Fluke 189 Problem

[Insatman]

I have an early fluke 189 that I bought new ages ago.  After being in storage for several years I fired it up recently.   The unit powers up but all the readings are wrong.  I don't mean slightly off...I mean way wrong.   Volts just reads around zero.  Ohms is all over the map.   All the ranges are out.  I took the unit apart and everything looks perfect.   No damage in the front end.   I used another meter to check the signal paths up to the rotary switch and it all checks out.   The supercap looks good.   The meter draws about 8-14mA without the backlight.   The unit senses if the probes are not in position for current mode just fine.   The inside was very clean.   I'm stumped.

I don't have any documentation on the purchase...I lost that long ago.   Any ideas?

Dave

[Electro Detective]

Long shot:

1. Take out the batteries and leave them out for 24 hours

2. If that fails go into the meter menus (if possible) and do a factory reset
I think it's called FCTRY or something like that.

3. Delete the logging memory thing, the last dial setting gets you there I think

3. Invisible battery leakage, check everywhere!

4. Take out the batteries and leave them out for 24 hours, second time around

5. Good luck  the meter is worth saving btw,  very accurate
and has THE fastest continuity buzzer on Earth, kills all the other Flukes too!
I openly dare/challenge anyone to deny it, or demonstrate a faster continuity buzzer on a high spec multimeter 

[retiredcaps]

Volts just reads around zero.  Ohms is all over the map.   All the ranges are out.

My guess is that this meter was potentially subjected to high voltage/abuse.  I suggest checking the input protection components like the fusible resistor, PTC and MOVs.

Modemhead's blog shows you how to measure the fusible resistor (should be around 1k ohm)

http://mrmodemhead.com/blog/fluke-87-fusible-resistor/

A PTC should be around 1.1k ohm and the MOVs should read 0L, infinite resistance.

If you need help identifying these components, post a clear focused picture of your 189 pcb and we can point them out.

[retiredcaps]

5. Good luck  the meter is worth saving btw,  very accurate
and has THE fastest continuity buzzer on Earth, kills all the other Flukes too!
I openly dare/challenge anyone to deny it, or demonstrate a faster continuity buzzer on a high spec multimeter 

Modemhead has done some unofficial and unpublished tests, but the Fluke 8024B, 8060A, 8021B, 8020B are faster than the 189 in terms on continuity.

The Tektronix TX1, which later became the Fluke 183, is also slightly faster than the 189.  The TX3 later became the Fluke 185.  See

https://www.eevblog.com/forum/testgear/fluke-185-review/

As for high spec, the 8060A, though 30+ years old, is still a decent spec meter in 2017.

https://www.eevblog.com/forum/testgear/old-fluke-multimeters/

[retiredcaps]

I don't have any documentation on the purchase...I lost that long ago.   Any ideas?

If you bought this from an authorized Fluke dealer after 2007, you may qualify for Fluke's "lifetime warranty" defined as a minimum of 10 years.  If you bought it before 2007, you might get lucky and Fluke will repair/replace it for free, but it is the luck of the draw.

Regardless, it is easy to check the input protection components first and report your findings.

[Electro Detective]

5. Good luck  the meter is worth saving btw,  very accurate
and has THE fastest continuity buzzer on Earth, kills all the other Flukes too!
I openly dare/challenge anyone to deny it, or demonstrate a faster continuity buzzer on a high spec multimeter 

Modemhead has done some unofficial and unpublished tests, but the Fluke 8024B, 8060A, 8021B, 8020B are faster than the 189 in terms on continuity.

The Tektronix TX1, which later became the Fluke 183, is also slightly faster than the 189.  The TX3 later became the Fluke 185.  See

https://www.eevblog.com/forum/testgear/fluke-185-review/

As for high spec, the 8060A, though 30+ years old, is still a decent spec meter in 2017.

https://www.eevblog.com/forum/testgear/old-fluke-multimeters/

I have an 8060A still in near mint condition posted at the EEVblog 8060A pages, and yes it is FAST! it comes in second place only to same vintage Fluke 87,  and currently 87V and APPA 99111 in my humble inventory.

BUT, the 189 is so fast    I can tap percussive tunes on the probes (ala Buddy Rich, Gene Krupa) and will not miss a beat at any speed or repetition, each beep is distinct and not a latch effect due to its (sometimes very annoying) chirp tone buzzer.
Continuity buzzer checks are blazingly fast with this meter and on the money.

I think I got a freaky one TBH  otherwise I would not issue a 'day at the races' meter buzzer challenge 

[Insatman]

I checked all the front end components.  I inspected the board serveral times under magnification.  The board has been cleaned with tuner cleaner a couple of times as well.  No fuseable component as indicated was found.  Pictures attached.

It is possible the unit was cleaned with a more agressive solvent than tuner cleaner years ago.  Batteries in the compartment leaked when the unit was stored for several years.  After cleaning the battery compartment no contamination of the circuit board was noticed at that time.  A C cell battery kit from Fluke was installed...mainly because I just happened to have one on hand and the AA cell terminals were in bad shape.   I don't remember if I tested the unit back then...it was over five years ago.   Since then I stored the unit without batteries.   Now I'm setting up an electronics lab and was hoping to use the 189.

I tired leaving the unit without batteries overnight.  I tried the factory reset and clearing the memory.   No change in behavior.

[Insatman]

A couple more pics

[coromonadalix]

Use a flux remover, it will degrease the pcb and remove any film or residues left

the bottom casing  have the post cracked ??,  maybe you should try to find a spare casing, it would surely help,   and test the small backup batterie on the pcb   maybe it's empty

the board seems very clean, nothing  burnt or physically damaged ??

[Electro Detective]

All that comes to mind is checking more usual suspects:
the selector dial, tracks and contacts,
severe gunk in the lead inputs (try a different set of leads too) 
check the soldering on the inputs
the small memory backup supercrapacitor is shorted (?)

put on a magnifier visor and check for bits of crusty battery debris, you would surprised where the lil pos can find a new home 

or the unit battery acid shorted itself into a trashed 'calibration mode' state

[Muttley Snickers]

From memory the UCAL shown in the display is indicating the the calibration data has been lost or corrupted, I strongly suspect that an adjustment or calibration of the meter is now required, slightly strange for it to be lost for no apparent reason. I also have a couple of these meters and they are still as accurate as the day they were bought and probably one of my favorite and most trusted meters. I've posted the service manual below in case you do not already have it

[Electro Detective]

From memory the UCAL shown in the display is indicating the the calibration data has been lost or corrupted, I strongly suspect that an adjustment or calibration of the meter is now required, slightly strange for it to be lost for no apparent reason. I also have a couple of these meters and they are still as accurate as the day they were bought and probably one of my favorite and most trusted meters. I've posted the service manual below in case you do not already have it

Mine was purchased used, forgotten and unloved,
with a battery leakage invasion that would have made any camera lens shudder to take pics,

After a board clean up, had to file all the battery terminals to bare metal, to remove corrosion and get life back 

Lubed the terminals, switched it on, reset everything, compared it to an 87V and 289, all specs bang on the money    

I haven't touched it since, it just always works
(and if it didn't, flashing it around would still get me into any job site without presenting ID lol) 

It's a kickass meter with the lot, easy to use, and the display isn't too shabby either

Sort out the UCAL biz asap, see if its related to anything on the board, a short, corrosion etc 

[Insatman]

I'm to blame for the Uncal message.  In my desperation to get the thing working I pushed the cal button in the back.  Dumb I know.   

I've checked and rechecked the meter for damage and junk everywhere.  It's been thoroughly cleaned with tuner cleaner 3 separate times.  I can find no sign of damage.  I've even tried powering up the meter from a bench supply.   Yes I've tried alternate leads.   Aside from hanging a magic crystal over it, I've tried everything I can think of.   Yes the meter will need a calibration...where to get that done in the Philippines where I live now is not obvious.   I can buy a used 289 here for about $400 and a new one for maybe 5-600.   If the repair/cal is more than $250 it's likely not worth it.   Right now I don't even have a place to get a quote from.





 

[Electro Detective]

That's a tough one   

People like this model, maybe sell it here, or Ebay as it is, and move on.

[frenky]

I am interested in buying cheap Fluke 187 or 189 so I could write opensource PC software for it.
I don't care if all measurements are wrong...

[Fungus]

Yes the meter will need a calibration...where to get that done in the Philippines where I live now is not obvious.   I can buy a used 289 here for about $400 and a new one for maybe 5-600.   If the repair/cal is more than $250 it's likely not worth it.   Right now I don't even have a place to get a quote from.

I just had look at the calibration manual. Some Flukes have quite a simple calibration procedure. All you need is a good voltage reference and you can do a reasonable job at home. It's a fun weekend project for an EE. 

Not the 189 though, it has pages and pages listing different references needed for a full calibration. 

[KrzysztofB]

I would advise you to check if the sockets are ok.
On the one i have one of contacts was broken after some time. Yours is suppose to be almost new, but still worth checking.
And if you fix it, i'm working on 5$ bluetooth hack for it. Actually is a matter of polishing Android app.

[FrenchieRaf]

I would check the voltages on the caps,plus the output voltage of the voltage ref (ref43?)
I did a little bit of reverse engineering on it some time ago, and yes, I love this meter too.
There are 3-4 voltages I think, something like +5,-5,+3.3 from the transformer and ldos if I remember.

What is the 5usd Bluetooth ?

[MosherIV]

RetiredCaps suggesred you check the input protection components MOVs, fuseable resistor, PTCs etc

I checked all the front end components. I inspected the board serveral times under magnification.

Not visually, electrially. From the sounds of it, the MOVs may be dammaged. I think they should be in the order of 10K or more
Another possibility is the overvolt transistors, watch Daves latest video he does a tutorial on how back to back bjt are used to protect against over volt.

Good luck with fix it

[FrenchieRaf]

Hi, just in case it can help someone (and if Insatman is still around), the image is a little bit heavy but here are some voltages on the psu; (plus some of the ir interface).
If you are missing only the negative rail I guess it could behave strangely, Worth to check.
Else to make a little plan that would lead you closer to the error I would say that you should put say 5 or 10 volts on the input in volt mode, and see how much you get at the input of the adc, compare to a working one (I guess I could help). And check the ref43 output. For this we must find the adc input.
This to see if the problem is on the path to the adc or in the adc.

[FrenchieRaf]

Ah you must download and open the image to get full resolution it seems, else no full detail...
By the way, I do not think it is very likely to get only one missing voltage, but who knows.

Back then,I also did a bit of analysis of the backlight circuit onboard if someone is interested, powered between + and - rails, a few transistors and diodes in a quite unusual way, draws quite a bit of current (70 ma total on batteries if I recall with second level, could be improved I think)

KrzysztofB , if you  are around what is the 5 usd Bluetooth?

[frenky]

You can add 5$ bluetooth/serial module to transmit data to PC or phone.
This works with all Flukes that have IR transmitter (87IV, 89IV, 187, 189, 287, 289)...

https://www.eevblog.com/forum/testgear/fluke-87iv-please-whisper-in-my-ir-p/msg1193472/#msg1193472