2000 Legacy stalls when clutch is depressed

[2000 Legacy]

Hi everyone,

 

Update; last night at my Adult Auto Class we did a Manual Compression Test and the results were as follows:

 

Test 1

Cyl # 1 = 210 PSI

Cyl # 3 = 220-225 PSI

Cyl # 2 = 210 PSI

Cyl # 4 = 210 PSI

 

Test 2

Cyl # 1 = 215 PSI

Cyl # 3 = 225 PSI

Cyl # 2 = 205 PSI

Cyl # 4 = 210 PSI

 

I had forgotten the specs/instructions at home to see if this test was within range. When I got home I checked it and "ME(H4)-22" from the manual said that the "Standard" should be 176 PSI and that the "Difference" should be within 7 PSI. So mine is higher, is that a sign of any problems? The other thing I noticed that the Manual said to said to "Relieve the Fuel Pressure", we didn't do that. Would that have affected our results?

 

So with these results we decided that a "Leak Down Test" wasn't needed.

 

Any other thoughts or ideas?

 

I also checked the Check Engine Light; Codes P0301, P0303 and P0304, all Misfire Codes.

 

Thank you.

2000 Legacy

[Mike104]

Are Subaru OEM ignition wires installed? I have heard that non OEM ones can cause problems. Misfires seem to be a common difficult to fix issue on a number of Subaru forums. If the wires are not Subaru or have been installed more than a few years try changing them with Subaru ones. I got some from an online dealer for $50 + shipping, cheaper than the NGK ones locally.

 

Good luck!

[Mike104]

The High compression values have me wondering whats up with that. Doesn't seem right? Timing?

[2000 Legacy]

Hi Mike104,

 

Thanks for responding, as for your questions,

 

I assume that when the Subaru dealer replaced the plugs and wires that they used OEM. In fact when we replaced the plugs a few weeks ago, the plugs that they used were OEM (NGK). At that time we did a "resistance test" on the wires and they were all within spec.

 

As for the High Compression Values", I asked someone at work about it, they asked if I had added any additives. The last time I changed the oil, about 1700 miles ago (January 13, 2010), I used a thicker oil, Mobil 10W-30 along with a can of "Restore" hoping that this would solve the problem. It didn't. The guy at work said that this could be tghe reason for the higher than normal values.

 

I'm not sure about the "Timing", according to my records the Timing Belt should be good for another 20,000 miles.

 

Thanks

2000 Legacy

[Dickensheets]

I had a Mazda B2200 with 180,000 miles (all mine) that would stall with the clutch in. Ran great otherwise in all respects.

 

I pulled my hair out for years trying to solve this. I basically just went to nuetral at stoplights, it became 2nd nature.

 

I sold the truck when I started a family. I often wondered later if the clutch was somehow not disengaging all the way and loading the engine at idle.

 

Maybe try seeing if your car is doing something like this.

 

That's all I got...

[2000 Legacy]

Hi Dickensheets,

 

It is very frustrating and I too have gotten use to putting it in neutral however it doesn't always work and I'm not comfortable letting my son or wife driving it. But I often wonder if the clutch is related becuase that is always the one common element. It seems that the minute the clutch is depressed and held in for a period of time that the RPM's drop and everything shutsdown. Thanks.

 

2000 Legacy

[OB99W]

I've just re-read this entire thread, because sometimes it helps to have all the info "refreshed" in your mind.

 

I'm beginning to suspect a problem with the VSS (Vechicle Speed Sensor). A significant enough failure would typically trigger a code, but they can begin to have problems before the ECU reports it. With incorrect VSS data, the ECU won't properly control fuel delivery, etc. A scan tool capable of viewing live data might be able to catch that if the problem is consistent enough. Does the speedometer seem accurate, especially at low speeds?

 

Possibly less importantly, the higher-than-expected compression readings can have a few different causes. It could simply be due to an inaccurate gauge. Some fluid in the cylinders (gas/oil, which isn't compressible) will raise the reading by effectively reducing the combustion chamber volume a bit. Some carbon buildup could do the same. Valve timing (relative to piston position) has an effect, so if the timing belt is off by a tooth at the crank, that can result in an increase.

See: http://www.ultimatesubaru.org/forum/showthread.php?t=79759

 

The P0133 O2 sensor code is usually a reliable indicator of a failing sensor. The only other thing that could likely trigger that code is an exhaust system leak before (upstream of) the sensor. If there isn't such a leak, the sensor is probably bad. While it may not be related to the current problem, if the front (upstream) sensor hasn't been replaced within some reasonable number of miles, it probably should.

 

EDIT: Have you ever seen any codes besides the P0133 and P030x ones? (For example, P1507.)

Edited April 23, 2010 by OB99W
Added question concerning other codes

[2000 Legacy]

Good morning OB99W,

 

Thanks for taking the time to help me with my stalling problem.

 

To answer your questions from 4/23/10:

 

The speedometer seems to work OK or be accurate at low speeds, however I have noticed that when it does stall the RPM's drop real quick and everything just shuts down. I never really noticed the speedometer, but it too must drop just as quick. I'll have to check it out the next time I take it out. A friend of mine who turned me onto this website thinks he may have a "scan tool capable of viewing live data", so maybe that will help.

 

As for the "higher-tha-expected compression readings", could my adding "Restore" at the last oil change caused this? Also, how could the timing belt be off by a tooth? The belt was changed about 85,000 miles ago/4 years ago. Again, my friend said that we should be able to check this out as well.

 

The check engine codes that we've had on this car have pretty much been as I mentioned. In the last four years we had a "Knock Sensor" code, I forgot the number but I changed the sensor and haven't seen it since. The "P0133" code came on twice, once at 172,929 (1/29/08) and at 178,225 (5/7/08). Both times I cleared it and haven't seen it since. The only other codes have been the P030X, Misfire codes which have been more frequent and current.

 

Again, thank you for your efforts as well as everyone else who has taken the time to help.

 

2000 Legacy

[OB99W]

[...]As for the "higher-tha-expected compression readings", could my adding "Restore" at the last oil change caused this? Also, how could the timing belt be off by a tooth? The belt was changed about 85,000 miles ago/4 years ago.[...]

An engine oil additive that increases the viscosity might slightly increase compression readings, due to better ring sealing -- but probably not the degree that you reported.

 

Timing belts wear, and 85,000 miles is getting near the end of life for one. Also, the tensioner and idler pulleys, especially if they weren't replaced the last time the belt was, can develop some play. It's then possible for the belt to jump under certain conditions.

 

Using a scan tool, be on the lookout for anything that would indicate a wrong/erratic reading from the VSS or the transmission neutral switch. Either one can "misinform" the ECU about running conditions, causing it to improperly adjust the A/F ratio.

[Gloyale]

It might be a good idea to manually test the operation of the clutch switch.

 

I that is the only electrical device directly connected to the clutch.....might be worth looking into?

 

If the ECU doesn't know the clutch is in, it won't keep the idle up.

 

It will cut all fuel and rely on the car's momentum to keep the enigne turning. It's a fuel economy thing.....to not waste gas going downhill or approaching a stop sign.

 

A simple check with a test light would be all that's needed.

 

 

Also, put a piece of corrugated wire loom wrap (plastic tubing) around you're spark plug wires where they are near the injector wires. Inspect your injector wires for any nicks or damaged insulation and repair if needed. That may well solve your misfires (ussually does, 2000 models seem to be the worst with this, I suspect software)

[hankosolder2]

I see where you're going with this Gloyale, but I'm fairly sure the clutch switch does not operate as you mentioned. The fuel cut is based strictly on engine RPM and throttle position. I think the logic is something like if engine RPM is above 1500 RPM and throttle position is 0, fuel is cut until engine RPM falls below 1500RPM. I think the clutch switch has more to do with emissions management during upshifts.

 

Also, if what you say was correct, the engine would stall if you slipped it out of gear w/o using the clutch... which will not cause a properly operating car to stall. I do have to admit that my understanding of the fuel cut system is based on how Hondas work, but I'd imagine Subaru does it the same way. I'm prepared to be wrong though!

 

I'm more focussed on the rough, hunting idle the OP mentioned at the start of the thread. It would be interesting to do a cylinder balance check. You'd have to disable the idle air control valve, kill the signal to one injector at a time and see how much the RPM drops. He could have a bad fuel injector, something like that. A dribbling injector could screw up the idle without making too much difference in how the car runs above idle.

 

Have you checked the PCV valve?

 

Also, what if the EGR valve is intermittently sticking? You could just unplug the vac hose to it and plug it... it will set a CEL, but you could see if the stalling goes away.

 

Nathan

[Gloyale]

I see where you're going with this Gloyale, but I'm fairly sure the clutch switch does not operate as you mentioned. The fuel cut is based strictly on engine RPM and throttle position. I think the logic is something like if engine RPM is above 1500 RPM and throttle position is 0, fuel is cut until engine RPM falls below 1500RPM. I think the clutch switch has more to do with emissions management during upshifts. [/qoute]

 

 

The clutch switch absolutely is used in combination with the TPS, Neutral switch, and VSS to determine fuel cut at zero throttle.

 

I could search and find threads here about stalling subarus and clutch switches and the relation between the two. But then again, so could you:grin:

 

Also, if what you say was correct, the engine would stall if you slipped it out of gear w/o using the clutch... which will not cause a properly operating car to stall. I do have to admit that my understanding of the fuel cut system is based on how Hondas work, but I'd imagine Subaru does it the same way. I'm prepared to be wrong though!

 

Again.....neutral switch......the car knows when the shifter is in gear or out of gear. Interpretation of this switch is one of the main differences in the ECU's specs MT vs. Auto.

 

I'm sure honda uses more than just RPM and TPS for fuel rationing.

 

It seems that the minute the clutch is depressed and held in for a period of time that the RPM's drop and everything shutsdown.

 

Car stalls when clutch depressed = need to at least CHECK the clutch switch.

 

there are actually two of them. One for starter interlock (safety) and the one for the ECU and Cruise. You want the later, the one mounted so it's depressed while the clutch is out.

 

IIRC, Should be open when depressed and close when you push in the clutch pedal.

Edited April 27, 2010 by Gloyale

[OB99W]

By the way, many sensors/switches connect to the ECU via the multi-pin connectors at the rear passenger side of the engine. I've seen several cases of corrosion in those connectors, probably due to road salt from winter driving getting into them. It might be worth pulling those connectors apart and checking. I know for sure that the neutral position switch and VSS grounds go through the 16-pin one (B22/E3), for example.

[hankosolder2]

 

The clutch switch absolutely is used in combination with the TPS, Neutral switch, and VSS to determine fuel cut at zero throttle.

 

I could search and find threads here about stalling subarus and clutch switches and the relation between the two. But then again, so could you:grin:

 

Again.....neutral switch......the car knows when the shifter is in gear or out of gear. Interpretation of this switch is one of the main differences in the ECU's specs MT vs. Auto.

 

I'm sure honda uses more than just RPM and TPS for fuel rationing.

 

l.

 

Well, I also did say that I was prepared to be wrong, at least in the case of Subarus.

 

Honda doesn't even use a neutral detection switch on the '98-'02 Accord 5mt, I double checked the FSM, I'm positive. I'm also pretty sure that the clutch switch is also only used for the cruise, as it doesn't appear to have any connections to anything but the cruise ECM. This car will not stall if you slip it into neutral and the ECM has no way of knowing when you did that as it has no neutral SW.

 

I DID do a quick search for stalling and clutch switch...came up with nothing relevant in the new gen forum. Is it a known issue?

 

I'll tell you what. I'm curious about this. I'd be willing to try disconnecting the clutch switch in the OBW to see what happens. My prediction is no stalling. It might be a little while before I get around to doing it, but I'll report back with what happens. I'll either be eating crow or humble pie!

 

Also, note the the OP says that his car sometimes still stalls when he slips it into neutral rather than declutching (unless you're claiming that both the neutral sensing SW and the clutch switch have both failed!)

 

Nathan

[Fairtax4me]

I always figured that clutch switch was just for the cruise control. I'm also interested to know if disconnecting that switch will adversely affect engine operation if the switch is known to be good. I'll have to try it tomorrow after work and see what I get.

[frag]

Since the clutch starter defeat switch is ground for the starter relay, would'nt a bad switch prevent the car from starting ?

[Fairtax4me]

Different clutch switch. There are two. The one in question is at the top of the pedal and is actuated with only about 1/2" of pedal travel. Generally it's only used for the cruise control but it seems Subaru found another way to utilize it.

[frag]

There's only one switch on my clutch pedal stem. Brighton. No cruise control.

[johnceggleston]

I always figured that clutch switch was just for the cruise control. I'm also interested to know if disconnecting that switch will adversely affect engine operation if the switch is known to be good. I'll have to try it tomorrow after work and see what I get.

 

i thought the primary function was 'no start' if the clutch pedal was NOT push in.

[frag]

i thought the primary function was 'no start' if the clutch pedal was NOT push in.

 

That's exactly what it is. When you push in the clutch you complete the path from the starter relay to the ground. That's why i think the OP clutch switch is not defective cause his car would'nt even start.

[Fairtax4me]

i thought the primary function was 'no start' if the clutch pedal was NOT push in.

 

That's correct. The clutch safety switch prevents operation of the starter motor unless the clutch pedal is depressed.

But the safety switch is not the switch Gloyale is talking about.

[frag]

Where is this other switch? Not on the clutch pedal itself on my car for sure. 96 Brighton.

[DaveT]

Hi all,

 

Thanks for the help so far. I've driven the car in question a few times now. When it's cold, the engine dies at idle. 400' from the garage to street, dies. 2 miles (down hill mostly) to the on ramp, dies. Restarted just by letting the clutch back out that time.

 

After it's warm, it just idles roughly.

 

This behavior is pretty consistent.

 

The clutch isn't hanging up.

 

We will have to check the switches mentioned above, and the timing belt.

[DaveT]

A few more notes:

 

Idling in first gear, on flat road / driveway: cold, stalls out. Warmed up, rough at times, but doesn't stall.

 

Traveling about 40MPH:

Cold, let clutch out or slip shifter out of gear without clutch, dies. Smoothly to zero RPM.

 

Warm, not as often, only sometimes. Sometimes, RPM drops smoothly to about 1000, then hesitates, drops to <500. & dies.

[DaveT]

Well, we checked a few things tonight.

 

Timing belt looks ok, the 2 cam pulley marks line up. Couldn't get to the Crank mark in the alotted time.., so I suppose it's not a complete check.

 

There is only 1 switch on the clutch pedal, and it is the starter lockout, and works for that.

 

We are pretty much stuck at this point...

Short of getting an identical parts car, and swapping things...