Sho Bud amp blown power transistors

[Prattacaster]

I'm working on a unique single channel Sho Bud amplifier. This one has 4 power transistors mounted to the amp chassis where most single channels had two. Anyone know if these are supposed to be the same kind (RCA 40636) or are one pair the drivers. When looking at the amp the inner pair were blown but the outer pair are not. All four are T-03 cases. Two were coded WW284 0636, one was WW284 0310, and the last is WW284 0715. Thanks for any help.

[Enzo]

What model?

Looking at another schematic, I see 40636 outputs and 40409 and 40410 drivers. My inclinatiion would be the extra two are more outputs. Easy to tell. Are the bases wired together? Collectors wired together? Is there another 0.33 ohm ballast resistor in series with each extra one?

If the collectors are wored together, the bases together, and the emitters wired together through low ohm resistors, then they are in parallel and are all outputs. If two were drivers, then the base of the output would likely be wired to the emitter of the driver, just as the 40409 and 40636 on the drawing I see.

[Prattacaster]

They are definitely not wired for parallel outputs. I am using the same schemo as you, although since the amp has no model number on it I'm not sure it is the right one. I post pics soon.

[Prattacaster]

ShoBudoutputtransistors.jpg picture by Prattacaster - Photobucket
ShoBudOutputStage.jpg picture by Prattacaster - Photobucket
First pic is the output transistor sockets. The middle pair are toast. Outside pair diode drop tests good. Second picture:something got toasted at some point. I dont know when but it looks like the component got moved or is on the underside of the board. This amp looks to be a '75. I know some people say the Big Mallory 4100uF cans last a long time but is it time to replace. Mouser or Bdent? What do you think about the other electrolytics. I will replace all of them unless directed otherwise.

It seems like everytime I work on a vintage solid state amp there is always some HUGE road block that seems impassible. Thats why I love tube amps, you can always find the right parts or make necessary changes easily.

I digress...so on the schematic http://s1122.photobucket.com/albums/...3_ShoBud_1.jpg
there is this chicken scratch, and the chicken scratch seems to be off compared to what RCA states. RCA says the drivers for a 40636 should be 40594 not 40494 same mistake for the PNP driver. What the heck?!?! And then the PI or whatever you call it in solid state land; those components are chicken scratched too. I wouldnt mind except you can hardly read it. If someone could chime in and give me the proper replacements for these components I would be very thankful.

Why are the drivers in a T-03 case?

The amp I have does not have the HT fuses. What value would be prescribed here, 500mA?

[Prattacaster]

After a shot of JD, I realized that Enzo, your suspicions may be correct. The top two T-03 transistors have their collectors and bases wire together and the bottom two have their emitters and bases wired together..i think either way you can see i the pictures I posted.

[Enzo]

Juvenile delinquent?

Justice denied?

Just ducky?

Jeorge Dickel? (Hah!)

[Prattacaster]

Whichever one it was it worked. It became very obvious that they are parallel. SO, that being said one transistor died from each side. Also there was a .33 5w "ballast" resistor that opened up. I tested the drivers they test fine per diode drop. Other than the extra output it follows the schematic with the exception of a preamp out added where the reverb feeds back into the preamp. The reverb tank had been disconnected so we will see if that works.

For the output transistors that need replaced, should I just order 4 new considering there were a total of three mismatched transistors(look at original post)?

[Enzo]

Mismatched? They all four looked to be WW284 per your post. The other numbers look like lot codes.

SInce you are not going to find those transistors anyway to replace just the bad ones, yes, I would just replace all four with something.

SHouldn;t be hard to sub, you have 90v rail to rail if it is like the schematic we have, so something like an MJ15003 would handle it with ease.

[Prattacaster]

Thanks Enzo, I was getting thrown off by the coincidence that the date code 0636 was close to the transistor number 40636 on the schematic.

Another question: the series power resistors were .47 ohm instead of .33 ohm like the schematic shows. I am ordering .39 ohm 7w resistors. How does this change affect the bias of the output transistors?

I'm guessing that I should still bias at 20mA like the schemo notes. How do you go about calculating the bias? never mind a little research gave me my answer.

One more, what would be the correct HT fuse values. 250mA? 500mA? 50mA?

Another... I was looking at a Citation 12 schemo, should I incorporate some of the protection techniques. For instance they have 2A diodes in parallel with the collector and emitter of the output transistors. Also they have a .1cap and 27ohm resistor in series to ground on the output. http://www.diyaudio.com/forums/solid...tt-amp-10.html. Thoughts?

[Enzo]

Look up "zobel" network, that is the stability network of R and C in series across the output you mention.


Understand what "bias" does in an amp like this. I'll refer to the positive and negative sides of the output as top and bottom. The output to speaker comes from the middle between top and bottom. They take turns conducting and not conducting. This puts the positive and negatove currents to the speaker to make the cone move to the music. They do not conduct at the same time. Turning the top and bottom both on at once would simply allow current to flow from V+ to V-. AMp would get hot.

What we want is for the top and bottom halves to hand off smoothly at zero as the music waveform crosses the zero back and forth. To do that, we want both top side and bottom side transistors JUST on the brink of conduction. Any signal that comes their way will push them into conduction and the speaker will see what it needs. If the transistors are a little short of conduction, that means the signal must move positive a little bit BEFORE the transistors will start conducting. Same thing happens on the negative. SO each time the signal crosses zero, there is a little pause. This leaves a notch or glitch in the output signal. We call that crossover distortion. That is what we call a cold bias setting.

GOing the other way, if we adjust the volatges so the output transistors are already conducting a little, then there is no notch, no crossover distortion, but the amp is now conducting more current than it needs to. Ther is no additional output power from this, just heat. This is a hot bias setting.

When an amp maker says set current through the transistors are 20ma or 5ma or whatever, what they are doing is telling you that they think that small current represents transistors at a warm enough current to avoid crossover distortion.

Changing the ballast resistor doesn't change the bias of the transistor, it changes the voltage you would see the idle current they want at. Well there is a clumsy sentence... If they want 10ma flowing, then that will drop across a 0.33 ohm resistor as 3.3mv. If you had .47 ohm ballasts, then 10ma would read as 4.7mv.

Persoanally, I don;t use those current settings. I prefer to adjust the amps for elimination of the crossover notch.

[Prattacaster]

Thanks for the explanation, I dont know why I'm surprised that so many of the same topologies for tube amps also apply to SS. Complete ignorance, but I'm learning. I am adding series inductance along with the zobel network along with the repairs I am doing. It seems like good cheap insurance that doesnt affect tone in a negative way.

Do you gain more output by using parallel output transistors? Will the MJ15003 give me more output and would it be safe with a 4ohm load. I ask because there is an ext. speaker output that is parallel to the amp's speaker(8ohm). So I want to be sure that another 8ohm in parallel would be safe.

[Enzo]

The output voltage will be whatever it is more or less. Lower impedance will take that same voltage and demand more current from the amp. Parallel output transistors will allow more current to be drawn. BUT, the power supply must have it to draw FROM.

I don;t know what those stock transistors were. From that era, good old 2N3055 was so very common. The output stages are current amplifiers, not voltage amaplifiers, so adding transistors won;t make it louder, just make it less likely to blow up under load.

Read this, especially page 10, where it says, if the manula says 2 transistors, use 4.

http://www.peavey.com/support/techno.../chapter_1.pdf

In fact, all of Hartley's papers are interesting, at least to me. All under tech notes in the Peavey web site support section.

[J M Fahey]

Agree.
The workhorse in that era was the 2N3055.
The "premium" RCA 2N3055 , probably handpicked for higher voltage, was called 40636.
Today it would be called MJ15015.
Your sugesstion of MJ15003 is even more robust.
Donīt see short protection in that schematic, so ... just avoid it.

PS: donīt find a problem with old SS amps.
Any obsolete transistor can be replaced by a modern one, functionally the same or better.
Try that with an obscure tube

[Enzo]

Nothing against MJ15015. When I might want to specify those, I usually just go to MJ15003. It is in my experience a much more common transistor that is already in stock in many shops.

[Gerald kippola]

Sho bud studio stage 500--looking for schematic. Anyone?

Not easy to fix

TE=Enzo;246480]What model?

Looking at another schematic, I see 40636 outputs and 40409 and 40410 drivers. My inclinatiion would be the extra two are more outputs. Easy to tell. Are the bases wired together? Collectors wired together? Is there another 0.33 ohm ballast resistor in series with each extra one?

If the collectors are wored together, the bases together, and the emitters wired together through low ohm resistors, then they are in parallel and are all outputs. If two were drivers, then the base of the output would likely be wired to the emitter of the driver, just as the 40409 and 40636 on the drawing I see.[/QUOTE]