Tom M'Guinn

Yeah, sorry, I guess I was asking a rhetorical question (which are rarely helpful online...). Your comment about "setting the ceiling" with the 108 octane limits seemed vulnerable to misinterpretation as Doug and others try to figure out how much timing to run -- so I should have just said that.

67King

No sweat, Tom. Folks are learning, so I think they appreciate your bringing it up. I enjoy talking about what I used to do, as well, anyway.

Dougs951S

I just learned today after giving siemens a call that I had my injector dead time way off, at a supply voltage of 13.2v the dead time is more like .88ms vs the 1.14 I had programmed in. Thats a substantial difference. I am completely sure this is why I've had trouble getting the car to idle well at stoich. I'll have to retune the VE table and acceleration enrichments but I'm guessing the car will run a lot better now. I would like some help in understanding exactly how that programming error was effecting my injector performance in real life. By telling the software the opening time was much longer than it actually was, was the ECU lengthening the pulse width compared to what it actually should have been to supply the correct fuel? I understand how that effects my fueling but not how it effects the linearity of my injectors with respect to very low load conditions.

67King

Well, there is a lot of error in the non-linear portion. So ideally, the non-linear portion should be a very small portion of the time the injector is open. Just to throw meaningless numbers out there, if you get 90% of your fueling from the linear portion, only 10% of it is highly unpredictable. But, if you get 50% of your fueling in teh linear portion, that means that you are essentially unable to control 50% of your fueling.

One thing you could do to improve idle is add torque reserve. That is a lot of retard. You'll burn more fuel, but what you do is retard timing such that the engine is less thermally efficient, and you have to be moving more air through it to overcome internal friction. In industry, this is done for the purpose of adding robustness to stalling from unexpected torque hits.......you can add timing more quickly than you can adjust throttle (or the idle bypass valve), so it makes things less perceptible to the driver (things being A/C compressor cycling, steering input -> increased load from the PS pump, etc.).

Dougs951S

Despite being rich, the car idles really strong when warmed up, I can easily take off from a stop just by letting the clutch out. Its just the cold start that is bad, the idle is smooth but it lacks torque, its extremely easy to stall the car until its been running for a few minutes. I dont run an idle valve and dont think I should have to, the car ran perfectly on the factory dme with no idle valve. I understand what you are saying, its like a choke almost. I can slightly raise the idle via the screw to bypass more air and retard the timing, problem is im afraid of high idle egts and that this will result in too high of an idle warmed up. Will that not be the case? So if I understand you right, for a given idle speed of say 1000 rpm, and assuming afr is the same, the motor will make more torque at that speed with more air and less timing vs less air and more timing?

67King

Well, the car came with an idle bypass valve. I forget what Porsche called it. But it serves various purposes, one being to ensure that the idle speed won't vary in different conditions. Customers will complain if they see things change, as one example.

You are right in that EGT's will go up with more retard, but at idle, the mass of air you are moving isn't sufficient for the heat to make much of a difference.

At the same idle speed, the engine will make the same torque regardless. It will make enough to reach equilibrium with the torque required to overcome friction in the engine. I got off on a bit of a tangent with the torque reserve stuff. What I was saying is that the engine can react more quickly to a change in load (e.g. A/C compressor kicks on, driver winds steering rack all of the way to a bump, etc.) by adding timing to maintain idle speed than it can by adding airflow. Was just saying that that is one strategy you could take to maintain a more smooth idle without raising your idle speed.

Dougs951S

hmm, interesting food for though as always. I guess I should note that my car is a track/street car, so its got a whole mess of stuff deleted like the ICV, power steering, A/C, and most of the electronics. The megasquirt firmware does support cold idle advance and I can get the car to idle perfect on a cold morning crank by having the ecu add timing, the problem is that there is no rpm cutoff for this, only a temperature curve, so it applies the additional timing to the whole rpm band which makes me a little uncomfortable driving around on a cold motor with 4-5* more timing than it needs. It almost sounds like its just the firmware that is limiting me, maybe I'll need to fab new post intercooler piping and add a ford idle control valve. I was hoping to avoid that mess, they seem to always leak and my motor has always run better once warmed up than when it had the factory ICV. Or a manual choke!


update: After inputting the new injector data and retuning everything....the car runs exactly the same. It still doesnt like to idle any leaner than 13:1, has a sucky cold idle, and runs perfect once warm. I think I'm just going to leave this one alone and know when to pick my battles. I'm not concerned with chasing numbers.

Thom

Sounds like your warm-up parameters are off. Set your idle table in open loop before switching to closed loop. Redo the fueling in the same time but make sure closed loop lambda is off during the warm-up phase.

Dougs951S

The car runs open loop all the time. I really just think the engine isnt getting enough airflow through it to idle well when cold without idling too fast when its warmed up. I have to set the air bypass according to fastest attained hot idle speed, since without an ICV i rely entirely on the idle screw. I think I'm just going to have to bite the bullet and add an idle valve so that it gets proper airflow at all engine temps.

Thom

Use the stock ISV. Stepper motors are a pain in the back.

Pauerman

If your MS supports 3 wire ICV, I would recommend that you run the factory piece like Thom has mentioned. This would solve all your cold & hot running conditions. Instead of adding a timing offset to your complete timing table for warm up, idle speed can easily be tailored based on coolant temp.