Tel-ray Supernova

[teemuk]

I always use a light bulb wired in series (in the mains line) with the tested equipment. If the device introduces a high current draw (e.g. as happens during short circuits) the high current flowing through the light bulb will warm up its filament, which then causes the filament resistance to increase. This effectively limits the current and in most cases saves stuff from blowing up in the tested device. Also, if the bulb is lit up it is a nice indication of a fault somewhere in the tested device. In normal operating condition the bulb should not light up or should only have a dim glow. In amplifiers the lightness will "pulsate" according to driving signal (hence the amount of loading). This pulsating can nicely reveal oscillation.

http://www.vintage-radio.com/projects/lamp-limiter.html

[muttomatic]

Great suggestion! Forgotten all about using lightbulbs as a load. I'll try it in the next day or so.

[muttomatic]

The bulb has a steady glow and there's still the audible hum coming from the speaker. But there doesn't appear to be any oscillation. The glow is fairly dim. About that of a night light so it's pulling current somewhere. I'm really getting frustrated with this thing. It sounded so good too.

And we're seeing around 20V at the speaker output. At least that's cut in half :p

Here is a pic of the power board.

[Jack1962]

Check the transistors support components and check you suppy current to them, I never limit current while testing an amp, if it won't hang when you turn it on under normal condition , then it's not right you still have a problem to solve and limiting current just would make it harder to do so. Good Luck, Bro

[J M Fahey]

Dear Muttomatic: build the excellent lamp limiter suggested by Teemuk, and plug your amp into it *without* hooking the speaker, no signal, all pots on "0", etc.
Measure the voltage at the "hot" speaker out, probably you will read around +B (close to +40V) as you did earlier. Then measure the voltage across the power transistors (collector to emitter) , (probably either very low *or* close to 80V)  and across base-emitter (around 500mV). Write a table such as , let´s say as an example:
"Q 4 - 2Nxxxx - VCE:80V - VBE:760mV" and so on. This particular measurement would indicate an open transistor.
Measure on the transistor pins themselves, you might have a worn or corroded socket, you might also have a cracked mica or some other unseen short.
Even better, hand draw a schematic of at least the end of the power amplifier: output transistors, driver transformer, and very specially the string of biasing resistors (it might include some thermistor ) from +B to -B. Probably most of it is in the board you posted.  Bye.

[muttomatic]

Did some measurements.

Speaker output is still reading 47VDC. This voltage seems to be everywhere. there's a six pin molex connector that comes from the amp/transformer section that has that voltage on every pin but the ground wires.

All of these voltages are read with no load, i.e. no speaker connected. They don't really like the voltage.

All -# just represent that they are the same transistor, just different location. They are the same model transistors.

The power transistors.

TIP3055-1 30mVDC and 28mVDC
TIP3055-2 Pretty much have nothing reading on that transistor. Like nothing is feeding it any power.

The rest of the transistors.

2N3397-1 37.9VDC 38.7VDC
2N3397-2 47.5VDC 38.3VDC

2N2955    47.5VDC 0.01mVDC

MPSU57-1 30.2mVDC 0.01mVDC
MPSU57-2 0.01mVDC 0.01mVDC

I'm still not sure what the heck is going on with this thing!

p.s. The more I look at that schematic and the more I look at this amp... they don't really go together. Need the Sano 1000R schematic.

[muttomatic]

What voltage should be present at the speaker output? I assume something minimal.

I replaced half of the diodes by the transformer output and I'm now down to 13VDC at the speaker output with a 10ohm load.

[muttomatic]

Yet another update...

Replaced the rest of the diodes by the transformer. Got rid of the 47VDC at the speaker output.

But.... now the light bulb glows quite brightly and without the bulb in the circuit it blows the fuse. Now I have to figure out where this crazy current draw is coming from.

[teemuk]

In all cases you should read a maximum of few millivolts across the speaker terminals. As what comes to other readings you might want to read (especially page 84):

http://www.pacificrecone.com/files/Darr6.PDF

Rest of it here:
http://www.pacificrecone.com/JackDarrBook.html

If you desperately need a schematic why don't you just sketch it out? That doesn't look like an overly complex circuit at all. Once you get into it it will likely take only a bit more than half an hour. Couple of hours max. You can easily spend days in trying to find a schematic from somewhere while the circuit is right in front of your eyes. Even if you eventually find something you can't trust it to be correct. However, if you trace out the circuit you will have a diagram that is 100% accurate.

[muttomatic]

Looks like it wasn't the power output portion at all. I'm not sure where the short is. Need to discharge the caps and unwire them and test them.

Then after that the OT??

[J M Fahey]

Dear Muttomatic: please don´t get desperate and start changing parts at random. I guess you misplaced some of the power diodes at the power supply, that´s why now it blows fuses straight away. Return it to the original state. You need that power supply working before you can proceed with anything else.
I´m posting a refried schematic, the very poor one seen before. We´ll have to use it for now, we have none other. Please do: 1)Label the transistors with their real names. If in doubt, copy every inscription on them, not on the schematic itself (it will get very cluttered) but on the text of your post. Also label them on the board or chassis as "Q5" ... etc. with some Sharpie or Magic Marker to always be sure what are we talking about. 2)tell me the real world value of the resistors R15 to R22 ¿Did I mention I don´t trust the schematic very much? 3)Draw the Molex connector on the schematic, number the pins (they´re usally labeled somehow) and then measure the voltage to power supply ground from every pin. I strongly suspect that your -B (-47V) is missing, making *everything else*  measure close to +47V. Other possibility is that both R20 and 22 are open. 4)Please measure some voltages for me. The red ones (V01/2/3) should be measured with the black probe on the power supply ground itself and the red probe where indicated. I suspect V03 (-B) will be missing at the board, in that case trace its path , specially on both sides (male and female) of the Molex connector. IF that is OK, measure V04 to 09, placing the black and red probes where indicated by the respective colors. Good luck.

[teemuk]

Looking at the photo of the power board, the circuit doesn't seem to resemble the one in the schematic at all. There are only two TO-3 cased devices mounted at the sides of the U-profiled heatsink. I suppose there are no more output devices – at least I see no wires running to them. Also, I see only two power emitter resistors. The two un-heatsinked TO-218 devices on the board are likely the output drivers and in between them - on the board - looks to be two diodes (at least they seem to have that red-ish glass package typical to diodes). If they are indeed diodes then they are likely part of the biasing string. It doesn't look like this circuit would be using an interstage transformer at all. The rest of the transistors at left are most likely parts of the input stage and voltage amplifier (the one in metal package) and I see no wires running to a transformer from them. I see no wires running to the speaker either so I suppose the connection is through the molex. What the heck is that round thing next to the big capacitor at top left? So, all in all, this looks like a much newer design than the one depicted in the schematic.

This is of course just my first impression. Detailed photos would help a lot in determing what each part is and what the circuit likely looks like.

I really encourage that you trace out that schematic. The one with interstage transformer doesn't seem to be any good and it is very unlikely that you would find the one for this circuit from somewhere. Having the schematic would help you – and us – a lot. Plus, you would get a clear insight of the circuit's architechture and figure out which part does what. Also, there's never too much schematics to old transistor amps around so in the future your effort to trace out the circuit might help other people who are in a similar position as you are now.

[J M Fahey]

Dear Muttomatic: now I know that the schematic and questions I posted before are useless in this case. My browser only showed me part of the picture you posted, I wrongly assumed that the board (and U shaped heatsink) were *much* longer, possibly containing other two TO3 devices and some other stuff (and a chassis mounted driver transformer). Now I "pulled" the full picture, opened it into Irfanview and sharpened it as much as possible. I´m posting a sketch of what I think the power schematic looks like, you´ll have to correct it and "fill the blanks". I assume the input transistors are NPN and the "Class A" PNP but the reverse is also possible. I´ve also drawn both outputs NPN, but they might be complementary (NPN/PNP). Please redraw the schematic as you see it on the actual board; also a couple more sharper pictures will help a lot. If possible also include a bottom one (solder side).  To begin with, the amp has no short protection at all; not so bad within a combo, but if you add any external speaker, it becomes very dangerous. We still need that Molex drawing and voltages.Still suspect your -B is missing, also +-47V is too much for 2n3055/2955 devices.  Bye.

[Artspoke]

Argh!  The exact same thing just happened to me!!  I was playing through my supernova, and a huge buzz ensued.  The 50 Ohm 10W resistor (R14) was smoking!  I shut it off immediately, and now I too can measure about 40 VDC at the speaker output.  HELP!  Did you ever get to the bottom of yours?  It seems like we have different models slightly, based on your pictures, but maybe the same problem. . .

[teemuk]

Those symptoms are almost a sure sign of an output transistor/stage failure.

One or several of the output devices fails by short circuiting, hence basically hooking the speaker load directly to one of the power supply rails. The resulting massive current draw will cause supply voltage to sag and ripple voltage to increase. DC can nicely burn a voice coil but it can not motivate the speaker back and forth - however, the AC ripple voltage riding on top of the DC can and you hear it as hum coming from the speaker.

The repair procedure should be rather obvious and easy: Locate the failed transistors and replace them.