forward converter problems

Sep 25, 2009 4 Replies
V

Hi,



I designed a 2-trans. forward converter for a battery charger. It is designed for 8A at 30V. It uses a PIC 16F877A controlling a IR2110 driver and IR840 MOSFETs.



The problem is that at moderate currents (>4A) the ferrite transformer (ETD59) starts emitting a crackling noise. tho it continues to work OK upto about 7.5A, after which the overload circuit trips (possibly due to interference).



Thinking it may be a feedback stability problem, I ran it in open loop. But amazingly at about 4A, the PIC started behaving erratically (display went beserk), and this blew out the power circuits.



Not having a 'scope, I cant seem to figure out what the problem is, esp. since it worked far better in closed loop.



Anyone has any suggestions?



Thanks,



vkj.


Vote
Twitter Facebook LinkedIn
M

ver

to

Get a scope. Your PIC is not making a uniform square wave.

If you really can't afford a scope:

Disconnect the power section from the PIC. Connect a resistor and a headphone to the PWM output. You should be able to hear any nonuniformity.

The noise is likely caused by the core saturating.

Vote
Twitter Facebook LinkedIn
J

Sounds more like conducted EMI from the power section getting back into the PIC. Layout, circuit architecture, hard to say without some pics. Of course such interference can make the code execution go berserk, mess with duty cycles and so on.

Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.
Vote
Twitter Facebook LinkedIn
B

Or, maybe the circuit, with feedback, is unstable ?? Does it Sort of sound like "Frying bacon" ??

boB

Vote
Twitter Facebook LinkedIn
V

Thanks for all the helpful responses.

Yes, "frying bacon" sounds about right, tho its been a long time...

I forgot to mention that the switching rate is ~20Khz (50.2 us), and the sample rate is 100.4 us.

The transformer is wound on a ETD59 core. I originally designed it using the usual design equations, (worked out to 53 prim. turns. turns and n=3.5) But this apparently saturated the core, resulting in blown MOSFETs even at very light loads. I rewound with 140 T, n=3.5. That helped bring the current upto

7.5A, tho the noise remained.

What Im really unable to understand is why does it work better in closed loop and fail in open loop?

Im thinking the crackling sound isnt responsible for the open loop failure.

Probably the latter is due to the layout, as MooseFET points out. (I have mounted the transformer and choke off-board and wired up with twisted wires

Vote
Twitter Facebook LinkedIn

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required