Newbie level 3
Just wondering if anyone can help me -
I have a PWM design which switch 24 volts via a TL494 and two IR 2100s into a push pull FET bridge and a large step up transformer which gives an output of 600 volts - which is then recified and smoothed into a small bank of caps and then pulsed into load of about 500 ohms.
What I am wanting to do is to vary the output voltage down. I have done this by presenting a PWM signal (of about 5 Khz) to the 494 - obviously a low mark space ratio only turns the push/pull on for a small while and generates a smallish voltage in the caps - the bigger the PWm mark/space the more volts output . The output regualtes well any any voltage between 0 and 600 can be acheived - success - so far.
The FET bridge stays cool at 600 volts output - even though the power output is high. But as the output voltage is reduced - suddenly the voltage across the transformer changes ( from a quite nice squareish wave ) and the heat developed in the FET (and transformer ) increases dramtically - infact at any output voltage below about 500 the system produces lots of heat (in the FETS and transformer) even though the system should be working less hard.
I have tried playing with the pulses into the FET - increasing the Deadtime and also reducing the pulse width but without joy . Also changing the freq of the converter which is about 60 khz seems to do nothing
I am guessing the transformer just does not like dumping (even a few pulses 600 volts) into a capacitor load that is a few hundred volts lower - this somehow gets reflected into the primary of the transformer and "upsets" the FETS
Can anyone shed more light on the matter - and hopfully come up with a Solution?????
7th May 2009, 04:33
Full Member level 1
pulse width modulation transformer
Heating problem may be parasitic frequencies (autooscillations).Try with little decoupling capacitors in drain of the fets.After you have an stable output will be easy work changing output voltage.