12 volt DC to 400 volt dc converter design help needed

[speedshop]

I'm looking to build a replacement ignition for my motorcycle. It has a capacitor discharge unit that is powered by a coil in the generator.

I want to use a DC to DC converter to charge the capacitor to 400 volts instead of the 110 volts it normally runs at. The capacitor in the standard unit has on it 4k4, which I guess is a 4.4uF item. The capacitor will be charged by the converter and the discharge controlled by a SCR whose gate gets its signal from the timing side (I'm using the signal from the standard generator for now - changing this is another project, I want to get the cap charging sorted first!)

I have had some success building a direct replacement for the standard ignition box powered by the coils in the generator. My unit runs fine using a 1uF capacitor, but the spark is noticeably weaker. I think I will opt for a pair of 1.5uf capacitors in parallel in the DC powered unit.

Question is can i buy off the shelf parts to build a converter (i don't want to get in to winding inductors)?

The power supply must be be stable and relaible, so must be well within its design spec and mustn't need fiddling or contanst messing about with to keep it running. Robust is components are good as they will be exposed to a high level of vibration.

Can I pot the unit without problems from heat?

Also the inverter should run from about 7 volts to 18 volts (if possible - if not as wide an input voltage as poss).

Any pointers or resources i can look at?


Thanks for your help

 

[speedshop]

This is a circuit I've been looking at, but it doesn't give the required 400 volts, maximum is 300.

 

[GrandAlf]

Not sure what you are trying to achieve here, are you trying to get a shorter duration spark for timing accuracy, if so this may result in an incomplete burn. It is very likely that the manufacturer spent a great deal of time and money getting this just right to suit the bike, it seems very unlikely that you could improve matters. Remember all the ignition components would have to be much higher rated and it is not probable that you would get it anything like as reliable. Sorry I am not more positive, but having messed about with various ignition systems in my youth, I concluded long ago that it was a waste of time. By all means try, you may prove me wrong!

 

[macer_0001]

Have you tried adjusting the potentiometer? If you already did it, change the value of the potentiometer. use 50kohm. Then gradually adjust the potentiometer, until you reach the desire output. Also change the capacity of 1uF. Replace the capacity to 450volts.

 

[speedshop]

Ok this is Cdi system system so the spark duration is already short, but hot. The only way you can get more out of a Cdi system is to increase the voltage that you charge the capacitor to (according to research carried out by Bosch).

The higher the voltage the smaller coils you can use so you get a weight saving (I know of one system runing at 650 volts). The capacitor being charged to 400 volts is stage 1. The next stage will be to convert it to digital programable timing as meth needs far more advance then petrol.

I know exactly what I need the ignition to do, I just need to work out a reliable high power boost converter, build it and get it running reliably first. I'll add to it as I go. But the plan is for it to go on a methanol burning engine (hence in the future I will need scope to adjust the timing beyound the range of the standard set up)

I'll post some pictures of stuff I've built, I've got 120 mph form a 30 year old 500cc single running on the CDi ignition I made. (Timed on an official run!)


--I haven't tried adjusting the pot, but will do. Thank you

 

[GrandAlf]

Sorry about the previous post, I had assumed that you were trying to mess with your road bike to make it a bit faster, you are obviously clued up about this. Not really up to date on this sort of thing anymore. I will look and see if I still have any old drawings, but I think probably you may get a better reply from someone with more recent experience.

 

[speedshop]

Its an interesting point you make about the ignitions being over rated. I know of at least two mass producers of automotive ignition systems who use 3 amp diodes, but 1/4 watt resistors in the same part of the cicuit!

I use .6 watt resistors and 1 amp diodes and have pushed the system well over what the factory intended with no problems. The diodes, scr and the cap are 100% over rated.

I'll be using a 630 v cap on the 400v dc to set up

 

[xaccto]

i've mucked with cdi on a turbo engine in past.

how many cylinders is it firing ? 1 or 2 may be different case, but any more I think trying to go to max voltage like 400-600V is just asking for trouble and no gain.

You can make the ignition pulse a bit longer by including a small inductance in series with the SCR, this will limit the di/dt ( a good idea for SCR anyway)

I think you will have trouble satisfying your 7V supply requirement with the UC3845 too.

Have you thought about multiple sparks per ignition ?

might have a schematic for you...just have to find it.


also

https://www.siliconchip.com.au/cms/A_110499/article.html

siliconchip also have another one, just can't find the article, which is multiple sparks and a very good circuit (tho I haven't built it) only drawback of the project is they made no attempt to make it physically small - easier for kit building I suppose.

 

[speedshop]

I'm going to be using it on 1 and 2 cylinder engines.

400 volts seems to be the the best comprimise, going above this does give some gains, but also causes more problems too.