I am in the process of figuring out how the power supply changes the character of an amplifier, and have been playing around with this for the past couple of nights:
Up to now, I have been running each channel off a small 25-0-25 50VA transformer. I redesigned the PSU (rectifier and caps) board this past weekend and that made such a huge difference to the sound that I just had to find out what would happen if I provide increased current from the transformer side.
Now I don't have a 300VA lying around which would have been my first choice, but I do have these 40-0-40 1000VA big boys lying around, so I update the circuit to ensure the higher rail voltage will work without burning anything out, then hook up both power supplies to this one transformer.
All I can say is INRUSH CURRENT!!!
I have a 5A slow blow fuse on the primary of the transformer, and this thing kept blowing up, so I design and implement some inrush limiting using two 50W resistors in parallel to give me around 30 ohms worth or inrush limiting.
So this seems to work well until the switch is flicked which bypasses the inrush limiter and then all hell breaks loose. I actually managed to physically destroy the primary fuse - glass broken, filament gone!
Now I am left wondering what the hell is going on, so I go over everything again with a fine tooth comb. I finally figure out that one of my bridge rectifiers had been damaged by the large inrush current (this is about 4 or 5 blown fuses later), so I replace them with some higher current ones. I was running some smallish ones which were fine with the 50VA supply, so I am not too stressed at this point about it.
After that, the fuse still keeps blowing, so now I am left feeling less optimistic because it should be working by now, and I am running out of fuses.
Finally, I decide to disconnect the amplifier board, and try start this thing up with just the transformer and the power supply boards connected. Once again, the fuse blows. Now at this point I must say that what is worrying me is that these inrush limiting resistors are impressively hot after about 1 second of being in the line, so things are not looking good!
Next up, I check everything again, and now things are looking fine - bridge rectifiers are taking the current, so no issues there. Caps are all fine, so nothing wrong there, no solder bridges or anything else on the PSU boards, so I just can't understand what's cooking the fuses.
So I hook up only one of the PSU boards this time, and with much trepidation fire up the power, and to my amazement there is no bright flash or scary sound, or anything of the like. The onboard LEDs fire up, and everything looks good. I disconnect the board, and hook up the other PSU board, and again - everything just works fine with no drama. By now I am feeling a bit perplexed about matters :???:
Connect both supplies up again, and once more the 5A fuse goes poof!
So naturally now I am out of ideas, so I am thinking that the current draw of the PSU caps filling up must be too much when I have both PSUs connected, therefore even after I switch the inrush limiting off, they are not full yet and drawing heavily. In retrospect it doesn't make sense because the inrush should be pretty much done with after 500ms or so, but at the time it seemed like the only answer. So what do I do? I change the 5A fuse on the primary for a 7.5A fuse.
Safety goggles and gloves on, I power it up and guess what - the 7.5A fuse ruptures like it is a small nuclear bomb! At this point I am seriously feeling like throwing everything in the river! :walled: :angrywife: :RTFM: :boxing: :stop:
As a last resort, I try out a different wiring configuration to connect the secondaries to the bridge rectifiers (each PSU has its own), and fire up one more time. Guess what - it switches on, and there is no drama or explosions, or nothing. Suddenly it dawn on me - the secondaries must have been fighting each other somehow. I still have no idea how this can be, because the two PSU boards are not connect to each other in any way, each has its own rectifiers, fuses, caps, etc yet its the only thing that makes any sense :Ooooooh: :facepalm:
Great, now that's working I decide to plug in the amplifiers onto the PSU boards, and power up again. Once more everything is fine. Now I am sitting infront of the amplifier boards checking for heat on the components just incase something is taking strain with the higher rail voltages. Everything on the small signal sections looks good, but the power transistors mounted on the heatsinks are warming up at an alarming rate. Not long after that, the 2.5A per rail fuses on both the PSUs blow and the system powers down.
So now I am guessing that I had the amplifiers overbiased (probably by a large amount), and while running on the small 50VA transformers this was no problem because it could only supply so much current. With the 1000VA things are different, so they may have gone through a bit of a self-destruct sequence, so I will spend some time tonight figuring out what's cooked where and change them out, then set the bias levels correctly, after which I will try it again.
School fees - they suck, but you do tend to learn alot in a short time...
Ian.
Up to now, I have been running each channel off a small 25-0-25 50VA transformer. I redesigned the PSU (rectifier and caps) board this past weekend and that made such a huge difference to the sound that I just had to find out what would happen if I provide increased current from the transformer side.
Now I don't have a 300VA lying around which would have been my first choice, but I do have these 40-0-40 1000VA big boys lying around, so I update the circuit to ensure the higher rail voltage will work without burning anything out, then hook up both power supplies to this one transformer.
All I can say is INRUSH CURRENT!!!
I have a 5A slow blow fuse on the primary of the transformer, and this thing kept blowing up, so I design and implement some inrush limiting using two 50W resistors in parallel to give me around 30 ohms worth or inrush limiting.
So this seems to work well until the switch is flicked which bypasses the inrush limiter and then all hell breaks loose. I actually managed to physically destroy the primary fuse - glass broken, filament gone!
Now I am left wondering what the hell is going on, so I go over everything again with a fine tooth comb. I finally figure out that one of my bridge rectifiers had been damaged by the large inrush current (this is about 4 or 5 blown fuses later), so I replace them with some higher current ones. I was running some smallish ones which were fine with the 50VA supply, so I am not too stressed at this point about it.
After that, the fuse still keeps blowing, so now I am left feeling less optimistic because it should be working by now, and I am running out of fuses.
Finally, I decide to disconnect the amplifier board, and try start this thing up with just the transformer and the power supply boards connected. Once again, the fuse blows. Now at this point I must say that what is worrying me is that these inrush limiting resistors are impressively hot after about 1 second of being in the line, so things are not looking good!
Next up, I check everything again, and now things are looking fine - bridge rectifiers are taking the current, so no issues there. Caps are all fine, so nothing wrong there, no solder bridges or anything else on the PSU boards, so I just can't understand what's cooking the fuses.
So I hook up only one of the PSU boards this time, and with much trepidation fire up the power, and to my amazement there is no bright flash or scary sound, or anything of the like. The onboard LEDs fire up, and everything looks good. I disconnect the board, and hook up the other PSU board, and again - everything just works fine with no drama. By now I am feeling a bit perplexed about matters :???:
Connect both supplies up again, and once more the 5A fuse goes poof!
So naturally now I am out of ideas, so I am thinking that the current draw of the PSU caps filling up must be too much when I have both PSUs connected, therefore even after I switch the inrush limiting off, they are not full yet and drawing heavily. In retrospect it doesn't make sense because the inrush should be pretty much done with after 500ms or so, but at the time it seemed like the only answer. So what do I do? I change the 5A fuse on the primary for a 7.5A fuse.
Safety goggles and gloves on, I power it up and guess what - the 7.5A fuse ruptures like it is a small nuclear bomb! At this point I am seriously feeling like throwing everything in the river! :walled: :angrywife: :RTFM: :boxing: :stop:
As a last resort, I try out a different wiring configuration to connect the secondaries to the bridge rectifiers (each PSU has its own), and fire up one more time. Guess what - it switches on, and there is no drama or explosions, or nothing. Suddenly it dawn on me - the secondaries must have been fighting each other somehow. I still have no idea how this can be, because the two PSU boards are not connect to each other in any way, each has its own rectifiers, fuses, caps, etc yet its the only thing that makes any sense :Ooooooh: :facepalm:
Great, now that's working I decide to plug in the amplifiers onto the PSU boards, and power up again. Once more everything is fine. Now I am sitting infront of the amplifier boards checking for heat on the components just incase something is taking strain with the higher rail voltages. Everything on the small signal sections looks good, but the power transistors mounted on the heatsinks are warming up at an alarming rate. Not long after that, the 2.5A per rail fuses on both the PSUs blow and the system powers down.
So now I am guessing that I had the amplifiers overbiased (probably by a large amount), and while running on the small 50VA transformers this was no problem because it could only supply so much current. With the 1000VA things are different, so they may have gone through a bit of a self-destruct sequence, so I will spend some time tonight figuring out what's cooked where and change them out, then set the bias levels correctly, after which I will try it again.
School fees - they suck, but you do tend to learn alot in a short time...
Ian.
