Fixing a Fluke 87v

[jpatten]

I bought a Fluke 87V in 2006 and it's been great up until a few weeks ago when I accidentally connected it to 220VAC when it was set to measure resistance. After that it would read around 1.5M ohms without even having any probes connected. I sent it in for repair and they quoted around 260 USD for the repair. Faced with spending that much, I decided to try and fix it or just buy a new one. The repair quote said that varistors 1-3 were blown. So I found the service manual online (revision 5, which is for an earlier model) and it only lists 2 varistors, not 3. So, I'm wondering:

Does anyone have a new revision of the service manual that might be more useful for the 2006 version of the 87V? Or specs or part numbers for these varistors?

Does anyone have advice about fixing this problem in general? I wonder if the protection circuitry did its job or if I'm likely to find other issues after swapping out the varistors... Any advice appreciated!

[Lightages]

You can replace the varistors (MOVs) similar speced ones. They don't need to be the exact voltage that was original. You would want to install the highest energy type you can find that will fit. All you need to do is to check to see if these are in series or if they are two different circuits. If they are in series then you need to put in MOVs that are 470V each to maybe 800V each so that the total is between 940V to 1600V. If they are separate circuits then you need to install 1000V to maybe 1800V MOVs.

The only thing is that after doing this you capacitance range might not be accurate, nor may other measurements and the meter might need a re-calibration/adjustment.

[alm]

Connecting it to 220 VAC should not damage it. The MOV will most likely be close to the input, before any switching. This means whether they blow or not is not dependent on switch position. Either your meter contained multiple broken thermistors that blew at less than 20% of their rated voltage, or it was exposed to voltages well in excess of 1 kV. In the latter case, there may be damage on downstream components, depending on the exact type of overload and how well the protection circuit did its job. MOVs read essentially open circuit when not exposed to a high voltage, so just removing them from the circuit should allow you to test the meter. Just don't connect it to what you call 220 VAC.

[nukie]

If $260 repair cost includes calibration, I would go for it.

[AlfBaz]

Pretty sure fluke don't fix them, they just replace the board

[retiredcaps]

I bought a Fluke 87V in 2006 and it's been great up until a few weeks ago when I accidentally connected it to 220VAC when it was set to measure resistance.

Wow, that is discomforting to hear.  You pay top $$$ for input protection and we have all seen Dave's ohms overload video where he subjects the 87V to mains for at least a minute (8:15).  In fact, all the meters in the video survived.

http://www.eevblog.com/2012/03/04/eevblog-252-multimeter-ohms-overload/

I wonder if the protection circuitry did its job or if I'm likely to find other issues after swapping out the varistors...

At $260 repair cost, you have nothing to lose but to try some replacement varistors.

Did Fluke explain why "lifetime warranty" is not applicable in this case assuming you are the original owner and purchased it from an authorized Fluke dealer?

[eevblogfan]

hey

I am inviting you to sent the meter for re cal and perhaps adjustment if needed , if you'd like to fix it yourself , go for it , I'll charge you funny little 10$ , but the shipping is aside thing .

Pm me if needed

Cheers

[jpatten]

Thanks for the replies.

Did Fluke explain why "lifetime warranty" is not applicable in this case assuming you are the original owner and purchased it from an authorized Fluke dealer?

They said that the fault was not due to defects in materials/workmanship. (i.e. that the meter was abused)

Connecting it to 220 VAC should not damage it. The MOV will most likely be close to the input, before any switching. This means whether they blow or not is not dependent on switch position. Either your meter contained multiple broken thermistors that blew at less than 20% of their rated voltage, or it was exposed to voltages well in excess of 1 kV.

I guess it's only an assumption on my part that the 220 VAC zapped it. I noticed the problem shortly after but I don't know for sure that's what caused it. I don't (knowingly) work with anything near 1 kV though.

Thanks again for all the assistance. I'm going to try replacing the MOVs and see how it performs afterward, and take things from there.

Cheers

[retiredcaps]

They said that the fault was not due to defects in materials/workmanship. (i.e. that the meter was abused)

IMHO, that is a very disappointing answer.  If a Fluke 87V can't survive mains 220VAC on the ohms range, then that is really really poor.

If you can't fix the multimeter and you still want another Fluke, you may want to consider buying an used 87V in the $150 USD range on ebay (with patience).

Let us know the outcome when you replace the components.

[amyk]

Maybe it was just a fluke.

[Lightages]

I find this disappointing too. I thought Fluke was among the best protected and had the best guarantee in the business. I think I would write a letter to HQ and ask them to review the situation again and describe what the abuse was.

I just went through the whole manual and the only limitation put on the use of the meter as far as voltage goes is to not apply more than 1000V on any jack. I think you have a complaint.

[JimmyMz]

Maybe it was just a fluke.

It made me smile.

[gerrysweeney]

I would agree, I would go back to Fluke, there is no way the meter should be damaged by putting 230VAC on the input in ohms mode, I have two 87V meters and I just tried both meters on mains 240VAC in OHMS mode and no damage at all.  I agree with @Lightages, you have a valid complaint to make....

Gerry

[N TYPE]

^^ What he said.. I connect mine to 240v on ohms range all the time and have never had any problems..

[alm]

I will play devil's advocate. Three MOVs were apparently blown. All three MOVs are probably rated for something like 900 V. The meter was presumably used to measure 230 VAC before, so the MOVs could withstand 230 V. As far as I know, the protection circuit is not switched when switching functions. What, except for a severe spike, could suddenly cause all three MOVs to start conducting and kill themselves? From their perspective, it probably looks like a severe overload. Where any other components, like series PTCs or fusible resistors, damaged?

[SLJ]

I would contact Fluke again.  The meter should withstand 230V on any of the ranges.  It's designed to do just that.  If it's under warranty they should fix it.   If they still won't, post the question as to why they won't fix an 87V on their facebook page.

[jpatten]

Thanks again for the replies. I wanted to post an update: After some more back and forth with Fluke I wasn't able to convince them to fix it, so I asked them to send it back unrepaired. I just got it back late last week and took another look at it. I tested AC volts, DC volts, DC mA, and these seem to be reading accurately, as best as I can tell with my equipment. I then opened it up and took another look at the MOVs, and two of them are indeed fried. (They are ruptured, with black residue on the outside.) I guess I missed this before I sent it off.

Having recently watched Dave's videos about multimeter input protection and the design of the 87v, it seems worth a shot to replace those MOVs and see if the ohms range comes back to life. I haven't worked with MOVs before, so I'm not sure how to interpret the markings to find a replacement. Can anyone offer some insight on this? I will attach a picture. The MOVs are roughly 1cm in diameter and the markings read "05 K575 06 18". I'm guessing the 06 18 is a date code. I googled this of course but I'm not sure I'm searching for the right thing.

As for Fluke not repairing the meter, I was frustrated and disappointed about this, but at the same time I have to agree with them that taking another look at the meter itself it does look like it saw a big voltage spike. I can't remember anything that could have caused this, and it seems like those who said the meter should be able to handle 220V just fine are correct. I work in a shared space, so it is possible someone borrowed my 87v and zapped it somehow. I'll probably never know. But from now on I'll be keeping it under lock and key. 

If anyone can suggest replacement MOVs that would be very appreciated!

[SeanB]

That is probably a 575V varistor, you probably will be able to get it from the usual sources, or just place 2 275V ones in series, as that is what those are, 2 disks soldered together then leaded and dipped. Probably was either used on high voltage or was exposed to something like an arc on a large inductor fed from a heavy DC supply, where the back EMF was enough to overload the VDR. Possible also one just dies from old age, and gave out smoke when powered from 220V.

[Excavatoree]

I've got a huge bag of MOVs I bought to fix some 27s that I had - I can send you three, if you'd like.  PM me your address.

As others said, there are many that will work.  The ones I have may be a little large - as there's more room in the 27, and I purposely got larger ones.  (Purchased from a seller on a well-know online auction site)

EDIT: I did do proper research to chose the specifications - I just happen to have forgotten what I chose.  I didn't just buy the first MOV's I saw.  They are working very well in a couple of 27s that I have - all still met their specs. after the replacement.

[geostep]

jpatten -

If you can locate the correct MOV's I would replace all of them, not just the ones with visible damage.  You mentioned Fluke said that three of them were bad.  If you only replace the two obviously damaged ones your 87V could still have trouble.

I did a quick search for you and found these MOV's with a 575 VAC rating and a width of 10 mm.

Digi-Key:  http://www.digikey.com/product-detail/en/V575LA10P/F3771-ND/1009609

Mouser:  http://www.mouser.com/Search/Refine.aspx?Keyword=576-V575LA10P

I'm not recommending Digi-Key or Mouser.  The links are for information only.

Price was about 66 cents each.  I'd replace them all at that price.  A lot cheaper then what Fluke wanted!

- George


- George

[benemorius]

Well if you're wanting to know whether the MOVs are causing the problem with the ohms range, the first step is to pluck them off and see if anything changes.

[fpliuzzi]

Certainly it would be prudent to replace all three of the MOVs in your Fluke, as stated by the other posters, but if you have a bunch of "mystery MOVs" that you have collected over the years by parting out old PCBs you may find the following Littelfuse appnote interesting.

www.littelfuse.com/data/en/Application_Notes/an9773.pdf

It contains a very simple circuit for the measurement of some key varistor characteristics. It seems to work quite well for me. For the HV DC supply (E1) I just use a simple unregulated supply made from a tiny old neon sign transformer, plus an HV diode and HV capacitor from a discarded microwave oven (and a current limiting string of resistors for safety sake). I just crank up the DC supply using my Variac to achieve the output voltage I need for the MOV under test (about 1KV DC maximum at a few mA is available).

Regards,
Frank

[retiredcaps]

As for Fluke not repairing the meter, I was frustrated and disappointed about this, but at the same time I have to agree with them that taking another look at the meter itself it does look like it saw a big voltage spike.

I can see how the physical damage of the MOVs would lead any Fluke customer support employee to think it has been abused.

[mikepa]

It is very disconcerting and not good for Fluke's name to be arguing over what appears to be a failure that should be covered under warranty.  I would write a letter to HQ asking for exactly what abuse they believe occurred. 

[eKretz]

I then opened it up and took another look at the MOVs, and two of them are indeed fried. (They are ruptured, with black residue on the outside.) I guess I missed this before I sent it off.

Having recently watched Dave's videos about multimeter input protection and the design of the 87v, it seems worth a shot to replace those MOVs and see if the ohms range comes back to life. I haven't worked with MOVs before, so I'm not sure how to interpret the markings to find a replacement. Can anyone offer some insight on this? I will attach a picture. The MOVs are roughly 1cm in diameter and the markings read "05 K575 06 18". I'm guessing the 06 18 is a date code. I googled this of course but I'm not sure I'm searching for the right thing.

As for Fluke not repairing the meter, I was frustrated and disappointed about this, but at the same time I have to agree with them that taking another look at the meter itself it does look like it saw a big voltage spike. I work in a shared space, so it is possible someone borrowed my 87v and zapped it somehow. I'll probably never know. But from now on I'll be keeping it under lock and key. 

I would bet this is what happened. I don't think varistors fail out of the blue for no reason very often. I used to work in a machine shop where I left my tens of thousands of dollars worth of measuring tools and instruments in an unlocked toolbox. One day I injured my back and was out of work for some time. When I came back, there were a lot of damaged and even missing tools. Of course nobody would come right out and say they did any of the damage. As it always seems to go when these types of things happen, nobody knew anything.