Aem ems is time based. the fuel map directly changes pw, there is no VE. o2 feedback is just closed loop, you dont command an air fuel ratio

There is also no "latency" table.. only a battery offset table, which in his case wouldnt be doing anything. Since the aem is time based the latency makes no difference.

What rpm is you idle at? if you want to quickly narrow down if its the injectors or something else, raise the idle 200 rpms and see if it settles down.

Unless you let the car idle for very long periods of time idling in the 13's wont hurt your plugs.

Do you have the fuel map setup as a standard 3d table or are you using boost comp method?

 

Wadzii, thanks for giving some insight. I will try raising the idle, right now it idles at 1000, when warmed up, and see what happens. One thing to note, while the car is warming up, the idle stays steady, it only starts jumping around once its completely warmed up. The boost fuel correct table is zero'd out, and i know all the tuning was done just on the fuel table itself, so im guessing its the standard 3d table.

 

Make it display in whatever units you want, that's your VE table.

Correct, it is just closed loop. Sure, you don't "command" AFR's like you would on a MAF based car, however, table can still be considered your commanded, or desired target at which the closed loop O2 feedback will aim to maintain. I have to clarify exactly what the person is trying to change given there is a problem.

Battery offset IS Latency.

Completely incorrect on not doing anything. Your battery offset/latency is a constant based on voltage. No matter what you command via your VE table, your battery offset will ALWAYS (except when injectors are shut off) be in effect per the table's battery voltage (bigger numbers at lower voltage as it takes longer for the injector to open). If you're idle total PW is 1.2ms and your offset is at 1ms, most of your idle is being controlled by that offset.

Some OEM cars have a Minimum Base Pulse width that when they run large enough injectors (or crappy medium sized injectors), they cannot lean out the mixture enough to get a smooth idle because the offset/latency + pw (bottomed out) are too much to maintain a stoich idle. Finding a good idle has a large amount to do with injector latency/battery offset.

 

How old is your sensor? Are the AFRs steady under load?

 

Sensor is not old, maybe 6 months. AFR's are completely steady under load.

 

What is your resolution? microsec/bit can found under Flex Fuel tab.

 

on a 3d style map you loose resolution at idle. You may not be able to get it any better with what you have. Thats one of the big advantages of the boost comp method.

 

Since most of the idle [total] PW is coming from the offset, he shouldn't have any problem getting a good idle, ESPECIALLY on a set of ID1000's. I've gotten ID2000's to idle perfectly smooth, without any change between AC on or off on a Subaru running Speed Density on the OEM ECU (and without lower Miniumum IPW).

I'll quote a friend of mine, "the big knob for idle and cruise is dialing in the latency (offset)".

 

most of the idle pw is not coming from the battery offset table. The battery offset is applied as a correction to the main fuel map.

The aem uses the main fuel table as the basis for everything, everything else is a correction that's applied to that. AEM EMS works nothing like an oem ecu. AEM Infinity has some similarities to how some stock ecu's work but not the v1/v2.

 

Offset is not a correction. It is SOLELY there to account for the time an injector opens once it is energized at any given voltage. That's it.

Final IPW is commanded PW (+/- trims or compensations) + latency/offset... AEM, OEM, whatever. Virtually everything uses the main fuel table or scalar for everything fuel related. The offset/latency works the same regardless. Again, it is to account for the time it takes the injector to open at a given voltage, which is something the main fueling table does not account for.


The OP is welcome to try it. You can essentially remove fueling from that part of the table and ramp up offset to control idle. Chances are, it will idle better as it will now be a constant and high enough PW is required that the injector will fully open. Set your offset to zero, watch it lean way the hell out and probably die (or just start decreasing it in .1 increments and watch it lean out). Increase it, watch your idle richen up.

I haven't found offset/latency to be different than any OEM ECU and it shouldn't be considering what it is in place for. There is a reason battery offset is graphed in ms vs. [battery] voltage. Take a log of your final injector pulse width and compare it to your VE table for that region and you'll see it is higher than what is on the map, to the tune of what latency value it is using for the voltage the ECU is seeing.

Some reading (AEM relevant):
http://www.3si.org/forum/f42/aem-bat...ectors-274394/

http://www.dsmtuners.com/forums/tuni...set-table.html

Additionally, you can search on any tuning forum for "Battery Offset" or "Injector Latency" read up on what it's doing.