I'm working on a Carver M-4.0T power amp that I inherited, and I know nothing of its history other than it's in very clean condition and that it's one of the last revisions with the charcoal faceplate. It had not been turned on in years.
I did a safe power up and found that it was basically working, at which point I started checking for more subtle problems or time bombs. I found a number of smaller electrolytic capacitors on the boards on their way out with rising ESR, so I'm replacing those. The +/-11.4V supply caps were cooked by resistors beneath them, and the 3.3uF 160V caps showed what looked like leakage around their leads.
Last night, doing careful checks with the amp at idle, I discovered that two of the power supply rail steering diodes are shorted -- no diode drop across them. These are the diodes that prevent the higher voltage power rails from dumping into the lower voltage rails during rail switching. One is D28 in the Left channel, on the +67V supply, and the other is D29 in the Right channel on the +31V supply.
Finding two of these shorted makes me consider if these were under-spec'd parts (3A @ 200V). The originals are MR852 rectifiers, which were fast at the time in 1989, but not that fast anymore. I'm looking at the Vishay MUR420, which is officially rated 4A, but, in the datasheet, rated 6A at lower temperatures. It's also a lot faster than the now-obsolete MR852 with max recovery time of 35ns vs. 200ns with the older part. Discussions of Class G amps I've read elsewhere say you want fast diodes here to minimize power supply rail coupling, for obvious reasons.
You can get 6A fast rectifiers, but they are large enough to start interfering with other components on the boards.
Does the Vishay MUR420 sound like a good choice? I'm mainly a tube amp guy, so I'm not as confident with sand-state devices. Thanks in advance for any help.
I did a safe power up and found that it was basically working, at which point I started checking for more subtle problems or time bombs. I found a number of smaller electrolytic capacitors on the boards on their way out with rising ESR, so I'm replacing those. The +/-11.4V supply caps were cooked by resistors beneath them, and the 3.3uF 160V caps showed what looked like leakage around their leads.
Last night, doing careful checks with the amp at idle, I discovered that two of the power supply rail steering diodes are shorted -- no diode drop across them. These are the diodes that prevent the higher voltage power rails from dumping into the lower voltage rails during rail switching. One is D28 in the Left channel, on the +67V supply, and the other is D29 in the Right channel on the +31V supply.
Finding two of these shorted makes me consider if these were under-spec'd parts (3A @ 200V). The originals are MR852 rectifiers, which were fast at the time in 1989, but not that fast anymore. I'm looking at the Vishay MUR420, which is officially rated 4A, but, in the datasheet, rated 6A at lower temperatures. It's also a lot faster than the now-obsolete MR852 with max recovery time of 35ns vs. 200ns with the older part. Discussions of Class G amps I've read elsewhere say you want fast diodes here to minimize power supply rail coupling, for obvious reasons.
You can get 6A fast rectifiers, but they are large enough to start interfering with other components on the boards.
Does the Vishay MUR420 sound like a good choice? I'm mainly a tube amp guy, so I'm not as confident with sand-state devices. Thanks in advance for any help.