Greetings,

Hopefully this is the right place for this, and apologies in advance for the noob-ish question... I'm repairing a tube guitar amp (schematic attached) which uses a regulated 12VDC supply to power the preamp tube heaters, some relays, a fan, and some logic/switching stuff that is out of my depth of understanding (although I'm willing to learn).

Anyway, the regulator for this 12V supply is supposedly a 2N3055 in what appears to be a TO-220 case, bolted to the side of the chassis. The collector connects to unregulated ~20VDC, the base is connected to a 13V zener to ground, and the regulated 12V (I guess more like 12.4) is taken from the emitter. This part of the circuit is near the bottom left of the schematic.

So the 2N3055 in this amp is dead, shorted base to emitter.

What I'm having trouble with is finding a replacement in the same form factor. Seems like 2N3055 is only available in a TO-3 case. I found an alternative, TIP3055, which is in a TO-247 case, but I'm not sure if that would be an appropriate substitute...

I'm also wondering why they used such a high-power part (~100 watts, 15A) in what seems like a low-power application (600mA for the heaters, plus a little 12V fan and a couple relays and logic stuff). And if it was indeed overrated for this application, why did they choose it and why might it have failed?

I'm going to go ahead and order a few TIP3055's, but if anyone could give me some reassurance that that's an okay substitute, or suggest a better alternative, it would be very much appreciated.

Thanks in advance!

 

At one time they, 2n3055's, were cheap and available. The circuit you are describing is a very poor regulator and using a "beast" of a part was supposed to offer some form of "protection". Obviously, the original designers were misinformed. Chances are that when it went bye-bye, it took other parts with it. Are you sure you want to go down this rabbit hole?

 

At one time they, 2n3055's, were cheap and available. The circuit you are describing is a very poor regulator and using a "beast" of a part was supposed to offer some form of "protection". Obviously, the original designers were misinformed. Chances are that when it went bye-bye, it took other parts with it. Are you sure you want to go down this rabbit hole?

Hmmm...I suspected it might not have been the best way to get a +12V supply going... Leave it to modern designers to over-complicate a freaking tube guitar amp that doesn't even have any onboard effects besides reverb. Perhaps the part was overrated to handle the cold-start current of the tube filaments...?

I guess it depends on which rabbit hole you're referring to - finding a replacement part? Or figuring out what else got taken out when the transistor failed?

Re: the second rabbit hole, the transformer winding has 2x 5-amp fuses, which are both intact, and all the loads coming off of that supply are on separate boards, with a separate connector for the +12 on each one, so if there is a short downstream at least it won't be hard to isolate...

 

The part is having to dissipate ~8V X 600ma = 4.8-Watts.
The Transistor package by it's self, can dissipate maybe 1-Watt, and survive,
the other 3.8-Watts has to be removed by way of a proper Heat-Sink.

A Stamped-Steel-Chassis makes a poor Heat-Sink,
especially with other hot parts near by, and no Air-movement.

Cheap Heat-Sink-Goop gets hard and shrinks with age,
and deteriorates faster with increased temps,
rendering it useless at transferring Heat.
Doohh !!! there it is ........ the infamous "Blue-Smoke" escaping !!!
.
.
.

 

I found an alternative, TIP3055, which is in a TO-247 case, but I'm not sure if that would be an appropriate substitute...

It has a lower maximum power dissipation but should be okay in your case. The benefit is that TO-247 was designed to mount in places designed for TO-3

 

Hmmm...I suspected it might not have been the best way to get a +12V supply going... Leave it to modern designers to over-complicate a freaking tube guitar amp that doesn't even have any onboard effects besides reverb. Perhaps the part was overrated to handle the cold-start current of the tube filaments...?

I guess it depends on which rabbit hole you're referring to - finding a replacement part? Or figuring out what else got taken out when the transistor failed?

Re: the second rabbit hole, the transformer winding has 2x 5-amp fuses, which are both intact, and all the loads coming off of that supply are on separate boards, with a separate connector for the +12 on each one, so if there is a short downstream at least it won't be hard to isolate...

Actually both. One rabbit hole is as good as the next.

 

The part is having to dissipate ~8V X 600ma = 4.8-Watts.
The Transistor package by it's self, can dissipate maybe 1-Watt, and survive,
the other 3.8-Watts has to be removed by way of a proper Heat-Sink.

A Stamped-Steel-Chassis makes a poor Heat-Sink,
especially with other hot parts near by, and no Air-movement.

Cheap Heat-Sink-Goop gets hard and shrinks with age,
and deteriorates faster with increased temps,
rendering it useless at transferring Heat.
Doohh !!! there it is ........ the infamous "Blue-Smoke" escaping !!!
.
.
.

Heat sink goop? What's that?

Here is a photo of where it was attached - looks like just a piece of clear tape behind the part... perhaps a proper heat sink is in order with the replacement, if it will fit.

 

You might consider converting to the common LM7812 12V regulator.
It's rated for 1A and has both short circuit and over-temperature protection.
The 1A limit would reduce the surge current when power is first applied to the cold heaters, possibly lengthening the tubes' life.

 

Heat sink goop? What's that?

Here is a photo of where it was attached - looks like just a piece of clear tape behind the part... perhaps a proper heat sink is in order with the replacement, if it will fit.

Heat sink goop is a thermally conductive paste - usually a talc-filled silicone that is pretty thermally conductive. It is intended to be spread in a very thin layer and be just enough to fill the air gaps when the flat transistor body is set against a flat chassis or heat sink. It is not thermally conductive enough to be a heat sink, it is just way more conductive than air gaps between the metal flats.

as for the "tape" - it is likely a mica sheet. Mica is a mineral that exists in a crystal form that can delaminate in sheets - very thin and even (flat) sheets. The mica is there because it is non-conductive and keeps the tab of the TO-220 package (usually electrically connected to the collector) from touching the heat sink (commonly connected to ground).

 

I vote Voltage-Regulator, a much more elegant solution.
.
.
.

 

Short direct answer: MJE3055 35¢ each from my distributor.

 

Hello,

looks like just a piece of clear tape behind the part.

The clear piece is likely a piece of mica, an isolator with reasonable heat transfer properties.

Here is an example on Newark:
https://www.newark.com/multicomp/mk3306/insulating-kit/dp/50P9593?st=mica to-220

Bertus

 

Thanks for the replies all! I was joking about the heatsink goop, I know what it is, but I did not know about the mica sheet being an alternative...I do have some goop on hand; would that be preferable to the mica sheet? I mean it's probably what I'm going to use because the current mica is no longer intact.

I had considered a 12V regulator too, but that would actually require more changes to the existing board than just dropping in a TIP3055. I'm gonna order a couple TIP3055's and a couple 78S12 regulators (2A version because I'm honestly not sure what the current requirements of this circuit are, other than the heaters) as an alternative.

 

Hello,

There is also the 78T-series, Those are good for 3 Amp.

Bertus

 

A screw-less TO-247 package with a pressure clip and Kapton tape isolation is significantly better thermally than a TO-220 with mica insulation and thermal paste. There are much better BJT in a TO-247Plus case than the 3055 part.

If the heatsink is grounded then a 3-terminal regulator won't need insulation (tab = centre ground pin) just paste and is probably the best long term option, IMHO.

 

.I do have some goop on hand; would that be preferable to the mica sheet?

You use the goop either way to reduce the thermal impedance.
The mica is if electrical insolation is required, as the goop doesn't provide that.

 

Thanks for the replies all! I was joking about the heatsink goop, I know what it is, but I did not know about the mica sheet being an alternative...I do have some goop on hand; would that be preferable to the mica sheet? I mean it's probably what I'm going to use because the current mica is no longer intact.
(Some text removed for clarity)

They are used together. The heatsink compound improves thermal coupling. The sheet of mica is to provide electrical insulation between the transistor and the metal chassis or heatsink.

Instead of mica and heatsink compound sometimes silicon insulation pads are used, they are glass fiber filled so they are tough. What looks like filaments in the insulator in the photo may be glass fibers.

 

Hello,

You might want to read section 7 of the following article:
https://sound-au.com/heatsinks.htm

Bertus

 

I had considered a 12V regulator too, but that would actually require more changes to the existing board than just dropping in a TIP3055

You just connect the regulator input to the unregulated DC source, the regulator ground to circuit common, and the regulator output to where the 3055 output connected.

 

You just connect the regulator input to the unregulated DC source, the regulator ground to circuit common, and the regulator output to where the 3055 output connected.

Like so... really not that difficult. Some regulators like a 0.1uF - 10uF capacitor across eithe input or output pins if connected via long leads - check data sheet - you can solder one if need be direct to the pins.