Automotive Coilpack Coversion?

[mooseheadm5]

http://www.dainst.com/info/edis/edis.html

http://www.megamanual.com/ms2/EDIS.htm

Nothing special, just a pulse width output to the EDIS, and an RPM pulse going to the controller of your choice (like the engine computer, a Megajolt Lite Jr, or your own programmer.) It would be nothing to make a simple advanve curve setup that mimics a mechanical advance. The MJL adds support for manifold pressure or throttle position advance mapping, but you probably don't need that. You could probably do it with a much cheaper controller like an Arduino (or clone from Ladyada.net) or similar controller, which is usually smaller than a pack of gum.

[donspicer2]

looks like alot of fun!

[LikeMike64]

Sounds like a good plan!

[ElPiloto]

I looked on ebay, so far, the edis seems a little pricey, obviously i don't know what i'm looking for. ?

You can find all the EDIS components you need in the wrecking yard. Including the 36-1 wheel. They are used on many 4 cylinder Fords. It seems like people are trying to get rich on eBay sales. Everything costs more on eBay. Some of the prices are just plain absurd.

Do a search on Google for EDIS and go to a lot of the sites you will find links for. There are a couple of people that sell EDIS components at very reasonable prices. They are all from the junkyard, but cleaned and tested.

Even using the Chevy coils can help to keep prices low for your build. Compare the wrecking yard prices with the Accel coil prices.

Go to megaspark.com and get some good info.

Have fun and good luck.

[donspicer2]

in the boneyard? I haven't a CLUE which car to hunt for.

[FirstYearDeek]

in the boneyard? I haven't a CLUE which car to hunt for.

Look for any carburetted 4-cylinder Ford, Lincoln or Mercury 1982 to 1990. Truck, car, it doesn't matter.

Eithewr of the links Moose posted will have specific models.

-Deek

[donspicer2]

have either of built this for a goldwing yet?

[n5zhu]

Woud it be feseable to use feul and ingion sysyems from a 4 cly car?Useing computer and all parts needed from a late model auto.
73 Patrick

[FirstYearDeek]

Woud it be feseable to use feul and ingion sysyems from a 4 cly car?Useing computer and all parts needed from a late model auto.
73 Patrick

Not likely.

There are two major roadblocks here; 1) the triggering method and 2) engine displacement.

During cruise an EFI system will go "closed loop" and use the O2 sensor to determine mixture, and that would work out fine, probably. The issue is: what about acceleration? Deceleration? Wide open throttle? Essentially the 80% percent of the time you're NOT at cruise and the system isn't closed loop... those enrichments and fuel tables are programmed in the controller with a specific engine in mind.

Unless you can program those variables and tables, an OEM FI system just won't work. That's what makes MegaSquirt so popular. All those things (and more) are completely programmable by you.

-Deek

[donspicer2]

the megajolt site narrative discusses chosing one of two methods to discover engine dynamics.

1 tie all the manifolds together to a tank, to read a stable vacuum signal.
2 install a throttle position sensor.

I tied mine altogether, the bike DIDN'T like it. I didn't install a tank, just tee'd all four together. ran poor, lumpy idle, stumbling. it responds to idle screw changes, but just won't run worth a hoot. any ideas? I've heard there are some who've had good results doing this, I wonder what I'm doing wrong. I'm pondering putting restrictors in the hoses, that might help, I suppose. more news to come. a throttle position sensor might be a real pain in the a**... Deek, Elpolito, Moose, LikeMike, anyone else following this thread, whatchtinkaboutdis? I'm gonna finish the vellman installation this weekend, and see. but I'd REALLY like to do the ford system.

[sunnbobb]

Hmm, I would think a tank would serve an important function to balance out the system both in terms of vacuum and time of demand.

[donspicer2]

I'll make up some kind of metal tank in the next few, try again.

[mooseheadm5]

You might be able to do it straight up with car stuff, but I wouldn't. They don't cover the rev range that motorcycle engines do, they do not have the acceleration rates that motorcycle engines do, etc. They are basically programmed for the displacement, acceleration rates, RPM, and volumetric efficiency of the car. You would be changing all of these things with the motorcycle engine, so it would be a complete fluke if it worked even a little. The O2 sensor does not have enough control over the mixture to make up for all these differences.

Don,
I would prefer a throttle position sensor vs manifold pressure (of course Megasquirt uses both.) You would need one that is like a rheostat, which can be found on many modern cars and hooking it up is pretty easy. Just add an arm from the bellcrank that turns the TPS. They are not that special.
Motorcycles tend to have very unstable idle manifold pressure because of the short intake runners and lumpy cams. The tank and some mild restrictors in the lines should really damp out the fluctutaions if you wanted to go that route. In reality with a motorcycle engine, either system can work fine for ignition mapping. Manifold pressure is better for your every day car, but throttle position tends to be better for the lumpy cams and fast changes in manifold pressure that are difficult to track with quick acceleration rates of motorcycle engines (it is much to track fast throttle position changes.) If you only want to mimic the mechanical advance with a 2-D ignition map, of course you don't need any of this.

[LikeMike64]

Don, with such short manifolds and carbs that are designed for no other source of air, when you tie all of the manifolds together, you are creating what the carb sees as a vacuum leak.

You might try putting a vacuum check valve (one way valve) inline between each cylinder and the tee that connects them all together. Set it so that it only lets air into the manifold. This will allow the cylinder to pull air from the tee, but will not let air flow out of the other three manifolds into the one that is pulling air through its carb at that moment. This way you only have one air source (the tank) for each cylinder than the carb, not 4 (the tank, and the other three cylinders).

Also, make sure that you have the common line coming from the tee connected to the correct port on the MAP sensor. Some MAP sensors have a atmosphere port, and a vacuum port.

However, I am not sure about running advance based upon the throttle. Throttle input is usually used for fuel injection, to tell the ECU what the operator wants (accel, decel, cruise, etc). Ignition is usually based on engine speed. A tachometer based input seems more logical for ignition timing advance to me.

[donspicer2]

point well taken, good insights. I can't figure out one thing - how does the vac. tank lose pressure as engine load reduces? I agree about the tach data. much easier to do.