So this is going to be a school project. I am planning on using a TL494 to modulate the audio and drive an IRF260 MOSFET, and a flyback to provide the step up to the HV out. I was thinking about 24 to 30 volts into the primary of the flyback at about 30 to 50 kHz. (I might remove the ferrite transformer core)
Here's the question to start:
Can a switching power supply be used in this application? The DS resistance when the MOSFET is on is only .04 ohms. That's a lot of current even assuming there is a 1 ohm resistance in the primary of the flyback transformer. If I were to use a switching supply tht supplies less than this current, won't the power out be shut off by the protection circuits in a switching supply? What happens if I use a regular transformer/rectifier with less than the requires output current?
This is all hypothetical. I would use an appropriately large power supply, I just want to discuss really how each would respond to high frequency, high current demands.
Here's the question to start:
Can a switching power supply be used in this application? The DS resistance when the MOSFET is on is only .04 ohms. That's a lot of current even assuming there is a 1 ohm resistance in the primary of the flyback transformer. If I were to use a switching supply tht supplies less than this current, won't the power out be shut off by the protection circuits in a switching supply? What happens if I use a regular transformer/rectifier with less than the requires output current?
This is all hypothetical. I would use an appropriately large power supply, I just want to discuss really how each would respond to high frequency, high current demands.
Why would you want to remove the transformer core?
Use a supply that has an adequate amount of current for your design.
Not having enough current will limit your power output due to the overall load impedance of the switching stage and give you poor results
Using a linear supply will eliminate any/all issues you may have the using a switching supply.
But that doesn't say that a switching supply won't work.
Yes, your RDSon is very low, But the inductance of the primary vs the switching frequency is what will determine part of your load lmpedance for the switching FET section.
And as you start to draw an arc the supply current will go up quite considerably.
The more turns you use for the primary this raises the inductance and the Impedance, although using more turns on the primary does lower your turns ratio as well.
As you can see in these photos I used only 2 turns per FET on the output (green) core of my Variable HV supply.
That is a 22ga wire length of only about 6 inches and the switching frequency is about 180Khz using some IRFZ42's that has an RDSon of .032ohms.
http://www.diyaudio.com/forums/plan...tor-insulation-mylar-coating.html#post2773108
This supply draws about 3.8 to 4.5 amps at 13V to 18v or so under a full load drawing a continuous non stop ARC just to give you an idea.
It is more than what the main supply transformer is rated at and does overheat when it is heavily used.
It rarely sees that kind of continuous use, But is capable of it.
Make sure that you use a proper gate drive method as well.
The TL494 may be okay but a gate driver after it will yield better results.
The TL494 was originally designed to drive transistors (BJT's) and not FET's and its output drive current is rather limited.
Driving FET's with it also requires a different configuration as well, So save yourself the headache and use a proper buffer to drive the FET.
There are two things to look out for with the gate voltage,
1. Make sure that the Gate voltage is able to switch fully off to 0V and not partially!
You don't want the voltage to be floating above ground (0V) when it is supposed to be off!!!
This will cause the FET to not fully turn off and this will cause it to overheat and destroy itself and/or use more power than it should be using because it will be operating in its linear range instead of just a switching device.
2. Beware of overvoltage spikes on the gate as this will surely take out the FET's!!!
It is advisable to Do use a Zener Diode to protect the gate because of this!
I didn't do this with my HVsupply and it cost me about 8 blown FET's until I got the situation under control.
Since then it has been working reliably with no mishaps or blown FET's for the last two years.
At First I was using IRF510's and they are much weaker (lower power rating and higher RDSon)than the IRFZ42's but it was the gate over voltage that killed them.
Make sure that you use an FET that has a breakdown voltage of at least 5 times higher than that of your supply voltage.
Remember that this is an AC switching circuit and not a DC circuit even though it appears as such.
Hope that helps you.
jer 🙂
Use a supply that has an adequate amount of current for your design.
Not having enough current will limit your power output due to the overall load impedance of the switching stage and give you poor results
Using a linear supply will eliminate any/all issues you may have the using a switching supply.
But that doesn't say that a switching supply won't work.
Yes, your RDSon is very low, But the inductance of the primary vs the switching frequency is what will determine part of your load lmpedance for the switching FET section.
And as you start to draw an arc the supply current will go up quite considerably.
The more turns you use for the primary this raises the inductance and the Impedance, although using more turns on the primary does lower your turns ratio as well.
As you can see in these photos I used only 2 turns per FET on the output (green) core of my Variable HV supply.
That is a 22ga wire length of only about 6 inches and the switching frequency is about 180Khz using some IRFZ42's that has an RDSon of .032ohms.
http://www.diyaudio.com/forums/plan...tor-insulation-mylar-coating.html#post2773108
This supply draws about 3.8 to 4.5 amps at 13V to 18v or so under a full load drawing a continuous non stop ARC just to give you an idea.
It is more than what the main supply transformer is rated at and does overheat when it is heavily used.
It rarely sees that kind of continuous use, But is capable of it.
Make sure that you use a proper gate drive method as well.
The TL494 may be okay but a gate driver after it will yield better results.
The TL494 was originally designed to drive transistors (BJT's) and not FET's and its output drive current is rather limited.
Driving FET's with it also requires a different configuration as well, So save yourself the headache and use a proper buffer to drive the FET.
There are two things to look out for with the gate voltage,
1. Make sure that the Gate voltage is able to switch fully off to 0V and not partially!
You don't want the voltage to be floating above ground (0V) when it is supposed to be off!!!
This will cause the FET to not fully turn off and this will cause it to overheat and destroy itself and/or use more power than it should be using because it will be operating in its linear range instead of just a switching device.
2. Beware of overvoltage spikes on the gate as this will surely take out the FET's!!!
It is advisable to Do use a Zener Diode to protect the gate because of this!
I didn't do this with my HVsupply and it cost me about 8 blown FET's until I got the situation under control.
Since then it has been working reliably with no mishaps or blown FET's for the last two years.
At First I was using IRF510's and they are much weaker (lower power rating and higher RDSon)than the IRFZ42's but it was the gate over voltage that killed them.
Make sure that you use an FET that has a breakdown voltage of at least 5 times higher than that of your supply voltage.
Remember that this is an AC switching circuit and not a DC circuit even though it appears as such.
Hope that helps you.
jer 🙂
Last edited:
For Future Reference try to chose an FET that has a low gate capacitance as this will help to keep your gate drive current requirement at a minimum.
The gate capacitance of the IRF260 is about 4057pf or .00457uf.
You can use this calculator to roughly find your (drive) input impedance for your switching frequency.
http://www.electronics2000.co.uk/calc/reactance-calculator.php
At 30Khz this comes to 1.308k and at 50Khz it is 784.6 ohms.
This is not bad, But should you want to use a higher frequency the impedance goes down and the drive current requirement goes up by the same rate.
This is why some FET's aren't suitable for very high switching frequency's such as for RF use.
FWIW
jer 🙂
The gate capacitance of the IRF260 is about 4057pf or .00457uf.
You can use this calculator to roughly find your (drive) input impedance for your switching frequency.
http://www.electronics2000.co.uk/calc/reactance-calculator.php
At 30Khz this comes to 1.308k and at 50Khz it is 784.6 ohms.
This is not bad, But should you want to use a higher frequency the impedance goes down and the drive current requirement goes up by the same rate.
This is why some FET's aren't suitable for very high switching frequency's such as for RF use.
FWIW
jer 🙂
- Status
- Not open for further replies.