exploding TO-220...TO-247... but what about TO-3?

[eccerr0r]

We've seen plenty of TO-220s and D-PAKs and DIP8s exploded from abuse...also TO-247s...

But has anyone seen a TO-3 explode? Are they capable of exploding or is the hermetic seal sufficient to keep the carnage internal?

Of course there are several variants of TO-3's like the aluminum and steel variants, but has anyone seen either blow up? Frag bomb?

How about TO-66?

I suspect the cap and weld are the weak points, but don't know.

BTW, I'm trying to understand how someone blew up their TO-3 on reddit ...

[petehall347]

Re: exploding TO-220...TO-247... but what about TO-3?

[eccerr0r]

Re: exploding TO-220...TO-247... but what about TO-3?

I don't have any/there aren't any!
And the reddit picture is very suspicious, so I will not include it here.

[redwire]

Re: exploding TO-220...TO-247... but what about TO-3?

Metal can TO-3, TO-66 have nothing to explode really.
I have seen one where a bonding wire (must have blew off the die) inside started an arc to the lid and made a hole in it.

[eccerr0r]

Re: exploding TO-220...TO-247... but what about TO-3?

That's what I would have thought... it's not like a epoxy (usually organic material with carbon atoms) that gotten hot enough it would burn... the welding indeed is possible.

Makes one want to go back to metal cased TO-3/TO-66's no? heh.

[CapLeaker]

Re: exploding TO-220...TO-247... but what about TO-3?

No exploding TO-3 that I can remember. They just failed. Very very rare occasion that the lid comes off.

I think you should do another thread asking about which device explodes most violently.

[eccerr0r]

Re: exploding TO-220...TO-247... but what about TO-3?

Feel free to do so! (and maybe a howto!)

I just recalled my TO-3 smoking was no smoke. It just up and died without a trace.

If anyone was wondering about the failure, it was in the amateur radio subreddit - someone had their Yaesu FT-101 fail spectacularly and there was some carnage with their TO-3 inverter transistors. While I don't doubt these can fail, the way it did fail seemed suspicious until they dissemble it further. The photo looked like the TO-3 flange *bent* -- this is simply incomprehensible, but the angle of the photo shot was not helping. Perhaps something under the TO-3 exploded with enough force to... bend the TO-3? No, no, no...

[stj]

Re: exploding TO-220...TO-247... but what about TO-3?

there is no oxygen in them - so there is nothing to expand and burn
they go open or short often enough though.

[redwire]

Re: exploding TO-220...TO-247... but what about TO-3?

Ultimately the bonding wires are the fuse. TO-3 for inverters can have double or triple to get say a real 40A rating, see BUX22 die shots and it is dual die as well.
Really old power transistors from the 1970's did not use bonding wires, but they used tabs for high current. The PCB connections to the pins can also get hot and fail.

I say bullshit on the warping, TO-3 has an extra heat-spreader plate inside.
Crappy plastic packaged parts i.e. TO-247 can't compete.

[eccerr0r]

Re: exploding TO-220...TO-247... but what about TO-3?

Dammit, but I don't care: I just head another anecdotal story on face***k that someone melted a TO-3 transistor due to omitting the snubber network. I suspect this is a secondary failure where the flyback caused arcing and causing the transistor to turn on all the time...and then causing meltdown of the die and allegedly flange warping???!!! But did anyone see this posting and can verify? Pictures or it didn't happen?

I do not have face***k so I have no idea...

[stj]

Re: exploding TO-220...TO-247... but what about TO-3?

how could the base warp - it's bolted to the heatsink

[eccerr0r]

Re: exploding TO-220...TO-247... but what about TO-3?

agreed, but if we're wrong, hopefully someone here has seen that FB picture and prove it.

[stj]

Re: exploding TO-220...TO-247... but what about TO-3?

given the thickness of the pins and the usual termination methods - i call bullshit.

they are usually soldered to wires using leaded solder that would melt and dry-joint or they are in sockets that have current limits.

[eccerr0r]

Re: exploding TO-220...TO-247... but what about TO-3?

Okay. The face***k posting has been debunked, it was a plastic device after all, apparently a TO-220. These plastic units explode all the time, specifically TO-220s, and I've had my share of exploded TO-220 units

Now it's just the reddit post that needs to be squared away with an allegedly bent TO-3 flange. Again I hope the poster coughs up more details or it really did not happen the way that it was stated...

[CapLeaker]

Re: exploding TO-220...TO-247... but what about TO-3?

Did I already mention… I hate Facefook!