AnsweredAssumed Answered

LT3752-1 with LT8311: housekeeping flyback not regulating the primary side voltage

Question asked by eltobol on Jul 4, 2018
Latest reply on Jul 4, 2018 by W.G

Dear Power by Linear,

 

      I have a circuit running from 390VDC, and the housekeeping flyback is not working correctly. I see hiccups of three or four pulses at the HOUT pin, then a pause of about 500 us, then the pulses, and this repeats. The pulses are very short, maybe around 600 ns. The INTVCC pin of the LT3752-1 stays at 10V, but since HFB never reaches 1.0V, the SS1 pin never charges and the forward converter doesn't start up. The HK MOSFET is the IPN70R1K5CE and the HK transformer is the same 750817020 as used in the DC1929A demoboard. I have measured with a low-L test fixture from the HISENSE pin to PGND and the pulses are triangular and reach 50 mV, but I see pulses even when HOUT isn't pulsing. I have a 100 mohm sense resistor and 499 ohms between the source of the HK MOSFET and the HISENSE pin. 

     Attached are two scope plots in which Ch.1 is the LT3752-1 VIN pin voltage, Ch.2 is the LT3752-1 INTVCC pin voltage and Ch.4 is the LT3752-1 HOUT pin voltage. I can share the full schematic and BOM via email, but my customer won't permit me to post them in a public forum. I'm attaching a section showing the HK portion. There is an error that I have corrected: R38 is open circuit and R37 is short circuit - this makes the configuration identical to the DC1929A.

      I though that the IPN70R1K5CE might have too much gate charge for the HOUT driver, it has a total gate charge of 10.5 nC, but in other LT3752 and LT3752-1 designs or demoboards I've seen HK MOSFETs with only slightly less gate charge. The gate threshold of the IPN70R1K5CE also seems reasonable for the HOUT drive level. The switching frequency is 136 kHz, and that seems reasonable to me.

      I tried replacing the slope comp resistor, R39, with 100 ohms, but when I did that the HOUT pin stopped pulsing and the INTVCC pin only rose to around 3V. When I put the 499 ohms back, everything went back to the way it was before. I'm not sure what else to test.

     Thanks!

 

Chris

Attachments

Outcomes