Here is the revised version, although I still don't understand what is wrong around the 10R. As far as I know, Vcc and the 12V shown in the link image are the same thing.

Do you have any sense of its maximum potential power output?

 

The transformer is powered directly from Vcc
494' is fed thru 10 ohm resistor.
GND is missing on the top Mosfet.

The circuits around 494' should be reviewed, compared to the datasheet, I will look at it tomorrow or the day after

 

The transformer is powered directly from Vcc
494' is fed thru 10 ohm resistor.
GND is missing on the top Mosfet.

The circuits around 494' should be reviewed, compared to the datasheet, I will look at it tomorrow or the day after

I finally got my head in gear so this should be correct. I will also look over a datasheet and see if anything stands out. I'm going for the TL494CN, assuming it's readily available.

 

You need PNP transistors instead of the BD139's, with their emitters connected to the emitters of the BC337's, for those drivers to work correctly!

 

You need PNP transistors instead of the BD139's, with their emitters connected to the emitters of the BC337's, for those drivers to work correctly!

Thats right.

Another mistake on the circuit diagram. The link I provided showed PNP and NPN transistors, BC337and BD140, (not BC327)

 

Thats right.

Another mistake on the circuit diagram. The link I provided showed PNP and NPN transistors, BC337and BD140, (not BC327)

So both top and bottom pairs should be BC337 and BD140?



Thank you sarahMCML for pointing out the error.

Is it worth adding any over-current protection to this circuit?

 

And here is the first draft of the PCB schematic. I have added a 10A fuse on the supply to prevent any FET runaway.

 

In the transformer you have chosen, the primary and secondary windings are placed quite far apart from each other.This means that there is a large leakage inductance in the transformer.
In case the secondary winding is shorted, the current in the primary should not be very high.
Imho, overcurrent protection is not mandatory in this case.

Another error in the circuit diagram. Pin 6 .Wrong combination of resistor and potentiometer, because if you turn the potentiometer all the way, pin 6 will be shorted to GND.
Replace these two components with a single resistor, 6.8 kOhm

 

In the transformer you have chosen, the primary and secondary windings are placed quite far apart from each other.This means that there is a large leakage inductance in the transformer.
In case the secondary winding is shorted, the current in the primary should not be very high.
Imho, overcurrent protection is not mandatory in this case.

Another error in the circuit diagram. Pin 6 .Wrong combination of resistor and potentiometer, because if you turn the potentiometer all the way, pin 6 will be shorted to GND.
Replace these two components with a single resistor, 6.8 kOhm

I’m not sure of the actual spacing but in case of a fault with the FETs then it makes me more comfortable to have a fuse.

Will check out pin 6.

 

PS.

By the way, please note that the Mosfets will be mounted on heatsinks .And you don't want to use an IRFZ44 transistor?

 

Yes, I will add heatsinks later when I start the PCB itself from the schematic. Why are the IRFZ44s a good choice?

Pin 6 should be ok now.

 

It has been suggested by a friend that it would work even better (faster and cleaner switching) if I used a dedicated FET driver, such as the IR2110, instead of the pairs of transistors. So I will post later today a draft of that circuit for any comment.

 

Here is the revised circuit using a dual FET driver. However, I am not confident of some of the IR2110's output connections.

 

That wont work, that's the arrangement for a lo-side/hi-side driver.. Remove the diode and the capacitor between pins 5 & 6. Connect VS to ground and VB to VCC.

 

That wont work, that's the arrangement for a lo-side/hi-side driver.. Remove the diode and the capacitor between pins 5 & 6. Connect VS to ground and VB to VCC.

Like this?



and the schematic:

 

With no current sense on the MOSFETs how do you plan to control drain current at higher duty cycles? Why the IRFZ44N? Do you have any idea on the transformer?

 

With no current sense on the MOSFETs how do you plan to control drain current at higher duty cycles? Why the IRFZ44N? Do you have any idea on the transformer?

I plan to use this flyback transformer:
https://www.hv-experimental.com/product/flyback-transformer-ac

I was assuming that current drain was determined by the transformer impedance up to the limit of the circuit for which I have included a 10A fuse. The IRFZ44 seems as good as any. What would you suggest otherwise?

 

I was assuming that current drain was determined by the transformer impedance up to the limit of the circuit for which I have included a 10A fuse

Well sort of, depends on what frequency/pulse width you drive it with..and the actual inductance of the winding. Its DC resistance will be a few milliOhms - if you're using a 12v car battery which can supply 10's of amps the MOSFET will fry in under 1mS - long before the fuse will blow. A 'fast' fuse can take >0.1sec to blow at 10x overload - they are intended to protect cables not MOSFETs. Pin 16 to error amp 2 +ve input is duty cycle control for current limit.

Have you any idea of the number of turns on the primary (each side) and the resulting inductance?

What heat-sinks are you planning to use?

 

Well sort of, depends on what frequency/pulse width you drive it with..and the actual inductance of the winding. Its DC resistance will be a few milliOhms - if you're using a 12v car battery which can supply 10's of amps the MOSFET will fry in under 1mS - long before the fuse will blow. A 'fast' fuse can take >0.1sec to blow at 10x overload - they are intended to protect cables not MOSFETs. Pin 16 to error amp 2 +ve input is duty cycle control for current limit.

Have you any idea of the number of turns on the primary (each side) and the resulting inductance?

What heat-sinks are you planning to use?

The transformer secondary has 2400 turns on it, and each of the two halves of the primary might have 50 turns (wound in the same direction) and so an educated guess for each half of the primary might be 25 - 50uH.

Here is my suggestion for current sensing with the additions in blue, so please correct as required. I'm not sure if a high-sided position of the sensing resistor would be better.


I plan to use this heatsink: https://uk.farnell.com/ohmite/wa-t220-101e/heatsink-to-220-black/dp/2097668 of which I have several left.

Can you also confirm that the secondary output will be as shown here, rather than one lead being grounded and the other oscillating HV:



Thank you.