Timing weirdness with EDIS

JohnJL · 03-24-2007, 06:06 PM

I've got a twin-plug EDIS setup on my twin-plugged SC 3.0. All seems to function properly and I've gotten the fuel roughly dialed in. The static ignition timing seems fine and I am using a generic ignition curve from another Pelicanite.

I am just doing a shakedown of the engine after a few track days (oil change, check compression, etc) and wanted to verify the dynamic timing. My adjustable dial-in timing light shows the expected cranking and low-idle figures. However at around 2000 rpm the timing seems to jump to 40 (?!?!?!) according to the timing light. However my ECU datalog shows the ignition at around 19 (expected.) I've played with the master trim on the ignition and as expected advancing it and retarding it comes up with the expected reaction from the motor. It seems the ECU is reading and controlling it fine.

Is my timing light bad or is it normal for wires on an EDIS ignition to perhaps trigger an inductive charge (making the timing light fire) when the spark is occurring on the opposite coil post? That is, would the "wasted" spark trigger the light as well? I wouldn't have thought so sinceonly one plug is to fire at a time on each coil.

Any hints?

Thanks

JohnJL · 03-25-2007, 01:58 AM

Bumpitty bump bump for the US Sunday crowd. Thanks All!

munsonbw · 03-26-2007, 03:47 PM

I'll give you some things to chew on:
-first, since you are using an inductive light make sure you have good clearance between the wires, just because it is easy to check
-get another light and verify yours, not familiar with dial in lights... maybe you set it wrong
-what kind of ECU? Megasquirt with Extra code? If so verify all settings and then verify them again.
- post this in the msextra forum

You do know that the EDIS base timing is 10*, assuming you set the VR sensor in the "factory" position, right? If this is news to you, go here:
http://www.msextra.com/manuals/MS_Extra_Manual_Index.htm

jpnovak · 03-26-2007, 05:56 PM

Doing it the old fashioned way... What does the timing read on the crank pulley?

JohnJL · 03-27-2007, 01:39 AM

Thanks guys. I don't have another light to check this one with...

When the dial on the timing light is set to "0" I read the timing marks on the pulley as expected up until revs get over ~2000...then the marks go out of view until I "dial" them back and read about 40 degrees on the light!?!

ECU is Adaptronic...all the settings Megasquirt or MOTEC can do, so can this...

munsonbw · 03-27-2007, 04:13 AM

Hmm, you say all of the marks go out of view unless you dial in 40* to the light? Then are you saying that the timing is actually 40* PLUS what is read on the timing marks? If so, and you have run this on the track, I am betting your light is giving you incorrect readings. I assume since you are running twin plug, that your compression is pretty high? 40+ degrees at 2000 rpm would seem like a recipe for detonation to me... not to mention what it is doing in the higher revs. Have you checked the timing higher up the rpm range? Does the timing make a quick jump right at 2000, or does it steadily climb even beyond 2000; i.e. do you think that your timing curve is offset or the wrong amount of total advance is being applied?

A few more questions; Where are you getting the signal to feed the EDIS? are you using a VR sensor? If so, have you verified its position relative to TDC? I assume you are running two EDIS-6 boxes, have you hooked your light up to the other plug wire fed off of the other EDIS? The two boxes could be doing different things.

Finally, I am not referring to "features" when I talk about ECU settings. EDIS was never designed to be an add on device for aftermarket ECUs. As a result, you must adjust settings in the ECU to account for how it interprets the PIP and then sends out the SAW back to the EDIS. Megasquirt has special menu options that change the program some to account for the 10* base timing and the properly interpret the signal from EDIS. Did your ECU come with instructions for EDIS, or is this your creation?

munsonbw · 03-27-2007, 04:19 AM

One more thing when I re-read your first post. In the Megasquirt world, if you do not have sensors hooked up properly the ECU datalog will not be accurate. Just because the datalog says 19* does not mean that it is feeding that to EDIS. The ECU "thinks" it is based on the information it is given and then the ignition settings and map you have created for it. Remember, incorrect in = incorrect out.

JohnJL · 04-06-2007, 05:08 PM

Thanks for your help guys.

It turns out the timing and logs were in fact correct...I don't understand the theory behind it but I was able to verify with another timing light and with the help of an electrical engineer that the wasted spark was messing with the adjustable timing light. It seems that the slope of the advance calculated by the timing light is 2X the actual slope. So while the idle advance was "close" to expected, even at idle it was 200% of the actual advance.

I also diagnosed a loss of spark on the top plug of #1 and #4. I suspected a bad coil tower (since they are on the same tower) but it seems I had ANOTHER underperforming Haltech igniter. Replacing htat and cleaning up the fouled plugs solved the problem.

munsonbw · 04-09-2007, 04:29 AM

That's good news. I didn't even think about the wasted spark messing with the timing light. I have timed mine with a standard light and had no issues. If you continue to have issues with the Haltech coils, you might try the Ford Edis-6 coil. Should be really cheap to play with.

JohnJL · 04-09-2007, 01:46 PM

The EDIS coils have been fine. Its the Haltech units that have been a problem. First they shipped me a blatent returned unit (didnt work on one channel, missing the connector, etc) then the 2nd unit shows inconsistent resistance between the channels.