IRFP250 die size

[djQUAN]

I built a pair of N channel amps by a holton and was having a little trouble. I was able to get several IRFP250's so I used them in the amp.

a few months back up to recently, I was always having trouble with the amp blowing up. (at the moment, I have 14 dead mosfets and four dead emitter resistors in several occasions, modded the amp to use four pairs of mosfets per channel but still the same thing)

I used two brands. the ones I had were made by Intersil. the replacements were by International Rectifier. the intersil ones were the ones always blowing up. all the IR's I had are still in the amp. the weird thing is, when I didn't had enough money, I took a scrap and got some used IRFP254's (higher voltage, lower current) used along with the intersil 250's in the other rail. guess what, the amp blew up but only the intersils.

which made me curious. the intersils may be fakes. I cracked one open and the die size is 6mm by 4.5mm. is that the right one? I don't have any dead IR's so I can't compare.

some info on the entire system.....

the system has a DIY 5 input stereo mixer, dual 31 band EQ, compressor/limiter/gate, 5 way Linkwitz Riley crossovers (four of rod's P19 but only 3 are used), four P3A's, two 250 watt bipolar amps, and the mosfet amps (with EAS). I use 4 ohm drivers in everything. tweeters, mids, midbass and subs.

I also use rod's guitar preamp and plug it to the system. I play bass, another wierd thing is that the amp only blows up if I plug the bass guitar to it. (the signal passes through ALL my processors) even at low volume. if I don't use the guitar, the amp stays alive even playing at moderate to high volumes.

the only question here is "why" ???????

anyone know?

[djQUAN]

ok. I guess no one has answers?

[MikeB]

The diesize seems pretty large...

If your amp blows only when connecting the guitarpreamp, i think
the problem lies within this circuit. Maybe it outputs some RF,
what can easily destroy an amp. Or you have some strange feedback
through the guitar ?
Normally its not too easy to blow the irfdevices, as they don't
have a 2nd breakdown.

What is your supplyvoltage in the poweramp ?
the irfp250 is rated at 200v, giving some security of ~20%, this
would give a max supplyvoltage of +/- 80Volt.

[jneutron]

Originally posted by djQUAN modded the amp to use four pairs of mosfets per channel but still the same thing)

which made me curious. the intersils may be fakes. I cracked one open and the die size is 6mm by 4.5mm. is that the right one? I don't have any dead IR's so I can't compare.

The irfp250 and 254 are both die size 5, and are both 6.53 mm square.

If you still blow the intersils even though you doubled them up, then it is neither a current or dissipation issue.

VGS characteristics may be very different..

Maximum gate to source voltage may also be an issue.

Bass...hmmm..I wonder if you are exceeding device voltage from speaker reactance...I think the IR devices have a cleaner breakdown knee as a result of a better controlled diffusion process, and a better breakdown ring control structure.. allowing the absorbtion of more hv energy.

Schematic possible?? It's a little easier to make intelligent guesses with one, vs my normally unintelligent guesses...

Cheers, John

[djQUAN]

mikeb;

ok. then I guess I'll replace all devices into IRF ones. the preamp circuit I'm using is exactly the same as the preamp section of rod's P27 guitar amp. it is bandwidth limited so I gues it's not RF. also, the zobel on the amp output does not burn. (although I'm using 2W resistors, it might not get hot enough)

the amp is powered by a 500VA dual 48V secondary toroidal. and eight 10,000uF caps, (20,000uF per rail per amp)

jneutron;

that I think could be possible. speaker reactance. but I'm using car subs (two 12's, JBL LCS-1250W) in 1.1cu-ft sealed each. although I doubt that it is a difficult load.

can't put up a schematic since my hosting site is down at the moment and I don't have a scanner or something. but I do have the PDF from anthony's site. email perhaps?