Handheld meter robustness testing

[Fungus]

What's the failure mode of a PTC?

(apart from "bang!")

Maybe the two-PTC-in-series thing is in case one of them fails short-circuit.

[joeqsmith]

What's the failure mode of a PTC?

(apart from "bang!")

Maybe the two-PTC-in-series thing is in case one of them fails short-circuit.

Google is your friend:

https://www.pakeptc.com/resources/what-are-the-failure-modes-of-ptc.html
https://www.navsea.navy.mil/Portals/103/Documents/NSWC_Crane/SD-18/PDFs/Products/Archive/Thermistors/Thermistors.pdf

As for the DMMs, the only failures I have witnessed are where they arch over, normally blowing off the casing. 

[joeqsmith]

I could repeat this test with the BM235 if there is any interest.

It'd be interesting to see how the BM869s handles this in Ohms mode, but I wouldn't want you to potentially destroy a perfectly working meter by doing this. Maybe if you have a scrap board...

I did try a few tests using the 2kV DC supply, creating this streamer.  Care to guess the cause?

[Simmed]

I could repeat this test with the BM235 if there is any interest.

It'd be interesting to see how the BM869s handles this in Ohms mode, but I wouldn't want you to potentially destroy a perfectly working meter by doing this. Maybe if you have a scrap board...

I did try a few tests using the 2kV DC supply, creating this streamer.  Care to guess the cause?

the vomit from the fuse ?
no wait ... its too shiny
entrails of a capacitor ?

[joeqsmith]

Pushing MOVs to their limits...   



Meter friendly video
No meters were harmed during the making of this video

[Fungus]

Google is your friend:

https://www.pakeptc.com/resources/what-are-the-failure-modes-of-ptc.html
https://www.navsea.navy.mil/Portals/103/Documents/NSWC_Crane/SD-18/PDFs/Products/Archive/Thermistors/Thermistors.pdf

The first link didn't answer it. 

The second says:


So I guess that explains the two PTCs in series.

15% is quite a high number when you're talking about safety. 15% of 15% is much smaller (2.25%).

(and I assume Fluke would choose one with less-than-average chance of failing short)

[bdunham7]

Pushing MOVs to their limits...   

You showed the first 869S with the MOVs starting to conduct at 1.34kV.  That doesn't seem compatible with reading 1000VAC as the MOVs would be conducting at the peaks.  Is that an issue or is there some input configuration change in the 1000VAC range?

[joeqsmith]

Pushing MOVs to their limits...   

You showed the first 869S with the MOVs starting to conduct at 1.34kV.  That doesn't seem compatible with reading 1000VAC as the MOVs would be conducting at the peaks.  Is that an issue or is there some input configuration change in the 1000VAC range?

Agree.  I would expect it switch closer to 1.7 (say 20% above the peak).  I try to remember to state that these are not virgin meters when ever I take data with them.  It's possible that the 14kV transient tests (or some other use) caused the MOVs to degrade.   In conduction, both meters still zero out.   Also shown reading a 5% 1Gohm with the newer meter. 

I think I have some MOVs that would get us closer to 2kV.  This may be a good excuse to swap them out.  This meter is a bit of a sacrificial experiment anyway.   Let me have a look. 

[joeqsmith]

For fun, I pulled the MOVs from the prototype BM789 and swapped them into the newer BM869s.   Two of the original MOVs in the BM869s are physically much smaller.  Brymen double foot printed the MOVs allowing for a wider pin spacing.   The two that stand up are a little proud now but there is a lot of clearance in that area of the front case.  Note the height of the plastic insulator.   Because this meter has so many mods, I went back and marked the other changes to the battery cover.   

After swapping,  meter still zeros out on conductance and measure 1nS for 1Gohm as before.   Starts to clamp around 2020V now, similar to the BM789 we looked at.   

[floobydust]

I thought Brymen typically uses a 10K102 and pair of smaller 07K821's?
The 10mm 911's are 910V 1mA and 1,500V at 25A and down to around 1,400V at a couple amps. It's a log scale.

I've looked at realistic clamping voltages for the two MOV's in series, to see what the PCB and rotary switch are subjected to.
I find MOV's are rated (part number) either for 1mA DCV or AC RMS operating voltage, so it is confusing sometimes to know what they will do at a few amps of the 61010 test when the multimeter is on voltage function. On Ohm's/Continuity/Diode-Test/Capacitance the MOV's are not really in the picture, the transistor clamp is doing all the work. Goodbye little SOT-23 lol.

[Fungus]

I've looked at realistic clamping voltages for the two MOV's in series, to see what the PCB and rotary switch are subjected to.

If anybody wants a challenge: It would be interesting to see a graph of the voltage at various points on the PCB during a transient, eg. at the clamps, on either side of the PTC, etc.

Not easy to do though...

[joeqsmith]

I've looked at realistic clamping voltages for the two MOV's in series, to see what the PCB and rotary switch are subjected to.

If anybody wants a challenge: It would be interesting to see a graph of the voltage at various points on the PCB during a transient, eg. at the clamps, on either side of the PTC, etc.

Not easy to do though...

The challenge would be sorting out your vague post.  There are many DMMs available with various input designs.  Maybe YOU could try and use SPICE to model it.   I have made a few videos over the years showing something similar.  Maybe this is all you are looking for.   

[joeqsmith]

Someone had posted that they didn't know that there were deaths reported from working on microwave ovens.   I thought I would do a search and then realized Brazil has redefinded microwave deaths...

For the rest of us, here is a typical case from 2022.  Untrained person decides to attempt repairs.  It appears they as lucked out at least once.   From the photos, it appears maybe the Fluke 107 was being used at the time of death.  Showing that even a robust meters can't save an idiot from doing something stupid.   

https://assets.contentstack.io/v3/assets/bltdec2ded849740f4d/bltd2795286414b5929/63ffe6248acb15113e5fdc3e/II-1399412-2022-(_28552).pdf

[rsjsouza]

Pushing MOVs to their limits...   

Meter friendly video
No meters were harmed during the making of this video

Just for pyro effects, perhaps consider in the future trying to blow one of these... (Banana DMM for comparison)



I used these 40kA beasts to help protect a "princess" pool controller that insisted in blowing up a fuse at every thunderstorm.

[rsjsouza]

Someone had posted that they didn't know that there were deaths reported from working on microwave ovens.   I thought I would do a search and then realized Brazil has redefinded microwave deaths...

Common drug lord technique to get rid of rival gang members... A common term since the 1990s. A few car tires, some flammable liquid, an individual and a match.

[joeqsmith]

Looks like they rate the LS40K150QP for 200VDC and 1.4W    I have a supply that can drive an easy 50mA at  200V or about 10W dissipation from that little MOV.    I doubt you would get much of an effect, even if I drove it with a half amp DC.    I did try to damage some MOVs before by transient testing them with my generator.  It just doesn't put out enough energy to do much of anything. 

https://www.tdk-electronics.tdk.com/inf/70/db/var/SIOV_Strap_LS40.pdf

[rsjsouza]

Indeed the Ptot is relatively small for continuous operation, but a 40kA transient would be a much more interesting show... 

[joeqsmith]

Indeed the Ptot is relatively small for continuous operation, but a 40kA transient would be a much more interesting show... 

A 40kA transient would be pretty simple.  It's the time or energy that's a show stopper.   

The half cycle line simulator can provide upward of 600J.   More than enough to shatter a light bulb.   But due to safety, it is not a simple trigger to get a pulse out of it.   

With them being a low voltage MOV, the easiest thing you could do, if you wanted to try it, add a new test outlet with a contactor that you can remotely trigger.  Just use the 220 single phase.  Even our older home has a 200A service feeding it.  Looks like the MOV should fail before the CBs trip.