AMX Test Results

[ivconic]

I received the pcb that I was drawing, "through hole" with the IR driver. But I don't get to solder the components.
It's a lot of work to launch a flock into the stratosphere!

Last night's position above my place, they are moving towards Bulgaria.

Joking aside; I'm a bit disoriented with what you wrote last time because of the use of SIC transistors and diodes.
I was planning to put the IRF740 to start with.
Now where can I find SIC transistors?
There are some, but they are not cheap.
The whole story is slowly getting more and more expensive.
And I didn't understand the reason why you switched from 2v power supply for TX to higher voltages?
If the mosfets do not behave at 2v, then some logic level is better, conditionally up to 500-600v.
Unfortunately, all the logic levels I have are "high current" but with low voltages.

[ivconic]

As I write this, I realize that I made a mistake in my assessment,
with the use of IR drivers, "non logic level" mosfets should also be driven without problems.
So... I still have to solder the components and try it out first, and then see what it turns out to be.
You say you burned a couple of IRFs. Is it because of the tilt? How do you prevent them from going into an avalanche?
I guess the snubber is not a good solution?

[Carl-NC]

My first test was with IRF740s and that worked fine. Use those. To prevent avalanche, add some Cextra to reduce the flyback. I'm not entirely sure why the supply voltage was so high, but it's partly due to the high coil resistance.

[ivconic]

My first test was with IRF740s and that worked fine. Use those. To prevent avalanche, add some Cextra to reduce the flyback. I'm not entirely sure why the supply voltage was so high, but it's partly due to the high coil resistance.

"S" ? I never saw such one, or haven't payed attention.
I can get this one only:
https://www.mgelectronic.rs/tranzistor-irf-740

[Carl-NC]

IRF740s, plural of IRF740

[ivconic]

Ah so!

[ivconic]

... I'm not entirely sure why the supply voltage was so high, but it's partly due to the high coil resistance.

At 25kHz pps one would expect the coil shall be "fast", small inductance and small resistance is somehow expected.
However; due to fast transitions; impedance can be higher than anticipated.
I missed the talks and opinions on coil, if there was any on past topics?
But I would try with coil made of pvc coated litz wire, probably around 20 turns or less than that on 20cm diammeter.
Yet, with such specs I would expect high currents and probable issues with mosfet heathing.
That's why I would probably "secure" it with some serial resistance. Which is not best possible idea at all.
Cextra value is a bit intrigue for me, how to calculate the proper value?
Any suggestions?

[Carl-NC]

I would consider coil resistance to be the most critical as it set power consumption. Litz wire is probably necessary if you want to look at the first few microseconds of delay.

You find Cextra by treating the LC as a sinusoidal resonant circuit. You know the peak current, so pick a desired peak voltage and figure out what frequency gives you that. Then extract Cextra from that.

[ivconic]

[ivconic]

After much procrastination and delay... I found some time to work on this a bit.
The topic kind of died... but when I already have a pcb; I think it's not bad to try the concept.
I haven't decided on mosfets yet.
For now I just set up and checked the power supply. I have 2, 5 and 10V, stable and regulated.
The schematic and pcb do not match in the 5v regulator part.
The schematic shows LM317, but in the meantime I changed my mind and put 7805.

[ivconic]

My m.d. technology challenged fellas... what's new?
Anybody alive?

[Carl-NC]

No time to work on anything right now. Too many summer projects.