Picaxe or adruino programming

[Floris]

Hi guys,

I thought I gave this topic a kick, it's been a while since the last post.
Ok so I am working on a large cap dump setup for the secondary, I want to be able to pulse high capacity caps to the battery, say from 10.000uf to 100.000uf. Been trying to work it out with a 555/opto/2n3440/mj15024. I have been unable to get good switching of the MJ15024 and I'm not really sure why, although it seems the trigger signal to base is to soft I think. So far switching is more like a ramp than a quick discharge.
Then I remembered I had good fun with an arduino board as PWM to control a opto/2n3440/2n6509. With a pot I could easily adjust the frequency of the dump. That worked quite well, but since then I have scaled things up and so things can become more potent, it could be easy to end up frying stuff. I think I will try hooking the pwm signal of the arduino to the base of the 2n3440 which then hopefully drives the MJ15024. If this works it may prove a nice platform for experimenting with cap dumping small or large. Will let you know if progress hits.

One question I have though is this: from the 555 setup you usually get a 50% duty cycle which drives the opto and thus the dumping of the cap. I sort of remember that JB also said that that was our goal to have 50% duty cycle. But does that make sense? I'm trying to understand why you would want a high current impulse followed by spikes. This way we are not converting all energy by way of the cap, since as the transistor is on the cap negative and batt neg are connected and the spikes come through. I may have missed parts here but I have adjusted dump time to 5ms to limit that and get earlier charging of the cap. So very short ontime to only dump the cap then offtime to charge the cap to desired value or time. This is NOT JB's way but appears to charge better?

The frequency of the dump btw can be from 1/2Hz or lower to probably 500Hz but most pratical was around 40Hz. My best charging was seen with more frequent low voltage dumps, but have yet to do the cycles with that.
look forward to your experiences,

floris

[bluestix]

Then I remembered I had good fun with an arduino board as PWM to control a opto/2n3440/2n6509.


floris

I have an 8 circuit solid state SSG connected to a 30v 50000uf cap. I use a voltage sensor on the cap to stop the ssg pulses and trigger the cap dump. The dump just uses one NPN triggered by the arduino. It dumps down to the secondary battery voltage then switches back to charging mode.

I have read that this will burnout the NPN but so far it works fine for me. The cap takes longer then I expected to discharge.

I haven't tested that much yet.

I will try to make a video tomorrow so you can see what I am talking about. I will post it here when I do.


I am also going to do some tests with charging a 16V 150F ultracapacitor. I am not going to dump that to a battery though because it would probably be dangerous and melt the wires.

I tried using one of the ultracaps in a car audio system but it has such low resistance that it looks like a dead short in the circuit until it is charged. I took it out because I didn't want to do the extra work to make it safe to use in a car. I will just go dual battery with larger alternator instead. My headlights only dim when the bass is really loud anyway so it's not really a problem. It makes them pulse to the beat. Lol.