Glad to see this thread picking up some steam!

The coil in the pictures above looks a lot like the one John sent me. His coil is already potted in epoxy and seems to throw some very long killer sparks. I believe he has a good source for them.

@Gompy... Microchip's XC8 compiler seems to be working fine for me. I really don't have $249 to spend on the mikroC compiler, even though it looks pretty cool. I don't think the older free version of mikroC will support the 12F1840 chip I want to move to next. With the STM32 and Atmel chips I have worked with before the compiler choice was easy, GCC (free and open source). The closest thing to free seems to be XC8, and I have just kinda assumed that the company that makes the actual chips can also make the best compiler for those chips.

The XC8 compiler is a fairly large download, but it's free and easily available. Even pro mode will work for free for 60 days. I don't think pro mode is a big deal though, it only saved a couple % of program memory and RAM, and we have plenty to spare.

I looked for the source that your CDI was using, but couldn't find it. I had heard that there was trouble licensing it as open source.

Once the 12F686 code is nailed down I'll be happy to port it to the chip you want to use or help you get the 686 working in your application. The great thing about C is that it will be very easy to port to different chips as long as they have the same basic timer capabilities.

&gt;<span style="font-family: Verdana, Arial; font-size: 13px; background-color: rgb(251, 252, 255); ">Test with a pulsgenerator is not the same if testing with a real engine.</span>

For sure! I'm not even actually using a pulse generator. When I say "simulator" I mean the entire timer board circuit (and a pulse generator) is simulated on the computer. My problem at the moment is that I don't currently have a CDI engine working. My GF also won't let me run an engine in the house for some odd reason, and I haven't loaded my laptop with the compiler yet.


John told me that he got the "JVM Creation" error fixed using the fix I mentioned above. So hopefully the MPLAB X IDE problems are over with now!

@COM... Good to hear you're testing the code. The default curve I had set was just set at a flat 10 degrees across the board if I remember right. It's also entirely possible there'e an error in my math somewhere. Let me know what results you're getting as you do more testing.

I'm glad I rushed ahead and put out the beta code as it seems to be helping move things along!


-Jake