Hi everyone,
we're currently having trouble with a LM5041-based Isolated DC/DC Power Supply generating 24V from a 40V - 120V Input Rail.
Although it's probably not directly related to the problem, heres our setup:
It's a current-fed cascaded topology that consists of (power train in order of current flow):
1. Buck (Non-synchronous)
2. Full-Bridge driving transformer
3. synchronous rectifiers
The feedback mechanism is a secondary-side compensation conntected to the COMP pin using an opto-coupler, FB is permanently connected to GND.
A scope plot probably shows best what im confused about. Evers signal has been measured directly at the LM5041 Pins except the (isolated) converter output.
To further clarify, I added numbers to the relevant points in time.
1. The output voltage reaches its setpoint voltage, COMP goes to GND (which is probably not a sign of good compensation, but that is beyond the scope of my question)
2. At some point later, COMP shoots back up as far as the current SS-voltage allows it to.
Although this picture shows the startup behaviour (as can be seen by the slowly rising softstart-voltage), the signals are looking the same non-startup case (SS staying high then).
Now to my question:
What happens between point 2 and 3? COMP is high, so the converter should be running, shouldn't it?
The time from COMP going to 0V and the converter "retrying" is very close to 250ms.
Is that a coincidence or are there any undocumented protection features that might switch the converter off for this period?
Things already checked:
* No UVLO shutdown, UVLO is stable high
* No 0.6V current limit occured
* No Vcc Under-Voltage (is supplied by independant 12V aux supply)
Please forgive me if I should have forgotten any details, I'm happy to deliver them on request.
I can't publish any parts of the schematic though.
Greetings from Germany,
Simon
