1994 Nissan Sentra SE-R going VEMS, then turbo.

[danmartin]

After my launch Ben goes "No wonder you pull on me so hard in the beginning" 

[rob@vems.co.uk]

All good news then!  Part-throttle and transient tuning is the most difficult and time consuming stuff to get right.

[BenFenner]

I beg to differ. That stuff is a breeze compared to idle control, and maybe even cold start.

From my experiences so far anyway. =]

[rob@vems.co.uk]

I've never really had much of an issue with cold start.  And as for idle control I've often got away with setting my P value getting my I pretty close then shoving a large D number then using the ignition advance idle control to make things smooth.

[BenFenner]

Dan's managed to get his idle at 850 rpms, and very stable/reliable without using the PID control.

I've attempted a similar feat with less success. Part of my problem is that Nissan put three idle controls on my engine. One is constant, controlled by a screw that you set. The other is used to increase cold idle speed, which is a bi-metal device that slowly closes as the two strips of metal warm at different speeds due to the engine heat and constant electrical supply. Then there is the computer controlled idle solenoid that I get to control.

I feel I'm very close to having a great idle. It's okay right now, and does very well at warm temps. Cold idle right now is a mixed bag. My other problem is that my headlights like to drop the idle 150 rpm or so. So when I set things correctly with them off and then I turn them on, it bogs. When I set things correctly with them on, it tends to race a bit when I turn them off.

I know there is a sweet spot somewhere, I just have to find it. It involves getting that screw into a position where it works with the solenoid control, not against it.

I've changed my fuel cut values in Basic Settings to 1300 and fuel resume at 1000. My thinking was that I'd set the idle really high, but just cut fuel above a certain rpm. This worked a bit, but produced an idle that would surge and then dip heavily. So after fiddling for a while I came up with those values above that seem to help idle. They also make the car work much more like factory. With fuel cut set at 3,000 rpms, I'd cruise at 2,500 rpms and then let off the throttle and the idle control would kit in, keeping the engine from dropping rpms as quickly as I'm used to.

All in all my idle is all right. I just need to do some more tuning. If it comes down to it and I can't get it to work as well as I'd like, I'll give the PID control another honest try.

[BenFenner]

Oh, and the part in Idle Settings where you can set cold idle, and warm idle by rpm must be a joke, right? I mean, yes, changing those values can have an affect on idle speed, but really it doesn't seem to have a clue as to what to do. The only way I can set idle speed right now is with the PWM table. Are these rpm values only in use with PID control? If so, why do they affect my idle with PID disabled? And why is it so bad at it?

[rob@vems.co.uk]

Those cold and hot idle speeds are targets, you wont hit those targets without getting the PID spot on.
If you enable the ignition based idle control it will attempt to run the engine at those RPMs

[Tcal]

Nice project going on Ben.


Here´s a log i did take today, no PID control..
And you mean that´s impossible.?? Cold idle is 1100.

 

[BenFenner]

Tcal, do you have a link to your car so I can get an idea of what you're working with?

I'm not saying a good idle is impossible, I'm just saying I'm having quite the bit of trouble with it. =]

[Tcal]

Ben, i´ve been thinking to show my own project  here,i think it´s coming in the future..


My car is 2.0 16 valve turbocharged Opel Calibra. I use it as my daily driver.
I´m having same kind of issues with that idle too. Worst scenario is when IAT is near +50c and that big fan is on with other electrical equipment.
Then idle drops to 600-700 rpm. That PID is probably the answer to that...

[BenFenner]

Yah, it just very well may be.

Mine's a daily as well. =]

[BenFenner]

I just noticed Rob was posting on someone else's question about their idle valve solenoid audibly clicking. His solution was to put the idle valve frequency at 0 to stop the clicking.

Dan Martin and I both had clicking idle valves (initial values were 3 and 255 for his car and mine respectively). We noticed when we put the value at 0 the clicking went away but we thought that's because we'd effectively stopped it from working at all. I found out on my own that the clicking went away with a value of 64, and then Dan suggested 128 which worked as well. Right now we are both running with 128 as our value. Dan's is a stock Ford Mustang 5.0 idle control valve, and mine is a stock Nissan SR20DE idle valve. Will we see any benefit from moving that value to 0? Maybe 128 is better, and should be suggested to others with clicking valves?

Edit: Also, may I suggest a dedicated section for idle control in the "VEMS" section of the message boards? It would be nice to have I think, and allow the centralization of idle control discussion as it seems to be a big enough point to warrant it's own section. I've been trying to use the different sections of the forums for discussion, and I think this would help.

EDIT Nov, 28 2007: Idle Valve Frequency should be put at 0 if you have a clicking idle solenoid. Values 0, 64, 128 and 192 all stop the clicking, and produce the same results, so you should just put the value at 0. The idle valve clicking seems to be very common, so this solution is floating around the forums.

[BenFenner]

Dan and I went out the other night to do some more benchmarking.

We had three cars that were the same exact speed before we started our VEMS projects. The cars are:

Car                                Modifications                                          Owner

1987 BMW 325is                 Stock                                                    Dan Martin
                                      93 octane
                                      170 lbs. ballast (passenger)

1994 Nissan Sentra SE-R     Short ram intake                                      Ben Fenner
                                      Base ignition timing advanced 4 degrees
                                      93 octane
                                      170 lbs. ballast (passenger)

1994 Nissan Sentra SE-R     Stock                                                    Dan Fenner
                                      87 octane



Now we can use my brother's SE-R to benchmark our cars, and see how our WOT tuning is going.
The verdict is in, and I pull well enough on my brother's car. It's to the point where I think only very minor modifications to the air/fuel ratio will take place, if any. I have a video of the run I may post, but it's very dark (of course) so we'll see if I get around to it.

Check Dan's thread for his results.

[BenFenner]

Still working on idle and cold start. Idle is pretty good, and I'm getting the hang of how it works. I think I'm very close, and I'll post results once I feel I have numbers that I'd recommend.

Cold start is another issue. I'm starting to feel that upping or lowering the fuel amount is not enough. I think I just need more of everything, air as well as fuel. So I'm going to try opening the idle control much more when the car is cold, and see how that pans out. I certainly have many things I can try, so I'm not hitting a brick wall.

Stay tuned.

[Tcal]

That Cold start tuning gets much easier if you heat up that O2 sensor before starting..
Go to terminal program, Type Manmde02, that commands O2 to heat. When heated, turn datalogging on and start.
Remember, if you heated that lambda manually then it wont activate closed loop before restarting vems.