944NA starts and stalls after work under intake manifold

pjs.oregan · 02-27-2016, 07:16 PM

Car is an 84 chassis with an 86 motor swapped in.

I have just finished a whole lot of 'tidy up' work under the intake manifold - fuel injector reseal, throttle body reseal, replacing vacuum lines, cleaning ICV and replacing ICV hoses, cleaning grounds, etc.

The car now starts perfectly fine but will not hold idle at all - it instantly stalls unless I hold the throttle, and it revs irregularly (seemingly out of sync with the pedal) if I do this.

I have checked all vac lines, swapped DME, disconnected one injector (in case of FPR failure), and tried disconnecting the TPS.

Any ideas? I have that sinking feeling...

dgcantrell · dgcantrell's Garage

You'll need to adjust the Throttle Position Sensor. Apparently it is out of adjustment and the DME isn't reading that the throttle is closed. I'm sure someone will be along shortly with a link on how to adjust it. It's a very simple operation though.

DannoXYZ · 02-27-2016, 10:47 PM

You may also have introduced a vacuum leak somewhere between the AFM and throttle-body. Check all the intake plumbing and hoses.

Also make sure that the ISV is not connected backwards. There's an arrow on the body.

mytrplseven · mytrplseven's Garage

Easy to miss vacuum fitting underneath the J boot. Also verify the two vacuum lines on the TB are at the proper port.

pjs.oregan · 02-28-2016, 03:56 PM

Hi! thanks for the help.

I have checked and double-checked all vac lines, including the one under the J-boot. Swapping the vac lines to the throttle body makes no difference.

ISV is connected with the arrow pointing to the short S-shaped hose that connects to the intake manifold (I believe this is the correct orientation).

The TPS has an audible click with throttle closed - is this enough to say it is reading the closed throttle correctly?

Any more ideas?

DannoXYZ · 02-28-2016, 05:51 PM

Electronically test the TPS. Just because it clicks doesn't mean it's actually closing the idle switch. Use voltmeter to test continuity of idle switch. Should have zero resistance in idle position, infinite resistance (open) just off idle.

Rasta Monsta · Rasta Monsta's Garage

First, it isn't really possible to do a visual check for vacuum leaks. Take it to a shop and have the intake smoke tested. Second, there is a procedure on Clark's to adjust idle air bypass screw....you may want to double check that.

M. Murphy · M. Murphy's Garage

Very similar experience last year on my 83 NA.
Will the engine run off idle? If so, my money is on a large vacuum leak under the manifold. Mine was the large line just inboard from where the dipstick is (a real pain to get to) it had slipped off the nipple.

thomasryan · 03-01-2016, 04:50 PM

Sometimes I will step back from it and repeat the entire process to see what I messed up...

....so I am thinking that you had many small vacuum leaks over the years and the bleed screw may be closed to much. If your AFM wiper pile isn't scored and the tps is set, do the frwilk jumper on the iac and see what happens.

It's always good to leave just a little slack in the throttle cable so the micro switch in the tps engages well.

pjs.oregan · 03-01-2016, 09:58 PM

TPS tested - seems to be behaving as it should.
Engine does run off idle but only at more or less wide open throttle >4000RPM or so.

Can't really adjust the idle air bypass screw when the car won't idle at all

I think I'll have get the intake off again and retrace my steps, re-re-re-re checking vac lines etc.

djnolan · djnolan's Garage

If it won't idle the first thing to suspect is a leak or improper assembly in one of the large vacuum lines(ICV, ACS, brake booster). Make sure they are all hooked up and clamped, including the fitting up under the intake near the dipstick tube. Also make sure none of the plastic fittings are cracked such as at the brake booster or the y-tube venturi. There should be a restrictor in the line between the y-tube and the j-boot. The y-tube is two pieces and is joined by an o-ring that could also be leaking. Also make sure icv is installed in proper direction.

Here is a DIY smoke test you can also run using fluid pump and a cigar:

dgcantrell · dgcantrell's Garage

Ummm.. Check to make sure you plugged the AFM back up. I did this one time. The connector fell down behind the meter and a I never realized that it wasn't plugged in until I finally noticed it after banging my head for a couple of hours..

M. Murphy · M. Murphy's Garage

The way I located my leak was: I sat in the car keeping it running while a friend used an incense stick around the various areas of the engine. He finally saw the smoke drawn under the intake manifold. It was pretty close to the leak area. Just an idea.

pjs.oregan · 03-02-2016, 10:05 PM

Thanks for the ideas guys... I'll try a low-budget smoke test, but I'm a little skeptical that it's a vacuum leak because the car had BIG vac leaks before I did any of this - the FPR / fuel damper line was just completely open, and the only symptom was a high & hunting idle. Plus I can see all the large hoses and am 100% sure they're where they should be. Nonetheless worth a go.

AFM is definitely plugged in but boy I wish it was something like that, haha

djnolan · djnolan's Garage

A small vacuum line disconnected would be noticeable but the car would still idle as the computer would compensate. However if one of the large lines is open or not connected properly, the computer will not be able to compensate and the car will not idle. There are many seals, lines and clamps to be concerned about:

1987 Porsche 944 Base Coupe - Intake System - Page 2

M. Murphy · M. Murphy's Garage

I would agree with DJNOLAN. I felt the same, did not want to consider a large vacuum leak because I just did not want to take the intake manifold back off. On my engine as I understand it the line that came off was the one that provides additional air when the A/C is turned on. During re assembly I found the nifty German hose clamp pretty much impossible to tighten in the blind confines. I figured just pushing the hose onto the nipple would be good enough. Wrong!
What I did not take into account was the natural spring of the hose, oily surface and vibration. All these conspired to disconnect the hose five minutes into a test drive, no idle.
Replaced the clamp with a common one with a hex fastener I could deal with blind, problem solved. BTW the hose in my problem has an internal diameter nearly 1/2 inch.

pjs.oregan · 03-03-2016, 07:35 PM

M.Murphy you have hit the nail on the head! The car lives again; the ICV-to-manifold hose had come off the manifold... so much for being 100% sure they were in the right place. I thought I could see it going into the manifold but I was wrong.

Got everything off again and clamped it nice and tight, and everything is as it was, and I am a little bit wiser and humbler. Thanks everybody for your help and ideas!

Sadly the idle is still all over the place, which was the reason I did all this to start with...

M. Murphy · M. Murphy's Garage

Thanks, great to hear it helped.

dgcantrell · dgcantrell's Garage

Glad you got it sorted out, man. It's hard to see that crap when you get it back together.

M. Murphy · M. Murphy's Garage

Here is something to consider. The small vacuum lines on these cars are hard plastic with rubber connectors and plastic tees. With time and heat the lines crack ad are pretty difficult to troubleshoot. Lindsey Racing sells a silicon kit with metal tees, I bought a kit for mine and it was money well spent. The silicon handles the under hood heat very well.