T
treez
Guest
Hello,
We wish to do a Power factor corrected flyback LED driver for a 60W led load. (240VAC input)
We wish to use no electrolytic capacitors at all.
We also want full dimmability down to 5w, and want it to be power factor corrected all the way down to 15w.
We rejected the Boundary conduction mode flyback designs, since theiir switching frequency gets unfortunately high at light loading.
So we chose the Fixed off time control using an L6564 controller . This is described in the app note, AN1792 by st.com (at bottom)
We have it always in DCM.
First of all, in order to “check out” this fixed off time control method, we bashed together a prototype in the lab, as in the attached schematic. It uses fixed off time control, but this one is just 13W, as we just wanted to see how it worked (the fixed off time control).
It worked just as expected for an hour in the lab…worked very well indeed…nice sinusoidal input current from the mains, and no electrolytic capacitors used whatsoever. I was running it, turning it on and off, and happily soldering probes on for all kinds of scope tests.
Then I removed it from the AC power source, and instead connected it to our carroll and mennell isolation transformer, and the LNK302 blew up. There’s a short on the “14VPRI” rail, and after disconnecting the 5V regulator, the short is still there, so it looks like the L6564 has gone short.
Do you know of any “bad boy” stories about the connection of the L6564 controller in fixed off time mode? Maybe the ZCD pin just isn’t meant to be connected up like this?
AN1792…fixed off time control with L656X
**broken link removed**
L6564 datasheet
https://www.st.com/resource/en/datasheet/l6564.pdf
We cant get a response from st.com
---------------------------------------------
Corrected schematic submitted by OP:
Found at:
https://ibb.co/f2S8pT
We wish to do a Power factor corrected flyback LED driver for a 60W led load. (240VAC input)
We wish to use no electrolytic capacitors at all.
We also want full dimmability down to 5w, and want it to be power factor corrected all the way down to 15w.
We rejected the Boundary conduction mode flyback designs, since theiir switching frequency gets unfortunately high at light loading.
So we chose the Fixed off time control using an L6564 controller . This is described in the app note, AN1792 by st.com (at bottom)
We have it always in DCM.
First of all, in order to “check out” this fixed off time control method, we bashed together a prototype in the lab, as in the attached schematic. It uses fixed off time control, but this one is just 13W, as we just wanted to see how it worked (the fixed off time control).
It worked just as expected for an hour in the lab…worked very well indeed…nice sinusoidal input current from the mains, and no electrolytic capacitors used whatsoever. I was running it, turning it on and off, and happily soldering probes on for all kinds of scope tests.
Then I removed it from the AC power source, and instead connected it to our carroll and mennell isolation transformer, and the LNK302 blew up. There’s a short on the “14VPRI” rail, and after disconnecting the 5V regulator, the short is still there, so it looks like the L6564 has gone short.
Do you know of any “bad boy” stories about the connection of the L6564 controller in fixed off time mode? Maybe the ZCD pin just isn’t meant to be connected up like this?
AN1792…fixed off time control with L656X
**broken link removed**
L6564 datasheet
https://www.st.com/resource/en/datasheet/l6564.pdf
We cant get a response from st.com
---------------------------------------------
Corrected schematic submitted by OP:
Found at:
https://ibb.co/f2S8pT
Last edited by a moderator: