Help understanding flameproof requirement please

[guitarmike2107]

What is the requirement for flameproof resistors.

I have been looking at a few schematics that stipulate flame proof resistors in certain positions, my understanding is that flameproof resistors should be used where the shorting of the resistor to ground or the overload of the resistor could cause the resistor to catch alight.

So I can figure out that the lower value resistors (less than 25k say) in the power supply would have to be very high wattage to cope with a short to ground, so using resistors that are spec’d at a lower wattage for the application is fine as long as they are flame proof in case they get overloaded.

So the question is where is the line or how do we know when a resistor will catch alight?

For example if I have a b+ of 350v and 100k plate resistor off that that plate resistor would need to be rated at at least 1.25w to cope with a short to ground, but that probably wouldn’t cause a 1 watt resistor to catch alight…. would it?

thanks in advance

[Enzo]

Very difficult to short a 12AX7 plate to ground, usually.

Flame proof is not so much worried about shorting to ground, it is worried about what happens when excess current flows through it. For example, a pre-driver transistor in some Peavey powr amp might burn up when the transistor fails, but it may get nowhere near to ground. Under failure conditions we don;t want resistors catching the amp on fire.

[Steve Conner]

Yes, if you can imagine a fault that would grossly overload the resistor, then it should be flameproof.

I sometimes see schematics that suggest the flameproof resistor is used as a kind of ghetto fuse. This suggests that flameproof resistors should fail open before they get hot enough to set things on fire.

It is all uncomfortably close to the CE/UL/CSA type stuff for me. You shouldn't technically be offering amps for sale unless they've passed all the appropriate tests, which involves the lab trying to make it catch fire, and charging you a fortune for the privilege. If you don't offer them for sale then it doesn't matter, you can use resistors as flammable as you like.

[guitarmike2107]

Thanks guys

I have actually got the CE documentation, I got it in an exchange for some turret boards I made for someone, but I don’t build amps for sale so I have never gone through the whole lengthy testing process or paid someone to do it for me.
I have given a few amps away though so I still have some responsibility and still try and stick to the standard as much as I can understand of it.

If the only way to find out is through testing then I can do that, I suspect there will be some relationship between power overload and time to burn, if one knew this you could then know whether your fuses should pop before the resistor burns etc

I just thought it would be better if there was a more “academic” way to approach it. The other option would be just to make all resistors flameproof types, but that’s not what I have Thousands of….

[Steve Conner]

Oh, I see what you mean, you're looking for the data you would get from a fuse manufacturer if the resistor were a fuse. Well a flameproof resistor datasheet will include that, but a regular one won't.

In my experience no modern design of resistor will catch fire. The cement block type can "Chernobyl" - the housing splits in half and the red-hot core springs out - but I've only seen that happen in the lower values. The aluminium housed wirewound ones will shoot clouds of smoke and blow their end plugs out. All of this happens about 10-100 times the rated power.

Smaller ones will start to smell bad about twice their rated power. The coating visibly changes colour. Eventually they will drift up in value and fail open circuit. The small ones can run red hot for some time before failing: they might not catch fire themselves, but they can ignite other things. Some wirewound resistors (such as the green ones made by Welwyn) are almost red hot at rated power, never mind when overloaded. These can be a real hazard.

[guitarmike2107]

Yeah thats pretty much what I am after, I didnt see it on the data sheet I was looking at though. doing some basic failure mode testing is probably a good idea and a easy way to learn.

interesting summation on the resistor types though I have a few of those welwyn ones here, Need to make sure they are not too close to any caps or anything that doens't like heat.

thanks