I got a broken Cambridge Soundworks BassCube 12 someone was giving away. Powered on, and sub suddenly went all the way in/out (not sure which). Disconnected and I measure -49V on speaker output. Ouch.
I trace things and find that it has a DPDT relay that connects the +/-50V rails to the amplifier, and that one side is stuck closed, and the other stuck open, so the amp was only getting -50V. I had also noticed that one of the transistors was hot, but the other wasn't. With it off, I didn't measure any shorts between the rails and the speaker output, so the transistors aren't shorted.
Oddly there was a second relay inside whose coil was in series with the main one, but had NOTHING connected to the contacts. So I rewired the power rails to this one, and it promptly blew out the contacts for the +50V; they're open permanently now.
I found another SPDT relay and connected it in series to control that rail, and it didn't destroy it. I got ~0V at speaker output, so I connected it to the speaker and powered on. Loud pop, then the same -50V again, with the speaker not liking it at all. As best as I can tell, the protection circuit is kicking in when it's turned on, but only turning ONE relay off, perhaps just lowering voltage until both rails aren't on. So the speaker gets -50V again.
At one point, only the +50V rail was connected, and the OTHER transistor got hot. So the first time one got hot was clearly a result of having only one rail.
If I UNPLUG the speaker from the amp, power in, THEN plug the speaker in, it works beautifully. Been running it for hours now, with georgeous room-filling bass and no hum or anything. I checked and none of the amp transistors is getting hot now. I'm going to get a really beefy DPDT relay and see if perhaps the relays in it were just old or maybe got worn out by all the bass. They were 5A 250V, so it's hard to believe the contacts would blow that easily. There aren't any capacitors either on the amp side after the relay, so no inrush current to fuse the contacts.
Also a side question about improving the voltage regulator design: With this thing on for a while now, I'm noticing that almost all the heat comes from what seems to me a terrible voltage regulator design: power resistor to zener diode. They employ FOUR of these, for two +/-15V supplies from the +/-50V filtered/rectified output of the transformer. Would there be any downside to just putting some 15V regulators in place of this? The PCB is really toasted/brown from all the heat these wasteful regulators generate. I'd rather not subject it to more heat, and save some electricity as well. I'm thinking that there's one possible snag: regulators would put less constant load on the transformer, which might result in a higher +/-50V supply since it's unregulated off the transformer. I'm wondering why they don't use a switching supply in subs like this, since switcher noise is high frequency and therefore totally filtered out by the nature of the sub being very low-frequency.
I trace things and find that it has a DPDT relay that connects the +/-50V rails to the amplifier, and that one side is stuck closed, and the other stuck open, so the amp was only getting -50V. I had also noticed that one of the transistors was hot, but the other wasn't. With it off, I didn't measure any shorts between the rails and the speaker output, so the transistors aren't shorted.
Oddly there was a second relay inside whose coil was in series with the main one, but had NOTHING connected to the contacts. So I rewired the power rails to this one, and it promptly blew out the contacts for the +50V; they're open permanently now.
I found another SPDT relay and connected it in series to control that rail, and it didn't destroy it. I got ~0V at speaker output, so I connected it to the speaker and powered on. Loud pop, then the same -50V again, with the speaker not liking it at all. As best as I can tell, the protection circuit is kicking in when it's turned on, but only turning ONE relay off, perhaps just lowering voltage until both rails aren't on. So the speaker gets -50V again.
At one point, only the +50V rail was connected, and the OTHER transistor got hot. So the first time one got hot was clearly a result of having only one rail.
If I UNPLUG the speaker from the amp, power in, THEN plug the speaker in, it works beautifully. Been running it for hours now, with georgeous room-filling bass and no hum or anything. I checked and none of the amp transistors is getting hot now. I'm going to get a really beefy DPDT relay and see if perhaps the relays in it were just old or maybe got worn out by all the bass. They were 5A 250V, so it's hard to believe the contacts would blow that easily. There aren't any capacitors either on the amp side after the relay, so no inrush current to fuse the contacts.
Also a side question about improving the voltage regulator design: With this thing on for a while now, I'm noticing that almost all the heat comes from what seems to me a terrible voltage regulator design: power resistor to zener diode. They employ FOUR of these, for two +/-15V supplies from the +/-50V filtered/rectified output of the transformer. Would there be any downside to just putting some 15V regulators in place of this? The PCB is really toasted/brown from all the heat these wasteful regulators generate. I'd rather not subject it to more heat, and save some electricity as well. I'm thinking that there's one possible snag: regulators would put less constant load on the transformer, which might result in a higher +/-50V supply since it's unregulated off the transformer. I'm wondering why they don't use a switching supply in subs like this, since switcher noise is high frequency and therefore totally filtered out by the nature of the sub being very low-frequency.
Comment