2
\$\begingroup\$

For a while, I used hand wind transformers to make high voltage (250v, then several stages of cockroft walton to 750-1000v) from ~12-24v at around 10khz and max 100ma on the primary side. But of course it's a lot of work to handwind transformers, and the end result is really far away from a compact design.

I need the high voltage to drive PMTs so it's fortunately very little current.

My goal is to make a transformer that uses max 2W of power, and I think I can get away whit just 1W. Due to this it doesn't matter if the transformer is lossy at a non loaded state, because it uses so little power.

One day, I found small smd transformers on aliexpress what's usually used in flash lamp circuits. It has a 1:25 turn ratio that would be perfect for my needs. But it's usually used in high current applications to charge a capacitor, for a couple of seconds. Because of this the primary has a 8.5 uH inductance which means I need to drive it several MHz if I want below 0.1 amp usage, so it gets impractical pretty fast.

Is there a good way to reduce the primary current in a way that I don't need to drive it that fast. Say drive it at 200kHz max, and add series resistor or inductor, or some current limiting feature. Can that work? Or is it simpler to just go another route, and make say a high voltage boost converter?

\$\endgroup\$
3
  • 1
    \$\begingroup\$ I Think NO, there is no goo way to reduce the power dissipation in the trasformer by adding a resistor in series. What happes is instead of dissipating the power in the transformer you would do it in the resistor. \$\endgroup\$
    – arnisz
    Commented May 24, 2021 at 8:46
  • \$\begingroup\$ It's late for me and I may not be reading things well. But have you looked at CCFL circuits? There's a lot of them and they operate with 12 V to 24 V input and generate AC output that from memory seems not far from what you seem to me to be discussing. Just want to make sure you've excluded the idea already. \$\endgroup\$
    – jonk
    Commented May 24, 2021 at 10:13
  • 1
    \$\begingroup\$ Low duty cycle high frequency PWM may meet the need. It's still a MHz + effective frequency pulse but low repition rate. Dumping a capacitor into the winding so energy transferred is limited to what is in the cap may work. This allows low repetition rate and no issues wit h short pulse widths. \$\endgroup\$
    – Russell McMahon ♦
    Commented May 24, 2021 at 10:30

2 Answers 2

Reset to default
3
\$\begingroup\$

That type of flash-cap charging transformer will invariably be a flyback.

You can reduce the power throughput, and hence the average current from the power supply, by lowering the pulse repetition rate. With a low primary inductance, the pulses will still need to be short. However, you only need to repeat them as often as is needed to get your power throughput. It's not a question of simply reducing the frequency to 200 kHz for instance, you need to define or even control the pulse length and repetition rate.

With a decent input capacitor, you might be drawing a high peak current from it during the charging pulse. However if the charging pulses are few and far between, the average current into the capacitor from your supply would be quite low.

\$\endgroup\$
1
\$\begingroup\$

You could try a parallel or series LC circuit running at or close to resonance by putting a capacitor in parallel or series with the primary of the transformer. However, you will need to simulate the circuit.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.