cam signal error causing timing error/cut out
 

99 miata (bored to) 1.9l, BW EFR 6258 running about 10lbs of boost for 245whp on AEM ems4. ATI harmonic damper with 12+1 trigger wheel. 800cc injectors, 190 high pressure fuel pump on pump gas.

After finally getting my car all sorted out/tuned and finally able to drive it (yay) we were getting a pretty consistent error count on the cam signal. My AEM ems4 is being told to only read 2 teeth on the cams even though there are 3, two are too close together for it to differentiate 95% of the time. But I believe that sometimes it catches the 3rd tooth and that's what could be causing this cut-out.

I read using an evo 8 cam sensor (mag) and changing the ecu to read falling signal instead of rising can help the issue, I tried it and it did help, so instead of a consistent error count it is much less common/more sparatic.

It's a quick blip like cut out which it likes to do at idle and consistent highway speeds. Sometimes while doing a pull, which is much more noticeable when it cuts, but its only done that maybe 2-3 times while idle/highway it does it maybe once every couple minutes. Other times its fine, its very random.

Should I cut 1 or even 2 teeth off the cam gear and tell AEM that its 12 and 1, which AEM prefers anyway?
Any other ideas on how to fix this or possibly what else might be causing it?

Darby - I'm a little confused by a couple of things. It could very well be I just installed mine differently since I have a 1994.

The cam angle sensor is located on the back of the 1994 head, driver's side. It has two tangs on it, one of which needs to be shaved to work with the EMS4. You have to remove the sensor from the head to modify it. In your photos, you are referencing the cam gears. Can't remember if the 99 has the same sensor? In either case, there is no reason to modify cam gears.

Next, the 12+1 trigger goes on the crank pulley, to be read by the crank sensor located on the front of the motor at the bottom, just to the right of the crank pulley when facing the car. Again from your description, it sounds like you have a trigger on the cam? The trigger wheel may not work with the ATI damper, but it definitely does work with the SuperMiata damper. Trackspeed Engineering sold me both the trigger wheel and the crank damper pulley.

Do you have both cam and crank signals feeding the EMS4?

To determine if the IACV is feeding noise, unplug it and then run the car again. If you have eliminated the "misses" then it's not the cam/crank signals giving you problems - it's the IACV causing a dirty signal that the ECU interprets as a miss or an extra trigger.

First time posting so here goes. Your assessment of how the cam signal gets received by the EMS is incorrect. The EMS definitely sees all 3 teeth, not 2. You can't modify the cam to one tooth and tell the EMS you have 12 crank tooth with 1 cam because your crank is a 12+1 wheel. The extra tooth on the crank gives the EMS a crank rev start/stop reference. What the Sync Teeth setting of 2 does is tell the EMS to sync on the rev that 2 cam teeth occur. Hopefully your sensors have been setup so that one crank rev has the single tooth occur and then the next crank rev has the two cam teeth occur. The best way to see this is to use an oscilloscope to watch the crank and cam signals to verify that the cam teeth are happening in the right spots.
Right now, I don't think you have a full understanding of the problem and simply changing the pattern to something else may or may not make for a positive change. You could have a sensor noise problem which changing the trigger pattern won't affect. What you really need to do is get in touch with AEM's tech support and send them data logs capturing channels that show the problem occurring. This way they can properly troubleshoot the problem and get a real solution.

FWIW, on Jumbo's car, I'm pretty sure I had to change from the normal falling edge trigger on the Hall crank sensor to rising. Timing control wasn't very good when setup for falling edge. Made the change to rising and it was dead nuts on. Also FWIW, you can run without the cam sensor at all and run semi-sequential off the crank input only. You have to do some "stuff" to get it setup but if you have a problematic cam input, running on crank only will get you going again.

Hi Brad. :wavey:

On my car the cam sensor is on the front of the valve cover (not the back like yours) and reads 3 teeth on the cam gear. I do have a 12+1 trigger wheel on my ATI damper and my crank position sensor and trigger wheel are working and reading fine. The cam signal is the only problem, and yes they are both communicating to the ecu

Wire your cam sensor better, its marginal to begin with so having iffy wiring can make it seem to cut out randomly. And you'll probably need to add the filters that AEM left out of the EMS4 for the cam and crank. A few of us have dealt with that piece of shit ecu, most have given up. There's some old threads. From someone who owns one, you're way better off dumping that thing and buying a megasquirt than trying to make it work properly. Even once you get it to work right you're still stuck in the 90's with injector pulse width fuel tuning instead of VE.

EMS-4 has always worked great for me. I've got one on a Honda RVF400RR that's a 458 now that makes 100whp and revs to 14,500 rpm. You just gotta know how to use it right and command pulse width tuning works just fine too.