Hi Charles,
I am using a large bobin now, I would it this afternoon. It gets about +15deg above ambient just at idle. When pumped at 1300Watts for about 1/2 hour it get to about +50deg above ambient.
As far as i am concerned this is okay for the time being. However I am going to order some different toroid cores, as i found out yesterday the cores i am using are getting extremely hot, and really fast too...
I would prefer to stick to the toroids as they fit my pcb design.
Regards
Peter....
I am using a large bobin now, I would it this afternoon. It gets about +15deg above ambient just at idle. When pumped at 1300Watts for about 1/2 hour it get to about +50deg above ambient.
As far as i am concerned this is okay for the time being. However I am going to order some different toroid cores, as i found out yesterday the cores i am using are getting extremely hot, and really fast too...
I would prefer to stick to the toroids as they fit my pcb design.
Regards
Peter....
Ferrite toroid cores tend to dislike a DC current passing through them. I have heard of folks using a dremel and cutting disk to gap them. You might want to go with a toroid core with a built in distributed gap. These are usually iron powder in composition. I believe Sannor has used such a core, but I don't know what the exact composition type was.
Hello,
I am not having that problem with my output inductor overheating now. I just designed a little deflector in the case to provide a small amount of airflow over it, It works wonderfully, Even at long high power runs into a 2ohm load it stays pretty cool.
I also build a pcb with that provides balanced inputs into the amplifier, that solved nearly all the 100Hz humming when connected to some devices ( mainly my laptop ).
Now, all i need to do is build a second module and put it in the case, and try the amp bridged into 4ohms.. It should be able to handle it easily. I am guessing the bridged power will be somewhere in the order of 2000 -> 2500 Watts RMS continous.
Each Module has its own 1000VA transformer so i guess i'll just have to wait and see, I can't wait.
Regards
Peter.
I am not having that problem with my output inductor overheating now. I just designed a little deflector in the case to provide a small amount of airflow over it, It works wonderfully, Even at long high power runs into a 2ohm load it stays pretty cool.
I also build a pcb with that provides balanced inputs into the amplifier, that solved nearly all the 100Hz humming when connected to some devices ( mainly my laptop ).
Now, all i need to do is build a second module and put it in the case, and try the amp bridged into 4ohms.. It should be able to handle it easily. I am guessing the bridged power will be somewhere in the order of 2000 -> 2500 Watts RMS continous.
Each Module has its own 1000VA transformer so i guess i'll just have to wait and see, I can't wait.
Regards
Peter.
So what values are you using now for the gate driver? If I have followed you, you had 15V regulator and 10 ohm resistors, but you have slowed it down a bit. What resistors are you using now?
Do you have schottkys in antiparallel with that resistors?
BTW: Do you have another mean to provide dead-time?
Best regards and congratulations, your project seems to go very well.
Do you have schottkys in antiparallel with that resistors?
BTW: Do you have another mean to provide dead-time?
Best regards and congratulations, your project seems to go very well.
Hello Pierre,
Yes i am using shottkey diodes in anti-parallel with 10ohm gate resistors. I found the cleanest way to slow it down was to actually drive the Driver chip from 12V as opposed to 15.
No, I don't have any other means to provide dead time, as far as I can see there is no need. There is currently no shoot-though, or any noticable cross-over distortion in the drive. ( ie very small amount of dead time)..
Thanks
Regards
Peter
Yes i am using shottkey diodes in anti-parallel with 10ohm gate resistors. I found the cleanest way to slow it down was to actually drive the Driver chip from 12V as opposed to 15.
No, I don't have any other means to provide dead time, as far as I can see there is no need. There is currently no shoot-though, or any noticable cross-over distortion in the drive. ( ie very small amount of dead time)..
Thanks
Regards
Peter
That's good.
Your mosfets seem very rudged but the gate charge is high! You have very good switching times however. Who said that IR211x were not capable of driving big mosfets decently?
I am having problems with feedback and estability, I have submitted a question in the "help with feedback" thread, I hope to solve it soon. You had pre-filter feedback, didn't you?
Best regards,
Pierre
Your mosfets seem very rudged but the gate charge is high! You have very good switching times however. Who said that IR211x were not capable of driving big mosfets decently?
I am having problems with feedback and estability, I have submitted a question in the "help with feedback" thread, I hope to solve it soon. You had pre-filter feedback, didn't you?
Best regards,
Pierre
Hi,
Driving with less voltage makes alot of sense
.
This way the gate capacitances are less saturated, and so help it to turn off much faster as well. That same theory could possibly apply to the internal FETS of the driver itself.
It would explain the cleaner switching by itself.
Regards
Chris
Driving with less voltage makes alot of sense
.
This way the gate capacitances are less saturated, and so help it to turn off much faster as well. That same theory could possibly apply to the internal FETS of the driver itself.
It would explain the cleaner switching by itself.
Regards
Chris
Results to date.
Hello,
For all those interested I am posting some photos of the progress of my amplifier.
I have done some extensive testing of the amplifier. Testing was done with a friends set of speakers from a night club.
Each Channel of the amplifier was tested at 2 Ohms for a 2 hour run, Each channel had 2 x 600 Watt JBL 18" Subs, 2 x (12" & Horns) -> JBL's as well, totalling 2 ohms. It runs great, We compared it directly with a QSC power amplifier of his, and this module is definatly more powerfuly, and sounds at least as good. btw the QSC is rated at 1000W per side into 2 ohms.
Also I have done some static testing of the amplifier in bridged mode, running a 4 ohm dummy load. The RMS power output into the load has been measured with my Scope to be 2400 Watts RMS without any clipping, needless to say - I am really happy!!!!!.
Regards
Peter
Hello,
For all those interested I am posting some photos of the progress of my amplifier.
I have done some extensive testing of the amplifier. Testing was done with a friends set of speakers from a night club.
Each Channel of the amplifier was tested at 2 Ohms for a 2 hour run, Each channel had 2 x 600 Watt JBL 18" Subs, 2 x (12" & Horns) -> JBL's as well, totalling 2 ohms. It runs great, We compared it directly with a QSC power amplifier of his, and this module is definatly more powerfuly, and sounds at least as good. btw the QSC is rated at 1000W per side into 2 ohms.
Also I have done some static testing of the amplifier in bridged mode, running a 4 ohm dummy load. The RMS power output into the load has been measured with my Scope to be 2400 Watts RMS without any clipping, needless to say - I am really happy!!!!!.
Regards
Peter
Attachments
To be serious, I couldn't really believe the reliable operation of a multi-kW DIY classD amplifier. I am very impressed and have to take my hat off to you. 
The dummy load looks a bit dangerous, especially with the wooden(?) board behind it.
Thank you for these instructive pictures! If you don't mind, I would like to catch a glimpse of the final circuit and layout. And if you do some maesurements, the results would be interesting too.
Again: congratulations for the successful setup.
Regards, Timo
The dummy load looks a bit dangerous, especially with the wooden(?) board behind it.
Thank you for these instructive pictures! If you don't mind, I would like to catch a glimpse of the final circuit and layout. And if you do some maesurements, the results would be interesting too.
Again: congratulations for the successful setup.
Regards, Timo
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.