extra local heat sink on TO3

[indianajo]

I'm reworking a PV1.3K amp for Ky Oaks day, Friday. This basically has a 4558 driving a TO92 PNP-NPN pair, driving a TO220 PNP-NPN pair, driving a TO3 PNP-NPN pair, driving 5 NPN MJ15024 and PNP MJ15095. I'm out of 3 mm toothed lockwashers and nuts to mount the ******** metric TO3'ps, and it would take me 3 days to get some from Cleveland or North Carolina on my budget, it is the end of the month. I spent the weekend cutting up dead PCAT power supplies for the tiny toothed washer nuts and the TO3P insulating pads, because regular TO3 insulating pads don't work and OEM 6-32 screws don't work. SO, I used a $1.50 TO3 MJ15015 for the driver transistor instead of another TO3-P NJW21194. Peavey had the driver transistor mounted on a mica pad on a piece of 1/8" aluminum angle, bolted to the metal case. I'm worried I don't have enough SOA, though Hfe 20 O.T.'s would require only 1.1 amp out of the TO3 driver to produce the spec'd 22 amps. Will a local heat sink with the fingers in the air mounted above the insulating pad on the driver transistor make it cooler? There is a fan.

[mickeymoose]

Set it aside. Send me a pm with your needs and I will send you the damn parts. E

PS: It is no longer called a fan. It is a "Turbulance Generator"!

[indianajo]

Ordered $15 worth of tiny little nuts and internal tooth washers, be here Wednesday. On the TO3P have been capturing the collector (middle) lead between the head of a 4-40 screw and an I.T. washer, running through the other TO3 hole to an I.T. washer held up against the PCB metal of the collector trace.
Still wonder if a thermalloy finger heat sink above the mica washer would help or hurt a TO3 mounted on aluminum angle. The dynakit St120 also uses the 1/8" aluminum flange for a heat sink, attached to the case and it has blown the O.T.'s twice, besides melting the solder on the output cap once. I put fans on it but can't figure how to cram a 6x8x1 real heat sink in there.
Unrelated, I can't figure out how Peavey sets the 0 noise bias current on these. There are a couple of op amps, but they sense two diodes on the O.T. heat sink. There are no pots. I'm going to put test points on the emmiter resistors on the side of the case, and try to take a bias measurement. Hope it doesn't cause an osciallation with the extra wire.