Sunn Beta Lead/Bass - managing heat of transistors in voltage regulation circuit

[Drummer4gc]

Hey folks, fixing up a Sunn Beta Bass and I want to make sure it's bullet proof. A common failure seems to be the TIP29 and TIP30 transistors (Q1 and Q11) involved with +/-15v regulation. They definitely get super hot during normal operation. I found some larger clip on heat sinks, which should help.

I came across this post, https://groupdiy.com/index.php?topic=61479.0, here's the quote I'm curious about:

"they get hot, the glue ages, the undersized clip-on heat sink falls off, the transistors get hotter, collector trace pads start to cook ... here is what we do down at the shop, grab a 2 x 4, cut it down to 2 x 5 , cut up an old reverb pan and no need for insl kits as you are isolated, we even used wood screws for the transistors"

Can anyone make out what is being described? I think the idea is removing the transistors from the board and mounting to metal somewhere, but I'm wondering if anyone has done this? I'd love to see a picture!

Matt

edit: the two transistors in question can be seen center-left in this image. Schematic below.

[J M Fahey]

Show what you have , we may suggest extra heatsinking or whatever but you keep us in the dark.

[Drummer4gc]

apologies, post updated.

[J M Fahey]

Thanks.

If you mean those TO220 transistors engulfed in hot glue blobs, yes, they must marginally dissipate heat at best, not safe.

Remove hot glue , it cracks easily under cutting nippers, add a small aluminum heatsink, either by using a screw or clip, remember to use thermal paste between transistor and heat sink, which must be long enough to touch PCB, and glue it to PCB to avoid vibration.
I use contact cement which sticks very well and keeps some elasticity when dry, so stands vibration better; hard rigid glues such as pistol applied hot glue or cyanoacrilate "Krazy Glue" tend to crack away from PCB.

[loudthud]

Components are crowded together too close to those transistors for my liking. Clip-on heatsinks might short to one of those components. I would attach some kind of metal to the chassis and run wires back to the PCB.

[Enzo]

Another trick for hot melt glue: squirt it with freeze spray. That makes it brittle and it easily shatters, then brush it away.

[J M Fahey]

Components are crowded together too close to those transistors for my liking. Clip-on heatsinks might short to one of those components. I would attach some kind of metal to the chassis and run wires back to the PCB.

True, both transistors backs have nearby hot ceramic resistors, which definitely do not help.
Being voltage regulators, so no signal through them, they can be moved a few inches away and mounted to the chassis with mica, thermal grease and nylon nipples (to insulate screws) . Chassis metal itself is heatsink enough.



here transistors are mounted straight to project case:



many suppliers offer the whole kit: mica or silpad, nipples, proper size screws, so you get all together with no hassle.

[loudthud]

On some versions of the amp there is very little clearance between the metal chassis and the wood and tolex cabinet. You have to slide the chassis in edgewise. There might not be enough room for a nut and long screw to stick out.

I bought one of these amps once. Got it cheap because it wouldn't make a sound. One of the regulators was bad.

[Drummer4gc]

For closure:

I mounted the two transistors directly to the amp chassis with wire extensions to the pcb and a mica insulator kit, screws pointing inwards to allow for clearance into the cabinet. Turned it on and felt the transistors warm up, and they didn’t feel nearly as hot as I remember them being when pcb mounted. Ran the amp through my testing procedure (AC/DCs Highway to Hell album at full-ish power into a dummy load) and it ran like a champ, no issues. I’d say it’s a worthwhile procedure to consider!

Matt

[johnskoolie]

Hey Matt,

I have the same amp except my board is off-white instead of blue. My neighbor gave it to me for shoveling his snow covered drive way. It was sitting in his garage for years and he said it didn't work. So I took it apart and the resistor in the top left of your picture (125ohm 10% 5w) was burnt out. I checked with my multimeter and it is still good. It actually had burn marks on the board coming from under that resistor. I raised it and soldered it back in. Amp still doesnt work. I got a load popping sound when I turned it on and noise but no sound from my guitar.

Got any advice? It sounds similar to your problem. I am a beginner when it comes to this stuff. I mostly build pedals and guitars. I did make sure the caps were all drained before working on it, just so you know.

[The Dude]

As with any amp, check power supplies first. Check the two rail fuses. There are signal test points throughout. If power is good, I'd insert a signal and follow it through to see where it gets lost. Test points are P-P if you have a scope. If not use a DVM, but know that you will be most likely measuring RMS so voltages will be lower. You are most likely looking for a large discrepancy, so don't worry about slight variations from schematic voltages. Also, do not hook up a speaker to the unit until you verify that there is no DC on the output.

[johnskoolie]

Hey Dude,

"As with any amp, check power supplies first. Check the two rail fuses. There are signal test points throughout. If power is good, I'd insert a signal and follow it through to see where it gets lost. Test points are P-P if you have a scope. If not use a DVM, but know that you will be most likely measuring RMS so voltages will be lower. You are most likely looking for a large discrepancy, so don't worry about slight variations from schematic voltages. Also, do not hook up a speaker to the unit until you verify that there is no DC on the output."


I'm not a pro at this at all. I am brand new to working with amps. I mostly work with pedals and guitars. I own a Klein mm420 multimeter and that's it. I bolded what I was hoping you could explain a little to me. Thank you for the help!

[The Dude]

two rail fuses = There are 2 fuses for the positive and negative supplies labeled F2 and F3. Make sure those are good.

I'd insert a signal and follow it through to see where it gets lost. Test points are P-P if you have a scope. If not use a DVM = The schematic shows test points with signal levels at each. Insert a test tone at the appropriate level 2.2V P-P at TP1. Then follow through each test point and see where signal is lost. If you don't have a scope, use your DVM set to ACV. If you don't have a tone generator, you can find test tones on YouTube and use a phone, computer, etc. to run into the amp's input.

You are most likely looking for a large discrepancy = Don't worry about slight differences between stated schematic values and meter readings. You are likely looking for a place where there isn't much signal at all since the amp produces no sound.

I hope that helps.