Hi lindomar,
I don't know as well how to draw pcbs for class d amp !
maybe that's the reason why there are some overshoot !
I won't give the file because it's in beta testing and costed me 6*IRF4227 and 3*IR2113 so make your own board and let's share experience...that's the way it should go...
Fredos gave his experience, I gave mine also and it's up to the community right now ;-)
it was hundreds hours of work...
Bye.
I don't know as well how to draw pcbs for class d amp !
maybe that's the reason why there are some overshoot !
I won't give the file because it's in beta testing and costed me 6*IRF4227 and 3*IR2113 so make your own board and let's share experience...that's the way it should go...
Fredos gave his experience, I gave mine also and it's up to the community right now ;-)
it was hundreds hours of work...
Bye.
I did'nt experience any probleme because I run PSU at 50% duty cycle. Only remaining AC ripple at output capacitor. The only tweak I use is hickup mode under fews undred watts...Then normal operation over this.
Fredos
I dont like regulation in PSU, this will add lot of EMI to the equipement....
Fredos
I dont like regulation in PSU, this will add lot of EMI to the equipement....
You are right. And control is also can be difficult because of the EMI. But I would like to try it anyway. My PCB seems should be redesigned because of the large loop area around the IGBTs and the current measuring resistor. But at least I saw the 8 mm output track on the secondary side working like a smoking oven. 
I afraid of using a transformer with fixed duty cycle. The leaking inductance between the primary and secondary coil drastically limit the output current if there is a little difference between the rectified secondary and the capacitor's voltage. So the output voltage will fluctuate fairly with load. There is an alternative method to eliminate the coil at secondary, but it needs more transistors and control.
Why don't you use 100% duty cycle and achieve soft-switching?
I afraid of using a transformer with fixed duty cycle. The leaking inductance between the primary and secondary coil drastically limit the output current if there is a little difference between the rectified secondary and the capacitor's voltage. So the output voltage will fluctuate fairly with load. There is an alternative method to eliminate the coil at secondary, but it needs more transistors and control.
Why don't you use 100% duty cycle and achieve soft-switching?
I'm at fixed 100% duty cycle (50% on transformer). With good winding technic, you can get less than 10% variation. This is enought for a good designed class d amplifier. And it's better than conventional 60Hz xformer!
Fredos
www.d-amp.com
Fredos
www.d-amp.com
I use 6 separate 2 KW switching PSU in //. The core I use is ETD49 3F3. In my 12KHVI, I use a single 144XT500 3F3 core for 6Kw.
Fredos
PS I know that nobody like the use of a toroidal for switching PSU, but their a way to do it well! Think about that!
www.d-amp.com
Fredos
PS I know that nobody like the use of a toroidal for switching PSU, but their a way to do it well! Think about that!
www.d-amp.com
I remember to 3F3 that it has optimal frequency at about 400 kHz. Do you use frequencies over 100 kHz? Or is it because of the special winding technique? Now, with multilayer winding I think the optimal fSW is around 50 kHz. But my opinion is not authoritative.
A few years ago I bought TX63/38/25-3F3 toroids, because I calculated the full ring winding length enables to make the winding in one layer. Is this the way you are going on?
Now, with multilayer winding I think the optimal fSW is around 50 kHz. But my opinion is not authoritative.A few years ago I bought TX63/38/25-3F3 toroids, because I calculated the full ring winding length enables to make the winding in one layer. Is this the way you are going on?
It run at 125Khz, crystal controled, minimal dead time, one layer for each winding, ratio of 20V/turn, half bridge direct drive mosfet drivers, APT 500V 51A monster. You have all you need to do it now!
Fredos
www.d-amp.com
Fredos
www.d-amp.com
Hi unicproaudio
yes it works !
this inversion has been made to compensate the 180°phase shift...
look at fredos 1200 power stage and crest power stage and you'll see one inverts signal and the other don't.
I use Fredos front end and feedback loop with crest power stage so i had to invert these pins to compensate this phase shift...
don't panic it works ! and works great !
yes it works !
this inversion has been made to compensate the 180°phase shift...
look at fredos 1200 power stage and crest power stage and you'll see one inverts signal and the other don't.
I use Fredos front end and feedback loop with crest power stage so i had to invert these pins to compensate this phase shift...
don't panic it works ! and works great !
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.