Pelican Parts Forums

Pelican Parts Forums (http://forums.pelicanparts.com/)
-   Porsche 911 Technical Forum (http://forums.pelicanparts.com/porsche-911-technical-forum/)
-   -   At wit's end with high idle (http://forums.pelicanparts.com/porsche-911-technical-forum/355202-wits-end-high-idle.html)

mhackney 07-02-2007 03:21 PM

At whit's end with high idle
 
I have been trying to diagnose my 70 S's high (1600rpm) idle for what seems like ages. Everytime I discover something that I think "aha, that's it!" I turn out to be wrong! Case in point: Of Mice & MFI where I discovered a dead mouse in the hose running up to the MFI thermostat. Surely that would be the problem! But no, the idle speed is still high and the thermostat seems to be working fine. I measure the temperature on the thermostat housing at 150 degrees F when the engine was warmed up to 180.

I've replaced air filter, oil, oil filter, plugs, points, set dwell, checked throttle and mfi linkages (I had a problem there also where 1 linkage was too long and was preventing the throttle from closing all the way), adjusted valves, checked throttle air flow. I even did a Q&D advance curve which looks reasonable - except that I was not able to go down to 950rpms.

The good news is I've found and corrected a lot of little problems. But, I still have not solved the fast idle problem. The engine responds well when driving. Sudden decellerations sometimes results in some "rumbling but no back fires. I have not touched the MFI. I have read and tried to follow the Check, Measure, Adjust proceedure.

I really need some guidance on how to proceed before I go crazy!

Cheers,
Michael

tcesar888 07-02-2007 03:51 PM

Hi Michael,

I have a '78 SC. I had similar problems with my idle (more so upon warm-up). It was narrowed down to a faulty port on my WUR (Warm Up Regulator). Honestly I don't know if your model year has one...? Anyway, it may be something to look into; its easy enough to test... Hope this helps.

WUR Issues...

Good luck!

Tom

sig_a 07-02-2007 04:11 PM

Seems like you checked all there is to check, but perhaps thoroughly cleaning, lubricating and adjusting throttle linkage cable and bell crank where, I presume, it attaches to the driver side of the gearbox. In my case, the high idle would be temporarily solved by slightly pulling back on the top of the accelerator pedal. As a result of this temporary fix/diagnosis, I disassembled, cleaned and adjusted the entire linkage system from where it exits the tunnel to where it reattaches to the bell crank on the side of the transmission and then beyond that to the linkage leading thru and to the engine compartment. It took two attempts and has been fixed for months. Hope this helps.

mhackney 07-02-2007 04:16 PM

Thanks for the replies. sig_a, I have also cleaned and relubbed all of my linkages and the bell crank. I know that the problem is down stream of disconnecting the accelerator linkage from the throttle cross bar because when I do that, the high idle still exists and pulling back on the throttle crossbar does not have any affect.

cheers,
Michael

304065 07-02-2007 05:53 PM

Michael,

I've been following your issues. I've been chasing a 2000 rpm idle for about three years.

Set the advance to TDC (zero degrees advance) at idle. While this might not give you the required 35 at 6K, for diagnosis purposes it will help. The trouble is that the slightest overspeed kicks in the advance and that causes the idle to rise even higher.

The only other problem is that your throttle bodies are pulling too much air with the plates closed. Have you removed the throttle bodies to verify that the wear isn't excessive? It takes about two hours the first time and about 45 minutes when you have done it 10 times like I have.

The ONLY other reason besides advance you could be idling too high is too much air being pulled through the TBs at idle. This is so basic as to be axiomatic, yet it took me a lot of thinking to realize. Either dial back the air corrector screws or adjust the throttle stops for each plate. I made a homemade flow bench out of a shop vac and used my STE synchrometer to set the plates to approximately the same angle before installing them. Now that my idle is high again, I'm going to check the advance and then if that's not it, remove the TBs and zero out the plates.

Good luck! I plan to do this task on Thursday so I should have some Kung Fu to share next weekend.

mhackney 07-02-2007 06:01 PM

Thanks John. So, since I idle at 1600rpm, should I set the advance at TDC at 1600? I'll reset the advance again and then proceed with the throttle body dismantle. I am relatively lucky in that Eurometrix/Matt Blast is about 2 miles from me. If something is up, I can get them R&Rd.

cheers,
Michael

mhackney 07-02-2007 06:06 PM

Forgot to mention, what I actually did to set the timing was to pick off the advance from the advance curve at 1600rpm (which is +5 degrees) and set the timing. If you look at the advance curve I measured (link above) you can see the results. But the idle had not dropped after setting. But now with adjsting the throttle linkage that was out of whack and removing the mouse plugging the thermostat hose, I need to check and reset the timing.

cheers,
Michael

mhackney 07-02-2007 06:09 PM

Also forgot to mention, I did get new gaskets for the throttle bodies so I am all set to give that a shot. Sounds like a good project for the 4th!

cheers,
Michael

boxster03 07-02-2007 06:15 PM

Grady would be the man here to get all of the issues on the table that could be a problem. I would concur that too much air could lead to this problem which could be something as simple as a bad/worn gasket.

Howard M 07-02-2007 06:36 PM

I strongly recommend taking apart the distributor to clean/lube the advance weights and their bearing surfaces, if this hasn't already been done.

Howard

mhackney 07-03-2007 07:15 AM

Thanks Howard. I have not taken the distributor apart. A little hesitant to do so! But maybe I should just go for it. I was thinking of sending it in for a rebuild just to rule that out.

cheers,
Michael

304065 07-03-2007 07:51 AM

No sense tearing down the distributor (an all-day job, and figure three hours to remove the circlip holding the advance mechanism) until you have verified that the advance is sticking. Easy enough to check with a timing light.

mhackney 07-03-2007 08:00 AM

Thanks. Can you take a look at the post I linked to above on my distributor advance curve? Does that look normal? It does to my novice eyes?

thanks,
Michael

jstobo 07-03-2007 08:05 AM

A suggestion. back off your microswitch adjustment screw. Many owners mistakenly adjust the idle with this screw which is for activating the stop solenoid. If it is screwed down too far you will have a high idle. The idle on a MFI engine is adjusted with the throttle bodies air bypass screws on each cylinder. I am surprised to car idles at 1600. This is the range where the RPM Transducer should be cutting off fuel. The idle should be hunting betwwen 1500 RPM and 1300 if the microswitch, RPM transducer and Stop solenoid circuit is working. You probably have some of these electrical problems also. CMA has the procedure to check the stop solenoid circuit. I hope this helps. Good Luck.

mhackney 07-03-2007 08:39 AM

thanks jstobo - something else to look into. I know I haven't touched the microswitch but who knows what the PO did. I have looked at the microswitch but can't recall what I did. Probably best to go back and measure.

Probably the order to look into things at this point is:

1) the microswitch/RPM transducer/Stop solenoid circuit
2) timing
3) throttle bodies
4) dizzy (assuming others agree that the advance curve I linked to in a previous post is not horribly bad).


cheers,
Michael

304065 07-03-2007 10:07 AM

Michael,

Just loosen the lock nut and remove the screw from the throttle arm for testing. Put it somewhere safe. That way it can't hold the throttles open.

The "speed switch" circuit itself doesn't have anything to do with causing a high idle. Remember, the speed switch provides output power when the revolutions are above 1500 rpm and switches off when the revolutions fall below 1300 rpm. The output power flows to the microswitch, which is normally open, but if it's closed, i.e. the throttles are at idle and the screw is actuating the switch, then power flows to the stop solenoid on the pump which moves the rack to cut off the fuel. None of which has anything to do with a high idle: it merely cuts off the fuel when the throttle is closed and yet the engine is still at high revs, such as on overrun, to prevent backfires.

Timing and distributor are the same. If you can, set timing to TDC for testing.

http://forums.pelicanparts.com/uploa...1183479084.jpg

http://forums.pelicanparts.com/uploa...1183479256.jpg

http://forums.pelicanparts.com/uploa...1183479283.jpg

Setting it to zero should keep it from creeping up on you while you determine the airflow. If you raise the revs to do a synchronization test and the timing doesn't return to zero, well, then it's time to tear into the distributor.

mhackney 07-03-2007 10:18 AM

Ok John, I may be completely confused here! Maybe I am completely misunderstanding "timing". Assuming the engine is properly warmed up;

My idle is 1600rpm. So, using the timing light, I set the advance to 4 degrees. If I raise the revs and then let it back down, it will always come down to 1600 with 4 degrees advance. Is this what you are talking about? I am not sure how to set the advance to 0 since that requires an idle of 950 correct?

304065 07-03-2007 10:42 AM

Check the timing at 1600, what is the figure?

Now dial back the timing a few degrees and tighten the clamp gently. Turn off the engine and let it sit for a couple seconds, then restart. Idle still 1600 rpm?

What you're trying to do is determine whether the distributor advance is causing the engine to speed up. So you start by retarding it. Most distributors, when the springs and weights wear, show too MUCH advance, so by dialing back the idle timing, you probably don't bring in the 6K point that much. That's the idea, anyway.

mhackney 07-03-2007 10:46 AM

Thanks John, I can work with that. I know (but will double check) that the timing is 4 degrees at 1600. I can take it from there. So, if I turn back to say 2 degrees, then restart, if I still idle at 1600, it is probably not the dizzy, right? If the idle drops, then the dizzy may be a culprit.

Jeff Higgins 07-03-2007 11:23 AM

Re: At wit's end with high idle
 
Quote:

Originally posted by mhackney
I have read and tried to follow the Check, Measure, Adjust proceedure.

I really need some guidance on how to proceed before I go crazy!

Cheers,
Michael

Does "tried to follow" mean you were successful in completing the process, or that you tried and were unable to finish? You absolutely must go through the CMA procedure in order, completely, when working with an MFI setup of unknown history. If you are unable to complete it for some reason, find some one who can. It's that important.

That said, CMA was written for relatively new systems. It's amazing what you might find after some one has been "tuning" one for awhile. With that in mind, and recalling you mentioned you were going to remove the throttle bodies, it would pay to check them. Two things are typically going to give you problems.

One, worn throttle shafts that are leaking air. The butterflies should not rattle or show any play in the housings. If the shafts are loose, you will need to rebuild or replace the throttle bodies.

Two, the butterflies do not fully close. Some one has cranked down their individual stop screws, they are dirty, rods are bent, they bind, whatever. With the butterflies "closed", hold them up to a light. If you see any light around the edges, there is a problem. The butterflies should seal the bores completely. The fore and aft edges of the butterflies are angled in opposite directions so that they seal the bore at a hair less than 90 degrees to the bore.

The stop screws on each butterfly should stop the butterfly, not the butterfly itself bearing on the bore. This will wear a groove in the bore that will allow air to leak past. Check for this groove; if it's there in any of the bores, you might need new throttle bodies. The two short rods off the center butterfly should not stop the fore and aft butterflies, either. In other words, their cross shafts should rest on their stop screws with no preload on the two rods.

Once the throttle bodies are verified as good, it's time to put them back on and go through CMA in its entirety. Checking and adjusting things piecemeal, out of order, will only lead to frustration. MFI tuning rewards a methodlogical approach; there are no shortcuts.

911pcars 07-03-2007 01:33 PM

"I am not sure how to set the advance to 0 since that requires an idle of 950 correct?"

A process called "dead timing". This sets base timing with the engine OFF. There are several ways of accomplishing this. This is one method:

Rotate engine manually so the 5ºBTDC mark on the pulley (or TDC or whatever timing is desired) is opposite the split in the crankcase.

Loosen distributor hold-down/adjustment bracket.

Remove coil wire from distributor cap and position it close enough to ground for a spark to jump. You can also use a timing light in place of the loose coil wire.

Ignition ON.

Grasp only a portion of the distributor housing. Rotate clockwise about 20º, then rotate CCW slowly until spark jumps to ground. Stop.

Tighten dist. hold down clamp.

Timing is now set to 5º BTDC. If engine is ON and the idle speed is below 1000 rpm., the timing light should confirm the dead-timed adjustment.

Hope this helps,
Sherwood

mhackney 07-03-2007 01:37 PM

Jeff, I have tried to follow it as close to the letter as I possibly can! To wit:

0) changed oil and filter, cleaned screen

1) replaced air cleaner cartridge and checked all air cleaner gaskets
2) had compression tested, checked out fine
3) replaced plugs and gapped appropriately
4) replaced points, rotor, cap and set dwell
5) checked and set timing
6) fuel pressure seems fine but I have no way to check
7) have not checked
8) have not checked
9) had it professionally checked 3 years ago, rechecked last month and found one throttle link was too short
10) have not checked but have Gastester to check when appropriate

Other than this fast idle problem, the engine performs very well. I have found a number of issues but none of them have solved the problem. The mouse in the thermostat hose was the cake though. I have no idea who in the Boston area who could do the complete CMA. The few shops I have contacted have shied away. I have not touched the MFI at all.

I think my best proceedure is to check the distributor as per John's previous post since that is easy enough and then remove the throttle bodies and check those. Make sense?

Michael

304065 07-03-2007 01:45 PM

Michael:

Remember: Any competent Porsche driver tracks faster if brakes come smoothly.

Air filter
compression loss
plugs
dwell
timing
fuel pressure
injectors
belt
correlation
smog

mhackney 07-03-2007 01:48 PM

Hey thanks Sherwood. On my engine, the static timing is 0 degrees so I will go with that. So I'll modify my plan from the last post to:

set static timing to 0 degrees as per Sherwood's post

do I then need to test the advance as per John or head straight to the throttle bodies?

Michael

mhackney 07-03-2007 05:15 PM

Ok Sherwood, I went out for a drive to warm up the engine. Came home and followed your static timing proceedure. First I marked the distributor position so I could recover and compare to what I had "before".

I tried to use my timing light but it did not seem to be triggered (I verified the rotor was pointing at #1 with the TDC mark aligned with the case mark.) So I pulled the coil wire and used the spark technique. The spark happened about 5 degrees clockwise from my original position mark. I tried to start the engine but it would barely run. So I rotated the distributor about 2 degrees counter clockwise and tried again. This time the engine fired up and sounded good and idle at a hair over 1k. I measured the timing to be about 4 degrees ATC. I tried to get the timing to TC but the idle speed went up to 1500 or so. It seems to be very sensitive to small distrubutor movements. I took the car out for a spirited drive and it continued to idle at 1000+a bit. The engine seemed to pull stronger from idle and a dead stop. It felt pretty good in fact.

So, does this make sense? Do I now need to reset the dwell?

Finally, I tested the microswitch by reving up to 2500RPM. When I pressed the switch, nothing happened. The engine did not slow. I will use my multimeter anf the CMA proceedure for testing this tomorrow. I do not get backfiring whn I take my foot off the gas though so I am a bit confused.

Cheers,
Michael

911pcars 07-03-2007 06:13 PM

Michael,
As I mentioned in my last post, using the ignition timing to adjust idle speed is a stop gap measure. While you say you feel adequate power, if you're retarding the timing an appreciable amount (e.g. 4ºATDC instead of TDC), you might be leaving some power on the table.

If you adjust the timing per the factory recommendation (Xº @ 5000 rpm), what are the symptoms? Same high idle as before?

Work on locating that air leak.

Sherwood

mhackney 07-03-2007 06:30 PM

Sherwood, Ok. Factory is 30 degrees at 6000rpm. I'll check that tomorrow and see what happens.

cheers,
Michael

BURN-BROS 07-03-2007 07:08 PM

If you have a propane torch, turn it on but do not light it. run the tip along the area of the throttle shafts. If there is a leak there, your idle should change. I'm placing my bet on worn shaft bushings...

mhackney 07-04-2007 06:50 AM

Well, I did spend a little time this morning checking out the microswitch, etc. Pressing the microswitch button above 2500rpm has no effect but the switch checked out finw with my multimeter. So I went on to check the RPM transducer according to the test in CMA and it is not outputting 12v at 1500rpm or so. I took it off the car and cleaned the terminals but that had no effect. Are there other checks and are these known to go kaput?

Meanwhile, I'll need to wait until later this afternoon to go for a drive to get the engine up to temp to check the timing at 6krpm.

I will also use either the propane or wd40 test on the throttles to check for leaks. I probably need to remove the throttles anyway in order to clean and lube the throttle cross shaft.

cheers,
Michael

mhackney 07-05-2007 04:43 AM

Well it rained all day here in the Boston area so I did not have an opportunity to check the timing. But I did do some research on the RPM Transducer - there is a lot of info here! I took the transducer out and opened it up to see if anything looked amiss. One tip: you can remove the aluminum case simply by bending the 2 tabs on 1 side of the case - you don't need to bend all 4 tabs. AND if you bend the 2 tabs that are towards the front of the car (opposite side of the BOSCH black and yellow decal) you can not see those when the unit is installed.

Anyway, the board looked remarkably clean. Several of the components were stabilized with some sort of clear material - it looked original. However, flipping the board over, one of the copper traces - directly in line with the #3 terminal and centered left to right on the board - has about 4 mm of the green coating pealed off 1 mm of exposed copper on each side and about 2mm of what looks like the copper completely corroded away in the center. It looks to be a completely open circuit but I did not have time to test it. I think I should be able to solder a simple copper wire jumper across the gap or would it be better to solder to the posts of components on either side of the gap?

I'll take a photo when I get home tonight.

cheers,
Michael

mhackney 07-05-2007 03:42 PM

Ok, here is my disassembly and hopeful repair of the transducer.

removing the case by bending tabs on 1 side only:
http://forums.pelicanparts.com/uploa...1183678212.jpg

component side of board:
http://forums.pelicanparts.com/uploa...1183678389.jpg

back side of board showing missing copper:
http://forums.pelicanparts.com/uploa...1183678553.jpg

new lead soldered across the gap:
http://forums.pelicanparts.com/uploa...1183678699.jpg

installed and ready to test:
http://forums.pelicanparts.com/uploa...1183678851.jpg

Unfortunately, the transducer is still not working properly as tested on terminal 30b (at the microswitch). Even under magnification I could not find anything else wrong on the board. Are there any other vendors that refurbish these units or sell refurbished units? Otto is out until next week.

regards,
Michael

jeff m 07-05-2007 06:46 PM

Re the speed switch
Get a mag glass and look at the legs of the transisters
Try lifting them with a small screwdiver you may find the leg looks good, but have rusted away. It's fixable.
Check 12v at A and ground at D in the socket. (I think it's A check diag.)

I would pull the dizzy and clean it, no need to dissasemble, 10 minutes.
Check for air leaks, check advance tube is clear.
Pull off linkages and then replace them watching that each link does not lift your butterflies as you replace it. They should all be sitting on their stops.

If you still have a high idle you have to go one step back and set the butterfly stops as per CMA.

mhackney 07-06-2007 03:17 AM

Hey thanks Jeff. I did use a magnifying head set and probed each of the components and they all looked fine. I have not checked power/ground at the transducer socket - I'll do that today. I am building a test harness today so I can test the transducer on the bench while it is still connected to the car - ala EarlySMan. Not sure what you mean by the A and D in the socket? I was going to check for ground at post 3, 12-14v at post 2, distributor pulse at post 4. I already know there is no output at post 1.

I need to check the timing today now that the rain seems to have stopped. We had 2 days of serious downpours - all after work of course. I try not to drive my S in the rain unless I get caught out.

Cheers,
Michael

jstobo 07-06-2007 08:21 AM

Michael,
On my 69 E I would set the timing at 4 degrees "after" TDC at idle to get 30 deg at 6000 RPM. The spec for my 69 is 30-32 degrees BTDC at 6000 RPM-no idle setting. I found the above idle setting would give me the spec at 6000 RPM for my distributor. (made checking easier)Your distributor curve may be different. But do not worry about setting it after TDC at idle as long as the timing is correct at 6000 RPM. The timing curves on the early cars are much different than the SC because of more valve overlap on the early cars and more radical cams. For example your valve timing overlap is 5mm vs 1 mm for a CIS SC. Good luck on fixing the RPM transducer.

911pcars 07-06-2007 08:45 AM

Quote:

Originally posted by jstobo
Michael,
On my 69 E I would set the timing at 4 degrees "after" TDC at idle to get 30 deg at 6000 RPM. The spec for my 69 is 30-32 degrees BTDC at 6000 RPM-no idle setting. I found the above idle setting would give me the spec at 6000 RPM for my distributor. (made checking easier)Your distributor curve may be different. But do not worry about setting it after TDC at idle as long as the timing is correct at 6000 RPM. The timing curves on the early cars are much different than the SC because of more valve overlap on the early cars and more radical cams. For example your valve timing overlap is 5mm vs 1 mm for a CIS SC. Good luck on fixing the RPM transducer.

You could gain some torque at the bottom/midrange by modifying your current advance curve. For example, if you were to limit distributor advance 4º and add it to the static timing (thus, TDC), you'd still have 30º total. And if you were to set it "S" specs at 5ºBTDC, you'd have 9º add'l degrees on the bottom end. This will result in either noticeable bottom end response or borderline detonation, depending on the fuel, altitude, air temp., compression ratio, etc.

Early on, I installed an early S distributor in my 2 liter 911T engine and with some modest jetting/venturi upgrades, never experienced any signs of detonation. Of course, YMMV.

Sherwood
http:members.rennlist.org/911pcars

Eric_Shea 07-06-2007 09:33 AM

Are we sure this isn't simply a tired old dizzy? I had mine rebuild recently and it still has trouble coming off advance (big ole valves and cams) but it sounds similar to that...

mhackney 07-06-2007 09:38 AM

Not 100% ruled out yet. Right now, the engine is idling at 1000rpm but I have not checked the 6000rpm timing because I have not been able to drive up to temperature the last few days. I am waiting patiently to get home tonight and drive and test!

Plus, now I also know that my RPM Transducer is kaput - so I have that to sort out.

cheers,
Michael

911pcars 07-06-2007 10:01 AM

Michael,
This, from the tech section:

"What the heck is this thing called the RPM transducer?

The rpm transducer is used to activate the shut-off solenoid on the MFI pump. It prevents the pump from continuing to pump fuel into the engine when coasting at high speed, which will result in backfiring and other unpleasantness.

The rpm transducer puts out 12v above about 1600 rpm, and 0v below it. It’s connected to the MFI shut-off solenoid through a microswitch that’s activated by the accelerator linkage—the one on the left side of the engine, attached to the input stacks. With your foot off the accelerator, the switch is closed, and if the engine speed is above 1600 rpm, i.e., you’re coasting at speed, then 12v is applied to the relay and it shuts off the pump. If you’re below 1600 rpm, even though the switch is closed, i.e., you’re idling, the solenoid is not activated because the sensor is putting out 0v, and fuel is pumped to the engine.

The transistor leads corrode and break over time. Virtually any generic NPN, like a 2N2222, will work.

Bob Spindel
btindel@gte.net"

---------------------
This shouldn't be an issue at idle speed, yes?

Sherwood

mhackney 07-06-2007 10:05 AM

Thanks and you are correct, it should not be an issue. It is one of those sidetrack issues! Originally, I was idling at 1500-1600rpm and someone pointed out that I should be seeing the fluctuations between 1300 and 1500 due to the shutoff circuit described above. Curious, I started testing the microswitch and RPM transducer and discovered the transducer is kaput. Meanwhile, having fixed several other things (i.e. removed a dead mouse from the hose to the thermostat!) and adjusting the static timing, I have been able to get the idle speed down to 1000 but I still have work to do to make sure things are "right".

Cheers,
Michael

mhackney 07-06-2007 03:16 PM

Well, in diagnosing the RPM Transducer problem I discovered that the fuse (#2 in the engine compartment box) was blown! I cleaned up the box (that will be my next project - a removal and cleanup) and put in a new fuse. It blows as soon as the engine is started (8amp white body).

How can I daignose the RPM transducer with that fuse blowing? Seems my next step will be to see if the fuse blows with the transducer removed. Need to get some more fuses first.

Any suggestions welcome. Does this point to anything in particular wrong with the RPM transducer?

cheers,
Michael


All times are GMT -8. The time now is 08:09 PM.

Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2025, vBulletin Solutions, Inc.
Search Engine Optimization by vBSEO 3.6.0
Copyright 2025 Pelican Parts, LLC - Posts may be archived for display on the Pelican Parts Website


DTO Garage Plus vBulletin Plugins by Drive Thru Online, Inc.