Four cylinder transaxle cars
  • User avatar
  • User avatar
  • User avatar
  • User avatar
  • User avatar
  • User avatar
  • User avatar
User avatar
By ChrisZ
#183749
Got a weird one: The 87 944S has been stumbling when I floor it. Accelerates fine under light to moderate power but not happy at full power.

So far I have checked the air filter (ok), fuel vacuum lines (vacuum present at idle), pulled each injector power plug to verify operating (all operate), and the distributor cap (looks ok enough). Could it be a clogged fuel filter, DME issue, or a faulty sensor?

C
User avatar
By Scott at Team Harco
#183757
I had a failing DME (many years ago) that had similar issues. Doesn't sound like the same you are experiencing. Check the TPS and MAF.
User avatar
By Gregs04.5
#183793
Could be as simple as bad fuel. Had a similar response on some rental equipment, if we eased into the throttle, it would run, but throttle too fast, major stumble, same thing when put under load.
Had the rental co come out to try to fix, but no luck, let us use it for free for the day. After refueling with known good gas, it ran as it should.
YMMV,

Greg
User avatar
By ChrisZ
#184024
Good data.
SeanR wrote:Good chance that it is failing reference sensors. Trouble shot this on a 951 over the course of a week and that ended up being the issue.
Was wondering that. I might pop out the DME and put a bunch of 24 gauge wires on the pins to see if I can see what the various sensors are doing at the DME. I'm guessing crank sensor, cam sensor, and how many water sensors.

The car doesn't have a MAF, it uses the older Bosch barn door sensor. Guess I can look at that as well to see if it's generating a smooth voltage or voltage spikes.

Also ordered a fuel filter because this feels oddly familiar. If I remove the old filter will the gas tank just empty itself on my driveway?

C
User avatar
By Scott at Team Harco
#184033
ChrisZ wrote: Mon Nov 21, 2022 10:58 am Good data.
SeanR wrote:Good chance that it is failing reference sensors. Trouble shot this on a 951 over the course of a week and that ended up being the issue.
The car doesn't have a MAF, it uses the older Bosch barn door sensor. Guess I can look at that as well to see if it's generating a smooth voltage or voltage spikes.
Sorry. I'm a bit out of touch with the old stuff. Here's what I was thinking...

http://www.clarks-garage.com/shop-manual/elect-22.htm
User avatar
By sh944
#184034
Everyone likes to go for the esoteric stuff when troubleshooting problems like this, but many times, its just the basic stuff. How old are the plug wires ( I had the same problem on a 944 years ago, swapping the plug wires solved the problem)? Fuel strainer in the tank is another one that's relatively easy to solve and if you don't know when it was last done, its worth doing. Likewise, popping in a new DME relay is easy enough to do. The reference and speed sensors are a bit harder to swap out (and go with the BMW ones instead, the leads are a bit longer but they are a lot cheaper and work exactly the same) but once you have them done, you should be good to go for quite a while.

The other thing would be to pop the black cover of the of the air flow meter and look at traces, there is a decent chance that they have worn thin over the years, as I recall, you can bend the arm slightly to get the contact points to run over the areas that haven't been worn but its been years since I have actually done that so you might do some researching before you take my advice on that one.

EDIT: Seems Scott was thinking the same thing I was...
User avatar
By ChrisZ
#184076
Yep, both are great answers and I should have checked Clark's. I hooked a 24 gauge wire wrap wire onto the DME output signal at the AFM, connected it to the fluke meter and checked it out. Looks like smooth voltages from .234v to 4.8 or so (wide open) with no visible jitter on the voltages (dropping, then going up as the barn door moves). So the AFM appears to be ok.

Resistor showed 2k or so at 35 degrees F, so that seems to be ok as well. For the moment I'll flag the AFM as "working"

The speed sender was replaced about 10 years ago, when that goes bad the car shuts down hard. The one on the front of the camshafts (this is an S and I think it's the Hall sender) was replaced awhile ago because the cable end disintegrated.

Looking at the manual I can see these are the sensors at the DME plug:

Can be tested static (with car and plug off)
Throttle valve switch. Normally open 53. Normally closed 52, Ground is ground.
Air flow sensor Top 12, Wiper 7, Bottom 26, resistor 44,26 (checked, ok)
Coolant sensor 45,gnd

Should be tested with car running
Oxygen sensor 28, ground
Hall sender Ground 8, Hall + 31
Knock sensor 1 30,11
Knock sensor 2 30,29
Speed sender Pins 47,48

As long as these are 0-5 volt signals I can check them with my Salee a/d converter and Logic software to see the patterns. For the rest I can read the voltages with the fluke.

There's also the thought it could be that stupid spark amplifier again, that died 100k miles ago and made the car run *really* odd. It's the Bosch amplifier transistor up at the front of the car, behind the driver's headlight. Possible.

Something to do while waiting for the fuel filter to appear. Is that an under the car thing, or can it be swapped with the rear wheel off?

C



I could drag the scope out there and see if the slopes are smooth, but I think I'm going to
User avatar
By sh944
#184102
Fuel tank strainer is an under the car thing, so is the fuel filter.
User avatar
By ChrisZ
#184149
*nod* Pulling the strainer will require me to empty the tank. Which could be helpful anyway for the fuel filter. I'll see if I can find the adapter for my fuel rail pressure tester.
User avatar
By sh944
#184154
One last idea, as I am rummaging through my fading memory of how to work on 944’s…. Fuel pressure regulator? Does it run like it’s on three cylinders when it’s stumbling or is it more of a steady roughness?
User avatar
By ChrisZ
#186294
Ok, I spent $30 and got a super cheap fuel pressure gauge tester on Amazon. Mainly for the adapter to the 944S' fuel rail. Hooked it up, jumpered the wires, and checked it out.

Fuel pump jumpered (30-87b): 3.6 bar pressure. Came up at a reasonable speed from empty, no issues.
Went away, came back 20m later: 3.5 bar. Doesn't seem to have moved. Leakdown good.
Put the DME relay back in, fired it up: Car started immediately, about 3.1 bar.

Revving the engine showed the gauge jumping higher but never lower than 3.0 bar.

I don't think I have a problem with the fuel delivery system at this point. Engine did miss a bit on accel, and I haven't tried driving it up the driveway. But fuel delivery to the injectors at least seems reasonable.

Same time I checked the voltage coming from the AFM to see how it looked. Was a nice 1.0 volts at idle, going up to 2 and 3 volts as I gave the engine power. A check of the min and max values didn't show anything below 1 which means the signal is not dropping out. So at this point I'm flagging the AFM system as "working".

Barn door is not skipping that I can see. Smooth values there.
Fuel pressure seems ok.
Unplugging an injector causes a change in idle.
Engine idles fine
Engine sucks when under hard acceleration, esp at low RPMs (lugging)

Next up:
I'll take a look at the sensors. Want to see if temp, cam sensor, and speed sensor are reporting properly. For that I should be able to use my digital scope. From there I can go on to the next steps.
User avatar
By Gregs04.5
#186296
Still sounds like bad fuel.

How old is the gas? is the car driven a lot?

Can you siphon the tank, and throw in some fresh gas?

Do you have a bore scope? (roughly $30 on amazon) after draining you can visually inspect the strainer. If the strainer is plugged up with crud, you may get reduced flow at higher rpm. but sufficient flow at idle. Can you rig the pressure guage so you can see it while driving?

Greg
User avatar
By ChrisZ
#186477
Gregs04.5 wrote:Still sounds like bad fuel.
Indeed. But another odd data point....
How old is the gas? is the car driven a lot?
Fresh from about 2-3 weeks ago. I did toss in a thing of injector cleaner at about the 1/2 tank mark, I'll fill it later this week.
Do you have a bore scope? (roughly $30 on amazon) after draining you can visually inspect the strainer. If the strainer is plugged up with crud, you may get reduced flow at higher rpm. but sufficient flow at idle. Can you rig the pressure guage so you can see it while driving?
I do, but before I go there a thought:

When I fired it up this morning it was cold. Took it on the road, put it in 4th, floored it, and the car *did not stumble*. Worked fine for a few minutes, then started stumbling a bit. As the car warmed up on the highway it went back to its normal crummy ways.

This *could* be an issue with Temp sensor II, however reading the 944S suppliment manual if the Temp sensor 2 reads shorted or dead open it should be running the engine as if the car was warmed up, so starting it should be lean and running it should be fine. It's possible this is a too rich mixture as opposed to too lean.

The problem there is I forgot where the heck the Temp sensor II is. The 944S has two knock sensors in the head and I see one of those between cyls 1 and 2, then the idle air stabilizer between 2 and 3, and the second knock sensor between 3 and 4. I forget, where the heck is the thing?

Might go out tomorrow morning and check it at the DME plug. This is where OBD would *really* be helpful, but although the 88 944S can use the 928S4 hammer I'm not sure if the 87 can.

C
User avatar
By ChrisZ
#186612
Moving onward, I decided to unplug the DME and do some tests:
Testing Temp Sensor 2 at DME plug:
Reads 3.9k with engine cold, outside air at 48f
Reads 514 ohms with engine running and at lowest bar (60c)
Verdict: Temp sensor 2 to DME is good. Good, I don't have to pull the intake manifold to fix.

Tested Air temp sensor 1 (the one on the throttle barn door)
Reads 3.7k at 48f
Verdict: Air temp sensor 1 probably good.

Test WOT switch:
53-24 closes at 90% throttle
Open otherwise
Verdict: WOT switch works pretty much right.

Test idle switch:
52-24 closed at idle, 2 ohms.
As soon as I come off idle it goes wide open.
Verdict: Idle switch ok.

Going under the hood I found that the connector for the hall sender disintegrated as soon as I touched it, it's plugged in, needs a new socket (or a new one again), might be a problem.

I also checked to see if the water system was properly bled by cracking the 12mm bleed screw with the engine at idle. Coolant came out immediately, no bubbles.

Next stop: Maybe the injectors or the plug wires are bad. Will check plugs next.

C
amdavid liked this
User avatar
By dougs968
#186946
Don’t know if this is helpful but my 968 had developed a stumble. Had a bad crank position sensor. Replaced and it all cleared up. That was a few years ago.....
User avatar
By ChrisZ
#187135
amdavid wrote:Did you ever replace the filter...? I know you said fuel pressure was good, but...
No, it's cold and I'd have to work under the car and deal with the fuel tank. I think I would see a pressure drop while backing up the driveway but pressure was fine.
dougs968 wrote:Don’t know if this is helpful but my 968 had developed a stumble. Had a bad crank position sensor. Replaced and it all cleared up. That was a few years ago.....
Quite possible, the connector disintegrated and the pins are dirty. I will try cleaning it up and see if that makes things better. As soon as it stops raining...

According to the manual, if the knock sensors are out or the CPS is bad the computer should be retarding timing by 6 degrees. Not sure if that's enough to sink the engine like this.....
User avatar
By ChrisZ
#187300
Better day to work on car.

Started by cleaning up the contacts of the CPS wires, wrapping them with electrical tape, and inserting them for testing. I have a new male connector assembly on order from Ian at 944Online, should be here in a few days.

Also pulled the plugs. Very interesting, first none of the plugs were "tight", and 3 of them came off with a small amount of wrench pressure. #4 came loose with finger pressure, so that's not good. Normally I tighten them to snug then about 1/8 of a turn, but I might have just tightened them to snug.

More oddness, plugs 1 and 2 were Bosch WR7+ and 3 and 4 were denso 40's. Odd, I wonder what possessed me to do that. All had a bit of oil, nothing big but #3 had some carbon flakes on it. Checking the gaps I found that #3 was also too close (.7mm could not fit) but the other three were about right.

Got a set of new in box WR7's from the stockpile (I have a 928S as well) gapped them to .7, then popped them in. Turned each snug then 1/8 turn, when I backed one out to check it took a fair bit of force to get it turning, so we're good there. In retrospect I should have used the torque wrench but oh well.

Fired up the car, started right up. Ran better at idle, took it out on the road and it does have more power: Putting it in 4th at 30mph and flooring it does not cause the stumble so, progress? I'll have to get some new gas and drive it a bit to see how it's doing, but this makes a difference.
sh944, amdavid, dougs968 liked this
User avatar
By ChrisZ
#188173
Well, after a weekend of driving, car seems to be running *much* better. No more stumbling on accel, got the pep back.

I'm wondering if it was because the plug were loose or bad. Kind of surprised about the loose bit as I would expect the pressure to blow the rubber plug wire things off the cylinder but there you go. But it makes some sense, as the engine would have enough compression to run at low power but would decompress and do weird stuff when I floored it.

Likewise I might test the plugs themselves to see if the resistors are good or possibly opened.

Regardless, for the moment I'd say "check your plugs" and see how they look first.
Gregs04.5, amdavid liked this
User avatar
By ChrisZ
#214528
Well, back from Switzerland, and back to the 944S. Before I left Ian over at 944Online had one 944S injector, so I bought it and it arrived while I was on hiatus. Also checked the injectors under pressure with the fuel pump running and rail up and sure enough injector #4 was leaking fuel intermittently.

Got back, swapped in the new injector, cleaned out the other three, and put the car back together. Good news, the engine seems to be running like a clock, but we'll see how it runs for a week or so. Still, improvement.

Also changed the oil and decided to switch over to 20W50 oil. Car does have 250k miles on it, probably time for something a bit heavier than the normal 10w40.
User avatar
By ChrisZ
#216605
So this continues to be interesting. I can pretty much count on the fact that when I try to start the car cold it's exceptionally difficult, requires a lot of cranking, and "catching" it with throttle. I have a baseline of 4 30 second cranks with no throttle before it starts.

Yesterday while listening to the engine I noticed I could hear a hiss by the center of the intake manifold. Sure enough it's the idle air control trying to modulate the air around the throttle flap. Which is weird since the car was just idling. So I pulled the plug and disconnected the IAC.

Engine idles nicely. Runs nicely. Starts nicely. And this morning I did a cold start, 5 seconds crank with no throttle and the car started right up. Runs well, no problems.

I think something is either wrong in the IAC valve or in the signal from the computer going to it. Car idles fine with it disconnected, and in order to get to it I'm going to have to pull the whole intake manifold....
78 in Fort worth

Ed that would look good with new tan carpets and b[…]

Ineos Grenadier

It’s not an SUV. It’s a hardcore offro[…]

This upcoming DFW First Saturday Breakfast will be[…]

Looks like I have one tracked down. Thanks guys[…]