RB21DET Idle Tuning on AEM Infinity

Hey everyone,

I am in the process of tuning my RB21det that I built (yes it is a 21 not a 20) (also first tune I have ever done) and am having an issue with getting a stable idle. It is an aggressive build and I am thinking it is likely due to cams and large throttle body. Attached are a couple pictures of the idle table and my DBW setup. Can anyone tell me why my target idle is so high when my idle target table is showing it should be targeting an 800 rpm at the operating temp it was in? I have verified timing is synced correctly and all sensors are producing reasonable readings. This setup produced 377HP on 91 pump gas when tuned on my Power FC ecu but I didn't like the drive-ability of it and decided to upgrade to the Infinity. Thanks for your help! This program has been fantastic so far.

My setup

RB21DET

Tomeii 260 9.25 lift Intake and Exhaust Cams

LS3 DBW TB and Pedal

GT3071R

Solid Lifters

Port and Polish Head

ID1000 Injectors

LQ9 Truck Coils

3" Tanabe Turbo Back Exhaust

Adjustable cam gears set to 0*

First, my compliments for taking the plunge on drive by wire. When you get used to it you realize how much simpler it is than a throttle cable. Now what is that Idle TPS offset in the second screenshot? looks like it is adding to the target idle rpm based on the throttle angle?

Also, the first thing to do is to turn off all the feedback altogether and see what idle TPS percent is needed to maintain a stable idle with no accessory load (assuming no idle ignition feedback either). So use only feed forward/open loop control and get the idle base position tuned. Then you slowly add in the feedback features.

At a quick glance I'd say the reason your target Idle speed is sitting at 1200 is twofold. Firstly the ECT is sitting at 140 F so the base idle target at that temp is 1000 rpm. Secondly, this is also being modified by the values in your IdleTPSOffset table. In the picture you can see this is adding 200 rpm to your target. If you zero out that table during your initial setup you should be able to tune the idle target based solely off the IdleTargetTable.

At a quick glance I'd say the reason your target Idle speed is sitting at 1200 is twofold. Firstly the ECT is sitting at 140 F so the base idle target at that temp is 1000 rpm. Secondly, this is also being modified by the values in your IdleTPSOffset table. In the picture you can see this is adding 200 rpm to your target. If you zero out that table during your initial setup you should be able to tune the idle target based solely off the IdleTargetTable.

Thanks Raymond and Andre! I appreciate your feedback. I was able to zero out the IdleTPSOffset table and the target idle did drop to 1000. I then started the process of initial idle tuning listed in the AEM Manual. In the process it directs you to set the entire target idle to 2000 and base throttle table to 50 and adjust base throttle up until idleFB reads -3% to -8%. I am able to get the car to run smoothly at 2000 rpm however I am struggling getting the feedback to get near those numbers. I am suspecting that since the TB I am using is 90mm and as it is on a smaller displacement motor that the 5% throttle position is too high to idle much lower without a lot of FB. I also am seeing a correlation to the DBW Idle Control. When I run it at a lower percentage say .3% or .4% I gain better control but not enough to get the base position low enough to lower the rpm without using the FB loop.

I am thinking of reducing the resting throttle opening to say 2% and see if that helps me gain more control.

Any suggestions?

On a stock tune with electronic throttle you willtypically see the spark retarded from MBT at idle to intentionally reduce torque. With retarded spark, the base position of the throttle has to increase. Here's the benefit of that: when an accessory load or other drag to the engine comes on, the ECU can use spark feedback first to keep the rpm from dipping. This is because spark responds faster than throttle. Then the throttle will "catch up" to allow more or less air in as accessory load switches on and off.

So in your situation, here is what I would do. Again, turn the feedback off completely and ignore what AEM recommends for a minute. You need to map the relationship between engine speed, spark timing, throttle angle, and accessory load. So take your current spark timing which is 15 degrees or whatever. Move the throttle angle in 1% increments, then retard spark 2 degrees and do it again. Take a log for each one and make a chart. So your test plan looks something like:

Spark - throttle - rpm

15 - 5% - ?

15 - 4% - ?

15 - 3% - ?

15 - 2% - ?

13 - 5% - ?

13 - 4% - ?

Et cetera. Give it a few seconds to stabilize each time you change something. Adjust the range as needed - you get the idea. Do it again with A/C on or headlights on or whatever. You don't necessarily need to do every single combination here, use good judgment to save time, but you want to understand that relationship. Then when you can make a nice chart of spark vs thottle vs idle rpm, you will have a very good idea of how to set up your tables so that the thing doesn't stall when an accessory load come on, or the feedback doesn't go nuts because the base spark and throttles are too far off.

I know it sounds like a lot of work but you can do it quickly, you're just sitting there at idle (also be a little mindful of heat soak). Just pop rpm into a spreadsheet as you change spark and throttle.