Hunting idle made worse after Idle control valve cleaning

[Paul Ri-Chard]

I decided to  clean my '98 Throttle body and idle control valve following Advise on Pedro's Garage and Pelican Parts because of a intermittent idle problem, when the LOW idle would sometime cause a stall.

When I put things back together, the idle now has the opposite problem, hunting from 1100 to 2100rpm's. I replaced the ICV gasket, and the seal between the  throttle body to the  'T" plenum.  I'm 99.9% sure the rubber seals between the 'T" and the plenums are seated correctly...BUT I noticed two potential problems upon inspection and re-assembly. Number one is that both rubber connector seals has has a port that opens to the outside. the right side I believes goes to small black tube that goes to a check valve below the right intake plenum. The left side goes to ... no where... (creating a nice leak).  Does anyone know where the left side port connects to ? Is that factory Porsche or do I have it installed wrong ?

Second problem is that I noticed  some fine cracks on the ICV  itself. When I was cleaning it with carberator fluid, these would actual weep a bit. Does that mean I need to replace the ICV ?  

[flyingpenguin]

Opened up the engine bay on my 99 Boxster to verify your issue.

 

The hole in the rubber sleeve on the passenger side should have a piece of vacuum hose (27 in the diagram below) that routes to a Y piece underneath there that connects to the vacuum tank. The hole on the driver side has a vacuum hose (not shown in this diagram) that routes to the fuel pressure regulator right by the AC compressor.

 

 

As far as your ICV goes. It can be disassembled by removing a hollow pin and then twisting the top off the bottom part. In my case the pin did not survive the operation and was replaced by a phillips screw.

 

 

Mine had also cracked and a bit of super glue did the job.

 

 

I'd be surprised if your hunting idle wouldn't disappear when reconnecting the vacuum hose to the fuel regulator though. Keep us posted.

 

Joost

[Paul Ri-Chard]

Joost,

Thanks for actually spending your time to open your engine bay to verify my problem. Who does that anymore ? Mucho gracias!

I will take a look again and retrace the drivers side port. I certainly don't remember seeing any loose vacuum lines on that side. Perhaps my blind spot had enlarged that day causing me to miss it. I do have glaucoma so it's possible. I temporarily plugged that side with super sticky 3m tape this am before I went to work and it helped quite a bit. Still idles high 1020rpm and occasionally hunts up to 1200, but the Rpms don't fall quickly between slow ( lazy) shifts.

The hollow pin on the ICV... Did you drill it out ? Will I have to worry about damaging the electrical side of the ICV if I do that ?

I will try Q-bond or super glue to see if I can seal the cracks. Sunset Porsche quoted me $461 for a new Bosch ICV. Ridiculous !

You can purchase a Kia ICV in the same exact physical configuration but slightly different inductance values for about $35, not that I would substitute that. I'm not a Louie Viton type of guy, but I don't always shop at K-Mart either.

[mikefocke]

See if you can find the Bosch part number equivalent.  It isn't like Porsche made the all the parts or even had them made in a special design back in that era. They were mostly using off the shelf parts. (I've seen parts that still had the Bosch part number stamped into them. Don't know if this is true on a ICV.)

 

So I googled "Boxster idle control valve bosch" and seems you can reduce the price by ~50%,

[Paul Ri-Chard]

Thanks Mike. I did find it lower on Amazon, same part Bosch part # 0 280 140 572, porsche # 996.606.160.01 for about $161 free shipping.

[flyingpenguin]

Gotta love amazon!

 

Drilling the pin out is still worth the try though. The ICV operates on an electromagnetic signal, so not very sophisticated and the both units I fixed this way survived the operation. I have an intake with fixed but working ICV for sale.

 

See here: http://www.renntech.org/forums/topic/47455-garage-sale-many-porsche-parts-available/

Item #15. Make me an offer!

[Paul Ri-Chard]

thanks Joost. I found the small vacuum hose on the left side. It was tucked down there out of sight. Once I reconnected it, it purred like a kitten at about 750 rpm... For a while. I noticed that the hose seemed a bit short so it might have pulled out of the rubber sleeve again. So I will look tomorrow to see if that is the case. In the mean time, I've posted some views in case some other buffoon makes the same mistake that I did. Even the diagrams from the Porsche dealership had very poor sketches for the left side. So I hopes this helps. First picture shows location of fuel pressure regulator.

This is connected to a short (2") curved section of approximately 5/16 hose that then connects to a small vacuum hard plastic vacuum line that routes to the front then underneath the left intake manifold.. These 1/8" hard plastic lines (Look at my fingertips on the blue gloves) have a shoulder that is supposed to sit flush against both ends. It's important to position the nipple on the rubber sleeve on the intake manifold in the right orientation. If you have it flipped 180, your vacuum hose will not seat well and cause a vacuum leak. The last photo shows the connection to the left ( driver's side) but viewed from the cockpit behind the driver's headrest.

[Paul Ri-Chard]

Opps this is the last picture of the hose connected to the rubber sleeve on the intake plenum. This view is from the cockpit, looking over the headrest of the driver's seat.

Edited October 30, 2014 by Paul Ri-Chard

[Paul Ri-Chard]

Well the idle problem is not resolved: Here are my car's 'symptoms' :

When I do a cold start the car idles evenly at about 850-900 until it starts to warm up to its operating temperature. Then the idle increases to about 1140-1200rpms but never returns. I replaced the ICV with a new one... yes the part # was correct.  the vacuum lines appear to be all correct except I never had a non-return check valve as the picture to the side (it ran better than it did now but with occasional stall).

My Durametric gives me codes P1602 =  Supply Voltage Open and P0506 = IACV control problem.

Any help from the group ? 

[JFP in PA]

Well the idle problem is not resolved: Here are my car's 'symptoms' :

When I do a cold start the car idles evenly at about 850-900 until it starts to warm up to its operating temperature. Then the idle increases to about 1140-1200rpms but never returns. I replaced the ICV with a new one... yes the part # was correct.  the vacuum lines appear to be all correct except I never had a non-return check valve as the picture to the side (it ran better than it did now but with occasional stall).photo (55).JPG

My Durametric gives me codes P1602 =  Supply Voltage Open and P0506 = IACV control problem.

Any help from the group ? 

 

P0506: IAC valve is sticking, air cleaner is blocked.

 

P1602: DME fault, no power at terminal 30, wiring interrupted, connector loose or disconnected.  When this power is lost, even temporarily, adaptation values are lost.

[Paul Ri-Chard]

The ICV is  New, less than 40 miles, so  unless the new ICV is sticking,could the code could mean something else ? Also can a disconnected battery cause it to fault  to P1602 ? I was doing some work on the seats recently < 50 miles ago and disconnected the battery so I wouldn't  trip the air bag light.

Also maybe I didn't wait long enough  for the DME to 'teach' the ICV the new values. How long is a typical  adaptive 'break-in' period ?

Do I need to clear the old codes first ? Before the new adaptive values can be re-learned ?

[JFP in PA]

The ICV is  New, less than 40 miles, so  unless the new ICV is sticking,could the code could mean something else ? Also can a disconnected battery cause it to fault  to P1602 ? I was doing some work on the seats recently < 50 miles ago and disconnected the battery so I wouldn't  trip the air bag light.

Also maybe I didn't wait long enough  for the DME to 'teach' the ICV the new values. How long is a typical  adaptive 'break-in' period ?

Do I need to clear the old codes first ? Before the new adaptive values can be re-learned ?

 

If the DME code is real, it should return after being cleared. Disconnecting the battery could trigger this code, but usually does not.  I would clear both codes and see if either returns; if they do, you have some DME wiring diagnostics to do. If the DME loses that connection, it also loses all of its adaptive data, like the idle control, every time the power goes off.

 

The exact time it takes the system to relearn often varies, but it should have completed by 100 miles.  You can trick the car into relearning by inserting the key and moving it to the on position without starting the car.  Usually, you will hear a series of clicks, which is the ICV going through its routine.  Leave the key in the on position for about 30-60 seconds, then turn it off and remove the key.  Wait a moment, then insert the key and start the car, if everything else is OK, the car should idle.

[Paul Ri-Chard]

Thanks JFP.  I used my Durametric to clear the codes and  left the key on the on position for 1 minute. My little base Boxster ran a little better today but consistent idle at < 800 rpms hasn't happened yet. I'll drive around a bit this week and get some miles on it to see if the P1602 code returns  and/or if the idle smooths out. 

By the way if I have to resort to tracing the  terminal 30 wiring, is there a logical way or logical pathway on how to proceed ? I looked at my DME-ECU and not even sure  on how to disconnect it properly, or where terminal 30 is located on the DME-ECU. Do I disconnect my battery when removing the DME ?

Thx, Paul Ri-Chard

[JFP in PA]

Thanks JFP.  I used my Durametric to clear the codes and  left the key on the on position for 1 minute. My little base Boxster ran a little better today but consistent idle at < 800 rpms hasn't happened yet. I'll drive around a bit this week and get some miles on it to see if the P1602 code returns  and/or if the idle smooths out. 

By the way if I have to resort to tracing the  terminal 30 wiring, is there a logical way or logical pathway on how to proceed ? I looked at my DME-ECU and not even sure  on how to disconnect it properly, or where terminal 30 is located on the DME-ECU. Do I disconnect my battery when removing the DME ?

Thx, Paul Ri-Chard

 

Porsche manuals will tell you to use their pin out box, which is both ridiculously expensive as well as near impossible to find.  You will need a multi-meter and a set of wiring diagrams to trace what you need to look at.  If you can borrow a 986 Bentley manual, there are a full set in the rear of the book.

 

You should not have to remove the DME to do the testing, just unplug it.  There are step by step testing instructions for the P1602 code in the OEM OBD II diagnostic manual, which is another impossible to find book.  When I get a chance later today, I will try to summarize them for you (shop is a little jammed up at the moment).

[JFP in PA]

OK, the dust has finally settled a bit.  The following is the diagnostic tree for the P1602 code.  You will still need access to a wiring diagram to make sense of the wire connector description's.  The special tool mentioned is the pin out box, but armed with determination, you can do this with just a multimeter.

 

[Paul Ri-Chard]

Thanks, I will get my trustee voltmeter out on Saturday. I'll do a fresh post to see if any forum members in my area would want to loan their Bentley manuals out for the weekend.