Olden days "real" watts

[TriMan66]

Hi,

We are all familiar with the power ratting of original valve amplifers stated in "Real" watts and the modern equivalent of PMPO etc guff which is 10 times the origanl power rating.
Well on a related topic, I was testing an HT power supply circuit I'm working on (first pic just for reference) and needed a dummy load. I used two resistors in parallel, an old old 24W resistor and the other a new 50W resistor. They were both loaded to 4 times their rated power for about 30 seconds and the new resistor went bang and shot out it's insides like a bullet! The old resistor just got hot. See the second pic.
I know the new types are meant to be heatsinked etc (see last pic), but I didn't expect such a dramatic failure! Anyway it reminded me of power rating for the new audio amplifiers. Perhaps these new resitors are PMPO rated!!!!

regards
Craig

[AlanC]

Well I would love to know how they get the figures for audio. There were some inspired claims for output power around in the 80's with some of the music centres and so on you got then. But more recently the makers of computer speakers elevated it to an art form. For example, I had some '4500 watt' PC speakers once that cost about £9.99 and had a small 8-pin DIL device like an LM386 in each channel. I wonder if they left the 'milli' in front of 'watts' out? A real 4.5 kW of audio would probably demolish your house.

Yes they do explode those heatsink resistors if they get too hot. There also used to be the dark green Welwyn wirewounds that were coated in some glass-like ceramic substance. These could be hot enough to melt the coating and they would live to resist another day!

Cheers, Alan

[davegsm82]

Unfortunately, the old old resistor relies on convection cooling and has a rather high physical mass when compared to the aluminium clad one. You'll find that if you heatsink the Alu one (as you've done) it will handle more than its rated power.

We use these alu clad ones at work in load banks, one of the 'engineers' did the same as you did and thought it would be ok for a short period. Bad mistake as the same thing happened to him, however the resistor in question was a 500W one from RS. It blew its end cap off and demolished some rather expensive equipment on its way out!

Dave.

[GrimJosef]

I'm afraid this is one of those cases where reading the instructions really matters. The datasheet for a typical 50W resistor http://www.farnell.com/datasheets/438609.pdf clearly says it can dissipate 50W if it's fastened to a heatsink which keeps it at 25 deg C. It also tells us what it can take if it's not on a heatsink, which is 12W to start with (while it's still at 25C) but falls below 10W once it's up to 70C. And it seems it can take 5 times its nominal power for 5 secs. I'm afraid that 30 secs at 200W is well over this though and I don't think we can reasonably grumble if it goes pop. In defence of this type of resistor I fried a much smaller one (5W maybe ?) a couple of weeks back by connecting it to a source which was depositing nearly 20W in it. Fortunately I'd just soldered the connecting leads to the resistor's tabs without threading them through the holes (it was a very temporary arrangement) and when the temperature of the tabs reached the melting point of the solder the leads duly fell off. As far as I can tell, the resistor itself has survived completely unscathed.

Cheers,

GJ

[AlanBeckett]

Craig,
What a splendid and graphic desription. I've never tried one like that. I've always bolted them to a chunk of Al and run them within their ratings. However, I'll bear that in mind from now on.

Alan,
The Welwin resistors are W2x, where x starts from 1 - 2.5W and goes up. In the bad(?) old days you could put one of a PSU and get it hot enough to light your fag once the glass coating had dripped off.

Alan

[glowinganode]

They do make a pop alright. I used one as a replacement part in a DC injection brake on a machine tool. The original fault was still present, the 25W resistor dissipating around 5kW briefly. The noise was like a shotgun going off, if looks could kill I'd be long gone...
Rob.

[G8HQP Dave]

For the same reason (thermal capacity) an old carbon comp resistor can cope with brief overloads which will kill a modern film type. It doesn't mean the old one is better, just that it has more thermal capacity. Think in terms of a low pass filter - it smooths out spikes, but a big enough and long enough spike will still get through.

[kalee20]

The aluminium-clad types don't have much thermal capacity, so they definitely need a heatsink for even short-duration use. (And yes, I've had one shoot out its end, too!)

However, with heatsink and operated as recommended, it's very likely that the new resistor performs better than the old resistor. Both are (hopefully) OK at their rated power, indefinitely. With 5% overload, I'd trust the aluminium-clad type better than the old-type free-air resistor, which I can see reaching a final steady temperature high enough to slowly oxidise the solder on the terminations.

[Skywave]

Originally Posted by AlanC

There also used to be the dark green Welwyn wirewounds that were coated in some glass-like ceramic substance. These could be hot enough to melt the coating and they would live to resist another day!
Cheers, Alan

On my, yes! When badly - and I do mean badly - overloaded, don't they get hot! Not by design - but by accident, I hasten to add - but I have known these get so hot that they have glowed bright orange! Not that they're much good after that, of course!

Al. / Skywave.

[Diabolical Artificer]

With these high wattage resistors,if for example I had 5 x 10w resitors in series,would the power handling then be 50w or 10w? Wasnt sure how ohms law applied.Andy

[davegsm82]

Originally Posted by magnetic

With these high wattage resistors,if for example I had 5 x 10w resitors in series,would the power handling then be 50w or 10w? Wasnt sure how ohms law applied.Andy

Regardless of how they are connected, you have 5 x 10 watt resistors, therefore the total power handling (if properly heatsinked!) is 50w.

Connect them in series and you'll need 5x the voltage to make full power, connect in parallel and its 5x the current.

Dave.

[julie_m]

If resistors of equal value are connected in series or parallel, the power dissipated will be shared evenly between all of them; since either the current (if they are in series) or the voltage (if they are in parallel) will be the same in all of them; and since the resistance is the same, then so will the other value (voltage or current) be the same.

If the resistors are not equal in value, the power will not be evenly shared and it will be necessary to use Ohm's Law to work out the power dissipated in each individual resistor.

[G8HQP Dave]

I think I once read on another website that the aluminium-clad (and cement-clad?) wirewounds cannot handle short-term overloads. On the other hand, the dark-green ones can cope with a lot of abuse. I think (don't rely on my memory!) that it was in the context of surge-limiting resistors to help an RF amplifier survive an arc in a big tetrode. The last thing you want is resistors exploding, when what should happen is that they survive long enough to protect the mains transformer until a fuse blows.

[GrimJosef]

Well it depends what you mean by 'cannot handle' and 'a lot of abuse' but it seems that the vitreous enamelled chappies are indeed better, by about a factor of two, than the aluminium clad ones. Welwyn make both types and have a helpful tech note on their pulse and overload capability here http://www.welwyn-tt.co.uk/pdf/appli...verload_AN.pdf The W20 ones (vitreous) should be OK for ten times their rated power for 5 secs while the WH (Al-clad) are only spec'd at five times. The WA cement ones are as good as their vitreous brethren though.

Cheers,

GJ

[Biggles]

I had an aluminium clad resistor go o/c in a circuit I built recently. The resistor was operating well within its power limits but thinking about the circuit now, it was subject to a very brief period of higher current on every switch on. I wonder if this caused the failure.
Alan.

[Denis G4DWC]

In a past life I was involved with a project that had a fair number of hefty linear power supplies which had a very high inrush current. The manufacturer decided to retrofit some 50W Alloy clad resistors of a few ohms to limit the surge current but sadly they kept failing. Of course at the instant of switch on these resistors of only a few ohms probably had the best part of the mains directly across them for a few cycles and they objected. Quite strongly in fact!!

Denis