Rob,
The formula is 1.414*AC=rail for a capacitor input filter. The one caveat here is that you'll have to add a volt or two for diode losses in the rectifier. So for a 45V rail, you're looking at roughly a 33-0-33V secondary. For a 48V rail, about 35-0-35V.
Note that the voltage need not be exactly 45V or 48V. The cicuit will be perfectly happy with anything within shouting distance of that range. Just be sure that you've got adequate heatsinking if you choose to run the circuit a bit hot.
Grey
The formula is 1.414*AC=rail for a capacitor input filter. The one caveat here is that you'll have to add a volt or two for diode losses in the rectifier. So for a 45V rail, you're looking at roughly a 33-0-33V secondary. For a 48V rail, about 35-0-35V.
Note that the voltage need not be exactly 45V or 48V. The cicuit will be perfectly happy with anything within shouting distance of that range. Just be sure that you've got adequate heatsinking if you choose to run the circuit a bit hot.
Grey
Grey the equation you provided is that for a single secondary or a dual secondary or does it even matter? So if I would use the transformer with the dual 30 volt secondaries would I get 42 volt rails before subtracting the volt or two for diodes in the rectifier? How will this affect the amps output power, and approximately to what degree? Thanks.
Rob
Rob
The engineer at plitron was trying to tell me that I needed a different set for that...I'm building an Aleph 4 and want roughly 48V rails but he said that to get that I should use the one with dual 27V secondaries ....he said this would give me what I wanted....
I however, initially thought that I needed 35V secondaries too....
Could you please confirm that I really do need 35V instead of 27V?
I however, initially thought that I needed 35V secondaries too....
Could you please confirm that I really do need 35V instead of 27V?
The 1.414*AC is per rail, i.e. a 30-0-30VAC transformer would give 42.42-0-42.42VDC, not including diode losses, capacitor leakage, and a bit of resistive loss here and there. The actual rail would work out about 40-41V or so.
Please note that a sufficient current draw will pull the rail down even further, but if you keep it reasonable (50-60% of the VA rating of the transformer or less), the rail will actually end up pretty close to the calculations. Don't sweat a half volt difference. We're also assuming a fair sized bank of caps. Try to save money on the caps, and the rail will collapse sooner, not to mention ripple problems.
Grey
Please note that a sufficient current draw will pull the rail down even further, but if you keep it reasonable (50-60% of the VA rating of the transformer or less), the rail will actually end up pretty close to the calculations. Don't sweat a half volt difference. We're also assuming a fair sized bank of caps. Try to save money on the caps, and the rail will collapse sooner, not to mention ripple problems.
Grey
It actually consumes a bit more than 500W ....its closer to 570W if you have 48V rails....even though it is an SE design you should still also allow for significant headroom...I'm personally using a 1000VA toroid but you should at least use a 750VA one...they're not much more expensive than 600VA ones
Hallo jduncan,
because it is an SE design you have to use an 1000W
transformer. You are drawing permanently 570W. In other
design (class AB you need only peak power). An 600W
transformergets too hot. Better to take two 600W transformer
and stack them together. You can flip one and cancelout the
scalarvector. (my next try)
because it is an SE design you have to use an 1000W
transformer. You are drawing permanently 570W. In other
design (class AB you need only peak power). An 600W
transformergets too hot. Better to take two 600W transformer
and stack them together. You can flip one and cancelout the
scalarvector. (my next try)
So if I use a 1010VA toroid with dual secondaries of 37.5 volts multiply that times 1.414 gives me 53.025 volts subtract 2 volts for diode capacitive and resistive losses and I end up with 51 volt rails. Am I correct in thinking that this is acceptable for an Aleph 4 which specs. 48 volt rails as long as I've got enough heatsinking on the output devices. Will it signifigantly increase needed heatsinking on the other semiconductors? What about an Aleph 2 which specs 45 volt rails?
Rob
Rob
Guys,
don't forget that the plitron and all transformers are rated for voltage under full load. At and above 500VA this voltage drop is about 4 %. So if one uses a 1kVA transformer for the aleph 4, which draws a constant 500VA, the and the voltage drop is assumed to be linear, one would end up with (35+(35*0.02))*SQRT(2)=50.48V, again without accounting for the diode drop.
When building class AB amps I tend to be conservative in the voltage rating meaning I buy transformer with a couple of volts less than they should have. Plitron makes good transformers, with low resistance large section wires, however, I sometimes bought transformers in which thin gauge wire is used and the drop is compensated by adding a whole bunch of extra windings that bring the voltage way above what it should be. Once the problem was so severe that I had to remove windings from the transformer.
[Edited by grataku on 11-28-2001 at 01:49 PM]
don't forget that the plitron and all transformers are rated for voltage under full load. At and above 500VA this voltage drop is about 4 %. So if one uses a 1kVA transformer for the aleph 4, which draws a constant 500VA, the and the voltage drop is assumed to be linear, one would end up with (35+(35*0.02))*SQRT(2)=50.48V, again without accounting for the diode drop.
When building class AB amps I tend to be conservative in the voltage rating meaning I buy transformer with a couple of volts less than they should have. Plitron makes good transformers, with low resistance large section wires, however, I sometimes bought transformers in which thin gauge wire is used and the drop is compensated by adding a whole bunch of extra windings that bring the voltage way above what it should be. Once the problem was so severe that I had to remove windings from the transformer.
[Edited by grataku on 11-28-2001 at 01:49 PM]
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.