Hello.
C24 should be at least 22-47uF, for 78xx stability reasons.
C26 is too large, 1uF ceramic cap should do just fine giving your frequency ( it should be much smaller than C25 indeed, because it charges from it ).
The primary total series capacitance seams rather small for 3kW of needed power, even with your frequency, it still needs to be larger, i think a total of at least 2uF, maybe more , otherwise the voltage drop on it could be too large, and you do not wan that, also only 3 parallel caps seams not enough giving the high ripple current they have to endure.
IC1 should have some local decoupling on it's power, as close as possible to it's pins.
I am not sure about the protection circuitry, it seams too complicated and possibly a bit slow. Discharging 100uF with MMBT5551 without peak current limiting seams to me unsafe. There could be quite a bit of delay between the moment of the short circuit's high current demand on the power devices, and the moment they are turned off, that delay is magnified by the response time of the IC's as well. You could try to act on IC2, that way you take out at least a few hundreds of nS of delay, and also reduce C15, should be safer this way.
All the best.
C24 should be at least 22-47uF, for 78xx stability reasons.
C26 is too large, 1uF ceramic cap should do just fine giving your frequency ( it should be much smaller than C25 indeed, because it charges from it ).
The primary total series capacitance seams rather small for 3kW of needed power, even with your frequency, it still needs to be larger, i think a total of at least 2uF, maybe more , otherwise the voltage drop on it could be too large, and you do not wan that, also only 3 parallel caps seams not enough giving the high ripple current they have to endure.
IC1 should have some local decoupling on it's power, as close as possible to it's pins.
I am not sure about the protection circuitry, it seams too complicated and possibly a bit slow. Discharging 100uF with MMBT5551 without peak current limiting seams to me unsafe. There could be quite a bit of delay between the moment of the short circuit's high current demand on the power devices, and the moment they are turned off, that delay is magnified by the response time of the IC's as well. You could try to act on IC2, that way you take out at least a few hundreds of nS of delay, and also reduce C15, should be safer this way.
All the best.
Thanks for reply ��
Yes good point sir, later i will update the schematic with your recommendations.
Yes good point sir, later i will update the schematic with your recommendations.
How about describing how you would set up and test the supply - perhaps in quite some detail. Then it would be good to see your design equations and design for the main transformer, as schematics don't really broach that aspect.
Your oscillator setup gives it about 142kHz, Fosc=1/Ct*(0.7Rt+3Rd), thus the power transformer frequency would be half that, 71khz that is.
Frequency has little to do with the achievable power, i have mentioned it just so you know it.
Yes i think you can get that power if you pay close attention to the theory, to the design phase, choose wisely and all should fall in place. I am not saying it would be easy, 3kW of power it's never easy nor cheap, but it can be done.
Yes i think you can get that power if you pay close attention to the theory, to the design phase, choose wisely and all should fall in place. I am not saying it would be easy, 3kW of power it's never easy nor cheap, but it can be done.
Warning‼
It seems to me that you have little experience in such a equipment. They are very expensive and manage potentially lethal voltages and currents.
It seems to me that you have little experience in such a equipment. They are very expensive and manage potentially lethal voltages and currents.
How many SMPS have you built and debugged? What power were they? What test equipment do you have? What knowledge do you have? You need to answer these questions before we can answer your question.klevismuka said:
Is it necessary to be so FU**** mean to this member? He simply asked that if he should build this design from Yamaha. If you cannot give the man a decent answer do not post at all.
Last edited:
Thank you very much Nico Ras.
Yes, i don't know so much about smps, but i'm learning amd the best and fastest way to learn is by contructing smps. But i don't want to build something easy (like detex audio smps). i wanted to build something that look professional and act professional. Also need to add some protection to it. And yes, i'm not a professional engineer like qsc and yamaha enginners, like most of the menbers here, and the best we can do is to copy from they.
I just have a e-55 transformer, and thougt that i can draw a pcb learning from other members, i simply can buy a pcb for 70$ but i want to buld something, and also share with you.
Thank you very much
Yes, i don't know so much about smps, but i'm learning amd the best and fastest way to learn is by contructing smps. But i don't want to build something easy (like detex audio smps). i wanted to build something that look professional and act professional. Also need to add some protection to it. And yes, i'm not a professional engineer like qsc and yamaha enginners, like most of the menbers here, and the best we can do is to copy from they.
I just have a e-55 transformer, and thougt that i can draw a pcb learning from other members, i simply can buy a pcb for 70$ but i want to buld something, and also share with you.
Thank you very much
I do not think they intended to be mean, it is true that this project would be expensive and complicated, and also potentially dangerous if not handled properly.
I did try to help, and would keep on helping where i can, if i'm needed.
I did try to help, and would keep on helping where i can, if i'm needed.
That does not answer any of my questions. Therefore I cannot answer your question.klevismuka said:
That answers the most important of my questions. A multi-kW SMPS is not a good or safe place to start.
I am trying to keep him and his family alive, with all his fingers intact etc. There is not much difference in safety terms between a kW SMPS and a kW transmitter. Neither are good places to start learning about power electronics at high frequencies. I think I have given him a very helpful answer. More helpful than your comment.Nico Ras said:
Yes.
Instead, build and debug a 20W SMPS. Learn from your mistakes. Then do a 200W one. Then start thinking about the 2kW one.
Instead, build and debug a 20W SMPS. Learn from your mistakes. Then do a 200W one. Then start thinking about the 2kW one.
a 500 to 700W will be enough for your needs? go here, take a look, it is well documented and pcb is available : http://www.diyaudio.com/forums/power-supplies/69111-offline-full-bridge-smps-need-help-62.html
it is only 2 rail voltages and you can learn about how to design a good smps.
hope it helps.
it is only 2 rail voltages and you can learn about how to design a good smps.
hope it helps.
- Status
- Not open for further replies.