CAT ratings and interpretation

[EEVblog]

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

It does not seem to be available in a simple paragraph or table etc.

Dave.

[EEVblog]

14.101  Circuits or components used as  TRANSIENT OVERVOLTAG E   limiting devices in
measuring circuits used to measure M AINS
If control of  TRANSIENT OVERVOLTAG E   is employed in a measuring circuit used to measure
MAINS, any overvoltage limiting component or circuit shall have adequate strength to limit
likely  TRANSIENT OVERVOLTAG ES  in NORMAL USE.
Conformity is checked by applying 5 positive and 5 negative impulses with the applicable
impulse withstand voltage of Table 102, spaced up to 1   min apart, from a hybrid impulse
generator (see  IEC  6 1180- 1). The generator produces an open- circuit voltage waveform of
1,2/50   µ s, a short - circuit current waveform of 8/20  µ s, with an output impedance (peak open -
circ uit voltage divided by peak short - circuit current) of 2 ? . Resistance may be added in
series if needed to raise the impedance.   The test impulse is applied in combination with the
MAINS  voltage. The   MAINS  voltage is the  highest  RATED  voltage of the  measuring circuit
TERMINALS,   but  no more than  400  V  a.c. r.m.s.
The test voltage is applied between each pair of TERMINALS,  used to measure  MAINS,  where
voltage- limiting devices are present. 
NOTE   This test can be extremely hazardous. Explosion shields and other provisions can be used to protect
personnel performing the test. 
No HAZARD  shall arise due to the operation of overvoltage limiting component. The component
shall not rupture and shall operate as intended during the test.  If the component is heated  as
a result of this test, it shall not heat other materials to their ignition points.   Tripping the circuit
breaker of the  MAINS  installation is an indication of failure.  If the results of the test   are
questionable or inconclusive, the test is to be repeated  two   more times.
Table 102 –  Impulse withstand voltages
 
                                                              V Impulse withstand voltage 
MEASUREMENT
AC VOLTS LINE NEUTRAL OR DC        CATEGORY  III      CATEGORY  IV 
300                                                          4 000                   6 000 
>   300    <=   600                                   6 000                   8 000 
>   600    <=   1 000                                8 000                   12 000
 

[mikeselectricstuff]

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.

Is it not the combination of fuse type and casing that determines whether or not it meets the  rating? A glass fuse in a suitable case could be as safe as a sand-filled one, providing there was no arc-over risk across the fuse (would probably need to be bigger than 20mm!.

Just wondering - do the CAT specs define that it must be user-safe, or also not damage the instrument apart from the fuse blowing?

[EEVblog]

Is it not the combination of fuse type and casing that determines whether or not it meets the  rating? A glass fuse in a suitable case could be as safe as a sand-filled one, providing there was no arc-over risk across the fuse (would probably need to be bigger than 20mm!.

It is broken down into two types of protection (fuse), certified (presumably certain brand/type HRC fuses), and un-certified:

101.3.3   Protection by uncertified cur rent limitation devices or by impedances 
Devices used for current limitation shall be capable of safely withstanding, dissipating, or
interrupting the energy that will be applied as a result of short- circuit current in the case of
REASONABLY FORESEEABLE MI SUSE .
An impedance used for limitation of current shall be one or more of the following .
a)   An appropriate single component which is constructed, selected, and tested so that safety
and reliability for protection against  relevant  HAZARDS  is   assured. In particular, the
component shall 
1)   be  RATED  for the maximum voltage that may be present during the  REASONABLY
FORESEEABLE MISUSE  event ;
2)   if a resistor, be  RATED  for twice the power   or energy  dissipation that may result from
the  REASONABLY FORESEEABLE MISUSE   event ;
3)   meet the applicable  CLEARANCE  and   CREEPAGE DISTANCE   requirements  of Annex K for
REINFORCED INSULATION   between its terminations.
b)   A combination of components which shall :
1)   withstand the maximum voltage that may be present during the  REASONABLY
FORESEEABLE MISUSE  event ;
2)   be able to dissipate the power  or  energy  that may result from the  REASONABLY
FORESEEABLE MISUSE  event ;
3)   meet the applicable  CLEARANCE  and   CREEPAGE DISTANCE   requirements  of Annex K for
REINFORCED INSULATION   between the termi nations of the combination of   components,
the test is repeated with the function or range controls in every combination of positions.
During and after the test, no  HAZARD  shall arise, nor shall there be any evidence of fire,
arcing, explosion, or damage to impedance limitation devices or any component intended to
provide protection against electric shock, heat, arc or fire, including the ENCLOSURE  and traces
on the printed wiring board. Any damage to a device used for current limitation shall be
ignored if other parts of the equipment were not affected during the test. 
During the test, the voltage output of the source is measured. If the source voltage decreases
by more than 20 % for more than 10 ms, the test is considered inconclusive and is repeated
with a lower impedance source.

So yes, I guess in theory you can use any type of fuse or other scheme provided that it meets the interrupting capability.
But multimeter are relatively small, and you have to meet clearance and creepage distance as well. So it is very unlikey that any glass fuse could fulfil both requirements in a standard size meter.
So as a meter designer it's easier and more beneficial to use an already approved HRC fuse, so the independent test authority does not have to do additional test and inspection of an "uncertified" device.

Just wondering - do the CAT specs define that it must be user-safe, or also not damage the instrument apart from the fuse blowing?

For the voltage CAT tests, it seems so.

101.4  Functional  integrity 
After the voltage of  4.4.2.101  has  been applied to the  METER, the  METER  shall continue to be
able to indicate the presence of HAZARDOUS  LIVE  voltages up to the maximum RATED  voltage.

For the Amps test, no, there seems to be no requirement, it just has to not cause a hazard.

Dave.

[Lightages]

1)   be  RATED  for the maximum voltage that may be present during the  REASONABLY
FORESEEABLE MISUSE  event ;

From this alone, it seems that any fuse must be rated for the CAT rating the meter is listed to have, regardless if it is sufficient or not.

[T4P]

Exactly. My point is clear now, the UT71x is a dangerous object waiting to blow up on your face

[Wytnucls]

It seems to me that most multimeters equipped with fuses will fall under 101.3.2, as fuses are certified overcurrent devices and not current limiting devices.
The requirements for misuse protection are then as such:
1.The fuses have to have a rated voltage of 1000V for meters with that CAT voltage category. (Both A and mA)
2.The rated breaking capacity (Amps) must be equal to 1000V divided by the impedance of the circuit, including test leads, up to a max specified in table AA.1 (?)
3.Traces and fuses are checked for arcing, by applying twice the rated CAT voltage (2000V) for 1 minute to the jacks, with a ruptured fuse in place.

If these are the new regulations, it would be interesting to see what has changed and how that would affect multimeters sold under the old rules.
If Dave could post table AA.1, we would have all the standards for fuse requirements.

[EEVblog]

If Dave could post table AA.1, we would have all the standards for fuse requirements.

I did above.

A further table shows a CAT-II source having <10kA short circuit current, with CAT-III <50kA, and CAT-IV >>50kA

So basically, any meter without a HRC fuse is CAT-I only. i.e. not suitable for any mains use at all, which requires CAT-II minimum.

with the added note:

The short - circuit current is calculated for a 1   000 V line- to - neutral voltage and the
minimum loop impedance. The values of loop impedances (installation impedances) do
not take into account the resistance of the probe assemblies   and impedances internal
to the measuring equipment. These short - circuit currents vary, depending on the
characteristics of the installation.

Dave.

[Wytnucls]

That could be the case, but remember if 1000V divided by the circuit and test lead impedance is less than what is specified in the table (50kA), that can be used instead.
It works out to 50kA without the test leads with 0.02 Ohms for the shunt and fuse.
So, it does look like HRC fuses are required.
I measured my leads and probes resistance at 0.15 Ohms (not sure if the impedance would be very different), so in this case a breaking capacity of 6kA would be enough. Still in HRC territory, I think.

[EEVblog]

So, it does look like HRC fuses are required.
I measured my leads and probes resistance at 0.15 Ohms (not sure if the impedance would be very different), so in this case a breaking capacity of 6kA would be enough. Still in HRC territory, I think.

A typical M205 250V glass fuse (as used in cheap meters) has a breaking capacity in the order of 60A at rated voltage, up to several hundred A at lower voltages.

Dave.

[Wytnucls]

The fast blow ones used in some Uni-T meters, like the 71series, have an even lower breaking capacity: 35A! 
http://docs-europe.electrocomponents.com/webdocs/058a/0900766b8058a7dd.pdf
Is there a date on that document, Dave?

[Lightages]

So after all of this can you agree that the UT71x series do not meet their supposed CAT rating?

[Wytnucls]

Absolutely, and so do not another myriad of multimeters, unless these are brand new regulations.
I never said the 71 met its CAT rating by the way, just that I thought it would be unlikely that a big company like Uni-T would disregard these rules, when it wouldn't be that expensive to comply with them. That is why independent testing is so important, but few people worry about approval stamps on the back of the meters.
We certainly have learned a lot with that document, especially about the amps ranges, which do not require high voltage spikes testing.
The meters are still reasonably safe, if you make sure your leads are in the right positions and you don't connect to potentially low impedance high voltage circuits.
Just in case, I will install 30 kA HRC fuses in my meters, including a GDT on the Volts/Ohms range, since we can't trust the CAT rating on the Volts/Ohms ranges either. Not a garantee of safety, but better than nothing.
 

[T4P]

You do remember the mA fuses in the UT71 is a tight fit right? No HRC fuses are going to fit in there.
That's why it's a dangerous meter even after modifying

[Wytnucls]

You do remember the mA fuses in the UT71 is a tight fit right? No HRC fuses are going to fit in there.
That's why it's a dangerous meter even after modifying

It is a tight fit, but there is space, SIBA makes a small 1000V 500mA 3AG fuse. No sure what the breaking capacity is yet, but that will do, if it is good enough for Amprobe.
http://www.testpath.com/Items/Fuse-500mA-1000V-6x32mm-Pkg-of-4-120-328.htm
The breaking capacity is also 50 kA, so it is an HRC fuse.
http://www.siba-fuses.us/upload/documents/catalogues/G_section_Miniature_Fuses.pdf

[alm]

Keep in mind that the maximum short circuit current through the fuse also depends on the impedance of the circuit. Amprobe may have used a higher resistance shunt, for example. Agilent uses HRC fuses with a higher rupture current than Fluke, this may either be to offer better safety or to compensate for some lower impedance / lower power component in the rest of the circuit. Safety requires a systems engineering approach and actual testing, you can't just add random safety components that are used in other instruments and expect to achieve the same level of safety.

[EEVblog]

Is there a date on that document, Dave?

Yes, the latest June 1012 version. 61010-2-033:2012

Dave.

[EEVblog]

The meters are still reasonably safe, if you make sure your leads are in the right positions and you don't connect to potentially low impedance high voltage circuits.

Which is the definition of CAT-I

Dave.

[Wytnucls]

I hear you, but I'm just replacing 2 fuses and adding a GDT, with a ready footprint on the PCB. No real drama here.
I don't claim this will achieve the same safety level as a Fluke, but it will be better than the original.
That's all I'm interested in.
I seldom work with 250V circuits anyway, so this is mostly a fun soldering exercise, but I will test it as far as I can go.
I know already that it can take 250V on the Volts/Ohms input jack, with any range selected (I dont have a pulse of 12,000V available yet).
I will test a short on the fuses with 250V (I don't have 2,000V available to check for arcing yet).
I'm going to London soon, maybe Electric Mike can help me here... 

[Wytnucls]

Is there a date on that document, Dave?

Yes, the latest June 1012 version. 61010-2-033:2012

Dave.

That's interesting! Do we know what the changes were over the last 4 years for instance?

[alm]

From 61010-2-033:

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.

I find this test very curious. I would think that the most likely moment for arcing to occur is right when the fuse opens. The inductive kick from breaking the circuit might cause the high voltage transient, which could ionize air in the fuse. Where would the supposedly two times the rated voltage come from after the fuse has opened?

[robrenz]

I find this test very curious. I would think that the most likely moment for arcing to occur is right when the fuse opens. The inductive kick from breaking the circuit might cause the high voltage transient, which could ionize air in the fuse. Where would the supposedly two times the rated voltage come from after the fuse has opened?

Maybe that is what the "sand" in the HRC fuse accomplishes, possibly hindering ionization after it supressed the arc flash of the element vaporizing?

[Wytnucls]

The HRC fuse is supposed to smother the initial arc without breaking.
I guess, they just want to make sure that there is no residual arcing across the fuse and traces, after the fuse has done its job.

[alm]

Yes, a HRC fuse is supposed to prevent this and the standard states that they assume the fuse ratings to be correct. But assuming they are not accounting for transients from external sources, the only reason the transient would occur is because the fuse breaks the circuit. Letting everything settle and cool down first and then apply a transient seems like a poor way to test this to me, although ramping up the voltage at the moment the fuse blows might be a tricky test to perform.

[robrenz]

Isn't it a way to access the quality of the physical and electrical circuitry design independent of the fuse?