AEM AEM EMS: Is it really this easy to switch to a different trigger disc?

[NickF509]

I've been looking at this for awhile now as I want help smooth out my idle and hard start with my AEM EMS. I've found a couple of trigger discs out there, the Kiggly 12 tooth sensor setup and the triggerdisc.com setup.

I'm looking more at the kiggly sensor setup as I might as well change the sensor if i'm doing this to begin with. Would the steps with this kit really be

1) Remove old sensor
2) Remove old crank trigger disc
3) Add new crank trigger disc
4) Modify cam disc
5) Plug in new sensor
6) Set AEM EMS to Kiggly settings
7) Retime
8) Profit?

Is there anything else I would need? Would I need to even retime?

Thank You

 

[LiquidX]

Upon searching I came across this PDF file for the 12 tooth AEM Crank Trigger from Kiggly. I don't know if you viewed this already?
http://www.kigglyracing.com/parts/Kiggly Racing 12-Tooth AEM Crank Trigger Directions.pdf

Since you are pulling out the CAS you would need to re time your ignition timing.

 

[91BlackLaser]

i bought this one from these guys cause it was way cheaper than the regular venders. It worked great with the instructions from trigger discs.com : Yoshifab.com - High resolution DSM CAS trigger disk

 

[NickF509]

Wow 18 dollars is pretty cheap. I've been really lazy (and it gets dark before i get off from work) so I need to see what sensor I have in there as from what I understand that would be for the optical setup?

And reading up on timing using the AEM EMS seems pretty straightforward as well. I just need to get a timing light. Does autozone rent them? I see they are about $100 otherwise for a really good one.

 

[eclipsegsx1736]

I've been looking at this for awhile now as I want help smooth out my idle and hard start with my AEM EMS. I've found a couple of trigger discs out there, the Kiggly 12 tooth sensor setup and the triggerdisc.com setup.

I'm looking more at the kiggly sensor setup as I might as well change the sensor if i'm doing this to begin with. Would the steps with this kit really be

1) Remove old sensor
2) Remove old crank trigger disc
3) Add new crank trigger disc
4) Modify cam disc
5) Plug in new sensor
6) Set AEM EMS to Kiggly settings
7) Retime
8) Profit?

Is there anything else I would need? Would I need to even retime?

Thank You

Yes, it pretty much is that easy. Kiggly's setup is the best since it actually measures the crank position on the crank as opposed to reading it off the cam like a 1g CAS does. This eliminates the possibility of spark scatter due to timing belt stretch/whip etc.

When you say smooth idle and hard starting, what exactly do you mean? The 12-tooth wheel won't do much for smoothing out idle. If you have a Series I EMS (Series II has special firmware to quick-sync with Mitsu trigger patterns - it'll start up faster than the stock ECU!), the syncing takes a bit longer than stock which may account for the long startup, but NOT for the "hard" startup you mention. Sounds like you need to adjust initial crank pulse/cranking fuel tables. Try using the "Crank inject ALL" option which batch-fires injectors during cranking. Doing so will require a reduction in cranking fuel to accommodate the ~4x fuel you'll be getting with this option on.

i bought this one from these guys cause it was way cheaper than the regular venders. It worked great with the instructions from trigger discs.com : Yoshifab.com - High resolution DSM CAS trigger disk

This will work as well and is easier - just take the cap off the CAS (ONLY works with green-top CAS, NOT black top) and replace the factory wheel with this one. This wheel has 24 teeth, but is on the cam which rotates 1/2 speed of the crank, so this is technically the same as having 12 teeth on the crank like Kiggly's setup. Just remember this is reading crank position of the cam so there is some room for spark scatter due to timing belt flex/whip - there's a reason why Mitsu went to a true crank-mounted trigger setup later in the years.

And reading up on timing using the AEM EMS seems pretty straightforward as well. I just need to get a timing light. Does autozone rent them? I see they are about $100 otherwise for a really good one.

Get the simplest (cheapest) timing light you can. You don't want one with any buttons or dials, it just makes things easier that way. I HIGHLY recommend checking timing before starting the car in case it's far off. Pull all 4 spark plugs out of the engine to allow it to spin faster. Disable all 4 injectors, put the timing light on cyl 1 spark plug wire (ground spark plug to valve cover) and check timing. During cranking, just get it within a few degrees. You can worry about getting it exactly right and setting timing drift once the engine's running.

 

[NickF509]

Yes, it pretty much is that easy. Kiggly's setup is the best since it actually measures the crank position on the crank as opposed to reading it off the cam like a 1g CAS does. This eliminates the possibility of spark scatter due to timing belt stretch/whip etc.

When you say smooth idle and hard starting, what exactly do you mean? The 12-tooth wheel won't do much for smoothing out idle. If you have a Series I EMS (Series II has special firmware to quick-sync with Mitsu trigger patterns - it'll start up faster than the stock ECU!), the syncing takes a bit longer than stock which may account for the long startup, but NOT for the "hard" startup you mention. Sounds like you need to adjust initial crank pulse/cranking fuel tables. Try using the "Crank inject ALL" option which batch-fires injectors during cranking. Doing so will require a reduction in cranking fuel to accommodate the ~4x fuel you'll be getting with this option on.

Thank you, I've been reading up on the one post (the one everyone points to for fixing starting issues, i don't have the link handy). I'm just waiting to get a few minutes to give the different settings a try including the crank inject all.

For the hard start it really is an annoyingly hard start if the car is cold. I'd say 20-30 cranks before it starts.

The idle issue I have is that at times (it seems random) my idle will start at 1200ish rpms, rev up to 2000 rpms then cut and drop back down to 1200ish rpms and just do that over and over and over. Sometimes giving it gas fixes it sometimes it doesn't. I was making the guess that there was an issue with the cam sensor and it had to do with the trigger disc. Maybe there is just an issue with the CAS all together?

This will work as well and is easier - just take the cap off the CAS (ONLY works with green-top CAS, NOT black top) and replace the factory wheel with this one. This wheel has 24 teeth, but is on the cam which rotates 1/2 speed of the crank, so this is technically the same as having 12 teeth on the crank like Kiggly's setup. Just remember this is reading crank position of the cam so there is some room for spark scatter due to timing belt flex/whip - there's a reason why Mitsu went to a true crank-mounted trigger setup later in the years.

I'll probably try this first and save the $200 . I have no clue what spark scatter is so i'll try and do some research on that.

Get the simplest (cheapest) timing light you can. You don't want one with any buttons or dials, it just makes things easier that way. I HIGHLY recommend checking timing before starting the car in case it's far off. Pull all 4 spark plugs out of the engine to allow it to spin faster. Disable all 4 injectors, put the timing light on cyl 1 spark plug wire (ground spark plug to valve cover) and check timing. During cranking, just get it within a few degrees. You can worry about getting it exactly right and setting timing drift once the engine's running.

By disabling the injectors your talking about just pulling the plug like you would do in a compression test correct? Not disabling them from within the software?

Thank you again for your very descriptive post!

 

[eclipsegsx1736]

Thank you, I've been reading up on the one post (the one everyone points to for fixing starting issues, i don't have the link handy). I'm just waiting to get a few minutes to give the different settings a try including the crank inject all.

For the hard start it really is an annoyingly hard start if the car is cold. I'd say 20-30 cranks before it starts.

The idle issue I have is that at times (it seems random) my idle will start at 1200ish rpms, rev up to 2000 rpms then cut and drop back down to 1200ish rpms and just do that over and over and over. Sometimes giving it gas fixes it sometimes it doesn't. I was making the guess that there was an issue with the cam sensor and it had to do with the trigger disc. Maybe there is just an issue with the CAS all together?



I'll probably try this first and save the $200 . I have no clue what spark scatter is so i'll try and do some research on that.



By disabling the injectors your talking about just pulling the plug like you would do in a compression test correct? Not disabling them from within the software?

Thank you again for your very descriptive post!

The "idle surge" you're describing is just that, idle surge. It's related to your idle settings that need to be fixed. Basically, the engine's RPM is increasing above the minimum RPM threshold for decel fuel-cut. The EMS then cuts fuel, causing the RPMs to drop back into idle range, and the cycle repeats. One of the following is the cause of the problem:
1) a combination of your base idle RPM + idle RPM trims (time after start, idle vs coolant temp, etc) is causing the idle target RPM to be higher than your decel fuel cut threshold.
2) the engine is overshooting its idle target because either the idle base % vs RPM is set up incorrectly, or the BISS is opened too much.

On our cars, it is so easy to unplug the injectors I'd suggest doing it that way rather than in the software.

Glad to help!

Beau

 

[eclipsegsx1736]

As far as spark scatter, it can be caused by many things, but I was discussing spark scatter due to the interface between the cam angle and crank angle.

Basically, in order to determine when to fire a coil, we need to know 2 main things. One, the position (angle) of the crankshaft (somewhere between 0 and 359 degrees). Two, what cycle the engine is in. On a 4-stroke engine, a complete engine cycle occurs every 720* of the crank (2 full rotations) or every 360* on the cam.

On a typical engine that has separate cam and crank sensors, lets say 12 teeth on the crank sensor mounted on the crank, and a single tooth on the cam (which is where you end up with when running the Kiggly setup) the cam sensor tells the EMS when the engine's at TDC in the compression/power stroke, and the crank sensor tells the EMS the actual angle the crank is in at any given time. Side note: you can see here why having more teeth on the crank sensor is generally a good thing – more teeth means more accuracy on the crankshaft’s angle at any given time.

In a perfect world, the camshaft should spin exactly half the speed of the crankshaft. In reality, due to many factors, that’s not always the case. Of course, if averaged out, it will be exactly half the speed (if it wasn’t, the timing belt would skip..). However, even if the engine is, say, at a “constant” 3000 RPM, the crank is constantly accelerating and decelerating every time a combustion event occurs. Think of pedaling up a very steep hill on a bike – every time you thrust with your legs the bike accelerates and then decelerates until you push the pedal with your other leg. The pistons do the same thing to the crankshaft. Not to mention the timing belt itself has stretch in it that will stretch more and less with each combustion event as well – it’s only Kevlar.

So what does this all mean and why do we care? When you have an engine that senses “crank” position off the cam sensor, as does a 1g DSM (and a few other unfortunate cars), it has to just assume that the crank is constantly a perfect ½ speed of the cam. From what we learned, we know this isn’t the case. This means the EMS will be “wrong” on the crankshaft position depending on the relation between crank and cam position. This will be reflected in ignition timing that “scatters” around your set goal. For instance, if you were shooting for a flat 15 degrees advance, you may see values from 12-18 degrees, for example. You can measure this by pointing a timing light at the crank pulley. Serious guys tuning a car to the very edge (where 2-3* scatter can destroy an engine) will use timing lights and high speed cameras to check for spark scatter.

Another side note: Now, doesn’t running the factory 2-tooth crank trigger sound really scary to you?? Think of only having 2 reference points every 360 degrees for the crankshaft angle. Of course the EMS can “guess” where the crank is at any given time based on the timing between each tooth count event, but what we learned about crankshaft acceleration and deceleration make this type of setup sound pretty terrible (and it really is). It’s amazing seeing people making 1000+ whp on these setups. Makes you wonder how much is left on the table if you can control spark accurately.

 

[NickF509]

Just a quick update on where I'm at with the car:

First thank you very much for the description of spark scatter and how everything works together.

I assume this is partially why the 2G dsm's switched to two seperate sensors, a Cam sensor and a crank sensor(am I correct that this even happened?) ?

Second for my update: I spent some time actually digging into the tuning aspect using the AEM software. Not that I know anything more about actually tuning the car than I did before but the interface is nice even in the series 1 software.

I then proceeded to follow the description for the E85 cold start from the evom forums. At first i reduced the values of the crank/fuel table like suggested if using pump gas. This didn't help at all.

Finally out of frustration I said what the heck lets try the values as is and it started instantly. I'm only slightly more rich on warmup than I was before as well (13.5 vs 13.8-9 AFR). I don't get it.

I also found some odd values in the tables that I think was causing my idle surge issues. I can't remember the exact tables (i have a backup still) but one was the RPM addition table on startup (I don't remember the exact table name). The values went something like 500, 475, 450, etc to 200, then there would be one odd value at like 350/400 in there. It make the curve spike somewhere in the middle of warmup if you were looking at a graphical representation of the table. Once i swapped the values with the ones suggested and made a nice high to low curve on warm-up I haven't had the idle issue again.

This is only my first day driving it though so I'm paying close attention to any issues I find.