CAT ratings and interpretation

[Jeff1946]

My way of looking at the CAT rating is no matter what you do, the meter is safe to test any system with the rated voltage.  For example a CAT 600 rating would mean you could use it safely on say a large motor powered with 440 VAC.  So if you touch a probe to 440 and hold the meter and have your feet grounded no shock.  Or you forget and leave it on amps and connect to 440 -- you just blow the fuse.  I assume that putting 440 would be ok (not burn out some circuit element) for all potential situations such as ohms, diode, capacitance.   Not that I would try it with my own meter.

[T4P]

wytnucls, you're still getting it wrong.
You banged on about the UT71E not going to have anything happen to it if you measure amps. Okay.
But did you know what happens if you attempt to measure voltage with the leads in amps? BOOM! If it was 600V? Your supposedly very safe glass fuse BLOWS UP! I have seen glass fuses break into two even with a ultra fast trip breaker (Yes, i have one, i paid good money for that, they trip in 10mS)
600V IS FLOWING THROUGH THE DAMN FUSE!
Okay then, what's the normal procedure of measuring amps? Break up a circuit right? So technically isn't the circuit's FULL voltage flowing through the fuse?
Do you know fuses are safely rated for the voltage before they ARC over even after breaking? Did you know a BS1362 style fuse whether glass or ceramic is 32VDC?
YES. 32VDC. 250VAC. What happens if you have 100amps at 600V flowing through the fuse? BOOM IT GOES and it shatters violently.
What's worse if it doesn't shatter, it will arc over and possibly meltdown your meter and well your hand too if the extremely loud pop doesn't scare you yet
But after that possibly causing a huge fire due to overheated glass from the plasma in the arc

You know why the safety folks are banging on about CAT ratings? This is the REASON. That's why i'm not too sure about anything UNI-T, but i can certainly the both the fuses in the UT61E to HRC fuse and sockets (like monkeh did, props to him) and install MOVs, i'm all set.
If CAT ratings were supposed to be with what? 100A breaking current fuses? If It wasn't neccesary fluke wouldn't bother and all meters will jeopardise somebody's life around the globe every second.
8000V across 2ohms is 40kA and you should know that.

You seem to bang on about not needing HRC fuses, but the truth is, your life depends on it. If you do say "below 50V" What happens if the seemingly not dangerous 32V circuit HAS a potential 500amps flowing through it? BOOM! Glass fuses shatter easily.

The breakover voltage for air might be 1kV/inch but you're wrong if you think the fuse will not arc from end to end, WHEN THEY BREAK they are effectively a few mm apart and that is a recipe for ultimate disaster, that's why you need big or really long fuses
Big fuse = more distance to separate the bondwires and well more sand
Long fuse = They might be held very tight, so if they break they just go off to the other end bu they have less breaking current
MORE CURRENT = FATTER ARC, MORE VOLTAGE = LONGER ARC
Alas, not all BS1362 ceramic fuses are safe. Some of them DO not have sand and thus are not 6kA (or so)
And some glass fuses are probably good for 20amps breaking current MAX
https://www.eevblog.com/forum/product-reviews-photos-and-discussion/beware-the-price-of-some-fluke-fuse-replacement/msg114936/#msg114936

[Lightages]

The document is password protected so cutting and pasting is not possible.

My findings from the document:

1. If there is a choice or range of tests that can be performed on a piece of equipment, the worst case test is to be used.

2. All multimeters need to have their maximum safe voltage and CAT rating marked at each terminal unless the capacity is generally stated or understood to be less than 50VAC or 120VDC.

3. All replaceable fuses must be marked with their voltage and type on the circuit board or holder.

4. Handheld equipment must survive a drop from 1m onto a hardwood surface.

5. Creepage distance for a coated printed circuit board at 1000VDC is a minimum of 3.2mm with a test voltage of 4000VDC

It appears that the document link provided is not a complete specification for what we are talking about here. But it is clear that unless a multimeter has clearly stated limits on each of its inputs, it is not correctly labelled.

Still no definitive answer from the documents directly but from what I have read so far on the Gossen site and the document linked here, all connections on the meter must meet the CAT rating and must be labelled so or the meter does not comply to the standards.

[Monkeh]

Did you know a BS1362 style fuse whether glass or ceramic is 50VDC?
YES. 50VDC.

NO. NOT 50VDC. And there's no such thing as a glass BS1362 fuse.

[Lightages]

Based on my point #2, it appears that even if the UT71E does pass a test for whatever category it may actually be rated at, it is no compliant to the standards because the current measurement terminals do not state any rating, never mind that they are only good to 250V.

[T4P]

well not glass BS1362, i mean 5x20 glass fuse. Sorry about that, BS1362 means 6x30 ceramic fuse (i think, because 6x30 ceramic fuses turn up when you search in farnell) but the UT71E appears to use 6x30 glass fuses

@Lightages
At least UNI-T has been kind to write "250V MAX" on the UT61E, BUT NOT THE 71!

[Monkeh]

well not glass BS1362, i mean 5x20 glass fuse. Sorry about that, BS1362 means 6x30 ceramic fuse (i think, because 6x30 ceramic fuses turn up when you search in farnell)

6.3x25.4mm. 1"x1/4".

Alas, not all BS1362 ceramic fuses are safe. Some of them DO not have sand and thus are not 6kA (or so)

Then it's not a BS1362 fuse.

[Wytnucls]

Based on my point #2, it appears that even if the UT71E does pass a test for whatever category it may actually be rated at, it is no compliant to the standards because the current measurement terminals do not state any rating, never mind that they are only good to 250V.

If I follow your reasoning, this one doesn't meet its CAT rating either, because it has the same markings as the 71D.

[Chasm]

Accidents happen in the real world which are similar to the test. Lets derail this thread even more.
(I think I told this one before, it still applies.)

Industrial setting, putting large machinery into operation. The paperwork for the completion certificate calls for measuring the input voltage. No big deal it's 230/400V the same as in most houses outside of 120V country. Just a bit more current in these busbars. (Like in any good horror movie you should now know will happen by now.)

So you grab your brand meter (genuine!) from the tool box, switch it to voltage and take a measurement. Then you notice that the test leads are getting warmer and warmer. A quick look at the meter tells you that you are using the Amps jack instead of the Volts one. Oops! Now you make the significant mistake. You flinch back and manage to draw an arc across the busbars. A few milliseconds later vapor state copper begins to impact your body, followed by large parts of the busbars in liquid form a little bit late. You are very lucky and wearing your personal safety equipment so the next stop is emergency room,not the morgue.  A few month and operations later (extensive work on hands and face) you can go back to work, the desk work portion of it.


So what did happen?
The voltage measurement was taken at the hook up point of the machinery. Which happened to be very close to the 20kV substation transformer. When the test leads created the short between two phases the current, for all practical purposes, was only limited by the resistance of the test leads and the internal resistance of the Amps range. A substation transformer is pretty much the definition od a really low impedance source. The current exceeded the rupture capacity of the fuse. Current continued to flow and things started to get warm in a hurry.

The meter was an original of Daves favorite brand. In full working condition, no tampering whatsoeve, original fuse, with current safety certificate. One of the older charcoal ones, maybe a 77 I don't know exactly. (The accident happened at the company my father worked for.)
As a result of the accident the company trashed all those meters mighty fast and went with Daves 2nd favorite brand instead. The mechanical jack covers would have prevented this accident.

[T4P]

No. That's different. Always check the manual.
UT71E's datasheet has nothing warning you about the very low capacity of the fuses or the 250VAC limit, but the UT61E does warn you, right in front of the meter.
So the UT71E is a stockpile of danger brewing, first it doesn't only probably kill you, secondly you might not even know if it will, if you didn't know they used glass fuses
As for the UT61E, you know you are in danger if you didn't check the voltage BUT at least they WARNED you. IT's written 250V MAX so you clearly the UT61E is a "much better" in terms of safety. Elsewhere too

Wait, let me paint another scene for ya. Look at those caps in my picture? Those are 35V 10000uF caps. What if you ACCIDENTALLY measure them in amps while in volts?

[Lightages]

If I follow your reasoning, this one doesn't meet its CAT rating either, because it has the same markings as the 71D.

It is possible, and I may be wrong too as I stated as it was what I could read from a related document. But the document I downloaded that you linked clearly states that ALL terminals must be marked with their rated capacity. On top of that the battery wire in the UT71E passes through a hole in the circuit board in almost direct contact with the input trace, before any input protection. The battery wire has an insulation rating of 600V, a bit short of the 1000V that it might see.

Until we have a full document from the 61010 standard it is still guessing.

[Wytnucls]

Ok, I can agree with you on this one. Dave may have received a copy. Let's wait and see.

[Lightages]

It is also possible that multimeter was designed and released before the latest standard made the markings mandatory, or the image you have found is from before that time and that the new ones have the markings required.

[Wytnucls]

No, all the Flukes are marked about the same, (check their site)
The link I posted points to pretty old regulations though, so the new ones could be quite different.

[Lightages]

And for all those installing MOVs or other over voltage devices in your meter where there are empty footprints, maybe you should consider the following possibility.

It is possible that the original design called for these components but under standards compliance testing the over voltage devices became THE hazard and were left out to avoid having little hand grenades inside the meter. It is possible that the meter passed the compliance tests after the over voltage devices were removed.

[Neilm]

As I recall 61010, the instrument should be tested between all combinations of compatible leads - so if you can plug the same lead into both amps and volts, you have to do the voltage test on the amps range. I falls under the forseeable misuse component and risk assessments at the back of the main standard. I cannot remember if it is explicitly mentioned in 61010-2-030. It certainly was in the old edition 2 where there was a specific test at the end (section 15.4 from memory) that covered accidental mis-connections.

Neil

[Lightages]

So from your experience Neil, the Gossen web page is correct?

[Wytnucls]

As I recall 61010, the instrument should be tested between all combinations of compatible leads - so if you can plug the same lead into both amps and volts, you have to do the voltage test on the amps range. I falls under the forseeable misuse component and risk assessments at the back of the main standard. I cannot remember if it is explicitly mentioned in 61010-2-030. It certainly was in the old edition 2 where there was a specific test at the end (section 15.4 from memory) that covered accidental mis-connections.

Neil

That would tie up, with the Gossen remarks, about applying the highest rating voltage (600V or 1000V) on all input jacks in turn, to make sure no bodily damage can result from misuse.

[alm]

The voltage measurement was taken at the hook up point of the machinery. Which happened to be very close to the 20kV substation transformer.

This most likely makes it a CAT IV circuit. I think it applies to circuits with less than 15 meters or so of wiring from the substation.

The meter was an original of Daves favorite brand. In full working condition, no tampering whatsoeve, original fuse, with current safety certificate. One of the older charcoal ones, maybe a 77 I don't know exactly. (The accident happened at the company my father worked for.)

Was the meter CAT IV 400V rated? Assuming this event does not predate IEC 1010. If it does, then it's a good example why CAT ratings were introduced.

[Chasm]

Was the meter CAT IV 400V rated? Assuming this event does not predate IEC 1010. If it does, then it's a good example why CAT ratings were introduced.

No idea, too long ago and 2nd hand. Chances are that this was before CAT rating.

I guess one of the things to take away is that it does not matter if the meter is rendered unusable in some tests. There are real life situations which can take it way beyond the normal operational parameters. The question is if it fails in a safe manner. Here the goal is interrupting the current and only after that containing the blast.

[Neilm]

As I recall 61010, the instrument should be tested between all combinations of compatible leads - so if you can plug the same lead into both amps and volts, you have to do the voltage test on the amps range. I falls under the forseeable misuse component and risk assessments at the back of the main standard. I cannot remember if it is explicitly mentioned in 61010-2-030. It certainly was in the old edition 2 where there was a specific test at the end (section 15.4 from memory) that covered accidental mis-connections.

Neil

That would tie up, with the Gossen remarks, about applying the highest rating voltage (600V or 1000V) on all input jacks in turn, to make sure no bodily damage can result from misuse.

The transient test for a CAT rating is done on the live voltage, usually at the top of the sine wave for maximum voltage. If you attach to an ammeter and apply volts the fuse might have already blown before this peak is reached. A good manufacturer might blow the fuse and then apply the transient to check that it is still OK but I don't believe that this is called for. However, I don't usually work with testing multimeters and the like so I won't say that my word would stand in a court of law.

Neil

Neil

[EEVblog]

I now have access to the 61010 standard and various attachments.
The most applicable part in this case appears to be:
From 61010-2-033:

101.3   Protection against  mismatches of inputs and ranges
101.3.1   General 
In NORMAL CONDITION   and in cases of  REASONABLY FORESEEABLE MISUSE , no HAZARD  shall arise
when the  highest  RATED  voltage or current of a measuring  circuit TERMINAL  is applied to any
other compatible TERMINAL, with any combination of function and range settings. 
NOTE  1  Mismatches of inputs and ranges are examples of REASONABLY FORESEEABLE MISUSE, even if the
documentation or markings prohibit such mismatch .   A typical example is inadvertent connection of a high voltage to
a measuring input intended for current or resistance. Possible HAZARDS  include electric shock, burns, fire, arcing
and explosion. 
NOTE  2  T ERMINALS  that are clearly not of similar types and that will not retain the  TERMINALS  of the probe or
accessory need not be tested. 
The equipment shall provide  protection against these  HAZARDS. One of the  following
techniques shall be used. 
a)   Use of a certified overcurrent  protection device to interrupt short- circuit currents before a
HAZARD  arises. In this case, the requirements and test of 101.3.2 apply.
b)   Use an uncert ified current limitation device , an impedance,   or a combination of  both  to
prevent the  HAZARD  from arising. In this case, the tests of 101.3.3 apply.
Conformity is checked by inspection, evaluation of the design of the equipment, and as
specified in  101.3. 2 to 101.3.3, as applicable.
These tests  shall  be performed with any probe assemblies  supplied by the manufacturer, and
repeated with the test leads of 101.3.4.

and

101.3.2   Protection by a certified overcurrent protection device
An overcurrent protection devic e is considered suitable if it is certified by an independent
laboratory to meet all of the following requirements. 
a)   The a.c. and d.c.  RATED  voltages of the overcurrent protection device shall be at least as
high as, respectively, the highest a.c. and d.c.  RATED  voltages of any measuring  circuit 
TERMINAL  on the equipment. 
b)   The RATED  time- current characteristic (speed) of the overcurrent protection device shall be
such that no  HAZARD  will result from any possible combination of RATED  input voltages,
TERMINALS,  and range selection. 
NOTE   In practice, downstream circuit elements such as components and printed wiring board traces  are
selected to be able to withstand the energy that the overcurrent protection device will let through.
c)   The a.c. and d.c. RATED  breaking capacities of the overcurrent protection device shall
exceed, respectively, the possible a.c. and d.c. short- circuit currents.
The possible a.c. and d.c. short- circuit currents  are  calculated as the highest  RATED
voltage for any  TERMINAL  divided by the impedance of the overcurrent- protected
measuring circuit, taking the impedance of the test leads specified in 101.3.4 into account.
The possible a.c. short - circuit current need not exceed the applicable value of Table   AA.1.
Additionally, spacings surrounding the overcurrent protectio n device in the equipment and
following the protection device in the measuring circuit shall be sufficiently large to prevent
arcing after the protection device opens.
Conformity is checked by inspection of the RATINGS   of the overcurrent protection device  and
by the following test .
If the protection device is a fuse, it is replaced with an open - circuited fuse. If the protection
device is a circuit breaker, it is set to its open position. A voltage of two times the  highest
RATED  voltage for any  TERMINAL  is a pplied to the  TERMINALS  of the overcurrent- protected
measuring circuit for 1 min. The source of the test voltage shall be capable of delivering
500   VA. During and after the test, no damage to the equipment shall occur.

So it seems that Gossen are correct about the all terminals test thing. But I cannot (yet) find mention of any CAT-like surge test requirement for the AMP range, hence I guess why the Amps  jacks are not CAT labelled. But still the testing must be done at the "maximum rated" voltage i.e 600/1000V presumably with a suitable high current capability source.
A further table shows a CAT-II source having <10kA short circuit current, with CAT-III <50kA, and CAT-IV >>50kA

As I said before, there is no requirment to survive this, it just has to "fail safe" and present to hazard to the user.

Dave.

[Lightages]

Thank you Dave.

It is then clear, very clear, that multimeters such as the UT71E that have 250v glass fuses on the current ranges do not meet their claimed CAT ratings. It therefore follows that they probably were not even tested for compliance by any 3rd party.

I am glad that the standards are as I had thought and that I had not gone crazy.

[Lightages]

Dave:

Could you post the actual specifications for voltages, pulses, and impedance used for each CAT rating level?

[EEVblog]

It is then clear, very clear, that multimeters such as the UT71E that have 250v glass fuses on the current ranges do not meet their claimed CAT ratings. It therefore follows that they probably were not even tested for compliance by any 3rd party.

Just check to see if it's got a third party testing stamp like TUV et.al on the back. If not then it hasn't been independently tested and certified.
BTW, there is no legal requirement to do this, you can either test it in-house, or just suspect it passes based on design. Anyone can slap a CAT rating on their meter, as there is no body to legally enforce it.
Of course, you need the CAT rating to sell into the EU now, but I believe still no law to say it has to be independently tested.

Dave.