Handheld meter robustness testing

[EEVblog]

That's my point. The model number is exactly the same, yet the meter differs greatly in it's safety specs.

Yes, I suppose that's true if the '-GS' is just an internal designation and the model that it was marketed as is just UT61E.  So under the jurisdiction of the TUV, they sold one 'UT61E' with protection and lowered CAT ratings, elsewhere they depopulated the protection and increased the purported CAT ratings.  I think other companies in other areas may do similar things between markets, but CAT ratings are a globally harmonized standard (or so I thought) so it does seem quite slippery of them.

With my Brymen meters that I resell under the EEVblog branding I have to have them the exact same model number BM235/BM786 and the Brymen name, otherwise if I change it they would of had to have got it entirely UL certified again which is of course lengthy and expensive process. I can't just whack another model number on it and use the UL logo and certificate for physically the exact same meter, it's that strict.

[gnavigator1007]

With my Brymen meters that I resell under the EEVblog branding I have to have them the exact same model number BM235/BM786 and the Brymen name, otherwise if I change it they would of had to have got it entirely UL certified again which is of course lengthy and expensive process. I can't just whack another model number on it and use the UL logo and certificate for physically the exact same meter, it's that strict.

Did the 121GW need any sort of recertification with board revisions?

[joeqsmith]

Especialy the low energy spikes will be hard to find any info about, because they don't cause imediate failure so it would be difficult to asess what would be the most common but already harmful energy and voltage level.But I think something well within Joes generators range.

You're going to have to define and quantify 'low energy' and fully specify the circuit characteristics and other test conditions for any further discussion to have meaning.  I wouldn't call the jqs^TM transients 'low energy'.

I've been meaning to respond to your comment.   I suspect many people share your thoughts which is partly why some relate them to safety and AC mains testing.   

I want to be clear that when I talk about my transients being low energy, I am making a relative comparison with the  61010 standards that I had loosely based them from.   Again, the standards call for a hybrid, or combo generator.   They specify an open circuit voltage wave form with a 1.2us rise and 50us FWHH  (1.2/50).    Consider that I have a faster rise and a longer delay.  This all came from my initial testing of the very first $50 meters.  I started out with a much narrower transient and worked my way up.  I also started out using much lower energy levels.   The initial levels were so low, I couldn't damage a meter.   By the time I started to test the $50 meters, I think I had settled on a 50us full width (not FWHH) at 10J.   When I decided to design a programmable generator, I doubled the energy to roughly 20J and settled on the 100us FWHH.  I also decided on a peak of 6kV where the Amprobe AM510 had been damaged during the $50 shootout. 

It may seem like my transients far exceed the standards, after they take twice as long to decay and have the same source impedance.   Makes sense until considering the current.   
 
The following was just the first TVS from Digikey. 
https://media.digikey.com/pdf/Data%20Sheets/NextGen%20Components%20PDFs/SM12.pdf     

Notice the text  "Peak Pulse Power per (8/20μs): 350 Watts"   That 8/20us defines the short circuit current waveform called out in the 61010 standard.   There are two waveforms, hence the name combo or hybrid generator. 

I use a 2 ohm source, so if I have a 1kV peak, the current will be 1kV/2 or 500A.  Where the combo generator has a 20us Full Width Half Height, mine may only have a ns.      No, it's not  that narrow but 20J isn't much.   I made a recent video showing what the current looked like through a meter.  Not much happened as expected.   

So maybe this will help a few of you understand.   Consider the attached data taken with my scope.   The scope's settings were kept the same throughout the test.  Channel 1 (gold) is looking at the voltage.  Channel 2 (pink) is the current.   

Tran1 starts with an open circuit.  Note that there is no current draw and the voltage is a bit shy of a kV (using my home made 100x probes again).   At 20us / div, half way up measures roughly 100us across.   No surprise.

Tran2, I have added a small load.  The peak voltage is still roughly the same but we can see some current flowing.  Notice now that the voltage FWHH is roughly  30us.   

Tran3, I have increased the load and we can see the peak voltage has dropped.  Where did the voltage go??  It's dropped across the generator's internal 2 ohm source! Notice that the current's FWHH is now only 10us or so.  Remember, the standard called for 20us with a short.  I'm sure a few of you can guess the effects of further increasing the load.

Again, the goal was never to make these meters explode like you will find in Fluke's test lab videos.  That should have been obvious by simply comparing the physical size of their 19" rack mounted Haefely compared with my tiny desktop setup.   Normally the meters will not break down and all that energy is dissipated in the generator's internal coupling network.  The 20J may be overkill but all you safety experts posting here already know where that number came from.   I wanted at much as possible but still work with a minimal risk.

You may not agree but hopefully you at least now have some understanding why I will continue to call my transients low energy.   

****
Rough reading in a few places.

[bdunham7]

The 20J may be overkill but all you safety experts posting here already know where that number came from. 

Hmmm.  Is it because that is about the energy used in a defibrillator or because it matches up with the output of an electric fence charger (a powerful one)? 

You may not agree but hopefully you at least now have some understanding why I will continue to call my transients low energy.

Thanks for the detailed explanation.  Low is a relative term, of course, and relative to MOV ratings, IEC transients, etc, 20J is obviously much lower.  The comment I responded to was talking about damage to semiconductors, etc and there 20J  will pop the lid on quite a few devices.   I think the low energy events he was referring to are the sorts of noise, ESD or spikes that might get through normal filtering for reasons other than that they overwhelm the protection systems by simply exceeding their voltage or energy limitations.  Piezo igniters and 220MHz RF come to mind....

With that in mind, given the mayhem your jqs IEC Lite^TM transients cause, it is difficult for me to see how some of these units could emerge from the full mains-connected IEC transient test unscathed, unless the criteria being applied are just that the device not explode or trip the mains limiting. 

[tautech]

Is it because that is about the energy used in a defibrillator or because it matches up with the output of an electric fence charger (a powerful one)? 

20J is a baby fence unit these days now energisers of 60J are available:
https://pel.co.nz/en-nz/node/11974

[rsjsouza]

Show me a photo of a Uni-T meter, a manual, or a website link that actually has a meter labelled "UT61E-GS".

My reply #4203 has them.  The meter is still labelled UT61E, but has the "GS" mark and has different CAT ratings.

That's my point. The model number is exactly the same, yet the meter differs greatly in it's safety specs and ratings.
Why would anyone trust a company that sells the exact same model number meter in different markets with different safety/protection components?
At the very least give it another model number.

They were indeed idiots when doing this to the UT61E but, in all fairness, that is exactly what they are doing now with their new UT61E+ / UT161E differentiation. Perhaps an attempt to redeem themselves? Even still, this does not excuse slapping false ratings in several of their meters.

[rsjsouza]

With that in mind, given the mayhem your jqs IEC Lite^TM transients cause, it is difficult for me to see how some of these units could emerge from the full mains-connected IEC transient test unscathed, unless the criteria being applied are just that the device not explode or trip the mains limiting.

In my interpretation of the 61010 is that the meter is not expected to survive the transients. The fact Fluke and others are not screaming at the top of their lungs about this is an indication I am not alone in my interpretation. Sure, Fluke has some spectacular videos of exploding meters, but they talk about containing the energy of the damage inside the housing to avoid high speed ejection of material and quenching arc flash.

[AndrewBCN]

Uni-T is doing the PCB copy'n'paste between UT61E+ and UT161E. I see the larger fuses and PTC's, PTC's moved slightly but nothing significant.
Looks like they want to charge a premium for a real 61010 product. I'll bet the BBQ lighter still makes it crash lol.

You can bet whatever you want, but thanks for digging out a picture of the internals of the UT61E+ and the externally identical, but internally quite different, intertek-certified UT161E (from a Russian forum?).

So UNI-T charge a 3€ "premium" for the larger fuses and PTCs, and proper independent testing and certification for the UT161E, which as you well know, has a cost.

I earlier wrote that for the extra 3€, I would rather buy the Intertek-certified UT161E rather than the UT61E+ that Joe is testing these days. The picture you posted just confirmed this.

[AndrewBCN]

Show me a photo of a Uni-T meter, a manual, or a website link that actually has a meter labelled "UT61E-GS".

My reply #4203 has them.  The meter is still labelled UT61E, but has the "GS" mark and has different CAT ratings.

That's my point. The model number is exactly the same, yet the meter differs greatly in it's safety specs and ratings.
Why would anyone trust a company that sells the exact same model number meter in different markets with different safety/protection components?
At the very least give it another model number.

The "GS" designation for the UT61E apparently stood for "German Safety". Here is a link to a reseller in Germany that specifically sells the UT61E GS.

https://www.pinsonne-elektronik.de/pi2/pd58.html

From that same page you can download the user manual (in English) for the UT61E GS, it has the safety standards compliance on page 6 and the GS mark explained on page 9.

I have no idea if the discontinued UT61E GS was ever sold in any other country than Germany (and I couldn't care less).

The Intertek-certified UT161E is available internationally, as I wrote before for a negligible premium over the non-certified, smaller fuses and PTCs UT61E+ which Joe is testing these days.

[Fungus]

So UNI-T charge a 3€ "premium" for the larger fuses and PTCs, and proper independent testing and certification for the UT161E, which as you well know, has a cost.

No, the UT161E is far more expensive, about double the price of the UT61E.

It still has the stupid transistor tester though. I'd never own one because of that "feature". I'd have to do a facepalm every time I turned the selector past that position on the dial.

[joeqsmith]

It still has the stupid transistor tester though. I'd never own one because of that "feature". I'd have to do a facepalm every time I turned the selector past that position on the dial.

You should consider getting this as it seems it would fit your skill set very well.

https://www.amazon.com/Fisher-Price-Laugh-Learn-Puppys-Remote/dp/B014KEEFO8

I doubt the group will think any less of your skills if that helps.       

[joeqsmith]

I often ask why people continue to discuss safety in a thread that has nothing to do with it.

One would need to reread the entire thread to see how that seeped in, but I suspect the similarities of your test transient levels to the ones specified for the CAT ratings are one factor.

Anyhow, by now I agree with you, if you are referring to safety as meters exploding or arc-flash type concerns.  My concerns are similar to what you have said about wanting meters to not die during normal, non-dangerous bench or other use.  Transients are one way to kill them of course, but accidental overvoltage is another.  I had one set of expectations about what a CAT-rating would mean about this, specifically that a meter with a CAT rating for any voltage--like a bench meter rated CAT I/1000V--would endure 1000VAC or 1000VDC on any range or input selection without damage. There's also the expectation that after the transient tests, the meter works.  If those expectations on my part are wrong and the CAT ratings don't indicate performance in that regard across the board, or if different manufacturers have different standards regarding these issues, then the CAT ratings themselves are meaningless to me and of questionable value to anyone else, IMO. 

If anyone wants to comment on my assertion that a CAT x/1000V meter should withstand a full 1000V on any input setting or jack, I can see about setting up a test.

We have.   If you take the time to go over it,  feel free to then let me know if there is something you did not understand.   

I am all for YOU running a test like this.  I think if you want to light it up, you are going to have to use the meter incorrectly and turn the function switch with it live.  If you pick a meter that I have already and find it survives, maybe I can attempt to repeat it.   I have that one 121GW that was used for the majority of my destructive tests.  It has been certified.  I could toss that into the mix as well. 

Some of the smaller PTCs are only rated for 500V.  Some meters have only a single PTC which if the low voltage clamp is active, will have well over 900V across them....  Maybe....  Turn the dial, I suspect you will get a light show.

Well that doesn't sound very robust!  I don't know if that is a failure to meet a standard or not, but it seems like a basic expectation to me.  Every CAT labelled meter I currently have that isn't known junk should pass that test.  Some of them already have by accident.

Maybe start reading here.  Dave chimes in.  So do I. 
https://www.eevblog.com/forum/testgear/eevblog-121gw-discussion-thread/msg1580530/#msg1580530

****
1kVDC with more than enough current to cause a major meltdown ready when you are. 

[joeqsmith]

Is it because that is about the energy used in a defibrillator or because it matches up with the output of an electric fence charger (a powerful one)? 

20J is a baby fence unit these days now energisers of 60J are available:
https://pel.co.nz/en-nz/node/11974

Don't take a whiz on that!   

[Fungus]

Here's a genuine CAT III 300V certified meter which has been discussed somewhere in the middle thread. It has two resistor trails leading off before anything gets to the PTC. The R31/R30/R29/R28 chain is 10MOhms, the same as the impedance of the meter in volts mode. It's not the only meter I own that's like this, hence the silly ideas in my head.

I think the input circuit is clear enough in this photo:


I lifted up one leg of the PTC as joe suggested:


Guess what? It still measures volts perfectly!


OK, maybe the Fluke 101 doesn't work that way (I should have looked at the resistor values before opening my mouth) but there's definitely some certifiable meters that do work that way.

[bdunham7]

I earlier wrote that for the extra 3€, I would rather buy the Intertek-certified UT161E rather than the UT61E+ that Joe is testing these days. The picture you posted just confirmed this.

Well, regardless of whatever else we disagree about, now that I see the correct meters that you are referring to I'd have to agree with that.  They seem to have finally made an attempt to build up the 'certified' version to the standards they printed on the UT61E+ rather than have it certified down to a much lower level.  It even looks like they are using the actual Bussmann fuse.

[tautech]

Is it because that is about the energy used in a defibrillator or because it matches up with the output of an electric fence charger (a powerful one)? 

20J is a baby fence unit these days now energisers of 60J are available:
https://pel.co.nz/en-nz/node/11974

Don't take a whiz on that!

Hell no !
Pic below of a short from 2.5mm (12g) wire to a tree 1 mile from our 40J unit where the CLACK could be heard a couple hundy yds away.

[joeqsmith]

Here's a genuine CAT III 300V certified meter which has been discussed somewhere in the middle thread. It has two resistor trails leading off before anything gets to the PTC. The R31/R30/R29/R28 chain is 10MOhms, the same as the impedance of the meter in volts mode. It's not the only meter I own that's like this, hence the silly ideas in my head.

I think the input circuit is clear enough in this photo:
...
I lifted up one leg of the PTC as joe suggested:
...
Guess what? It still measures volts perfectly!
...

OK, maybe the Fluke 101 doesn't work that way (I should have looked at the resistor values before opening my mouth) but there's definitely some certifiable meters that do work that way.

As I said

Keep those stories coming.  We're here all night folks.

I'm having a moment, aren't I?

I suspect you have been looking at too many low end meters like the UNI-T,  ANENG....    Your statement about "...generally multimeters have a separate 10MOhm voltage input.." could be correct as I suspect there are more low end meters being introduced and sold.    We love our cheap, disposable products which drives the market and I just don't want to admit it. 

Just another cheap meter.   These don't surprise me but I'm interested in what your Fluke 101 does.   Now that you have had some practice with the soldering iron, time to lift one more pin. 

Quick search, looks like you died from that meter 
https://www.eevblog.com/forum/testgear/heads-up-cheap-multimeter-lidl-uk-(080218)/msg1429596/#msg1429596

[bdunham7]

I am all for YOU running a test like this.  I think if you want to light it up, you are going to have to use the meter incorrectly and turn the function switch with it live.  If you pick a meter that I have already and find it survives, maybe I can attempt to repeat it.   I have that one 121GW that was used for the majority of my destructive tests.  It has been certified.  I could toss that into the mix as well. 

1kVDC with more than enough current to cause a major meltdown ready when you are.

OK, I finally got another project done after my parts went on a four day tour of North Dakota--so there's room on the bench.  Next projects are some calibrator overhauls, so just the perfect opportunity to test some meters.  Unfortunately, I discovered that I don't have all that much power, probably not even what the calibrator is supposed to have, so I'll have to look at that issue before I get too far.  I don't want to abuse the meters by switching ranges under power, just test them at max voltage on all ranges.  I'll leave exploding things to others for now.

First test was a Fluke 116.  You have tested a 115, I don't know if you have it still or how similar they are.  I gave it 600VDC and 600VAC/100Hz in all ranges, 10-15 seconds per range, then checked its calibration afterwards.  All good.  I did notice that when it clamps, it can clamp a lot more current than my calibrator can supply.  I had to use a different DC supply. For AC, I had to start at a lower voltage then work my way up as fast as I could push the buttons. 

The next victim was actually not a handheld, but an old Fluke 8842A bench meter.  No CAT rating, but it took the 1000VDC and 700VAC listed on the front panel without complaint, ohms range and all, but the calibrator started whining (literally) so I quit for the night.  I'm really not looking to blow that one up.

I can toast a Harbor Freight meter for giggles, but other than that I don't have a lot to throw at it right now.

[Fungus]

Just another cheap meter.   These don't surprise me but I'm interested in what your Fluke 101 does.

As pointed out: The resistor chain is only 1MOhm. I don't think it's going to be the voltage input based on that and your recent postings.

I'm more interested in the function of that circuit than simply confirming something that's already known - science doesn't advance that way. Corrrect procedure requires a theory before I barge in and do any experiments so I need to find time to sit down and trace out the 101 PCB as the next step.

PS: The Fluke 101 is a "cheap meter". For the  price of a Uni-T U61E I can get a Fluke 101 and an Aneng 870.

[joeqsmith]

Just another cheap meter.   These don't surprise me but I'm interested in what your Fluke 101 does.

..
I'm more interested in the function of that circuit than simply confirming something that's already known - science doesn't advance that way. Corrrect procedure requires a theory before I barge in and do any experiments so I need to find time to sit down and trace out the 101 PCB as the next step.

 
Lame, but that's alright.   

PS: The Fluke 101 is a "cheap meter". For the  price of a Uni-T U61E I can get a Fluke 101 and an Aneng 870.

I would have thought when I write use the word cheap,  generally people understood it was a shoddy product.  Another definition for cheap is low cost but I will typically write that out.   

[joeqsmith]

...
I don't want to abuse the meters by switching ranges under power, just test them at max voltage on all ranges.  I'll leave exploding things to others for now.

First test was a Fluke 116.  You have tested a 115, I don't know if you have it still or how similar they are.  I gave it 600VDC and 600VAC/100Hz in all ranges, 10-15 seconds per range, then checked its calibration afterwards.  All good.  I did notice that when it clamps, it can clamp a lot more current than my calibrator can supply.  I had to use a different DC supply. For AC, I had to start at a lower voltage then work my way up as fast as I could push the buttons. 

The next victim was actually not a handheld, but an old Fluke 8842A bench meter.  No CAT rating, but it took the 1000VDC and 700VAC listed on the front panel without complaint, ohms range and all, but the calibrator started whining (literally) so I quit for the night.  I'm really not looking to blow that one up.

I can toast a Harbor Freight meter for giggles, but other than that I don't have a lot to throw at it right now.

When cold, I am expecting with a kV applied, the initial draw is somewhere around 1kV/2000ohms or 500mA.   Higher for some of the cheap meters.   I doubt many electronics hobbyist have that sort of power supply on hand.   I would think the use case was you touch the leads to the live source with the meter set to each function, except for the current.  Allowing a minute or so for the PTC to cool between tests.   Creeping up on the voltage would seem like a rare case.   

I have had meters become damaged with a 1kV transient. These would certainly fail with DC applied.  These are not in the same class as your 116. 

I have a very bad habit of changing functions live.  I do this with my bench meters as well as the handhelds.  I've written software to changes modes as part of a test.  Commonly, I will move from AC to DC.  Meters with LowZ would be interesting as that should get you enough of a path to form an arc if you switch it live.  Mostly I'm working below where I would be concerned with arcing the switch or relay contacts. 

****
A short experiment with a 5kohm resistor, 1kVDC source and an air gap (could be the DMMs switch)
https://youtu.be/bgz-pqg0rKo?t=1271

[Fungus]

I got around to experimenting with my 101 as Joe suggested, here's the results:

I lifted up on leg of the surge resistor:


Meter now fails to measure voltage. The number on screen is a floating value that appears even if I connect only one probe to the live wire. The voltage hazard indicator doesn't light up:


Here it is back together and working again:


The resistor chain goes up to the selector and is connected across to the adjacent track when the switch is in the mV/Ohms/Capacitance/Hz modes. It isn't connected in  the VAC/VDC modes. That's my theory of it being used for hazardous voltage indication shot down in flames.


The back of the PCB showing the signal path. The signal jumps across the selector then heads off in the general direction of the IC. It goes into another via and I lost it there. There's no corresponding via on the front of the PCB so it must go into an inner layer.


Now I need a new theory. The signal isn't used on either of the main voltage ranges and it's not for the hazardous voltage indicator. 

[joeqsmith]

Nice job sir.   Do you have the tools/skills needed to remove one of the resistors that make up the chain R8-R2?   If so, I would just reattach R20 and pull one of the other and see which functions no longer work. 

[2N3055]

Now I need a new theory. The signal isn't used on either of the main voltage ranges and it's not for the hazardous voltage indicator. 

What do you mean another theory? It is used here like you said:

The resistor chain goes up to the selector and is connected across to the adjacent track when the switch is in the mV/Ohms/Capacitance/Hz modes.

There must separate path injecting current into DUT for resistance and capacitance, and this composite 1 MOhm resistor is measurement path for these ranges. There will be transistor clamp after it somewhere before going into DMM chip. ...

[joeqsmith]

People learn in various ways.  Some, like myself are very hands on.   I have found that seeing and experiencing how things work first hand can improve my understanding beyond just reading a book.    In this case, we are talking about removing a couple of resistors.  The idea would be for Fungus to prove what is happening to themselves,  not the rest of the internet world.