I recently purchased a used QSC CX302 amp. Switcher supply, 300 w/ch stereo, class AB. It worked okay on receipt, except the fan noise at 5 w/ch was incredibly loud. Has balanced inputs and screw terminal outputs suitable to replace the RMX850 at a church I attend with a dead B channel.
Cover off, production date was 8/10/98 and it was full of dust balls. A high mileage survivor, IMHO.
As the heat sinks are bigger than my CS800s, the fan noise I take to be a fault. I certainly won't donate it to a church that noisy and with 25 year old high mileage e-caps and fan. Perhaps one of the 3 10 k temp sensors, or more likely the e-cap parallel to it, is leaky. I can't test Q88 the temp sense transistor base to ground, it is invisible. Probably mounted under one of the 4 huge heat sinks.
If I pull the main board & daughter boards to replace these & other caps ( like the 2200 uf 200 v mains caps or the 10 470 uf 80 v miid-rail caps) the switcher supply comes out with it on the coffee table. I suggest to all that when re-e-capping, a sound check be made after every two caps to check one's (my) work. I don't want to be horsing those 200 v high energy parts right side up & upside down every 15 minutes. Nor will they probably work with a 60 w tungsten bulb in the AC feed.
What I propose is to power the mid-rails with +-47 v from a 2 amp 95 vct transformer. That transformer can reside in a nice grounded box and can be powered by the light bulb box. OEM the midrails are +-67 volts, and I see regulators that take that down to +-15, +18, +5, and other voltages much lower than +-47. (High rails are for the class H cousins with 5 times the wattage in the same box). I can leave the high energy switcher parts disconnected from the AC power feed.
I don't see a good spot to tack solder feed wires, D21 D22 D46 D47 are all under the big heat sinks. Those heat sinks have to come off to replace the invisible parts like C33, a temp sense filter cap.
It appears all control voltage feeds to the switcher supply are ahead of the final mid rail feed diodes D74 D75 D80 D81. Unless I'm wrong.
Is this a dumb way to test the boards under repair? Frequently? Has anybody else done this?
Cover off, production date was 8/10/98 and it was full of dust balls. A high mileage survivor, IMHO.
As the heat sinks are bigger than my CS800s, the fan noise I take to be a fault. I certainly won't donate it to a church that noisy and with 25 year old high mileage e-caps and fan. Perhaps one of the 3 10 k temp sensors, or more likely the e-cap parallel to it, is leaky. I can't test Q88 the temp sense transistor base to ground, it is invisible. Probably mounted under one of the 4 huge heat sinks.
If I pull the main board & daughter boards to replace these & other caps ( like the 2200 uf 200 v mains caps or the 10 470 uf 80 v miid-rail caps) the switcher supply comes out with it on the coffee table. I suggest to all that when re-e-capping, a sound check be made after every two caps to check one's (my) work. I don't want to be horsing those 200 v high energy parts right side up & upside down every 15 minutes. Nor will they probably work with a 60 w tungsten bulb in the AC feed.
What I propose is to power the mid-rails with +-47 v from a 2 amp 95 vct transformer. That transformer can reside in a nice grounded box and can be powered by the light bulb box. OEM the midrails are +-67 volts, and I see regulators that take that down to +-15, +18, +5, and other voltages much lower than +-47. (High rails are for the class H cousins with 5 times the wattage in the same box). I can leave the high energy switcher parts disconnected from the AC power feed.
I don't see a good spot to tack solder feed wires, D21 D22 D46 D47 are all under the big heat sinks. Those heat sinks have to come off to replace the invisible parts like C33, a temp sense filter cap.
It appears all control voltage feeds to the switcher supply are ahead of the final mid rail feed diodes D74 D75 D80 D81. Unless I'm wrong.
Is this a dumb way to test the boards under repair? Frequently? Has anybody else done this?
Last edited: