FLUKE 189 DC reading problem [with video]

[joeqsmith]

So pretty much everything is bad and in the low side ... the schematic for this meters I don't think it's available but you should try to find and measure the precision voltage reference and power supplies for the analog part ,  before condemning the ASIC chip itself ...
The input protection can't explain the current measurement being off ...

After seeing the second video, it behaves different from what I first thought.  With the data you posted,  I concur  with the above post in that probing the reference would be the first step.  I wasn't able to find a schematic for it but it shouldn't be too difficult to trace this area out.   Just takes time.  Maybe just have a look around that REF43 under a microscope.   Someone had posted some decent pictures here:


https://www.eevblog.com/forum/testgear/fluke-187-189-or-87-iv-schematic/

[kamcm]

So pretty much everything is bad and in the low side ... the schematic for this meters I don't think it's available but you should try to find and measure the precision voltage reference and power supplies for the analog part ,  before condemning the ASIC chip itself ...
The input protection can't explain the current measurement being off ...

After seeing the second video, it behaves different from what I first thought.  With the data you posted,  I concur  with the above post in that probing the reference would be the first step.  I wasn't able to find a schematic for it but it shouldn't be too difficult to trace this area out.   Just takes time.  Maybe just have a look around that REF43 under a microscope.   Someone had posted some decent pictures here:


https://www.eevblog.com/forum/testgear/fluke-187-189-or-87-iv-schematic/

I had check out the REF43 area and cannot find anything wrong under a microscope.



I have also test how much current my Fluke 189 draw from the battery. Here is the result.













With the Fluke 189 power off, I still able to measure 55.9M between COM and V socket under 1000V testing.   
Anyway, I just order some IPA from RS and I am going to clean the board completely with IPA.

[joeqsmith]

I doubt cleaning the board will do anything for you.  The meter's case looks in very good condition, at least it doesn't appear to have been abused too bad.  I would get another meter and start probing that reference.  Obviously, you will need a decent meter to do this.   Data sheets are on-line for it. 

[joeqsmith]

For a reference, the attached graph is looking at the current draw of a working Fluke 189 under various conditions.   Full disclosure,  this data was collected with a home made meter I have been playing around with but I don't see anything strange with yours.    If you want more details about this testing, goto the link in my signature.  There is a TOC on the first page. 

Interesting enough, I am playing with this home made source meter now and am using a low cost Brymen BM869s to detect the trends.   It has a 500,000 count mode which has worked out fairly well for this.   

[kamcm]

I just cleaned the whole board (both side) with IPA 





But the problem remain the same. 



I also noted that when the reading is not correct, if I press the range button to get the reading under manual range, the reading will go crazy out of range.

Which components are likely failed ? Is it the resistance network? how can I test it out ?

Please advice,

Thanks!

[windsmurf]

Just some simple things to check on a used meter if it has measurement issues...

Inspect all solder joints
Clean Dial contacts - remove old dielectric grease from the dial contacts with IPA and swab, both sides (board and slider)
Check the PTC, MOV, and input protection resistors
Clean input jack with IPA/swab, check solder for solid electrical connection.  The pin under the jack are sometimes broken.
Check the resistor network for solid contact (both the board and on the resistor itself).  I repaired one that had intermittent connection issues.

[windsmurf]

Is your 87V calibrated?  I measure current draw of about 3-5mV less on my 189 than what you get on yours.
Also, 55MOhms at 1000v I think is normal... I get about 48MOhms on mine at 1000V.


I would thoroughly clean your dial contacts on the board as well as the spring contacts under the rotary... and clean the spaces between the dial contact traces.   Not just a rinse with IPA but rub well with a QTip.  Don't lose those springs as they can pop off and fly away...
I might even try swapping the spring contact placements to see if that changes things.

Do you get strange reading with no input/nothing connected to the meter when switching modes?

[kamcm]

Just some simple things to check on a used meter if it has measurement issues...

Inspect all solder joints
Clean Dial contacts - remove old dielectric grease from the dial contacts with IPA and swab, both sides (board and slider)
Check the PTC, MOV, and input protection resistors
Clean input jack with IPA/swab, check solder for solid electrical connection.  The pin under the jack are sometimes broken.
Check the resistor network for solid contact (both the board and on the resistor itself).  I repaired one that had intermittent connection issues.

Thanks a lot. I have checked all of the components you mention. Everything looks good.

I want to check the exact DC voltage go in FLUKE 669918, but I don't know which pin I should probe because no datasheet is available.
It would be helpful if someone can tell me the exact pin of DC voltage go in FLUKE 669918.

Thanks!

[CDaniel]

So , just to be sure , there is any range for volt , milivolt , ohm , cap , current that is 100% ok ?
I have a Fluke 187 with ( as far I can see ) identical board so I could check some voltages .

[kamcm]

So , just to be sure , there is any range for volt , milivolt , ohm , cap , current that is 100% ok ?
I have a Fluke 187 with ( as far I can see ) identical board so I could check some voltages .

At the time it behave normally, all the function works well and very accuracy.
But most of the time. The reading just off the spec. 
https://www.eevblog.com/forum/testgear/fluke-189-dc-reading-problem-(with-video)/msg2548869/#msg2548869
The testing was done with fluke 702. I knew My 702 is accuracy and works well, I verify it with a few meters including 289 and 87V 87III etc.

I think the incorrect reading of mA has somethings interesting. In some range the reading just higher than actual.



Actual: 10mA
Measure: 10.072mA  (Get Higher than actual)

Actual: 5mA   
Measure: 5.214mA    (Get Higher than actual)

Actual: 4mA
Measure: 4.062mA   (Get Higher than actual)

Actual 3mA
Measure: 2.763mA (Get Lower than actual)

Actual 2mA
Measure: 0.076mA (Get Lower than actual)

Actual 1mA
Measure: 0.038mA (Get Lower than actual)

[windsmurf]

It really doesn't look like its anything in the digital electronics.

I still suspect something around the dial, as Joe mentioned, and possibly something in the input. 

You might try swapping out the PTC to make sure it isn't having strange intermittent issues.... swap out with resistor temporarily if you don't have a spare.


   

[CDaniel]

Current input should be completely independent from V/ohm/cap and has no protection ... so it's not the input PTC , MOV and so on , as I thougth myself initially
I asked before if the analog voltages are ok , besides the precision reference that ASIC should have dual power supply for the analog internal Op Amps . Inside is the ADC and the True rms converter  . Other Fluke multimeters have those as external IC's , much easier to troubleshoot even without the schematic.

[IconicPCB]

In post number nine on the first page of this topic a photo shows some MOVs.
I need to identify and source a MOV from the manufacturer of the RED MOV.

Please advise the name of the manufacturer if known.

[BigB_52]

Hi,
I have a Fluke 189 which was in storage for a long time with batteries in, when I took it out the battery was corroded, I cleaned the battery cartridge, placed new batteries and turned it on the meter was running for itself without measuring anything, probs were in the air and no matter where the selector switch was. I opened the meter to inspect inside found nothing wrong, no battery liquid leaked and it was supper clean, I changed the supper cap, nothing changed in the result, I sent it for calibration they reported “the meter can not register calibration constants.
I tested the eprom 93c56 that holds the calibration constants, it is fine,I can read from and write to it with no problem with an external eprom programmer.
Since it came back from calibration facility, the UNCAL mark is on the screen All the time, and the mere shows 0000 in all conditions, even when I activate calibration mode the measurement is zero, the measurement does not follow the input to match the required voltage calibration menu,
Any one has any thought in that?

Thanks
BM