Saving yet another Fluke 189

[joeqsmith]

Picked up a cheap, dead, Fluke 189.  The plastic is in poor condition.  Looks like it was dipped in chemicals and abandoned.  I suspect this weakened the plastic and it there are a lot of cracks and chips.   I have a 189 I had bought dead for the purpose of transient testing.   A few of the plastics parts on that meter were in better condition, so I swapped them out.   All of the standoffs on the front cover were broke off and split.   I used an inverted cone cutter with the dremel tool to cut these down to a flat surface and inserted brass standoffs into them, which I then covered with heat shrink.  The ones towards the leads, I glued to the sidewall of the case.   I replaced the supercap and bad battery clips.  There was a small crack in the lens was all, so just some light polishing.     

[joeqsmith]

Didn't get so lucky.  The terminal block was damaged as well.  All 4 pins for the common and voltage inputs were broke off.   For now, I stole the terminal block from my beat up meter until I can get a replacement.   The new meter is the one reading a count low.   Doing a basic checkout, it doesn't appear anyone tried to muck with the alignment. 

[joeqsmith]

After posting about the instability problem I ran into with the 121GW, someone commented how the BM869s gives the wrong result for I assume ACmA (never heard of resistance voltage units).   

Showing the repaired 189 with the 789 and both BM869s meters.  None of these are in cal but suggests the new Fluke is alright.  Also that our viewer is clueless.  I would imagine if we used a signal with high frequency, they would read different.   Maybe they don't understand that.  I'm showing a 60Hz sinewave. 

[joeqsmith]

Appears there are none of the OEM terminal blocks available.  There are two different red ones.   Has anyone purchased one of these?  Do they use rubber like the OEM parts?  Sealed?  Crap metal terminals from China?   

https://www.ebay.com/itm/234863848071

https://www.ebay.com/itm/265685003138

[daisizhou]

Have you carefully checked and cleaned the 4 terminal blocks?
You must carefully clean the bottom of each terminal,It is recommended to soak with 95% ethanol after cleaning and rinse each terminal with a syringe.After all steps are completed, they must be dried and then checked to see if the data is consistent.
If there are still errors, follow the steps above and try again.

I think it makes no practical sense for you to replace 4 terminal blocks.
First of all, it is difficult to buy brand new terminal blocks, and secondly, counterfeit terminals are not as reliable as repairing old ones.

[joeqsmith]

It wasn't too hard to spot all the broken pins.  All of the fractures were flush with the board.  These are I assume welded or molded together and I don't see a good way to repair them and maintain the seal.  You would need to locate the entire post assembly.  The metal pins are molded into the post.   Press out the old ones and press in the new ones.  Then maybe glue it together.   Would rather just replace the entire assembly then try and cobble something together. 

I went ahead and pressed out a good and bad pin to show you how they go together.   

[bdunham7]

Appears there are none of the OEM terminal blocks available.  There are two different red ones.   Has anyone purchased one of these?  Do they use rubber like the OEM parts?  Sealed?  Crap metal terminals from China?   

https://www.ebay.com/itm/234863848071

https://www.ebay.com/itm/265685003138

fluke-usa is a 100% legit seller.  When Fluke ceases support, they clean out their inventory and somebody gets it.  I have tracked various Fluke bits and bobs through surplus channels and I haven't found any counterfeiting yet. I have bought other stuff from fluke-usa but the 189 terminal block I bought through Newark, of all places.  As always, YMMV--you pay your money and you take your chances.

[joeqsmith]

Fluke USA it is.  Thanks.

[daisizhou]

It has many imitations in China.
If you replace it without the original soldering material, there may be errors--I mean measurement deviation

[joeqsmith]

It has many imitations in China.
If you replace it without the original soldering material, there may be errors--I mean measurement deviation

Please provide references and details.

[daisizhou]

picture

[bdunham7]

It entirely possible that the current OEM part is made in China.  The old ones were apparently black while the new ones are red.  The part is still used, AFAIK, in the 787B and 789 products.  I can't find the one I got from Newark a while back, once I'm mobile again I'll dig it out.  I'm pretty sure it was red.  I see the two eBay sellers have pictures with slightly different appearance--the fluke-usa version doesn't have the machining marks the other one does. I'm assuming the Newark-sourced version, which is NLA through them, would not be counterfeit.

[joeqsmith]

picture

These do not appear to be the correct replacement for the 189.   See the ebay links I provided.   

It has many imitations in China.
If you replace it without the original soldering material, there may be errors--I mean measurement deviation

Please provide references and details.

I was more curious about your comment about the original soldering materials.   I doubt anyone is mixing solders or not soldering them in.   My interpretation of what you wrote is that if the original solder is not used, it may cause errors.    This is what I am asking for details on. 

[joeqsmith]

I mentioned I had stole the best parts from my worse meter to repair the new one.  One of the things I stole was the rubber keypad.  I noticed when I reinstalled the terminal block and functional tested it, only two of the buttons worked.   It looks like what ever chemicals this was exposed to damaged the rubber contacts.    

I had used some of MGChemicals conductive paint for some 3D printed microwave horns I was testing.  I sprayed a small amount into the cap and used a toothpick to apply a small bead on the end of the rubber.  Let it dry overnight.  All of the buttons works fine again.   

https://www.eevblog.com/forum/rf-microwave/experimenting-with-waveguides-using-the-litevna/msg4638982/#msg4638982

[daisizhou]

The error I am referring to is that there may be deviations in the measured values.

As far as I know, the soldering material used in the original factory FLUKE 187/189 contains silver，And ordinary soldering tin rarely contains this silver material.
Another issue is that if the solder contains silver, its welding temperature will increase. If you replace the seat, you will find that the solder pad will become focal, which can also lead to measurement deviation.

So I generally do not recommend that everyone replace this seat themselves unless it is truly beyond repair

[daisizhou]

Black rubber material for the end of your button----

One approach is relatively simple, using a conductive metal film with adhesive force on one side and conductive on the other side，You only need to cut a small disc and paste it accurately in the original black position.

Another method is to use conductive rubber adhesive, but this method is complex, but it can last for a long time after repair

[joeqsmith]

I pulled both terminal blocks at 640F.  No trouble at all and both appear original.   Do you have a document from Fluke that talks about the need for a special silver solder?  Something you were told?   I'm interested in hearing more about this. 

I am not sure what the seat is or what is meant by pad becoming focal.   Plan will be to replace the entire terminal block.  Again, the terminal block is molded into one part.    The pads stay the same.   

I use copper foil tape which sounds like what you are suggesting.  I'll let you know how the conductive paint holds up. 

Video showing same repair I plan to do. 

[daisizhou]

I drew a schematic diagram,When you weld, you increase the welding temperature of the soldering iron,So it will cause discoloration of the PCB insulation layer,If the temperature is too high, it will cause two welding points to conduct and have resistance.

I'm referring to turning brown,To be precise, the PCB insulation layer is affected by temperature,Causing the previously insulated part to become conductive now

[joeqsmith]

Attached are photos of the new meter.  Consider I had pulled the original terminal block, then installed the one from my bad meter, then removed it again.  So 4 heat cycles so far when you include the factory original install.   

I  wonder if someone used lead free on the boards you worked on.   Normally I solder around 620-640F.    Not enough to damage the board.   

Pins for the com and V are shorted on the board.  The current inputs are open for the sense.  Because you are not measuring through the two pins, it shouldn't effect the measurement.  With the damage you describe, it could however cause the LEAdS error.   

[daisizhou]

You have good temperature control, but few people are able to do so

[joeqsmith]

In the video I linked, it didn't appear they did much damage to the board.  I'm using my old Pace ST45 with a PS-90 iron and a large chisel with a lot of thermal mass. 

https://paceworldwide.com/116-chisel-thermo-drive

Maybe your seeing a combination of lead free and cheap irons?   Then again, I demonstrated some SMD soldering with a 100 year old blow torch.   

[Shock]

Most input problems are more likely residual contamination (leakage) problems first, not soldering by a beginner. Which is why you need to clean the area like your life depends on it before the oem part is replaced.

I'd solder it with close to the original alloy though if it matters. Do a sanity test and pay close attention to settle times, repeatability and current draw. If still a problem then I'd ditch the super cap and thoroughly clean and dry the pcb before anything else. Knowing it's not been poorly calibrated is really helpful as it sets a baseline to repair to.

Dave just did a video on board leakage btw.

[joeqsmith]

Most input problems are more likely residual contamination (leakage) problems first, not soldering by a beginner. Which is why you need to clean the area like your life depends on it before the oem part is replaced.

I'd solder it with close to the original alloy though if it matters. Do a sanity test and pay close attention to settle times, repeatability and current draw. If still a problem then I'd ditch the super cap and thoroughly clean and dry the pcb before anything else. Knowing it's not been poorly calibrated is really helpful as it sets a baseline to repair to.

While I did show photos of the broken pins, I'm guessing you did not see them.  No cleaning is going to help this meter. 

So the new terminal block arrived from FlukeUSA.  I'm not impressed.  It looks like a Chinese knockoff rather than the quality I see with the OEM parts.   

[joeqsmith]

Shown with a 1k, 1Gohm and 100uADC applied. Ran through all the basic tests an it appears in good order now. 

Would be nice if there was a way to repair the rubber on these but I guess the scars on all of mine give them character. 

[daisizhou]

This seat doesn’t look very “exquisite”.Maybe it works, but it's not perfect