a reading is obviously off.  They can all be done "in circuit" and if any readings are suspicious, then I verify them out of circuit.

[Robomeds]

Yes, you are right, I missed that when I was trying to figure out my resistance readings.  With a 1k ohm temporary resistor all is well so I just need to get a new PTC for the thing.

Thanks for all the help!

[retiredcaps]

With a 1k ohm temporary resistor all is well so I just need to get a new PTC for the thing.

When I was looking for a replacement, I tried to find a 1.1kohm, 1000V PTC and was unsuccessful.

I believe the GE YS4020 (1.1kohm, 1000V) is a suitable replacement, but there is no stock.

http://datasheet.octopart.com/YS4020-Thermometrics-datasheet-11237.pdf

The closest I found was a Epcos (1.1kohm 550V) in stock at mouser, but 550V is too low.

http://ca.mouser.com/ProductDetail/EPCOS/B59774B115A70/?qs=sGAEpiMZZMvQRzVR7o8pId1XpME%2fcxbqDjzSCXaTPOE%3d

My "replacement" came from a wrecked unrepairable Fluke 177.  If you find a replacement, let us know.

[Excavatoree]

Yes, you are right, I missed that when I was trying to figure out my resistance readings.  With a 1k ohm temporary resistor all is well so I just need to get a new PTC for the thing.

Thanks for all the help!

If you or retiredcaps can get a picture of the PCT you need, I may have one.  I'm not sure which meters use which, so I'd have to identify by sight.

[AndersAnd]

I'm assuming the large green resistor at the edge of the board (brown-black-red-silver) is my fusible resistor.  In my meter (as installed) it reads 965 ohms.  I'm assuming the error is just a parallel circuit.

There doesn't have to be any parallel circuit for that 1 kohm resistor to read 965 ohm.
The silver band means it's only a ±10% resistor. Everything between 900 - 1100 ohm is within specs for a 1000 ohm ±10% resistor, so 965 ohm is well within specs.
Even for a ±5% resistor 965 ohm would still be within specs.

[retiredcaps]

I'm not sure which meters use which, so I'd have to identify by sight.

For my own learning purposes, I went back to the original 8020A manual/schematic/BOM and it has a RT1 (1kohm thermistor).  What is not clear or explained is the rated voltage of the thermistor.

The same goes for the original 70 series and series II.  It isn't until the 77 III that the schematic actually says that RT1 is a 1.1kohm 20% 1000V.

By that reasoning, I would say any current selling multimeter like the 113-117, 77 IV/170 series, 80 series V, etc are likely to have the 1000V rated thermistor.

I found a link, but I keep getting redirected to the .ca site instead of the .com site that may have the YS4020 in stock for $3.07 CDN (I have no idea what is minimum order/quantity)?  USA residents may want to try and see?

http://www.plccenter.ca/en-CA/Buy/THERMOMETRICS/YS4020

PS. The thermistors found in the 8xxx, original 70 and series II look physically different from the one used in the 87V.  They are smaller in size and more rectangular vs the fat blob disc.

[Robomeds]

I was able to pull a PTC out of a dead 113 at work.  Problem solved!