Don't use pullup or down resistors, they are set enable into the pic.
Do you set the fuses like the HOWTO from Charlie ?
I think the copy past is the problem.
Copy the code into a real txt-file, no notepad or something like this.
Open a DOS-sesion, typ *edit code.txt* without the *'s, past the code into this file, edit the code and safe this file somewhere.
Now you have a real file with the code ono to one.

Hi Rallyfinnen,

Open Micro C and goto project /edit project and make sure the fuses are set correctly.

I guess that fuses is the same as config bits?

I have two sets of instructions (one old and one new?) and Master Clear Enable is ON in one, and OFF in the other. I tried both settings, which is the correct one?
The other settings seem to be the same.

Now I have also checked the strobe output (not using that one either, so it's not connected), and that one seems to be dead too.
I have two PIC's, so I have tried both of them just in case.

5V is present on pins 5 / 14, and I have pulse on pin 7, that should be enough to get the output on pin 6, right?

If I use the test function in the code, does it enter test mode any time, or only at start up?

It should not be a copy/paste problem, since the downloaded hex is not working either.
I will remove the pull down resistor again, just from looking at the data sheet it seemed like it was a pull-up output, so I thought thet it was needed.

If the hex isn't work too, the hardware is wrong / broken.
The fuses / configbits are set into the hex when it was create.
MCLR = off
I make two HEX-files for you ready to program
File 1 - incl testfunction
File 2 - excl testfuntion

www.test.gompy.net/rallyfinnen.zip

Incredible support guys!! Thank you!

I got my first output signal! Now i will play around compiling my own files to see if it works too!

Wat do you have todo.
Start the exel curve generator and draw your curve.
Copy past the code into MicroC and now you can edit the code.
(DON'T SHUTDOWN EXEL BEFOR YOU HAVE PAST THE CODE)
Edit only the line; Delay_us(1650); // Calculated delay
Don't forget to set the fuses into the configfile as scribe into the HOWTO.

The files I compile don't work for some reason. Only the hex-files you made before..

I'm not sure i have the right instruction, because the instruction says MCLR = on..
Maybe there is something else wrong too?

(picture removed)

This is how it has to be set.

Please resize your picture or make it a attachment, it far to big to see the picture on screen

Ok, thank you for the confirmation!
I had the brown out active, but other than that i had the correct settings.
Maybe the brown out can be active since the processor is running on batteries?

Now, for some reason, I get an error when compiling, even with the example c-file..

I think I will take some time off this project for a while, I get too frustrated banging my head against tha wall..It's really testing the patience I don't have

In 2005 we (JMJ) start the CDI-topic, now 5 years later we still working on it and you will quit within 2 days ??????????
I don't think so mister, your project isn't ready yet - you can't quit

http://www.test.gompy.net/rallyfinnen2.zip
This file contain the complete project.
Unzip it in some directory, set the project in MicroC, open the project and everything has to work.

Hi Rallyfinnen,

I know Mikro C can be finicky to use. Post your code here maybe we can help you sort it.

Thank you again for your support, you're the best !

I'm definitely not giving up, I was just going to give it a rest for a while and do other stuff to gather my thoughts (and calm my nerves )
I've got a lot of projects in the pipeline, so I figured I should switch to something else in between.

Bigboat, I will download the complete package and give it another go tomorrow! I suddenly feel inspired again

Rob...
I finally figured out the trigger problem with cdipwr-4 board.. The hall resistor was in the wrong place..has to be before the cap on the trigger pulse input trace to reference the signal to ground. Made the changes to the printout I did and added what I changed to get it to work. Also figured out the diodes D-2, D-3, D-4, D-5 and placed the corrections on this sheet.
Thanks again for your constant support with my problems, sorry for the frustration I caused with my board not working...
It now functions great! In fact I built 4 more of them.. Confidence is high right now on a completed project!
Rallyfinnen has the right idea, you are the one I look to for support!

John

Now the PIC worked with the code I compiled

I started to adapt it to the engine now. Using the built in magnet to trig the hall sensor, but i need to make a bracket for it etc.

Thank you again for the help!

John,

With those winding values what voltage are you getting at C4 and Ground?

A specialy for John, rest on the webpage.

Charlie, with the windings I am using, I get approximately 450 volts at C-4. My digital VOM jumps all over the place but if I am not in test mode, a solid 450 volts. I am considering going to a 600 volt .47uf cap. Physical size limitations however prevent me from doing this right now..

Rob, thanks for making the revisions and additions to the PCB and component overlays. It looks foolproof now..Well, for this old fool anyway! Ha!..
Since I made the changes on mine, it has been firing non-stop for 16 hours on my bench supply without a problem at 7000+ rpms. I regulated the voltage down to 5.5 volts for the test..Actually on a battery pack, it will still operate at a voltage of 4.4 volts.. then cuts off... Nice! Lets you know that the battery needs recharging for sure.

Charlie, I wound some toroids to 450 turns and get 450 volts plus..higher and the caps blow..not immediately but eventually..so I would say, between 400 and 450 is about right.

John

Wat have you do wrong ?

Your not ready yet, we meet again - don't where - don't no when.....but we will be there

Charlie, with the coil connected loading the circuit, not firing, both jumpers on the timer board, I get only 166 volts at the cap..sufficient voltage to fire my efficient CDI coil at a gap of 3/16 inch at 7000 rpms.
What figures are you getting?

John

I put a new exelsheet online to test.
http://www.electronics.gompy.net/cditci/cditest.xls

Hi John,

one of my setups is 20 on primary,3 turns on the feedback and 245 on the secondary. The voltages I get around 408 volts. just wondering sense you have 400 turns on the secondary.

It depent of the ferrit wat Voltage you get from the transformer.
I think I've try 20 or more ferrits with all the same size and the same distance of wire.
All ferrits are given a differend Voltage and some ferrit don't ocsillate at all.
There are HF-ferrirs, LF-ferrits and everthing between it.
Ferrits from powerwires don't do anything and ferrits from switchpowersupplys working great.
In the topic from JMJ you can read the tests I've make.

Same with e-cores, they don't work all.

Charlie, if you noticed I am using much thinner wire for my secondary windings...#36 and it only takes about 21 feet of it to complete 450 turns.
My primary is #26 and evenly spaced..secondary is even the first time around the core, then kind of scatter wound till I hit 450..really works well for me..tried evenly wound 450 and gained nothing, so in hopes of repeatability, I do it the same for every one of them now.
I also solder insulated leads on the ends to prevent breakage, wrapping them with tape a few layers around the core. Haven't had any shake loose even when mounted beneath the engine block. Either way, the newest board Rob presented now is underway..etching and will drill and mount components to at least one tonight sometime. Testing tomorrow.

John

John,

do you think there is a advantage to using toroids over e cores?

I think there was a formula somewhere on the forum about the efficiency of toroids..maybe in the range of 90% +.
Ecores, maybe they are 80%..not sure but I know I never had luck with ecores..besides being bigger in size and easier to wind. I have a few I got out of the flourescent light bulbs that Brett posted a link for..Or was it Rob.. whoever it was...anyway, they are alot smaller and you'd have to use really fine wire for the secondary or getting the 400-450 turns would be a real chore. Actually I use a 1/8 inch x 1 foot length of hardwood dowel notched in a v on both ends for a shuttle for the toroids..makes it a breeze to wind em.. 20 minutes tops and I have 450 on one.
Tonight I have finished 2 of them while watching TV.. Ready to wind the primaries..

I'll have to look up that rating for the toroids.. we can compare notes and see what is better to use.. Beginners would have a tough time with em.. so maybe E-cores are best for most..

John