When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Cold start stumble and runs rough like on 3 cylinders
My '86 944 has done the cold start stumble thing since I bought it in May of '17. Reference this thread: Cold Start Stumble (2009 thread) For reference I am in central Texas. The weather when I ran it the other day was mid 50's F.
I have read that thread a few times and no one seemed to positively figure out what caused it. Maybe we all just live with it. I would like to find the answer to that, but for now onto other problems...
I let it sit for 60 days (Thanksgiving to Mid January) and it liked to have never started. It took maybe over five minutes and the battery was starting to slow down. But it did start.
Drove it for 30+ minutes. Will idle fine at 900 rpm, will drive, doesn't die, doesn't overheat, all gauges reading normal. But, feels like it is running on 3 cylinders. In October it felt like this a few times, for a few seconds, and then would "get over it." Now it runs like this constantly, but will "get over it" for a second or two now and then. The car will restart but not as easily as October/November.
Been reading lots of threads. Not sure where to start. A lot say vacuum leaks. I am not sure if this is the best place to start and if so, how to test for a leak. Any advice on how I should methodically test?
It's tough to do this via the internet where you will get 30 suggestions and no one can see and hear your car to really get a sense for what is going on. However the best things to do first IMO are the sort where you can get clear info and results and are also the normal stuff people do when their car is not running right.
Check the plugs, after idling badly for a few minutes. Are they all the same? All clean or all dirty? Wet or smell like gas? One in particular?
If they are grungy but not really showing you anything interesting, clean or replace them.
Run the car with plug wires disconnected one at a time.
Does it run just as badly on any given 3 cylinders or does it run "less worse" with one disconnected?
Check the spark, use a spark checker if need be. The spark must be able to jump 10mm as a test or it's not strong enough.
Check the cap and rotor, check the plug wires.
- Loose rotor, broken or burned cap and rotor, bad plug wires.
- Idle the car in the dark and look for sparks.
Best way to test for a vacuum leak is a smoke test.
You can go to a shop and have them use the expensive machine that puts smoke into the intake to see where it comes out, or you can buy a cigar and unhook a vacuum line.
I will also agree with the 'unhook one spark plug wire at a time' technique. In theory, three wires will make it run worse (on 2 cylinders instead of 3) and one will make no difference.
You would still have to figure out if it's spark or fuel or something else (bad valves?), but at least you'd have somewhere to start.
Or none would make any difference, and then it's something 'system wide', like a vacuum leak.
After I did the video I pulled the plug wired while it was running. Pulling #2, #3, and #4 (one at a time) made it run worse. Pulling #1 made no difference at all. But! When I plug #1 back in, for a second or two, it seems the misfire is gone. The spark jumped well over 1/2 to 3/4 inch also. The way this is behaving, it makes me think the injector may be clogged or partially clogged.
While it may be a few different things, the fact that yanking and replacing it seemed to "fix" it for a second or two would make me think plug, plug end or wire is at fault.
1/26/18 2:40 pm, 66°F, 0% Precip, 66% Humidity Hard to start again today, but not as bad as yesterday. Gave it some throttle and it started after about 4-5 tries. Smooth idle. Drove it 4 miles to Hopewell Cemetery Road and back. Ran perfect. No misfire, no stumble. When I got back, used mechanics stethoscope, listened to all 4 injectors, they sound all the same (click click click). Heater still doesn't work so I am thinking I still could have a vacuum leak. I want to keep testing because I am not convinced this rough running won't resurface. I feel like testing for vacuum leaks is a good idea anyway. I don't smoke or vape, so I will find someone who does and have them blow smoke into a vacuum line.
1/26/18 3:49 pm, 66°F, 0% Precip, 66% Humidity Drove it 2 miles to Post Office to drop off mail. Running rough again at below 3,000 RPMs. Above 3K, misfire/stumble disappeared and it ran normally. When I got back home I let it idle. It sounded better than the day before, but still not smooth. Ordered new NGK plugs. Planning on putting some injector cleaner in too; people seem to say good things about Techron. After new plugs and Techron, run some gas through it - see how it goes. Will update later. Thanks Jfrahm, Wisconsin Joe, and StoogeMoe for the replies and direction.
1/28/18 5:10 pm Spent a lot of time yesterday studying the diagrams of hoses and vacuum lines. Did visual inspection today. Found out there is a missing piece of vacuum line from the small check valve on the brake booster that goes to the heater controls at the firewall. The vac. hose was disconnected and hanging loose by the firewall. It was too short to reach the place it is supposed to go, so I will have to get some replacement line. Guess that explains at least one reason why the heater doesn't work.
Last edited by Chalt; 01-28-2018 at 09:12 PM.
Reason: Added more
1/31/18 5:30 pm, 60°F, 0% Precip, Fixed the vacuum line. Found out it was the heater valve line. made no difference in how it ran. And the heater still isn't working. But the heater pipes inside by the passenger footwell are hot, so hot water is making it into the heater core.
As far as how it ran today, still same - hard cold start - 6-7 tries. Drove for about 15 minutes. Ran clear and strong above 3K RPMs, but in 5th gear stumbles when you give it more throttle. Still stumbles at idle. When I got back, with the engine still running, I hit each header with infrared heat gun. pipe #2, #3, and #4 all close to 700°F, #1 was 398°F. Feeling like #1 isn't getting enough fuel. If it was spark, it wouldn't clear up at rpm, right?
Got a endoscope camera yesterday to look in my cylinders - been running rough and hard to start. '86 944 NA. Camera works good. I think this will be useful for more than what I bought it for. The two things I don't like about it: 1) the head is so long I can't bend it to 90° but it did come with a mirror for sideways. 2) The depth of field is shallow. For $38 though, no complaints. Auto mechanic endoscopes cost hundreds.
Well there ya go, blown head gasket. The fire ring in the head gasket is probably steel and rusted a bit. Gaskets can get eaten if your coolant is not changed. The engine and cooling system becomes a battery. You can actually check the voltage of your cooling system, it is a good way to keep an eye on the condition of the coolant.
Still working on my car with the cold start. I got busy and the 944 has been on the back burner for a while. Been testing things rather than throw parts at it. I found out that the rough running is a coolant leak in the cylinder, but the difficulty starting when engine is cold is a separate issue. It had gotten worse to where it wouldn't start without starting fluid, but once it did, it ran normally.
Here is my latest find. The DME Temp Sensor (NTC II) tested bad, which would create a lean condition when engine is cold.
(Pictured Above) OLD Sensor: Tested at 17.7 Ohms cold. (200 Ohms Range on meter)
(Pictured Above) OLD Sensor: Tested at 0.01K-Ohms at room temperature. (20K Ohms Range on meter)
(Pictured Above) NEW Sensor: Tested 2.17K-Ohms at room temperature.
So Clark's Garage says the Resistance should be 3.3 k-ohms at 59°F or 1.46 k-ohms at 86°F. So at 70°F, 2.17 k-ohms is within range for the new sensor. I went ahead and got a new J-Boot when I ordered the DME Temp Sensor.
Someone suggested new spark cables, i just discovered my cold start probable issue. in complete darkness i can see very little spark escaping at some points on cables so thats quite sure indication that needing new cables i guess.
Symptoms are also similar running rough at first, starts also at third try, second try almost gets it going. When warmed up starts right up like healthy horse (or should i say like 200 healthy horses - S2)
And AFM that has also given me some rough idles, i just moved the circuit board 0.5-1mm lower so the contact arms are on the fresh surface.
Yes. After I replaced the DME Temp Sensor, no more starting problems. The intermittent rough running after warmed up was the head gasket starting to fail as shown in the pictures above. At first I thought the hard starting on a cold engine was the slight coolant leak into the cylinder, but it wasn't – it was a bad DME Temp sensor, which caused the fuel delivery to be way too lean starting a cold engine (Cold meaning not having been driven overnight - I live in Texas so we don't have anything remotely cold like the northern states, our winters are very mild.)
Are you having similar problems?
01-25-2018, 12:50 AM
Cold start stumble and runs rough like on 3 cylinders
