Handheld meter robustness testing

[Fungus]

If you think that Uni-T 135C is bad (and I agree with you), just be glad you haven't had the misfortune to encounter one of these turds:

"The ultimate electrician's digital Multimeter!"



Does "CAT III 500V" even exist as a rating? 

[bdunham7]

Yes, there are high resolution pictures of the top and bottom sides of the UT125C PCB on HKJ's website.

Can you pop the fuse out and confirm which end the positive lead is connected to?

Edit: it appears from one of HKJ's other photos that it is connected to the top (near the PTCs) as one would hope.

[2N3055]

That UNI-T UT125C looks like it was designed to guarantee arc flash accident if someone would connect it something with enough energy and there was a overvoltage event.....

It would arc over gap, evaporate copper trace, and draw arc towards cables that are mangled on the input in the case (insulation is badly damaged on cables on that photo..) All of that before fuse can blow.. Or it would arc over damaged cables first, distance might be even less than spark gap.
If that happens, arc would exit the case following the cables. Arc moves towards the source of the current (Jacob's ladder style).
All in all, pretty much death trap...


In attachment nice material from Stäubli. It is marketing material, but Germans being traditionally thorough, lots of good info. Also it shows perspective that cabling and accessories should be up to task too.  Very safe Fluke or Brymen means nothing if cables are mangled..

EDIT: Attachments are better if you actually attach them. 

[Fungus]

If that happens, arc would exit the case following the cables.

It might.... if you used it outside the CAT III environment that it's rated for.

[bdunham7]

It might.... if you used it outside the CAT III environment that it's rated for.

And where do you think CAT III is?  It's not your house.  AFAIK, a subpanel serving a 100HP 480VAC 3-phase motor with 200A breakers would (or could, depending on the exact setup) be CAT III.  CAT III definitely includes arc-flash hazards.  CAT II does not, IIRC.

[Fungus]

And where do you think CAT III is?

On the safest side of a distribution panel with circuit breakers in it.

We still don't know for sure if what we're looking at is a spark gap or where it is in the meter's input circuit, eg. Is it in series with the fuse?

To me it looks like there could be a current shunt across that gap in some variant of the mater (between the two vias where the vertical white line is painted).

[bdunham7]

On the safest side of a distribution panel with circuit breakers in it.

I don't think that is how it is defined (a single-phase circuit on a 20A breaker can be CAT IV) and even if it generally works out that way, AFAIK there's no limit on the size of those breakers, or indeed any actual requirement that they exist--although electrical codes other than CAT would usually require them.

So here is a plausible example.  Suppose you have a large machine that uses 10 of these motors:

https://www.ebay.com/itm/352974263362?hash=item522ee7d442:g:BJoAAOSwdLJfS9qn

And there is 575VAC 3PH service panel dedicated to this machine and it is located 100 feet from the main service entrance.  Inside the panel there is a main breaker feeding 10 individual breakers for each motor.

https://www.grainger.com/product/SQUARE-D-Molded-Case-Circuit-Breaker-6NHE7

Note that the breaker will trip quickly (magnetically) at 12,000 amperes.  You hook up your UNI-T (I'll stick with the Fluke, thanks) meter to the output of one of the phases of the main breaker and there is a transient of 6000 volts.

That, AFAIK, is CAT III/600V.  Someone (not me!) should make that video. 

We still don't know for sure if what we're looking at is a spark gap or where it is in the meter's input circuit, eg. Is it in series with the fuse? To me it looks like there could be a current shunt across that gap in some variant of the mater (between the two vias where the vertical white line is painted).

It does appear to be a spark gap, at least in this version, and it is inline with the fuse, not directly across the inputs.

[2N3055]

If that happens, arc would exit the case following the cables.

It might.... if you used it outside the CAT III environment that it's rated for.

CAT III is already a serious place. It is circuit just after or on fuse panel. High fault currents are still possible.
For instance you have an 200kW elevator motor, that one would be CAT III because it is behind junction box and has a fuse.

A phase-to-phase fault on a 480-V system with 20,000 amperes of fault current provides 9600000 watts of power (9,6 MW). If the fault lasts for 200 milliseconds before the overcurrent protection clears it, the released energy would be 1,92 MJ, which corresponds roughly to a stick of dynamite.

At 20000A current, at a distance of 0.5 m, light intensity might reach magnitude of about 1,8 M lux! 

Fault currents in CAT III can be up to 25kA... I assure you that is a serious arc fault accident.

[AndrewBCN]

Yes, there are high resolution pictures of the top and bottom sides of the UT125C PCB on HKJ's website.

Can you pop the fuse out and confirm which end the positive lead is connected to?

Edit: it appears from one of HKJ's other photos that it is connected to the top (near the PTCs) as one would hope.

Yes, the positive lead is connected to the side of the fuse that meets the PTCs. So I would guess if ever there is a transient that arcs across the gap, it should blow the fuse?

[AndrewBCN]

...
We still don't know for sure if what we're looking at is a spark gap or where it is in the meter's input circuit, eg. Is it in series with the fuse?
...

Yes, it is indeed a spark gap (the edges are plated) and yes, it is in series with the fuse.

... (insulation is badly damaged on cables on that photo..) ...

I had noticed that in HKJ's photo, and I checked mine: the insulation is not damaged in the unit I received.

[bdunham7]

Yes, the positive lead is connected to the side of the fuse that meets the PTCs. So I would guess if ever there is a transient that arcs across the gap, it should blow the fuse?

That's the idea, I'm sure.  As long as the fault current is less than the interrupt rating of the fuse, it should all work out.  If not there likely won't be enough left of the meter to determine who to sue.  Of course by using a spark gap instead of a sealed GDT, you've now ionized the interior of the meter, which may not be good in a dicey situation.  But UNI-T isn't alone in using intentional or unintentional spark gaps.

[AndrewBCN]

...
So here is a plausible example.  Suppose you have a large machine that uses 10 of these motors:
...

...
A phase-to-phase fault on a 480-V system with 20,000 amperes of fault current provides 9600000 watts of power (9,6 MW).
...

Excluding the apocalyptic scenarios, we are talking about a $15 10cm tall DMM for home use, certainly not an electrician's tool for use in industrial settings. All I am interested in is that it is independently certified to a CAT III 600V rating. And of course I am curious about the spark gap. Btw if you look closely it also has a PCB cutout under the (beefy?) PTCs. So it seems that UNI-T spent a few extra cents to try to meet the CAT III rating, but was the effort and money well spent?

[2N3055]

Yes, there are high resolution pictures of the top and bottom sides of the UT125C PCB on HKJ's website.

Can you pop the fuse out and confirm which end the positive lead is connected to?

Edit: it appears from one of HKJ's other photos that it is connected to the top (near the PTCs) as one would hope.

Yes, the positive lead is connected to the side of the fuse that meets the PTCs. So I would guess if ever there is a transient that arcs across the gap, it should blow the fuse?

It will blow the fuse, but they are slow devices. If current trough the fuse creates plasma near input terminals, it can resume taking current there, directly, and keep sustaining there, taking fuse out of the circuit.

In permanent installations, the only way to stop it is to cut power closer to the source, via protective switch that is designed for such fault current disconnections, otherwise it can arc over too... Unstopped, arc can travel hundreds of meters down the cables... It is a scary thing.

In this case, input cables would fail, and if there is a bit of luck they are connected to two sides far enough, it might extinguish when cables gap gets to big..

But all that is just talk.
People working on these kinds places where there is so much energy, should have proper tools, protective equipment and training in proper procedures and safety.

These meters are not what should be used in that case. No current range should exist on meter. Only noncontact current measurement....

[2N3055]

...
So here is a plausible example.  Suppose you have a large machine that uses 10 of these motors:
...

...
A phase-to-phase fault on a 480-V system with 20,000 amperes of fault current provides 9600000 watts of power (9,6 MW).
...

Excluding the apocalyptic scenarios, we are talking about a $15 10cm tall DMM for home use, certainly not an electrician's tool for use in industrial settings. All I am interested in is that it is independently certified to a CAT III 600V rating. And of course I am curious about the spark gap. Btw if you look closely it also has a PCB cutout under the (beefy?) PTCs. So it seems that UNI-T spent a few extra cents to try to meet the CAT III rating, but was the effort and money well spent?

As I said, I would be reluctant to use it in CAT III environment.

And that apocalyptical scenario is requirement connected with CAT III.  Without it, it didn't pass... CAT III is industrial use category. Claiming it has CAT III makes it professional range meter..

And CAT II and CAT I you'll be fine with pretty much any meter. Sure it will be loud (or not so loud) bang. But you'll be safe. Inside house and in electronics lab, I agree with Joe: it bothers me more that you can damage those with static electricity. This is more likely to happen to you. Other apocalyptical scenarios can't happen inside your home. There is not enough energy. Except MOT.. those are nasty..

But a elevator technician can get in lot's of trouble in same building where your apartment is. 

[AndrewBCN]

...
Yes, the positive lead is connected to the side of the fuse that meets the PTCs. So I would guess if ever there is a transient that arcs across the gap, it should blow the fuse?

It will blow the fuse, but they are slow devices. ...

I believe there is a PTC+MOV in parallel with the spark gap + fuse across the input leads, shouldn't that react faster than the fuse?

So the transient energy would be absorbed in two steps, first by the PTC+MOV combo and then eventually second, by arcing across the spark gap and blowing the fuse?

I haven't measured the width of the spark gap but it seems to be about 1mm to 1.5mm wide, so the transient would need to be above 3kV to 4.5kV to arc across it, or am I getting it wrong?

[bdunham7]

All I am interested in is that it is independently certified to a CAT III 600V rating.

That is actually my concern as well, but from perhaps a different angle.  You really don't need CAT III around the house and and in most cases, there is not an actual arc-flash hazard anywhere in home wiring.  That's not to say you can't electrocute or burn yourself.  But people like the reassurance of having those labels because they see it as an accurate, impartial indication of quality and I don't think it is--this meter being pretty strong evidence of that.  So the casual user gets the reassurance of a (fake, IMO, certification or not) safety rating at a bargain price and the professional user knows to avoid them--so everyone's happy?  If that's good enough, then I guess that's just the modern world. 

The UNI-T is probably an OK $25 meter for hobby and household use, although the cable damage in HKJ's photo doesn't look good.  There's no way it should be wearing a CAT III/600V label, IMO, because that implies that it will withstand those 'apocalytic' scenarios.  How it got that label is what interests me, not whether or not it is a cheap piece of crap--as if that was a question.  And if it turns out that it got that CAT III/600V rating through entirely legitimate means, then the CAT ratings themselves are not a very strong indication of anything.  And that is bad for everyone, IMO.

[2N3055]

...
Yes, the positive lead is connected to the side of the fuse that meets the PTCs. So I would guess if ever there is a transient that arcs across the gap, it should blow the fuse?

It will blow the fuse, but they are slow devices. ...

I believe there is a PTC+MOV in parallel with the spark gap + fuse across the input leads, shouldn't that react faster than the fuse?

So the transient energy would be absorbed in two steps, first by the PTC+MOV combo and then eventually second, by arcing across the spark gap and blowing the fuse?

I haven't measured the width of the spark gap but it seems to be about 1mm to 1.5mm wide, so the transient would need to be above 3kV to 4.5kV to arc across it, or am I getting it wrong?

With arc accidents, it is not transient that does damage. Transient creates plasma that creates short fault.
And power grid powers the explosion

Also in current measurements , PTC or MOV is not in a circuit. Only fuse and shunt...

[bdunham7]

I believe there is a PTC+MOV in parallel with the spark gap + fuse across the input leads, shouldn't that react faster than the fuse?
I haven't measured the width of the spark gap but it seems to be about 1mm to 1.5mm wide, so the transient would need to be above 3kV to 4.5kV to arc across it, or am I getting it wrong?

Looks like 1 PTC +2 MOVs in series from the + lead to the large  (ground?) plane, but I don't see a direct connection to the - (COM) lead and in any case it is a pretty circuitous route. CAT III/600V test transients are 6kV/2 ohm, so 3000 amps?  That long, thin trace up the side of the board to the COB isn't going to be carrying that and the other route seems to go through the switch and 'stuff'. 

Yes, to jump the gap the transient would need to be ~3kV/mm on a dry day with no contamination of the inside of the meter. 

[joeqsmith]

Btw if you look closely it also has a PCB cutout under the (beefy?) PTCs.

    As I stated, there's nothing there to save the PTCs except the PCB and leads.  Maybe the MOVs would open up first. 

Want to see what I consider a beefy PTC that can handle some shit?  Look at Fluke, Gossen, HIOKI.   See something else that's missing in your low end meter?   

...., but Germans being traditionally thorough, ....

I'm sorry that shield shown on the German designed Gossen looks like crap but hey, when they overlook such a basic problem there's not much else to do.  That's what you want, your customers solving your problems.   And when your distributors start spreading lies and corporate listens and propagate it without even asking the customer,  I don't consider that thorough.   That sounds like our news media.  I  see this as an even bigger problem giving us a glimpse into the culture.    I'm sure their engineers could have addressed the problems I found, but the best solution they could come up with was to change the name of the meter.  That's certainly being thorough. 

I feel sorry for the engineers that designed the electronics.  The basics are all there.  Move up the food chain, I'm not impressed. 


****

Those surge rated resistors are expensive.  Lets just put two 5mm PTCs in series.  What could go wrong?

[Fungus]

A phase-to-phase fault on a 480-V system with 20,000 amperes of fault current provides 9600000 watts of power (9,6 MW). If the fault lasts for 200 milliseconds before the overcurrent protection clears it, the released energy would be 1,92 MJ, which corresponds roughly to a stick of dynamite.

Sure, but would you even trust your CAT IV Fluke set to amps mode in that situation? I'd pay to watch you do it. 

My domestic 220V, 20A distribution panel is also CAT III. It seems a more likely scenario for somebody to use a $15 Uni-T there. I've had domestic short circuits before now and the walls of the house are still standing, not even any windows broken.

[Fungus]

if it turns out that it got that CAT III/600V rating through entirely legitimate means, then the CAT ratings themselves are not a very strong indication of anything.  And that is bad for everyone, IMO.

CAT ratings have many grey areas. It's easy to invent worst-case scenarios for any category.

I'm not going to pretend to understand them but I'll happily point out that a CAT III 600V rating isn't automatically a CAT IV 300V rating even though they both say "6000V @ 2 Ohms" in the cute little chart.

ie. It's a lot more complicated than multiplying volts by amps.

[Fungus]

With arc accidents, it is not transient that does damage. Transient creates plasma that creates short fault.
And power grid powers the explosion

Correct, but we have no idea how many volts are needed to jump the spark gap in this meter

(if it even is a spark gap - why would you have a spark gap in series with a fuse? To make sure the fuse blows faster?)

[AndrewBCN]

...
    As I stated, there's nothing there to save the PTCs except the PCB and leads.  Maybe the MOVs would open up first. 
...
Those surge rated resistors are expensive.  Lets just put two 5mm PTCs in series.  What could go wrong?

I think I understand: a surge rated resistor is needed in series with the PTC, to absorb part of the energy of the transient until the PTC has enough time to heat up and see its resistance value increase. If there is no surge rated resistor, the PTC absorbs all the energy of the transient before its resistance value has enough time to increase and could eventually fail catastrophically.

IIRC the CATIII 600V rating means the DUT has to withstand 10 repeated transients, so if the PTC fails catastrophically after the first or second transient, the DUT does not get the rating?

[Fungus]

IIRC the CATIII 600V rating means the DUT has to withstand 10 repeated transients, so if the PTC fails catastrophically after the first or second transient, the DUT does not get the rating?

The way I understand it is that the standards only say that the meter has to fail in a safe manner.

ie. the PTC can explode during the first transient and it still gets the rating if no shrapnel penetrates the case.


(Edit: For those who still don't get it... I sometimes exaggerate a tiny bit to get points across)

[joeqsmith]

...
    As I stated, there's nothing there to save the PTCs except the PCB and leads.  Maybe the MOVs would open up first. 
...
Those surge rated resistors are expensive.  Lets just put two 5mm PTCs in series.  What could go wrong?

I think I understand: a surge rated resistor is needed in series with the PTC, to absorb part of the energy of the transient until the PTC has enough time to heat up and see its resistance value increase. If there is no surge rated resistor, the PTC absorbs all the energy of the transient before its resistance value has enough time to increase and could eventually fail catastrophically.

IIRC the CATIII 600V rating means the DUT has to withstand 10 repeated transients, so if the PTC fails catastrophically after the first or second transient, the DUT does not get the rating?

It seems you feel that the transients I apply cause the PTCs to switch.   I have no idea why you would feel this way, but I've made whole videos on this subject.   You could also look at the datasheets for various PTCs and see what are the effects of various package sizes.

****
As far as the safety ratings, I have provided various quotes from the standards along with two large companies interpretations of them.  I've also tried to make it very clear that I am not an electrician and have little interest in the subject.  I've also tried to make it very clear that my transient tests have nothing to do with these standards.   I'm actually surprised how the majority of these threads become polluted with chatter on safety.  It wouldn't be too bad if you actually had people who worked in this industry chiming in.   

As far as what gets approval, hard to say.  Again, anymore I see the standards as outdated and more about trade than safety.