ok, i know there are some electronics geniuses on this forum that could make a ms plugin for rotary ignition control...lets find one of em!!!

pat

 

well, you'll need to know assembly for the motorola microcontroller. I don't think it would really be all that bad if you knew how to do that. There's already a daughter board made for extra drivers.

 

Could you use info here, http://www.jsm-net.demon.co.uk/megas...s/install.html and then use an EDIS-8? It has outputs for 4 coils, does that mean it could control each of our coils independantly? Then you just tell the MS-n-EDIS when to fire each coil.

 

leading fires waste spark so you really only need 3. wouldn't that mean that edis-6 would work as well?

 

yes but does it allow you to set the timing independently? or just set the timing as a group?

pat

 

http://www.bgsoflex.com/mjl/mjl_edis_summary.html After reading this (understood about 75%) it sounds like it won't do timing independently like I was hoping.

 

I wasn't going to get into this, but it appears there's enough interest...

EDIS 8 and 6 won't work properly to control leading and trailing, as the spark events are evenly spaced over 720 degrees. The EDIS system is nice, though, in that it will run at fixed timing (typically 10 degree BTDC) if no advance signal is received. This fail-safe mode is nice, as well as the knowledge that these units have been firing away reliably in oem installations for well over 10 years.

One EDIS-based rotary option is to use two EDIS-4 boxes (each contains 2 coil drivers per unit). One EDIS4 controls leading, the other trailing. Each EDIS4 module fires two coils, either coil-on-plug, or coil-near-plug. FC trailing coils would be a likely candidate, as they contain two separate single-post coils.

Setup options are as follows:
1.) Trigger both EDIS4 modules from the same VR sensor. Then sending identical SAW (advance) signals from MS results in simultaneous firing of leading and trailing plugs on each rotor face. This can be done now with MSnEDIS code. With some code modifications, send out two SAW signals to provide timing split.
2.) Use two VR sensors mounted the desired timing split apart. Now timing split is fixed, and sending the same SAW signal to both modules will result in leading and trailing advancing together. Again, MSnEDIS code should be able to run this today. To get variable timing split, code mods are needed.
Either of these options require mounting a 36-1 wheel and associated VR sensors.

Another option: Use MSnS to control an FC leading coil to fire both plugs wasted spark style. Leave the trailing firing through the distributor. Now full control over leading advance curve is possible, while trailing is fixed at whatever the distributor provides. Not the greatest solution, but should be very simple to implement.

Neither of these options are ideal, but will provide at least some computer control over ignition timing. For the average automotive hobbyist who's not buying blocks of dyno time, the drawbacks are likely nil from a performance standpoint. Option #2 is well-suited to the 1st gen crowd, as they already have the distributor. A few wiring changes, an FC coil, and a Bosch ignitor, and you should be all set.

 

That is some really great info Renns.....so there is a solution, it's just not as simple as on a piston engine.

 

so a 12a Distributor will run 13BT trailing coils? I thought there was some problem with this...

If that will work, sounds like the way to go to me..

pat

 

No. The EDIS-based solution I mentioned discards the distributor/CAS altogether, instead getting its signal from a 36-1 wheel on the eccentric shaft. The MSnS option mentioned allows for programmable control of leading timing only using an FC leading coil controlled by MS, with the distributor providing conventional trailing spark via the trailing coil, cap, rotor as in stock 1st gen setup.

 

Sorry about the dumb questions coming below:

By how much can you improve gas mileage with such a gizmo on an NA ?

I assume you have to keep the ECU will all the connections to sensors/solenoids for emissions + ignition and have the MS just controlling the injectors ? Is that how you do it ?

Hugues -

 

"The MSnS option mentioned allows for programmable control of leading timing only using an FC leading coil controlled by MS, with the distributor providing conventional trailing spark via the trailing coil, cap, rotor as in stock 1st gen setup."

isnt that what I said? I didnt mean controllable timing, just that it will run it as stock? If the 12a dizzy can run the trailing timing as stock, and the MSnS can run leading, then whats the problem?

(not trying to be smart, just not sure that i understand)

pat

 

how much are these compared to the microtech and haltech systems??

 

megasquirt= around 200
microtech/haltech/etc= 1200-4000?

 

No worries. Looks like there is still some confusion. The trailing coil mentioned above is not a 2nd gen trailing coil, but the traditional single post coil used in 1st gen distributor applications. In fact, the entire trailing igntion is stock 1st gen, with the MS only controlling the timing of the FC leading coil. Obviously with this method, timing spilt is whatever you end up with subtracting trailing from leading. Trailing is determined by the distributor curve (vacuum, mech, and base advance), while leading timing is fully controlled by the MS.

Keep in mind these ideas I've posted are just that; ideas. No one that I'm aware of has run a rotary with this setup, but by posting info, maybe some innovative folks will implement these options, or come up with another.

Obviously from these and earlier postings, if you want a proven full standalone ecu with spark and fuel control, the MS is not your choice. Go with a commercial unit and save us all the grief. On the other hand, if you have a decent understanding of these systems, and get a kick out of building this stuff yourself, then the MS is an excellent starting point.

 

how much of a pain in the *** would be configuring this for a half bridge motor running larger secondaries?

i know i will have to keep the stock ecu to control timing, would that be a problem running a halfbridge?

 

Took my 87 TII to the dyno friday with my Megasquirt to tune in the ve table. Took about a hour and did a couple of pulls at the end. Ended up with 239 rwhp at 9 psi with 12.5:1 AFR all the way across the board at WOT. This is compared with my pulls last fall with the stock ecu contolling the fuel of 233 rwhp at 12psi and seeing 14:1 at the higher rpms. I think the drop in boost was due to it being hotter friday than last fall when I went.

 

impressive. i may seriously try this.

 

JC that is awesome!! You've got my psyched about the MS again. Maybe once I get the motor broken in then I'll throw my MS back on and take her straight to a dyno instead of trying to tune on the street again. Would you mind sending me your VE tables? I'd love to take a look at them and see how you got it tuned in so well.

 

Congratulations on a job well done! Can you provide some details on your engine configuration? I assume you've removed the afm?

Do you have a datalog snippet I could take a look at? Actually, posting the datalog, with setup files (.msq, .vex) on the Yahoo site would be a great help to others. Every week or two there's another TII owner looking for info.

 

mmm I wouldn't consider 12.5 "safe" necessarily. I'd be doing 11:1 or so to be safe... maybe even lower

 

I'm just implementing MS on my rx-7 and I am about to go the way with letting the stock ecu controlling ignition. However, about removing of the afm and temp sensors from the stock ecu, the workshop manual for my 1990 TII states that in the absence of afm the controller will switch to failsafe mode with fixed ignition timing. This doesn't sound to great, have you (jkrueger) somehow tricked the ecu to think the afm is there?

/Martin

 

he has an S4. If you want to keep factory ignition; you have to convert to S4. I'm throwing a dizzy on my N/A S5

 

I do believe there is a way to trick the ECU...

 

bump..
im waiting for the new megasquirt ecu...sposed to be out this fall...

pat