Page 1 sur 1

Software PWM, and proportional cuts

Posté : mar. sept. 22, 2009 2:56 am
par Simeon
Hi Jerome,
My board doesn't have wire heat control built in (early Hobby CNC), and the usual PIC solution seems a little excessive when we already output square waves from the software to the board. My board didn't have timer either, but that is a simple 555 timer circuit.

Would it be possible to have Jedicut output a PWM signal instead of the existing heat signal (I'm not sure what it is at the moment)? That way, all we need to do is connect an inexpensive power transistor between the supply and bow, and feed the PWM from Jedicut to the transistor input. No need for heat aquisition as it is set from inside the software already.

I have manual heat control at the moment, but it would be much better for to use your dynamic programming I think.

I am just starting to use Jedicut, but so far I like it a lot!

Regards,
Simeon.

Re: Software PWM, and proportional cuts

Posté : mar. sept. 22, 2009 12:52 pm
par Jerome
Hello !

I think it's possible but the signal will not be precise enough... What frequency do you need ? Can you describe the signals you need to receive ?

Re: Software PWM, and proportional cuts

Posté : mer. sept. 23, 2009 3:18 am
par Simeon
Hi Jerome,

The transistor I have looked up has a complete cycle time (off, rise, on, fall, off) of 190ns. (BUZ-71) It can handle 17A at 50V which is more than enough for cutting model wings.

What I have in mind is a PWM signal which is ON for a proportion of it's phase (between 0% and 100%). This will cycle the power to the bow very quickly with the result that the wire will be at 0%-100% of it's maximum heat for the supplied power.

The transistor can respond to signals up to around 5 MHz, but I would think that using the timer input as a carrier would be more than sufficient (4KHz) and this would allow a wider range of transistors to be used around the world (and may make your coding easier!). My bow currently runs a 50Hz A/C current without any visible marks to the foam, so probably 50Hz or faster signal is fine.

Thanks for your thoughts.
Simeon.

Re: Software PWM, and proportional cuts

Posté : jeu. sept. 24, 2009 2:05 am
par Simeon
Actually, the Buz 71 is not a good choice as it is a FET, and won't do quite what I need, but the principle is identical for a bi-polar transistor.