I made a HV flyback transformer circuit using IR2153 with 50% duty cycle , the transformer makes loud noise when i connected to load.
1.tell me how to reduce the sound.
2.which type of sound absorbing materials should be used i.e,insulating of sound. so that i can encapsulate flyback transformer.
I think that instead of trying to hide the noise you should suppress it at the origin. I suppose your switching frequency is higher than 20 kHz. The IR2153 is not intended in principle for flyback operation. Because your duty cycle is fixed at 50%, probably you have saturation problems in the transformer. The IR2153 does not have current control. If you send a diagram of your circuit, it will be easier to help you.
I switched ir2153 at 15khz,i think problem is due to core saturation.
i used ir2153 as a switching circuit to drive the gate terminal of mosfet,i applied 12v to flyback first terminal and second terminal to drain terminal of mosfet and source terminal is grounded.
(i tried this circuit not by internet)
I think problem is due to core saturation.
I propose to reduce the charge to connect or connecting another same type of transformer and to operate both items in parallel.
If you just increse the switching frequency you will reduce the flux in the transformer (which is proportional to ton (and put the noise out of the audible range), so saturation will be reduced.
My objective is to maintain high voltage,at 15khz the voltage level is maximum,if i increase the frequency the voltage level is decreasing.
So,i am asking about how to reduce the noise.
To reduce emission, vacuum impregnate the windings with electrical grade epoxy to prevent any movement of the wires and windings, After assembling the winding(s) to cores and setting any air gap(s), inject epoxy into the air gaps and between the windings and core to prevent any differential movement.
If this does not eliminate the noise, you're likely getting magnetostriction from the core material itself, so soundproofing and/or vibration mounting may be necessary. Some suppliers:
http://www.soundproofcow.com/
http://www.soundproofing.org/
If your switching frequency is 15kHz, then any problem at that frequency will still be in-audible to average humans. I think the problem is originating from oscillations in the closed loop PI controller, which is creating much lower frequency oscillations. these low frequency oscillations are making the cores get attracted and repelled, creating the audible sound. Of course, saturation will aggravate the sound and impregnation will reduce it, but the cause will be corrected by tuning the PI controller carefully. I suspect the P-gain is too large.
If increassing the switching frequency the voltage drops this means that your transformer design is too critical, I mean that the indjuctance is too high. But I insist, don't try to impregnate the transformer an hide the problem, change the transformer design and increase a little the swithching frequency. The power you can get is 0,5 (V ton)2 fs / L, where V is the primary voltage, fs the switching frequency and L the primary inductance. In your case, as duty cycle is 50%, ton = 1/ (2 fs). So you have to decrese L, using less turns. But then you have to check that the induction is not too high and the transformer does not saturate: check Bmax = V ton / (N S), where N is the nuber of turns and S the effective area of the ferrite. If you have decreased N to increase L, the induction B will increase. With standard ferrites (3C90, N87) do not go beyond 0,32 Tesla. If this is not possible you should have to use a bigger ferrite core. Then another problem will be the losses in the ferrite core, but this is another story ...
- Badges
- Science method