Flaky Fluke 175

[torch]

Is there a way to force zero adjustment on the ohms setting for this meter?

Lately it's been acting weird -- the minimum ohms displayed seems to creep upwards. Today it was at 140. I tried a new battery. Then I decided to try some contact cleaner on the rotary switch. I took the board out and it was sitting at <1 ohm when I reconnected things. I assembled it and it gradually started creeping upwards again. That is to say, every time I turn it off then back on, the new minimum is higher. At about 90 ohms, I took it apart again. Same thing -- touching the back of the circuit board seems to reset things, but on reassembly it starts creeping upwards. Almost like a cap was holding a charge, but it will continue to creep even when shut off and the battery disconnected.

I hate to toss a Fluke, but this is ridiculous.

[retiredcaps]

Is there a way to force zero adjustment on the ohms setting for this meter?

The Fluke 17x series is all closed case calibration.  There is no pot to turn to zero and there is no relative function.

Lately it's been acting weird -- the minimum ohms displayed seems to creep upwards.

Suggestions.

1) A dirty or contaminated pcb could be causing leakage somewhere and affecting the readings.  You can give the pcb an IPA bath.  Let it dry thoroughly before re-assembly and testing.

2) Clean out the input jacks with a q-tip dipped in IPA.  Clean until the q-tip is white.

3) Resolder the COM and Volt/ohm input jacks.  Sometimes they develop solder cracks which are not easily detectable to the naked eye.  Modemhead, a member here, has reflowed the input jacks on a number of these.

4) Check the PTC.  It should be 1k ohm.  It might have been subjected to high energy and might read significantly higher than 1k ohm.

I hate to toss a Fluke, but this is ridiculous.

If you are going to "toss" it, send me a PM first.   

[retiredcaps]

I forgot to ask

5) Does DCV work properly?
a) If you short the probes, do you get 0.000V?
b) If you measure a 9V battery, do you get roughly 9V?

6) Does resistance work properly?
a) If you short the probes, do you get a stable and steady 0.2 ohms?
b) If you measure a 1k ohm resistor, do you get a stable and steady 1k ohm reading?

[torch]

I forgot to ask

5) Does DCV work properly?
a) If you short the probes, do you get 0.000V?
b) If you measure a 9V battery, do you get roughly 9V?

5a) Yes.
5b) Yes. Compared to a cheap Mastercraft (Canadian Tire store house brand -- made somewhere in China) autoranging meter, the two agree within 0.005volts @ 1.5 VDC and within 0.03 volts at 31 VDC from my bench power supply. On a fresh 9v battery, they are within 0.02 volts. No difference if I swap the probes from one meter to the other or exchange + and -.

6) Does resistance work properly?
a) If you short the probes, do you get a stable and steady 0.2 ohms?
b) If you measure a 1k ohm resistor, do you get a stable and steady 1k ohm reading?

6a) No. That's the problem. At the moment shorting the probes starts off at about 24 ohms and slowly (over +/- 30 seconds) drops to stabilize at 19.4 ohms. Next time I turn it on, it might read 40 or 50 ohms. I've seen it read as high as 140 ohms with the leads shorted.

One other oddity: when it's behaving (ie: immediately after dissassembly) open leads will show "OL" on Mohm scale. When it's acting up, open leads will show a high, but not infinite, floating value (at the moment, floating from 48 to 54 Mohm).

6b) No. At the moment a 1 ohm resistor is reading 20.4 ohms, and again, it took about 30 seconds of slowly dropping to stabilize at that. (The cheap meter shows 1.1 ohms, rising from 0 within 2- 3 seconds.) A 1Kohm 1% resistor (1.000 Kohm on the cheap meter, 1.005Kohm on a Karl Heinze type component tester) reads 1.087Kohms when shorting the leads produces .091Kohms (manual range selected).

I'll try the alcohol bath. PTC? As in a thermistor?

[rsjsouza]

A while ago my 8060A was showing a similar behaviour on lower resistance ranges a short while after turning the DMM on, and it turned out to be a small crack on one of the leads of the Caddock resistor (heat was probably causing the contact failure). Perhaps the same is happening to you?

This agrees with the behaviour where the reassembly will make it fail. That and a cold solder joint or a small crack on the PCB.

You could also try to plug your Mastercraft as a voltmeter into the 175 jacks and see how badly the voltage fluctuates, which could confirm the error on the resistor or perhaps on the DMM's voltage regulator. 

My 179 reads (just for reference, but it may help):
600ohm: 7,315V
6kohm: 7,304V
60kohm: 7,273V
600kohm: 3,498V
6Mohm: 5,276V
60Mohm: 2,776V 

[torch]

My 179 reads (just for reference, but it may help):
600ohm: 7,315V
6kohm: 7,304V
60kohm: 7,273V
600kohm: 3,498V
6Mohm: 5,276V
60Mohm: 2,776V

Mine are considerably lower:

600ohm: 1.247V
6kohm: 1.245V
60kohm: 1.232V
600kohm: 1.112V
6Mohm: 0.680V
60Mohm: 0.678V

Odd, as I thought the 17x series were essentially the same, differing only in small features such as backlight and temperature measurement.

(Interesting aside: the cheap Mastercraft ranges from 0.390 to 0.440, making it a better choice than the expensive Fluke for in-circuit testing as described in Dave's recent episode on reverse engineering the Rigol DS1054z)

[ModemHead]

Definitely check if those input jacks are secure in the PCB. Try hard to rotate them with your fingers, if you feel the slightest bit of rotation, they're loose.

I've run into this problem quite often with the "tapered" style Fluke meters.  Seems like a weakness to me.  I believe the jacks are swaged in before wave soldering, but somehow they get loose with age.  Hardly affects voltage readings, but shows up intermittently on low ohms readings.

A full IPA bath wouldn't hurt either.  Just dunk the whole thing minus the LCD assembly and brush it, then allow to dry thoroughly.  Check the fine-pitch chips to make sure there's no foreign material between the pins.  Be on the lookout for contamination like in the attached photo of a 77-III board.

A observation on the OC voltage on Ohms: I think it goes with the board revision (175, 177 and 179 use the same "17x" board.)  Here I have a 179 with a rev 7 board, a 179 with a rev 12 board, and a 177 with a rev 15 board.  The two older ones have OC voltages that match torch's low readings, while the later rev board in the 177 matches rsjsouza's higher readings.

[retiredcaps]

Compared to a cheap Mastercraft (Canadian Tire store house brand -- made somewhere in China)

Some Canadian Tire multimeters are made by Colluck.

http://www.colluck.com.hk/diyhp/9359/enus/l25520-0/Multimeters.html

PTC? As in a thermistor?

Yes, it is right above the volt/ohm input jack. It should read around 1k or 1.1k ohm.

[rsjsouza]

As Modemhead said, my board is Rev 15 - check the teardown I did below, where you can perhaps also compare other details.

https://www.eevblog.com/forum/testgear/fluke-179-teardown-photos/

(off-topic) Modemhead, I left an entry at your homepage about the LCD replacement procedure that helped me fix my dad's 8020A LCD. The final assembly was not as tidy as yours, but it is working like a horse! Thanks again.

[retiredcaps]

A while ago my 8060A was showing a similar behaviour on lower resistance ranges a short while after turning the DMM on, and it turned out to be a small crack on one of the leads of the Caddock resistor (heat was probably causing the contact failure).

Modemhead found a very obvious problem with the Caddock resistor network in this repair below.

http://mrmodemhead.com/blog/beckman-industrial-hd110-multimeter/

[retiredcaps]

Odd, as I thought the 17x series were essentially the same, differing only in small features such as backlight and temperature measurement.

As modemhead notes, we have seen 3 revisions of the PCB for the 17x series since its introduction.  The biggest changes between rev 12 and 15 are the rotary switch and the main ICs.

My guess is that rev 15 can be identified when you turn the meter on and see it say F175.  Rev 12 won't display that.

[torch]

Some Canadian Tire multimeters are made by Colluck.

http://www.colluck.com.hk/diyhp/9359/enus/l25520-0/Multimeters.html

Hmmm. Well it does look very much like their HH2205L.

I'm away from home until Wednesday, but I'll check that thermistor out and try the bath then.

[torch]

Epilogue:

The PTC was fine. The posts all secure with 0 resistance or movement when poked, prodded and torqued. I cleaned the heck out of them with Q-tips. There was some sort of contamination around 1 via in the board. Following the IPA bath, brushing and rinsing, the contamination was gone, together with the green lacquer covering a neighbouring trace that the contamination had extended over.

After thorough drying, I reassembled it and it seems to be functioning fine now. Resistance is consistently infinite with no leads connected (previously floating Mohm value) and 0.2 ohms with the leads crossed (previously variable, up to 140 or 150 ohms). 0.1 ohms when the posts were shorted with a piece of silver solder.

So the IPA bath was apparently the cure. Thanks to all for your suggestions.

(Sorry retiredcaps. No free Flukes here today  )

[retiredcaps]

(Sorry retiredcaps. No free Flukes here today  )

Sometimes, I am my own worst enemy at getting used Fluke multimeters at a very low price!    Especially in Canada, they are very difficult to find at a decent used price.

I am glad you got your Fluke 175 working again. 

[torch]

Especially in Canada, they are very difficult to find at a decent used price.

Unfortunately just about everything is difficult to find at a decent price in Canada. Especially shipping. It probably costs more to ship a dead Fluke across Canada than it does to buy a new one south of the border.

(Ok, maybe that's a slight exaggeration. But only slight.)

[markce]

Do you by any chance happen to use TL175 test leads? If there is a 4mm spring contact
on it, you need to fasten the screw. This caused the same kind of problem on my Fluke 175.

[torch]

The ones surrounded by the fake/Chinese market controversy? No. I'm using the leads that came with the meter. I'm not sure of the model number but they're Fluke brand labelled "CAT III 1000v" and a certification emblem (UL? Don't remember off hand).

Why? Is there additional controversy about that model?

[retiredcaps]

I'm using the leads that came with the meter. I'm not sure of the model number but they're Fluke brand labelled "CAT III 1000v" and a certification emblem (UL? Don't remember off hand).

They are likely the TL75 leads.

[markce]

I had to replace the original test leads. They became unreliable after stumbling over them.
There seems to be a crimp connection inside, and if you press the leads close to the probe the contact
changed (at least with mine). I replaced them with what was priced and advertised as genuine Fluke
test leads (TwistGuard red-black TL175E from reichelt.de).