Cafe Racer Forum banner

Micro Controller Techno Geek needed

2436 Views 27 Replies 9 Participants Last post by  jbranson
Anyone here have decent experience programming and interfacing devices with Microcontrollers such as the "basic stamp"?

I've just got a little prototyping/programing board with a basicstamp...looking for someone with lots of experience playing with them. Seems fairly straightforward so far...just modified Basic Language. I'm going to eventually need help with the I/O. I'm looking to build a programable CDI ignition based on these chips or similar.
Besides not wanting to spend $1000+ on a store bought unit...most of them I've seen would not be suitable for my application anyway.

I want to intercept the trigger signal of the CDI unit that dumps the capacitor to the coil and delay it based on a rpm map.
JohnnyB
1 - 11 of 28 Posts
Your coffee machine probably has a micro-controller very similar to the PIC16C57 that I'm talking about.
It's not rocket science....if you can comprehend Ayn Rand you can do this.
JohnnyB
Look my bitches...this ain't no comedy shop!!

Where's the brains I ask you!

JohnnyB
Very similar to what I'm doing.
The programming isn't that big a deal..it's designing the hardware that I'll need help with. I can do things like program a microcontroller to move servo motors, light shit up...etc..based on an internal "map" of values. What I need is someone that knows both Microcontrollers and CDI ignitions. So I can mate the two of them.

The basics are...the rotor/stator charges a big capacitor...that holds the charge until the "brain" tells it to discharge into the main ignition coil...and from there it goes to the plugs. What I have to do is grab that signal from the brain, hold on to it a predetermined amount of time that is called from an rpm table....( like if the engine is turning 5,000 rpm the timing is 35 degrees...if it's 10,000 rpm it's 45 degrees). Which means I need an rpm input too.
The fundemental system is not really complicated....but hooking it up to the right components in the right place is where I need help. I mean if you hook up a MicroController the size of your little fingernail to the wrong wire in a 40,000 volt ignition system...well you can imagine that it would become a $50 little ball of glass real quick.
JohnnyB

The upside is....if I can perfect the system I'm looking a street bike grunt at 6,000 rpm...and max hp at 12,000 rpm.
See less See more
Oh snap...I can imagine.
I remember the first time I took my 97 Roadking in for servicing...I told the Harley "wrench" it was idling a little high...he took out a thing like a big calculator...pluged it into the harness, punched a few buttons and it slowed right down.
I can imagine that tweaking maps at the track could be some good money. Even on bikes 80% of the people with Power Commanders can load maps but have no real idea how to create or modify the ones loaded in the box.
If someone made a $500 programmable ignition compatible with my PVL I'd be ordering it right now.
In fact...I think I'll go look around before I get too deep in this.
JohnnyB
Yep, Rand gets a little tedious...but I'm a fan of Objectivism.

I with you 100% on kerouac....it's like...well...I guess you had to be there :)

JohnnyB
Rob,
Thanks, I'll give them a read.

JohnnyB
I've joined one of the yahoo groups dealing with the megajolt stuff...thinking maybe some of those brainaics can help.

If the PVL had a trigger like the mag/cdi on Mary's little bike it would be no problem...you just pick up the signal from there...delay it based on an rpm input and you are good to go.

The PVL has no trigger...so I'm not sure where to pick up the signal...or maybe it doesn't have one...maybe just the collapse of the stator field causes the capacitor to dump to the ignition coil.

It probably wouldn't be too hard to just build a new CDI box that plugs into the PVL wiring. The programming would be pretty simple...lots of math...but not much programming. Using something like:

LOOKUP index ( delay1, delay2, delay3) etc.

Where "index" is a variable calculated from rpm. The LOOKUP command would tell it to pick delay 1 or 2 or 3 etc. The index variable just tells LOOKUP where to jump to in the list to find the right delay value.

The microcontroller could probably handle the rpm sensing too...based on the frequency of the signals it's trying to delay. I'd probably build a timer circut using a 555 IC timer though and have that dump a byte to some inputs on the microcontroller so it could just read the rpm values without having to do the work itself. The BasicStamp doesn't like to do two things at once.

I think I can handle the microcontroller end...what I gotta find is someone REAL heavy in CDI design. I see lots of fancy schematics for CDI out there...but I just need a bare bones design.

I'm trying to avoid having a trigger pickup on the rotor...just another thing to have to mount, wire and adjust.

But....I got a feeling about this...I think our 175's would really like this. I'd love to have my timing at around 30-35 degrees when I'm grunting up a hill at 6000-7000 rpm...and then up to 47 degrees at 11,000 rpm. I bet you'd pick up 2-3 hp in the mid range. The engine just can't like 47 BTDC center when it's lugging WOT at 6,000 rpm.
JohnnyB

PS. What's weird is if I had mechanical advance...I could program the microcontroller to move the advance plate based on rpm with a servo motor and write the program in one night. Rube Goldberg would love it.


Edited by - jbranson on Feb 15 2006 03:17:16 AM
See less See more
Phase one is complete. I've built and tested a signal generator circut to simulate pulses from a crankshaft.

Uses a very simple circut that includes an integrated circut timer (555) and a few other modifications I made to allow for adjustable frequency, pulse width and signal strength.

Here's a teaser shot of the scope during one of the tests.



More details and pics can be found here:
http://www.jrbranson.com/showandtell-2.htm

Any pointers from electronic pros or microcontroller geeks is of course welcome...I'm just burnin my way through this without soaking up as much as I should.
JohnnyB
See less See more
I'll be stunned if the shit ever works on a bike.

Actually....anything is possible with enough time and effort.

Give me a year or so and I'll have my laptop at the track programming my bike before a race.

Next gadget might be a servo motor that varies intake velocity stack length depending on rpm....kinda like an EXUP valve on the intake instead of the exhaust.

All in the spirit of vintage racing or course.
JohnnyB
Just tested a software microcontroller tachometer. Reads accurate up to about 48,000 rpm...based on a 800 hz input from the signal generator.

Now just program it output 1.0-2.0 ms pulses to a servo motor and you have a super quick, stable, accurate servo tach.

Hehe...this is step one to building a rpm sensitive variable intake track....sliding velocity stacks, run by a servo motor like you see in an RC car. Long stacks for low rpm torque...short stacks for high rpm. Don't worry...it will be like or similar to a fuckin UFO I saw in 1967!

ALL YOUR FORMULA ONE IDEAS ARE BELONG TO ME !!!
JohnnyB
1 - 11 of 28 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top