Diode slection question

Regarding this flyback/ignition coil driver circuit again:

**broken link removed**

When I replace the ignition coil with a tv flyback transformer, clearly it is generating kick back current/voltage that is exceeding the power rating of 4 x 75V zener diodes.

Does anyone have some other suggestions? Zener diodes of more than 5W rating don't seem easy to come by.

Or could you do something like "The Zener Diode Regulator" according to this site: https://www.electronics-tutorials.ws/diode/diode_7.html, to limit current through the zeners enough to keep power dissipation within their limits?

TV flyback transformers are designed to have voltages up to maybe 1KV on their primary sides and they should be driven at a frequency of around 15 - 16KHz to be efficient.

The Zener Diode regulator circuit is not suitable because it is designed to prevent too much current flowing through the Zeners, in the Jacob's Ladder circuit , you want them to draw as much current as possible so they clip the voltage across the transistor and prevent it rising too high.

I would suggest your first change would be to speed up the 555 frequency so it drives the flyback transformer at a more suitable frequency. At the moment it should oscillate at around 80Hz, you want about 15500Hz so try dropping the 330nF capacitor to about 1.5nF to increase the frequency.

Brian.

betwixt said:

TV flyback transformers are designed to have voltages up to maybe 1KV on their primary sides and they should be driven at a frequency of around 15 - 16KHz to be efficient.

The Zener Diode regulator circuit is not suitable because it is designed to prevent too much current flowing through the Zeners, in the Jacob's Ladder circuit , you want them to draw as much current as possible so they clip the voltage across the transistor and prevent it rising too high.

I would suggest your first change would be to speed up the 555 frequency so it drives the flyback transformer at a more suitable frequency. At the moment it should oscillate at around 80Hz, you want about 15500Hz so try dropping the 330nF capacitor to about 1.5nF to increase the frequency.

Brian.

Click to expand...

Sorry betwixt, perhaps I should not have been lazy and posted the actual circuit I am using. The one above/below was easier to link and has essentially the same layout.

Duty cycle is 90% and I have found that, with my own primary winding on the exposed part of the ferrite core, I can get the frequency up to 68nF timing capacitor/between 19 and 20 kHz.


I presume this is what has caused the zeners to suddenly fry as I have increased the efficiency of the flyback substantially and as a result getting much larger kick back that is beyond the power rating of my zener snubber.

I have adopted ths zener snubber from a purchased jacobs ladder kit as per the previously posted spec were the duty cycle is only about 10% and 75Hz frequency. And clearly I need to modify it accordingly to dissipate the larger kick back.

It appears I may be able to put a resistor in series with the zeners to limit the current and therefore the power dissipation. Along the lines of the zener voltage regulator in here: https://www.electronics-tutorials.ws/diode/diode_7.html Would you suggest that this is indeed an option?

The only other solution I can think of is to have use 20 x 15V zeners in series instead, so that the clamp voltage is still 300V but the power rating of the series is 20x5W = 100W.

I suppose I should try puting my mutimeter across the mosfet and try and get some idea of what the kick back voltage really is - it can't be more than the 1000VDC max range of my multimeter.

When the primary side has that much overvoltage, at that many amps...

Then maybe it's time to install a spark gap on the *primary* side too!

BradtheRad said:

When the primary side has that much overvoltage, at that many amps...

Then maybe it's time to install a spark gap on the *primary* side too!

Click to expand...

Mmmm! Had not thought of that angle.

But at most likely signficantly less than a kV of over voltage, it would have to be a very tiny spark gap. I doubt that I could acheive the required prescision with a couple of bolts similar to what you would use with a tesla coil.

How would you suggest I produce such a spark gap? Are there specific components available with spark gaps of specific voltage ratings?

BradtheRad said:

When the primary side has that much overvoltage, at that many amps...

Then maybe it's time to install a spark gap on the *primary* side too!

Click to expand...

Surely these would do the trick: **broken link removed**

600W peak power dissipation would have to be enough surely. There is also a 1500W version.

Be careful, if I read your second schematic correctly you have an IRF631 as the driver, it is only rated to 150V. The Zeners must clamp the voltage below that.
In this circuit it is more normal to use a very high voltage transistor of the type designed for TV horizontal deflection. Typically these have a maximum Vce or Vds of 1KV or more and you would not use the Zeners at all.

Brian.

boylesg said:

Mmmm! Had not thought of that angle.

But at most likely signficantly less than a kV of over voltage, it would have to be a very tiny spark gap. I doubt that I could acheive the required prescision with a couple of bolts similar to what you would use with a tesla coil.

How would you suggest I produce such a spark gap? Are there specific components available with spark gaps of specific voltage ratings?

Click to expand...

Just two wires with the ends barely separated. It might fire at 100 or 200V. Just for the sake of experimentation to protect other components. There seems to be a small selection of protective devices rated to absorb the spikes.

But absorbing the spikes absorbs energy at the moment when you need it to create a spark between the ladder wires.

I have not built a Jacob's ladder so I don't know exactly what forces are at work. What I have noticed in animated simulations of a flyback converter, is that a load should be on the secondary at all times.

The reason is because a major portion of the current flows (or wants to flow) in the secondary circuit during the switch-Off cycle. This is counter to what we might expect. But that is when the flux field is collapsing.

It's the identical principle as in an old-fashioned ignition system. The spark plug fires when the points open, not when they close. (It seemed incredible to me when I learned about it.)

So I believe it is during times when there is **no** current flowing across your ladder wires (in other words when there is no load), that the high voltage spikes (and heat) are generated in your primary circuit.

The problem might be reduced by moving the lower ends of your ladder wires closer together, so the spark can start to jump more easily, and give the primary an outlet for all that energy.

You may even be able to add a second Jacob's ladder.

betwixt said:

Be careful, if I read your second schematic correctly you have an IRF631 as the driver, it is only rated to 150V. The Zeners must clamp the voltage below that.
In this circuit it is more normal to use a very high voltage transistor of the type designed for TV horizontal deflection. Typically these have a maximum Vce or Vds of 1KV or more and you would not use the Zeners at all.

Brian.

Click to expand...

Ignore the IRF631. It is only in the schematic because the actual mosfet I am using is not available in this simulation software. The one I am using at present has a voltage rating of 500V.