AEM EMS Tuning?????

[Chatzvr4]

I was browsing through as well as doing a search and could not find anything on tuning an AEM EMS. So i figured it would be a nice thread to have going here especially because I have one and would like to learn how to tune with it. So if anyone has any info please share

[Austin@STM]

Just from watching Emery tune all the cars here, i can say its a fairly involved program.

I would suggest starting with just playing with the software and learning where everything is and realize just how many things there are to change in it, before even starting to tune it at all.

-Austin

[Chatzvr4]

Just from watching Emery tune all the cars here, i can say its a fairly involved program.

I would suggest starting with just playing with the software and learning where everything is and realize just how many things there are to change in it, before even starting to tune it at all.

-Austin

Yea thats what i was thinking of doing. Car is not ready as of yet anyhow

[J. Fast]

Post up your questions

[ICPREY]

Ok, I have a question.

If I jab my throttle the car seems to go lean (on my wideband gauge) and it has a hesitation before it responces. Even if I ease into the throttle, it'll still hesitate. What settings/adjustments should I be looking at in order to eliminate the hesitation? Car runs great at WOT and through out the RPM range, but the hesitation makes it a PITA to drive in stop and go traffic...especially on hills. Car has 16Gs, 720 injectors, AEM EMS,MSD coils,Dual Walbros, Meth and all the supporting mods required. Any thoughts? suggestions? Thanks

[J. Fast]

Lots of little things can cause Tip-in problems. First thing I would do is confirm your injector dead time. On crappy injectors or weak drivers you can have a delay as much as 1ms when the injector is flowing no fuel at all. Make sure you have those injectors dialed in first.

If you're confident you have the latency dialed in you can work on the fuel tables.

Post up a screen shot of your base fuel table and a your accel enrichment tables (Accel Warm Up Enrichment, Accel Modifier table, dTPS Accel table, TPS Accel factor table).

It sounds like you might need to do some work on the base fuel table, and accel enrichment dTPS table and TPS table.

[ICPREY]

Sorry for the noob question, but what do you mean by "injectors dialed in"? I thought a set up wizard just had you select the size injectors. The car has been dyno tuned by an AEM tuner, but I don't think he worked that hard on driveability. He's a good two hours away and I figured I'd try to give it a shot. I'll work on those screen shots...probably be Monday until I get a chance to post em.

[J. Fast]

By dialing in the injectors I mean being critical of The wizard values and templates used to build a base calibration. You use the wizards for loading base calibration info that's close and then you fine tune off the presets. Not all injectors have the same dead time, not all injectors have the same nozzles, not all injectors flow linear at different voltages, there will be areas where you need to add fuel to the base table and then fine tune with the modifier tables. So let's answer this "how do you dial in injectors question?" In order to do that we have to really understand injectors.

Basic Injector Operation

"At first glance, the operation of a fuel injector seems quite simple. We apply voltage to a coil, and the magnetic field generated by the coil pulls a disc, pintle, or ball off its seat allowing fuel to flow. In other words, it is an electronically controlled on/off valve.

To find the flow rate of the injector, we can supply pressurized fuel at the inlet, apply voltage to the coil, and measure the amount of fuel that flows through the injector in a given period of time. This is called static flow testing, and the result is the static flow of the injector at the stated test pressure." Paul Yaw (EFI 102)

Well it's great when AEM, RC Engineering, or any other injector cleaning and flow service provides static flow numbers but we're not holding them in that position on our vehicles. We're using our fuel injectors dynamically. On,off, short pulse, long pulse, varying voltage, and etc.

For a visual standpoint lets say you have someone flow test your injectors. Lets say they tell you they flow 1000cc of fuel in one minute. Now lets say were using them dynamically and were going to turn them "off" 30 times in one minute, and "on" 30 times in one minute. We use up the whole minute cycling the fuel on and off so we can effectively say our injectors were flowing fuel 50% of the time which means they were flowing at 50% injector duty cycle. Well in theory since the injectors were "on" 50% of the time we should have 500cc of fuel, right? Well let me tell you, you won't have near that amount of fuel and that's where the wizards get you in trouble.

There are several things the AEM fuel injector wizards or any standalone injector wizard can't estimate. The first is the initial magnetic charge on the pintle of the injector which opens it (there's absolutely no fuel flowing when this takes place). The second is the time it takes for the injector to move from closed to full open (you'd think it moves very quickly but in electronics it actually takes a long time).

We have several dynamic conditions which effect our fuel injector flow and the answer as to why we don't have 500cc of fuel if we reference our static fuel flow charts at 50%. Instead of having an injector "on" and flowing fuel 50% of the time, and "off" 50% of the time as well, we now know there's some other factors that affect the raw fuel flow of the injectors. The conditions that we know exist are an injector that is "on" but charging and flowing no fuel, "on" and opening but not flowing full flow until the pintle has moved to full bore, and third, "on" and actually flowing full bore. By "dialing in your injectors", that's what I mean... dialing in the dead time compensation.

So now lets apply this to basic EFI principle. Let's say we're on the dyno using AEM and our target AFR is 11.5:1 but our actual AFR is 12.5:1. We'll convert it to lambda and divide our ratios to calculate our injector pulsewidth increase.

.85 lambda for 12.5
.782 lambda for 11.5

.85/.782 = 1. 0869 or 8.7%

so we select the range of cells we need to increase fuel to in the base table by examining their current injector pulsewidths and multiply them by 1.087 to make the correction right? Well, lets take just one cell and evaluate this. Lets say were in the base fuel table and the cell we're working on happens to have an IPW of 3 milliseconds. The existing and target afr's are the same as above and you multiply it by 1.087 so 1.087 x 3 = 3.261. We enter the correction in the cell and for some reason the AFR still doesn't read 11.5:1. Well, when you don't get adequate scaled AFR corrections, that means your injector latency is incorrect and you have to modify the wizard template information.

So let's look at this latency thing again and think about it and see if the corrections are linear to the previous. Lets say it takes .1 milliseconds to process the call, .4 milliseconds to charge the injector and .25 milliseconds to open full bore, well we now have an injector dead time compensation of .725 milliseconds. Let's tie this back into the last fuel correction and we'll see what happens.

3 milliseconds - .725 (dead time) = 2.275
(1.087 increase) 3.261 milliseconds - .725 (dead time) = 2.536

With the dead time compensation the correction is 1.11
Without the compensation it's 1.086

Now we have a 4% error in fuel calculation that will increase exponentially as the injector supplied voltage decreases or increases based on load. That makes for problems with closed loop fuel tables because a 4% fuel compensation error demands a 4% correction.

You will need to adjust your latency table until you get the smallest percent deviation possible before you tune the compensation tables. This is the biggest tuner mistake of all. Lots of cars with hunting idle's, hesitations, and non linear fueling throughout various ranges in the fuel map especially when the compensation tables reference them for lookup corrections. This is also the same reason why you get shitty reliability on on piggyback setups with injectors larger than say 550cc.

Latency fuel corrections are not linear as compared to standard percent fuel corrections. There's a 4% difference in the example I provided and that's just one cell. The smaller the IPW the larger the percent error which means the more time you spend in closed loop (smaller effective injector pulse width) the larger your fuel problems get if you do not compensate for them properly.

Hopefully this answers some questions

[Chatzvr4]

Thank you Jeremy. This was the type of info i was talking about. it could save people money by at least helping them get a base to start tuning from.

[zapman2000]

Great info im getting ready to buy an AEM EMS2 to replace my blown ECU and am just starting to learn about how to tune this is very well explained.