Be careful with wire-wound resistors

[armandas]

Until today, I had no idea these resistors can explode violently when grossly overloaded.

I was doing some testing today, where I put 2A at 24V across two 24R 10W resistors in parallel. I put them on a big heatsink and while they were getting hot, they were doing quite well. Then someone came and messed with the PSU thereby putting 5A at 60V through the two buggers. In a matter of seconds they exploded (with a flash) and emitted nasty-smelling smoke.

PS: don't mind the mounting hole, that was done by me

[Ajahn Lambda]

Nice work. 


Given enough oomph, almost anything will go BANG!


Did you manage to at least punch the person responsible for overloading the resistors?

[Mechatrommer]

resistors can explode violently when grossly overloaded

change "resistors" to "almost anything"

[armandas]

Given enough oomph, almost anything will go BANG!

resistors can explode violently when grossly overloaded

change "resistors" to "almost anything"

Surely, normal 1/4W resistors would just melt? I'll try this tomorrow

Did you manage to at least punch the person responsible for overloading the resistors?

That wouldn't be good for my career

[IanB]

I was doing some testing today, where I put 2A at 24V across two 24R 10W resistors in parallel. I put them on a big heatsink and while they were getting hot, they were doing quite well. Then someone came and messed with the PSU thereby putting 5A at 60V through the two buggers. In a matter of seconds they exploded (with a flash) and emitted nasty-smelling smoke.

You should obviously beat the person who fiddled with your power supply with a big stick (except if it was a child you should beat yourself with a big stick for leaving the apparatus unattended).

But I don't get your numbers. You have some redundancy here. If you put 24 V across a 24 ohm resistor then 1 A will flow and the power dissipation will be 24 W. This is rather in excess of the 10 W rating. Of course if you put two resistors in parallel then the total current will be 2 A but each resistor individually will still exceed its power rating by a large margin.

A good rule of thumb is to de-rate by 50%, so for a 10 W resistor don't rely on dissipating more than about 5 W continuously. That will give you a nice margin for lack of air circulation, higher ambient temperatures, radiation blocking and reduced stress on the component.

[SeanB]

1/4 W resistors just smoke and flame out, some glow red for a few seconds during this, but some just pop with a few brief flames. Carbon composition resistors, on the other hand, actually burn nicely as the wax used in the mix and cover burns. Wirewound white ceramic boats glow white before cracking.

[armandas]

But I don't get your numbers. You have some redundancy here. If you put 24 V across a 24 ohm resistor then 1 A will flow and the power dissipation will be 24 W. This is rather in excess of the 10 W rating. Of course if you put two resistors in parallel then the total current will be 2 A but each resistor individually will still exceed its power rating by a large margin.

Not quite sure what numbers you don't get. I actually started with just a single resistor and was putting 24W through it. The second resistor came it when I wanted to double the current.

A good rule of thumb is to de-rate by 50%, so for a 10 W resistor don't rely on dissipating more than about 5 W continuously. That will give you a nice margin for lack of air circulation, higher ambient temperatures, radiation blocking and reduced stress on the component.

I'm sure it's a good rule for production, but in this case, these resistors were just a part of a test device. I reckoned that with a heatsink, I could run them at 2-3 times the power rating for a while. I'm not saying that what I done was good in any way, the heatsink was pretty big and even that was getting warm after 10 or so minutes.

[SeanB]

Best way for short term overloading is to drop them in a small pyrex dish filled around 3/4 with water. Works well and you can go to extreme overloads, provided the case does not boil the water too much. Of course low voltage and short term only.

[IanB]

Not quite sure what numbers you don't get. I actually started with just a single resistor and was putting 24W through it. The second resistor came it when I wanted to double the current.

I eventually figured out what you were trying to say, but it was your use of words that was the problem. You described things in an imprecise way with words used out of context so I had to pause and scratch my head for a bit to figure out what you meant.

For example, you said that you "put 2A at 24V across two 24R 10W resistors in parallel". But it doesn't make sense to specify the current and the voltage and the resistance at the same time since one of them is fixed by the other two. This leads to a double-take and moment of confusion while trying to figure out what is going on.

For instance you can put 24 V across a 24 ohm resistor (in which case the current is known to be 1 A).

Or you can feed 1 A through a 24 ohm resistor (in which case the voltage is known to be 24 V). When you feed current through something exactly the same current goes in as comes out.

Or you can feed 24 W into a 24 ohm resistor (in which case we know both the voltage and the current). When you feed power into something it gets absorbed or transformed, it doesn't come out the other side.

It may seem like I'm being pedantic, and maybe you will say that I am, but precise use of words and giving just the right amount of information helps tremendously with understanding. Giving too much information can often be as puzzling as giving too little information.

[T4P]

I don't actually get anything , unless the voltage was 10V 1Amp and 10R resistance then will you hit max rated dissipation ...
You put 24V 1amps into a 24R 10W resistor ... that's not 10watts that's 24watts.
By the way is the 10W rating open case? If it was on a heatsink i understand what you are getting at, which i just read upwards.
But yeah 60V @ 5Amps means 2.5amps per resistor and that is 150W , any 10W resistors will blow it's guts out at that sort of power dissipation .
That's was the max of your power supply right ?
But anyway, i am pretty stoked to see a metal case wirewound spit out it's guts ... 

[wkb]

1/4 W resistors just smoke and flame out, some glow red for a few seconds during this, but some just pop with a few brief flames. Carbon composition resistors, on the other hand, actually burn nicely as the wax used in the mix and cover burns. Wirewound white ceramic boats glow white before cracking.

Try a 12V halogen car headlight bulb when discharging a reasonably size flash capacitor (Metz 45CT hammerhead flash for the curious)..  The very robust crystal(?) glass of the bulb simply exploded in tiny fragments..  with an impressive BANG.. wow..

Next time I used some mains bulbs :-) instead of a random grab from what was on the table available and what I assumed to be a 220V bulb...

[IanB]

Try a 12V halogen car headlight bulb when discharging a reasonably size flash capacitor (Metz 45CT hammerhead flash for the curious)..  The very robust crystal(?) glass of the bulb simply exploded in tiny fragments..  with an impressive BANG.. wow..

Let's see now. A 12 V 60 W bulb has a filament resistance of 2.4 ohms at working temperature. By the normal rule of thumb that would be about 0.25 ohms when cold. If we apply 300 V across 0.25 ohms we get an instantaneous power dissipation of 300^2/0.25 = 360 000 W. So you just turned a 60 W bulb into a 360 kW bulb. No wonder it exploded...

[mikeselectricstuff]

Been there, done that!

It's tempting to assume that this type of resistor can tolerate a huge dissipation for a short time due to the aluminium, but not the case.

It can get even more fun - in my case I was using one to ballast a big argon laser, the end shot off, trailing the resistance wire begin it like a taser. Just to add to the fun it had about 340V of rectified mains on it.....

 

[Psi]

I was using one to ballast a big argon laser, the end shot off, trailing the resistance wire begin it like a taser. Just to add to the fun it had about 340V of rectified mains on it.....

hehe, that sounds both awesome and scarey

[Mechatrommer]

Surely, normal 1/4W resistors would just melt? I'll try this tomorrow

try 10ohm 1/4W on mains live to neutral. wear eyeprotection beforehand.

[SeanB]

Try putting a fluorescent starter across the mains. Never did find the case, glass or capacitor.......The dog was looking funny at me for days........ I did have the bakelite base though, with some rather melted pins still attached.

[aargee]

I've never seen inside these 'heatsink' power resistors before, so it relies on air conduction to the case to dissipate the heat? Or is there some intermediary (oil like) between the resistance element and the case?

[amyk]

It blew open probably because of the pressure generated by the heating. If it was vented it would just get hotter until it fuses or glows brightly (or both, usually the latter followed by the former).

[mikeselectricstuff]

I've never seen inside these 'heatsink' power resistors before, so it relies on air conduction to the case to dissipate the heat? Or is there some intermediary (oil like) between the resistance element and the case?

It's some kind of solid potting compound.

[robrenz]

In a reasonably high quality resistor it should be a high thermal conductivity ceramic potting compound like these.
The 810 in this http://www.cotronics.com/vo/cotr/pdf/801.pdf
or http://www.sauereisen.com/adhesives/sensors_resistors.aspx
or http://www.aremco.com/product/a4/

[T4P]

Try putting a fluorescent starter across the mains. Never did find the case, glass or capacitor.......The dog was looking funny at me for days........ I did have the bakelite base though, with some rather melted pins still attached.

Classic inductive storage or a electronic starter ?

[IanB]

Classic inductive storage or a electronic starter ?

I would guess the traditional gas discharge tube. As soon as it switches on the contacts will close and make a short circuit. A short circuit across the mains is nearly always exciting.

[metalphreak]

I remember seeing a video a long while ago where someone made a video of "blowing up leds". He was putting a not-too-outrageous voltage through it (but clearly well past what its rated to) and the top part of the led blew off straight towards the camera, giving him quite a scare 

[Ajahn Lambda]

Hehe, yes, it's always the shrapnel that scares me, not the fire and smoke.

[T4P]

Classic inductive storage or a electronic starter ?

I would guess the traditional gas discharge tube. As soon as it switches on the contacts will close and make a short circuit. A short circuit across the mains is nearly always exciting.

Right i just remembered , it's that starter one uses with preheat fixtures?