I am having some big problems when developing hardware using the LT8304 IC. The converter specs are:
Vin = 42Vdc
Vout = 110Vdc
Pout = 20W
I initially ran through all the calculations and simulated the circuit. The simulation looks very good and appeared stable including after load transients. I went ahead and designed the PCB and assembled them, i also had a second PCB designed by a colleague. I wound my own transformer with a 1:2 ratio, Lp is roughly 50uH and the coupling is about 0.98-0.99 for the various ones I have wound.
Firstly the switch node waveform looks terrible. If you see below the switch is really not switching on and off as it should. Is worth noting I have been doing all testing with a resistive load that should give about 12.5W Pout with the desired 110V Vout. Vout also doesn't regulate correctly when changing Vfb. I tried to capture the primary current using a 10mOhm sense resistor in series with one of the transformer legs. It is shown below in yellow and as you can see it doesn't look good. Please ignore the green and purple traces, they aren't representative of what is going on.
I know the output voltage will not regulate correctly if the primary current is so bad as the IC relies upon the primary current to be converted to a voltage representation in which it uses to drive the internal FET on and off (through the internal current loop). I wound some other transformers with different gauge wires and the switch node did change but the duty cycle went really low and the output voltage was lower.
When looking over a longer timebase the output voltage is collapsing, it looks like the IC is stopping and restarting.
Please note I see the same occurrence with both designed PCB's so I don't think its the layout.
I have tried many other investigation leads such as swapping components etc.
I am 95% sure it is down the the transformer as all other components are pretty standard and the layout of the PCB adheres to good design rules regarding clearances and routing orientation. If the transformer is poor I guess it has a direct affect on the primary current therefore the IC struggles to regulate. I know I'm sort of answering my own question but surely the IC can't be so sensitive to transformer variations!!
Has anyone had similar problems with this IC and what was the solution? Considering it's suppose to be a very simple design it has proven to be difficult.
