Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

flyback transformer noise

Status
Not open for further replies.

seyyah

Advanced Member level 2
Joined
Oct 7, 2001
Messages
646
Helped
8
Reputation
16
Reaction score
8
Trophy points
1,298
Activity points
6,233
I made a flyback converter and it worked. But the transformer makes loud noise. The switching frequency is 80khz. I also tried 40khz. How can i cut this noise off?
 

uc3842 transformer noise

This is caused by magnetostriction. There must be a low frequency component to the drive current to make the frequency that you can hear.

You can try changing the core material, making the core larger, or putting some material around it to attenuate the noise without limiting the air circulation required to remove the heat.
 

uc3842 transformer noise

If you can hear the noise, I suspect the power supply loop is oscillating at some audio frequency.
To check that, connect a scope to the output and see if there is some ripple at low frequency; almost sinusoidal, at perhaps a few kHz.

You can connect the scope to the output of the error amplifier to see this signal; normally the output of the error amplifier should be just DC, but if the loop is oscillating you will see the signal described above.

Or, connect the scope to the switching element and check if you see a large jitter on the leading or falling edge. Some is normal, due to the 50 or 60Hz input ripple, but if it's excessive, that is an indication the loop is oscillating.

If you do not have a scope, try slowing down the loop, by connecting a fairly large cap across the feeback network. If the noise goes away (or at least changes its pitch dramatically), you know the loop was oscillating. You can then redo the loop compensation. You can post the schematic and I can suggest the changes.

Normally, when you design a power supply, you should measure the gain-phase characteristic, but that requires a gain-phase analyzer, an expensive piece of equipment. If you do not have it, you need to calculate the loop and check the transient response to see if you get oscillatory response. That would mean the loop is close to being unstable.
 
noise on flyback power supply

Look at this topic. I implemented the circuit of an application note with a few changes.
 

switching power supply transformer noise

OK, the capacitor you want to increase (try even 100nF) is the 100pF capacitor between the compensation output (COMP) and the feedback input (FB) of the UC3842. If the noise goes away, then you know the loop is oscillating.
Then we'll find a way to remove the oscillation.

What changes did you make in the circuit?
 
transformer noise

Could you post any link on how to calculate a flyback transformer?
I’m planning to use a pot core.

Thanks in advance
 

noise flyback

To VVV: I'll try it and post the result here, now i am away of the circuit.

To Bomba: It's not easy to explain it here. Core is the first important thing. After you determined it, you determine the turns of primer winding. Auxiliary and seconder are then easy to calculate like a normal line transformer. You only decide conductor size. Then you should determine airgap size etc.

Look at these books

Electronics - Switchmode Power Supply Handbook 1st Edition, Keith Billings (should exist in this forum)

Switching power supply design, Abraham I. pressman.

First book explains the subjects without going very deep in a practical and efficient way.

If you plan to use uc3842 you can find some info on application notes also.

If you want to go deeper i think you have a long way because power supply design really needs expertizing on several different subjects i think.
 

fly back transformer noise

When i changed 100p to 100n noise significantly reduced. What do you suggest me? By the way the differences between the original and mine are only transformer, mosfet and snubber values, current sensing resistors etc. Shortly I made it to fit my situation.
 
bomba said:
Could you post any link on how to calculate a flyback transformer?
I’m planning to use a pot core.

Thanks in advance


++++++++++++++++++++++++++++++++++++++++++

www.powerint.com

application notes & design
 

I packed the transformer more strongly and loaded the converter, now i can hear the noise only if i listen carefully near the transformer. But i think there can still be problem for example i get these waveforms from uc3842? Is this usual or what may be the problem? By the way Rt=10k and Ct=2.2nF. It's running at ~80kHz.
 

u can use " magnetic components for smps by Dr. L. Umanand and Bhatt" to get the information for designing transformrs for smps
 

UC3842 WAVEFORMS.GIF say error amplifier gain too high. Also compensation elements have wrong values.
 
I had face such type of problem in one design where i had changed capacitor in clamping circuit by ceramic to polyester type of cap.and problem solved.
 

Re: switching power supply transformer noise

OK, the capacitor you want to increase (try even 100nF) is the 100pF capacitor between the compensation output (COMP) and the feedback input (FB) of the UC3842. If the noise goes away, then you know the loop is oscillating.
Then we'll find a way to remove the oscillation.

What changes did you make in the circuit?


Thank you!

It solved the problem.

There still is a little noise but it's because the load is minimum,when I increase the load the noise goes away.

Thanks again!
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top