Good Day All,
I have just blown my prototype this morning. As you can imagine, I am ecstatic!
This SMPS is half-bridge technology and uses an L6599 from ST Microelectronics.
I realized I made a mistake in the measurement of the leakage inductance and consequently increased the size of the resonant cap. A quick screech was heard followed by the explosion of both MOSFETs.
I did not have the time to see what was going on on my CRO.
The resonant frequency has been calculated to 150kHz.
I think I also had a bad regulation loop. The frequency was stuck to the minimum (100kHz). Max f is 200kHz.
How do you compute/calculate the current in the opto-coupler? We must also take into account the variation of coupling coefficient from opto to opto (same type. I.e. 4N25).
During the successful run, the current was partly sine wave. Partly, because it was only portions of sine wave in synchro with the square waveform of the chopper. The frequency of the current was much higher than the f of the chopper. Typically 300-400kHz.
The IC also died. I will protect the next chip by inserting series diodes with the gates of the power MOSFETs.
Any comments or advice will be much appreciated. Thank you.
Serge
I have just blown my prototype this morning. As you can imagine, I am ecstatic!
This SMPS is half-bridge technology and uses an L6599 from ST Microelectronics.
I realized I made a mistake in the measurement of the leakage inductance and consequently increased the size of the resonant cap. A quick screech was heard followed by the explosion of both MOSFETs.
I did not have the time to see what was going on on my CRO.
The resonant frequency has been calculated to 150kHz.
I think I also had a bad regulation loop. The frequency was stuck to the minimum (100kHz). Max f is 200kHz.
How do you compute/calculate the current in the opto-coupler? We must also take into account the variation of coupling coefficient from opto to opto (same type. I.e. 4N25).
During the successful run, the current was partly sine wave. Partly, because it was only portions of sine wave in synchro with the square waveform of the chopper. The frequency of the current was much higher than the f of the chopper. Typically 300-400kHz.
The IC also died. I will protect the next chip by inserting series diodes with the gates of the power MOSFETs.
Any comments or advice will be much appreciated. Thank you.
Serge
Hi Workhorse,
Thanks for the advice. I will test this topology next year.
I am going on holiday so .... Wooohoooo
I will repair my prototype and test it thouroughly, but with reduced voltage on the primary. I need to see what happens at resonance.
With a current source, I will experiment with the opto regulation. This is something I am not familiar with.
Somehow, I do not think that making the xfromer resonate is not a good thing. The other designs have an external choke for that purpose. Although, the L6599 applications do not have external choke; and have only one high voltage cap.
Anyway, this project is to power up vacuum tubes with 350 and 1kV.
Very dangerous stuff.
and I may not need a perfect regulation.
But it is a good experience.
Thanks again. I am going to look at datasheet on quasi-resonant circuit right now.
Serge
Thanks for the advice. I will test this topology next year.
I am going on holiday so .... Wooohoooo
I will repair my prototype and test it thouroughly, but with reduced voltage on the primary. I need to see what happens at resonance.
With a current source, I will experiment with the opto regulation. This is something I am not familiar with.
Somehow, I do not think that making the xfromer resonate is not a good thing. The other designs have an external choke for that purpose. Although, the L6599 applications do not have external choke; and have only one high voltage cap.
Anyway, this project is to power up vacuum tubes with 350 and 1kV.
Very dangerous stuff.
and I may not need a perfect regulation.But it is a good experience.
Thanks again. I am going to look at datasheet on quasi-resonant circuit right now.
Serge
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.