Multimeter fuses

[Fungus]

My new Fluke 37 has two fuses in series on the 320mA range. A standard size 630mA glass fuse and a big-ass 2A HRC fuse.

I'm guessing the 20 cent glass fuse is there to protect the expensive HRC fuse from silly mistakes like leaving the leads in the wrong holes. This seems like a really good idea to me, why don't modern meters do it?

[Hydron]

Some of them kind-of do, at least bench meters. The DMM6500 has little 3A fast-blow glass fuses accessible from the front/rear 3A sockets, while inside it has a 3.5A HRC fuse on the same range. You'd have to void the cal by opening up the unit to get to it, but at least it might avoid serious damage if you f**k up bad. The 10A range (rear panel only ) has a HRC fuse only (changeable without opening unit this time), but that kinda makes sense given it's a much lower burden-voltage range, which 2 fuses in series would compromise.

[Hexley]

The Keysight 34465A family has series fuses in the 3A path. There is a little 3.14A 500VAC/400VDC time delay fuse that is accessible from the rear panel, in series with the 11A 1000V fast acting fuse inside the case.

For minor overloads, the little 3A fuse handles the job and is easily replaced.

After my first oopsie, I laid in a supply of the 3A fuses and taped a baggy full of them to the back. That worked like a charm -- haven't blown one since. :-)

[Edit: typo]

[dcac]

I'm guessing the 20 cent glass fuse is there to protect the expensive HRC fuse from silly mistakes like leaving the leads in the wrong holes. This seems like a really good idea to me, why don't modern meters do it?

Modern meter manufactures perhaps believe selling expensive HRC fuses is good for business.

[dcac]

This - or similar - could serve the same function.

[HKJ]

This - or similar - could serve the same function.

No, it is a safety hazard.

[bdunham7]

No, it is a safety hazard.

Why? 

[dcac]

Perhaps, one reason I can think of is you might be physically closer to the fuse (i.e. with your hands) in case of it rupturing when blowing in a (very) high energy circuit.

[joeqsmith]

..
I'm guessing the 20 cent glass fuse is there to protect the expensive HRC fuse from silly mistakes like leaving the leads in the wrong holes. This seems like a really good idea to me, why don't modern meters do it?
..

I looked at an old Fluke that has something similar.  I wonder if they arc flash test the meter, and the smaller one breaks down and creates lots of ionized gas, where is the next flash over point.  Does the large HRC fuse even come into play every single time or are there unintended paths.    I agree that from a layman's standpoint, it makes a lot of sense but I am not sure how it plays out in practice.   

I am looking at another 40 year old Fluke that also uses this arrangement.  There is a small fuse that is user accessible from the front of the meter.  Then there are 4Xlarge HRC fuses buried deep inside the meter that your can pull after pretty much taking the meter all the way apart.   This board and others use GDTs.  This thing is a bench meter and would have been used by scientists, not electricians.   It shows the early signs that Fluke was starting to see what it took to protect a meter.

[Fungus]

I looked at an old Fluke that has something similar.  I wonder if they arc flash test the meter, and the smaller one breaks down and creates lots of ionized gas, where is the next flash over point.

Yep. The small one can arc over. The arc would then reach the HRC fuse.

Does the large HRC fuse even come into play every single time or are there unintended paths.    I agree that from a layman's standpoint, it makes a lot of sense but I am not sure how it plays out in practice.   

FWIW the fuses in the 37 are on a small daughterboard that stands away from the meter's main PCB. Hopefully Fluke will have laid out the fuses appropriately. I guess I could take a pic of that board but I'm not sure if it would prove anything.

It shows the early signs that Fluke was starting to see what it took to protect a meter.

I believe the Fluke 37 was made before CAT ratings were a thing but the manual has this to say on safety:



Both of those standards are pay-to-read. 

[dcac]

Fluke 45 has this fuse configuration.




Where the 100mA input is protected by 500mA 250V whats looks like an ordinary 5x20mm glass fuse - but searching the Fluke part number it costs about $8.00.

[joeqsmith]

I looked at an old Fluke that has something similar.  I wonder if they arc flash test the meter, and the smaller one breaks down and creates lots of ionized gas, where is the next flash over point.

Yep. The small one can arc over. The arc would then reach the HRC fuse.

That's the layman's point of view.  The reality is the ionized gas cloud isn't confined to area neat the small fuse.   Rather it can create a large pressure wave.  That expanding gas may cause other areas to break down.    The HRC fuse will contain the arc.  Imagine having the outside of the fuse surrounded by a cloud of conductive gas.   

I ran a small experiment once to demonstrate the gas expansion.   No where near the energy of an actual arc flash event.   

[HKJ]

No, it is a safety hazard.

Why?

Two reason:
1) You measure a voltage and it shows 0, you then assume it is safe to work on it, but the reason the meter showed 0 was a blown fuse.
2) You have the leads plugged into the 10A socket and put the probes across some high power, you may get a very bad hand from that (I doubt the fuse can break kA).

[Fungus]

That's the layman's point of view.  The reality is the ionized gas cloud isn't confined to area neat the small fuse.   Rather it can create a large pressure wave.  That expanding gas may cause other areas to break down.    The HRC fuse will contain the arc.  Imagine having the outside of the fuse surrounded by a cloud of conductive gas.   

Maybe the glass fuse is after the HRC fuse.

[Hydron]

No, it is a safety hazard.

Why?

User checks for live mains voltage using fused lead, sees nothing (due to blown fuse), touches stuff, becomes ex-user. (HKJ beat me to it).

[joeqsmith]

That's the layman's point of view.  The reality is the ionized gas cloud isn't confined to area neat the small fuse.   Rather it can create a large pressure wave.  That expanding gas may cause other areas to break down.    The HRC fuse will contain the arc.  Imagine having the outside of the fuse surrounded by a cloud of conductive gas.   

Maybe the glass fuse is after the HRC fuse.

Before/after, I don't see your point.    If the small glass fuse were to rupture from an arc, I don't think you can with out a doubt say what the path is. 

Let's assume I am full of shit and you and I both agree that this is a 100% safe and controlled method.   (I'm not saying that)   Ask yourself, why did Fluke stop this practice. 

I suspect they helped to shape the safety standards we have today. 

[dcac]

No, it is a safety hazard.

Why?

Two reason:
1) You measure a voltage and it shows 0, you then assume it is safe to work on it, but the reason the meter showed 0 was a blown fuse.
2) You have the leads plugged into the 10A socket and put the probes across some high power, you may get a very bad hand from that (I doubt the fuse can break kA).

These are CAT IV rated fused test leads: http://www.extech.com/products/TL900

"(1000V, 11A, 20kA breaking capacity)"

But of course these fuses are probably just as expensive to replace as any DMM HRC fuse - but still probably easier accessible.

But I agree on your notion that fused test leads could be dangerous for voltage measurements - especially if you don't happen to know they are in fact fused test leads.

[Fungus]

Maybe the glass fuse is after the HRC fuse.

Before/after, I don't see your point.

If it's after the HRC then not enough energy can make it through the HRC fuse to make the glass fuse arc.

Edit: Or, at least, not enough to sustain an arc - the arc would have to be fed through the HRC.

FWIW I looked at the PCB and the glass fuse is indeed after the HRC:

[bdunham7]

1) You measure a voltage and it shows 0, you then assume it is safe to work on it, but the reason the meter showed 0 was a blown fuse.
2) You have the leads plugged into the 10A socket and put the probes across some high power, you may get a very bad hand from that (I doubt the fuse can break kA).

I suppose anything is a safety hazard if you don't use it right.  Scenarios like your #1 can happen with a variety of meter faults, which is why there is a TBT (Touch Before Test) protocol for any type of high energy circuit.

Fused leads and probes are made by well-reputed companies and are not a shady product.  The leads in question have an IEC CAT-IV 1000V rating, for which the manufacturer states a specific fuse is required. 

[Hydron]

RE: joeqsmith's posts:

The bench meter I mentioned (and probably the KS one too) is only CAT II 300V rated, so I guess there is probably less concern about high energy faults (though still enough to include the HRC in series). The fuse is also built into the input jack, so not next to the PCB, though as you say a high enough energy event may not be deterred by an itty little bit of plastic around the fuse (I would also be concerned about the fuse holder leaving the instrument panel at a rate of knots, as amply demonstrated in the video!)

Seems to me like a reasonable trade-off using the dual fuse design in the CAT II bench meter which is not going to be poking around in a switchboard but is likely to have a few over-current mistakes made while measure low voltage DC circuits (it is after all the same type that you'd find on the mains input of most equipment). For handhelds you just have to live with the pain of costly fuse replacements in the name of safety, again it seems reasonable given they are much more likely to be used in situations where high fault current is possible (also teaches the use to be careful - I've blown a few glass fuses in the past but haven't popped an HRC yet thankfully!).

[Fungus]

Ask yourself, why did Fluke stop this practice. 

Penny pinching? Not enough space to do it inside handhelds?

There's other reasons to do things apart from safety.

[dcac]

@Fungus

Is that fuse at the top of your picture for the high current input? It's contact surfaces looks a bit moldy, what kind of burden voltage are you getting?

[Fungus]

@Fungus

Is that fuse at the top of your picture for the high current input? It's contact surfaces looks a bit moldy, what kind of burden voltage are you getting?

Yep, that's the 10A fuse. I hadn't looked closely before but it certainly looks a bit crusty in that photo. 

I just pulled it out and it smells OK, I don't think it's mold.

(seems more like old grease than anything - I gave it a bit of a polish with IPA just in case, it came up a bit shinier)

Edit: Here's a pic. I don't think it's the original Fluke fuse.

[bdunham7]

Before/after, I don't see your point.    If the small glass fuse were to rupture from an arc, I don't think you can with out a doubt say what the path is. 

Let's assume I am full of shit and you and I both agree that this is a 100% safe and controlled method.   (I'm not saying that)   Ask yourself, why did Fluke stop this practice. 

I suspect they helped to shape the safety standards we have today.

Well I agree that there is no before or after unless we're talking about actual lightning or something--the fuses are in series, so the same current must flow through each, at least at first.  You aren't going to rupture the small fuse into an ionized plasma with a small current, say 10 amps.  However, that 10 amps will blow the big fuse very quickly and open the circuit.  The only way I see a problem is if there were a high voltage DC applied with limited current.  Say you had a 1kV DC supply limited to 3 amps--you would blow the small fuse but not the big one and now you have a 3kW arc across the little fuse.  There probably are other scenarios I haven't thought of.

On my version of your meter, the 8505A, I found two blown fuses and the 39.1K X-Ref input resistors were melted.  I've wondered how that could have happened. I looked up those fuses and one thing I noticed is that they are only rated for 170VDC.  I replaced them with a a 1kVAC/DC 10kA rated fuse, which I think is better overall protection than 600VAC/170VDC 100kA.   

https://www.mouser.com/datasheet/2/643/ds-CP-0ADBX-series-1664157.pdf

As far as why Fluke would discontinue the practice (while apparently some other manufacturers continue it) it probably is just a case of consolidating their designs--all of their meters appear to use the same 0.44A/11A pair of fuses--and eliminating both known corner cases and possible unknowns.  IOW, they have a known solution (the particular fuses) so why add parts and expense and increase risks just for a user convenience?

[Fungus]

IOW, they have a known solution (the particular fuses) so why add parts and expense and increase risks just for a user convenience?

The only thing we know for sure is that Fluke doesn't worry about how much people have to pay for fuses.