Seeking help from Electronic Engineer part 2

  phill1959 12:06 21 Mar 06
Locked

Paddyjack has now set me up so I can answer questions directly.
Firstly i would like to say thank you for your help also i am only an amature at this although hoping to get better.
I am using an sg3525 pwm to drive a transfomer which is is driving two irf840 fets in a half bridge config i have zeners clamping the the voltage at 15v however i seem to be driving the inductors instead of the load, The load being 12 peizo vibrators. The coils seem to resonate instead of the transducers, i know that the frequency has alot to do with the matching of the inductors and have tried to play with both without any satisfactory results. I have occasionly had problems with shoot though and am not sure if i have a parasitic problem with stray EMPs ?

  terryf 12:15 21 Mar 06

It seems to me (as an expired electronics engineer, I blew all my fuses a long time ago) that you may get help with this on a different sort of forum. I suggested in my last post that you perhaps spend time getting used to the web and then post a question into the Google search engine, it is surprising how typing a straight forward question will come up with the answer or you could search for 'electronic design forums'. Don't despair, some-one out there will know the answer to your problem

  octal 12:50 21 Mar 06

It sounds like the inductors are resonating. Are the indictors acting as a choke? If the are they will have to be designed so that resonances fall outside the frequency range, no mean feat if you are hard switching the FET’s which will be rich in harmonics.

Have you thought of using non-inductive thick film resistors in place of the inductors? They can be obtained from Farnell Electronic Components.

Without seeing the circuit its difficult to visualise. Maybe upload a copy of the circuit to here click here and someone might be able to advise.

  phill1959 13:46 21 Mar 06

To be honest i did not expect replies to my question so quickly, paddyjack has put me straight on how to communicate with people such as yourself. I will do as you suggested there seems to be a few forums. Thankyou for your help.

  phill1959 14:08 21 Mar 06

Yes the inductors are resonating if i play by adding as well as altering the wind ratio i get some improvement on the tranformer and also the inductor with some ultrasonic action. However the fets seem unstable when i do this and after a brief 30 secs or so the circuit collapses and i get shoot through i presume i am getting saturation in the coils.I get the impression that the root of the problem could be the frequency as well as the dead time from the PWM as the circuit will only start if pin 8 is disconnected. I will do as you say and try to upload a Circuit diagram to the site you gave me thankyou for your time and help.

  ACOLYTE 14:15 21 Mar 06

Just a link to the other post paddy started
click here

did you check out the sites i linked to they have drawings on them.

  phill1959 14:39 21 Mar 06

Finding my way around the sites, there is so much to look through Thank you 4 your help

  octal 17:28 21 Mar 06

If you are winding the coils on ferrite cores it could be the ferrite saturating, when this happens the ferrite heats up and the magnetic properties get destroyed until it cools down again, as I have found from past experience.

You mention PWM, if you are using an IC to generate the PWM, which I take it you are doing because you mention pin 8, you should have a dead space at the zero point where both transistor are off until the next pulse is generated, if this is not done both transistors can conduct at the same time which will very likely destroy them.

We really do need to see the circuit.

  octal 19:34 22 Mar 06

I have uploaded your circuit for you click here

It looks like an ultrasonic cleaning circuit, I can puzzle out how it works. I think the first thing to look at is the output waveform from the ic, you will need to disconnect the FET gates from the transformer and leave the FET's disconnected then put an oscilloscope across one of the output windings and make sure the waveform is stable and continuous. I suspect it will be OK, but its worth eliminating it.

If the waveform looks OK then the problem is in the output section, my thoughts go back to that ferrite core that's driving the quartz, looking at it its fairly high power and you're going to require a big lump of ferrite to handle the power otherwise it will saturate for sure. What is the ferrite? Is it a ferrox cube type?

  phill1959 08:32 23 Mar 06

Appreciate all your help!!
For some reason could not get on the site last night page would not display? > Yes it is a U.S.C circuit or would be if i can get it running!! The gate wave form seems to be good. until the circuit collapses on full load. (T1)I am using fairly heavy ungapped e cores rated 100hz> 100khz I think? This area i am not 2 Familiar with however the cores used are also at 30 khz in another cleaner.
L4 is 2 ungapped e cores but i am shimming them to alter the flux density. At the present it does not alter as power is not getting past T1 The circuit makes the T1 rattle quite badly and nothing else. I would like to note though the transformer EG L1-L2-L3 has a high pitch whistle where T1 has a much lower pitch more of a rattle than anything Today i will recheck the wave form and freguncy on L 1>2>3 unfortunately i will be unable to let u know the result of my findings until tomorrow have to go out this evening thanks 4 your help and advice

  phill1959 22:05 26 Mar 06

yep wave form is good! don't seem to be getting anywhere Think maybe it is time to learn the math on transformers and ferrite core flux density cannot seem to get my head around it do you know where i could get info simplified for the dummy,I cannot find it on the net well "none that can tell me how to select the ferrite type and frequency??????? or even what type of ferrite i have now"

This thread is now locked and can not be replied to.

The Legend of Zelda Breath of the Wild review: Five hours with Zelda on the Nintendo Switch

1995-2015: How technology has changed the world in 20 years

How the painting-like animated sequences in A Monster Calls were created by Glassworks Barcelona

The 22 best Safari extensions | Best Safari plugins: Improve Apple's Safari web browser with these…