Leak Delta 30 problem

[PaulR]

I bought this amp some time ago and replaced the rather randomly chosen output transistors with 2n3773s. One of these blew for no apparent reason and I ordered more. Mistakenly I ordered 2n3055s. These are branded as multicomp and came from CPC so I imagine they are ok.

Before they arrived I tested the transistors in the amp board which were ok and then replaced the blown 2n3773 with a random TO3 transistor I found and that channel worked so I was pretty confident that it would be a simple fix, although the reason why the first one blew is still a mystery. I wound down the quiescent current a bit just in case.

Anyway the new transistors arrived and I replaced the ones in the previously working channel. When I switched it on the lamp limiter glowed and in the instant before I switched it off there was a hum through the headphones I was using. I checked the new transistors and they are ok.

I have checked the wiring against the notes I made and it is correct. I have to admit I have melted the insulation a bit on some of the wires but not enough to expose the wire underneath. I have snipped the cable ties bundling the wires together and there is no sign of a problem. I have also swapped the amp boards round with no result.

I daren't leave it on long enough to test the voltages and I was wondering whether there are any cold tests I can do or whether I can disconnect anything in order to diagnose the fault further. There is probably something staring me in the face but I just cannot see it!

Thanks

[G6ONEDave]

The original transistor type was RCA 16006 and the service manual also shows 2N3055 as being suitable for the Delta 30 with 48v supply. However if you actually have a Delta 70 the 2N3055 would not be suitable as it's vce is less than the 75v supply. If it turns out to be a Delta 70 then 2N3773 types should be OK.

I would confirm the supply voltage and then check the drivers and biasing, after that carefully check the dc voltages from T6 through the output amp. If you have a variac it might be worth powering up at a reduced input voltage to check for any abnormal voltages. Then slowly increase the input and keep checking for anything that starts to appear wrong. If need be remove supply from one channel and once you have the channel that you're working on fixed, remove it's supply and re-apply supply to the other channel to sort that one out. Once all is OK again then re-apply power to both channels.

Dave

[Radio Wrangler]

Originally Posted by PaulR

There is probably something staring me in the face but I just cannot see it!

My primary instrument is an oscilloscope rather than a multimeter. It's something I grew up with as a kid, building better and better scopes. Then when i got a job at HP in the R&D lab, where everyone reached for their scope first. They are less precise, but they give you many more clues and a better view of what's going on.

The Stereo 30 has a single rail supply and uses a DC blocking capacitor on its outputs. During turn-on the amp has to stabilise with the output transistors giving half the rail voltage to the output capacitor and to provide the current to charge the capacitor.

This is a very likely time for all sorts of transients and bursts of oscillation to occur. You might see something with a scope, you certainly won't with a multimeter.

I think there is also a bootstrap capacitor to charge-up.

The single easiest way to destroy a bipolar transistor is to reverse bias the base-emitter junction until it avalanches and takes a bit of current. This is worth looking for.

David

[PaulR]

Thanks for the replies. I will have another poke around

[SteveCG]

PaulR - remind me please - is this the design that uses edge connectors on its circuit boards?

If so, have you ensured that the sockets as well as the circuit board plug ends are clean? Bad contacts here can create all sorts of faulty operation.

[PaulR]

Yes that is the one. I did clean the contacts but will do so again.

I have in fact solved the latest problem. A bit embarrassing really. I took out the lower transistor yet again looking for inspiration and I noticed a small bead of solder between the screw that holds it on and the bottom of the chassis which was shorting its collector to ground. I removed this and it all seems ok now. I just have one channel running whilst I set up the quiescent current again. The manual says to set it for about 40mA but I think I will use 20mA which is what the Armstrong 600 series says for the same transistors.

They mystery of why the transistor blew, which started all this off is still unsolved. That was the upper one on the other channel.