Transformer producing high voltage on primary

[SilverWingedSeraph]

Hello, EEVBlog Forums! I'd like a little help with a problem I'm having.

I'm using a NE555 and a 2SK2765 N-channel MOS FET (sinking current) to drive a transformer at ~150 Hz for the purpose of producing a high voltage. That part works, with the secondary putting out about 600 volts from my 12v source, but when I probe the connection between the MOS FET's drain and the primary winding, rather than a 12 volt square wave at 150 Hz, I see a really weird waveform peaking around 50V. I assume this is back EMF from the primary, but I can't for the life of me figure out where to put a diode to absorb it.

Relatedly, how would I go about debiasing the square wave (the one actually being fed into the transformer)? I tried putting a capacitor in series, but it didn't seem to work.

Thanks!

[ali6x944]

This is Simple one , it's inductive spicking, what is going is that when current flows in the inducter a magnetic field is produced which energy is stored in, when the curren stop flowing the magnetic field collapse including a voltage aka back EMF.

[SilverWingedSeraph]

Just as I figured. Thanks!

The question remains, however... how do I prevent it from damaging my MOS FET, or do I not need to be worried about that?

[Connoiseur]

Just as I figured. Thanks!

The question remains, however... how do I prevent it from damaging my MOS FET, or do I not need to be worried about that?

You should be worried about it. It can easily fry the mosfet if run you it for long without load. Put a clamping circuit like the one attached.

[SilverWingedSeraph]

My resistor exploded.

[Connoiseur]

My resistor exploded.

Insert a power resistor NOT 1/4 watt one.

[SilverWingedSeraph]

How much current can I expect to be going through it? I have some large ceramic 0.22 Ohm resistors, will those do the job?

[prawncrackers]

if youre worried about efficiency (like in an inverter/backup power kind of thing) then a demagnetizing winding might help.
basically its another winding approx 10% the current capacity and wound at the opposite direction of the primary with a diode in series with anode towards the ground to dump the inductive kicks back to the primary rails. of course this makes transformer winding complicated.

another might be a "lossless" snubber but this is a bit more complicated as it requires a large-ish inductor.

the third one might be a an RCD snubber common for switched power supplies.

check the rise/fall of your inductive spike and adjust the RCD snubber accordingly.

[SilverWingedSeraph]

Thanks for the additional information, I'll look into that.

I'm still wondering how to remove the DC bias from the signal, if that's possible?

[ali6x944]

Thanks for the additional information, I'll look into that.

I'm still wondering how to remove the DC bias from the signal, if that's possible?

Maybe a cap or some thing like that would remove the dc offset and give u the pure AC change.
And  it's not that complicated to stop it producing high voltages just put a diode in the opposite direction of current flow in the inducter  as  prawncrackers point out it may get a bit complicated if u are searching for high efficiency but a simple reverse bias diode with a 0.1uf cap in parallel to reduce the amplitude of the spike but be ware that this might cause it to resonate and couse distortions in your secondary so be ware and Good luck with your project 

[Zero999]

Hello, EEVBlog Forums! I'd like a little help with a problem I'm having.

I'm using a NE555 and a 2SK2765 N-channel MOS FET (sinking current) to drive a transformer at ~150 Hz for the purpose of producing a high voltage. That part works, with the secondary putting out about 600 volts from my 12v source, but when I probe the connection between the MOS FET's drain and the primary winding, rather than a 12 volt square wave at 150 Hz, I see a really weird waveform peaking around 50V. I assume this is back EMF from the primary, but I can't for the life of me figure out where to put a diode to absorb it.

Relatedly, how would I go about debiasing the square wave (the one actually being fed into the transformer)? I tried putting a capacitor in series, but it didn't seem to work.

Thanks!

Don't worry about it. Adding a diode would reduce the efficiency and the output voltage. It's the inductive kick-back which is giving a higher voltage on the output, than what would be expected by just multiplying 12 by the turns ratio.

Look at the data sheet for the 2SK2765. It's rated to 800V and is more than capable of handling 50V pulses.
http://pdf.datasheetcatalog.com/datasheet/fuji/2SK2765-01.pdf

You could probably get better results using a MOSFET with a lower on resistance and voltage rating. If the highest voltage is 50V, then a 100V MOSFET will do.

[SilverWingedSeraph]

Thanks for the additional information, I'll look into that.

I'm still wondering how to remove the DC bias from the signal, if that's possible?

Maybe a cap or some thing like that would remove the dc offset and give u the pure AC change.

This is exactly what I tried - a capacitor in series. Putting it in series with the 555 output works, so I know the value is right (47 uF), but putting it between +12V and the transformer or between the transformer and the MOS FET or between the MOS FET and ground all failed to remove the DC bias - in fact, the current doesn't flow at all.

Don't worry about it. Adding a diode would reduce the efficiency and the output voltage. It's the inductive kick-back which is giving a higher voltage on the output, than what would be expected by just multiplying 12 by the turns ratio.

Look at the data sheet for the 2SK2765. It's rated to 800V and is more than capable of handling 50V pulses.
http://pdf.datasheetcatalog.com/datasheet/fuji/2SK2765-01.pdf

You could probably get better results using a MOSFET with a lower on resistance and voltage rating. If the highest voltage is 50V, then a 100V MOSFET will do.

Thank you for the reassurance. Unfortunately, I'm using scavenged parts for this, and I got the MOS FET from an old audio amp. It's the best I've got at the moment.

[highvoltpower]

Up to voltage 500 KV, normally single unit of high voltage transformer is employed except for higher voltage evaluation more than one transformer are connected in cascade to provide needed high voltage. Essentially for receiving high voltage, one electrical transformer is to be very huge in size that is not at all economical.