Vintage Audio: Measurements of Arcam Black Box 3 mk1

[SIY]

In theory, you'll get a bit less noise on the supply. I use soft-recovery routinely because they're not expensive, but you can accomplish the same thing with snubbing.

 

[restorer-john]

Anybody here have an idea why Arcam chose to use "Fast soft-recovery controlled avalanche rectifiers" and not just simple bridge rectifier diodes?

Fashionable at the time I guess and less noise, but it's a simple supply.

It looks like only three of the four diodes got toasty on the PCB. In a full bridge, if one diode fails, you get half wave rectification and all seems ok, although ripple goes through the roof. I've situations in amplifiers where a pair of diodes in a bridge have failed on a CT +/- supply and the thing still works, albeit with a significant level of hum and lack of power.

Something caused them to get cooked. Realistically, it can only be a shorted diode (that opened up later) or the 6800uF cap. The 7805 is not to blame- it won't allow more than 1A anyway, let alone when in free air- it would shutdown well before then from internal temp rise.

Toasting up a PCB like that takes time, so it's not a random event, more a long term slow-cook IMO. Strange.

 

[LTig]

Fashionable at the time I guess and less noise, but it's a simple supply.

It looks like only three of the four diodes got toasty on the PCB. In a full bridge, if one diode fails, you get half wave rectification and all seems ok, although ripple goes through the roof. I've situations in amplifiers where a pair of diodes in a bridge have failed on a CT +/- supply and the thing still works, albeit with a significant level of hum and lack of power.

Something caused them to get cooked. Realistically, it can only be a shorted diode (that opened up later) or the 6800uF cap. The 7805 is not to blame- it won't allow more than 1A anyway, let alone when in free air- it would shutdown well before then from internal temp rise.

Toasting up a PCB like that takes time, so it's not a random event, more a long term slow-cook IMO. Strange.

Thanks. I shall replace them. May I go with the 1N4004 (plenty on stock), or should I be conservative and use my last four 1N5404?

 

[LTig]

In theory, you'll get a bit less noise on the supply. I use soft-recovery routinely because they're not expensive

Which ones do you use? Maybe I can find them here.

but you can accomplish the same thing with snubbing.

Snubbing? Small caps parallel to each diode?

 

[SIY]

I use UF4007 (strictly speaking, fast recovery, but same effect). Stock item at Digikey and Mouser, so likely the same at the European equivalents.

Yes, small cap and large resistor in parallel for snubbing. A neat alternative method was worked out by Morgan Jones; you'll save the cost of the article in parts the first time you use his tricks.

 

[LTig]

I use UF4007 (strictly speaking, fast recovery, but same effect). Stock item at Digikey and Mouser, so likely the same at the European equivalents.

Yep, found them here, thanks.

Yes, small cap and large resistor in parallel for snubbing. A neat alternative method was worked out by Morgan Jones; you'll save the cost of the article in parts the first time you use his tricks.

OK.