Idle speed adjustment tricks on 964

[andreas naeslun]

Hi,
I have an older 911 with a modified 964 based 3,8l (mated to a G50) with LWF, RSR cams, headers, ported heads etc. which makes it a great car to drive fast, but idling at 890 rpm which I believe is the set target value in the original DME idle algorithm for cars with a manual transmission, is a little rough.

Raising the idle can be done in different ways:

One way is to adjust the sealed stop screw in the idle servo valve, as described elsewhere in this forum. This however mechanically limits the lower end of the regulation loop, effectively putting the closed loop idle regulation out of play once the motor is hot.
A more sophisticated way to raise the idle would be to change the target/set value in the DME, which can be done if you have the computer equipment, software and competence needed (I do not unfortunately). I know of at least one local company here in Sweden that has hacked the Bosh code, which means they can set pretty much everything in a Motronic system if you give them some of your cash.

I have an idea for a simpler way to raise the idle one notch and yet retain full idle servo regulation:

There are two jumper-connectors on the DME harness sticking out close to the DME main connector.

One three pin connector controlling mainly lambda closed loop feedback on or off.

The other connector, which has two pins(holes actually) will if jumped tell the DME that the motor is hooked up to a tiptronic automatic transmission.
The only difference that I am aware of (by rumor) after jumping this connector is that the target/set value for the idle algorithm is raised slightly to avoid stalling the engine when going into drive from neutral or coming to a stop while staying in drive in a tiptronic equipped car.
I am not sure if the idle control loop algorithm is changed in any other way or how this will affect idle in a manual car.

Has anyone on the forum tried to jump this connector in a manual car?
What will the higher idle set value be? (i stay clear of tiptronics)
Does the jumper lead need a resistor, or is it just a straight cable?
Does the enabled "transmission jumper" make any other alterations (such as idle algorithm) to the DME apart from raising the idle? If so, what affect will it have if it is applied on a manual car?
Maybe the tiptronic algorithm is tuned to the weight of a torque-converter and will not work well with my LWF?

The main thing I would like to know before trying this on my car is the resistance of the jumper if any. I do not want to fry my DME if I can avoid it.

I am curious on your feedback to above idea.

/Andreas

[Steve W]

If you change the jumpers to the Tiptronic settings, you will reduce the performance of your car, because you will be telling it to reference the Tiptronic ignition maps, which are retarded in the low rpm and low load region to make shifting smoother for an automatic. If you need to raise the idle or stabilize for a LWF, it needs to be addressed in the chip program, and raising the idle speed is not the answer. Porsche used a different idle algorythm in the factory Euro Cup cars, which involved tweaking the fuel injector cycles and revising the ignition timing. When we tested the coding in non cup 964s, solved the stalling, stumbling and drivability issues from installing a LWF.

[andreas naeslun]

Steve,

Thanks for your answer. It seems that the tiptronic jumper is not the way to go.

I currently use the RS maps which makes the idle stable with LWF and driveability is reasonable even at low rpm's. The only thing I would like is to raise the idle with 100 rpm or so because of the long duration "RSR" cams that I use.

Next idea:

How about the AC input to the DME? (telling the DME that AC is activated)
I do not have an AC installed in my car. Wouldn't activating this DME input raise the idle slightly?

/Andreas

[Steve W]

I never tried it, but you could try it and see. The chip you have, if it's any version of a factory chip, is not really programmed in anyway special for a LWF. The only chip that has any special programming for stabilizing a LWF and for a smoother idle, is the Euro Cup car chip, and those chips will not work in a standard 964 DME as they use a special DME. I think this was Porsche's way of preventing people from snagging for copying the Euro cup chips and installing them in standard 964s. But with a Motronic programmer that knows what hes doing, that should not prevent the programmer from modifying the Cup Car program to work in a standard 964 DME.

[kang]

I have heard that raising the idle by telling the DME that your AC is activated raises it too much. I’ve heard that this gives you an idle of around 1,500 or so. If you actually had an activated AC, the drag of the compressor would lower the idle to the appropriate range.

I haven’t tired this, because I don’t want an idle at 1,500. But I don’t have any confirmation that what I heard is true, except that the source is generally considered reliable. If you try, let us know how it works out.

[bazar01]

My compressor clutch connector got disconnected while the a/c was ON. My idle stayed the same. Perhaps the idle control valve maintained the set RPM.

[andreas naeslun]

Thanks all for your feedback.

Maybe the AC input just changes other parameters such as low RPM ignition advance?

Raising the set value to 1500 sounds unlikely since the whole idea of an idle control servo loop is to try to maintain an even idle regardless of outside influences. 1500rpm is also close to (or above?) the fuel cut-off limit, which would give a very fluctuating idle if the idle micro-switch would be actuated.

My idea was that maybe the set value may be raised just a tad, let's say by 50-100rpm, to give an extra margin for untimely stalls.

This is what I will do:
I will install a switch in the cockpit for the AC input circuit, so that I can turn it on and off manually and record any differences.
I will go on four weeks vacation starting today, so my feedback on this matter may take a while.