[Cloud9...68]

Under load, e.g. when lugging the engine a bit (running at too low an rpm for the gear I'm in), and quickly flooring it, the engine stumbles and bucks very badly. In my previous post on this topic, I had emphasized the fact that it happens under WOT, but after driving it a lot today, I've come to the conclusion that any situation where the engine is loaded brings the stumble on. WOT is one such condition, but others include pulling away from a stop, particularly uphill. The engine only pulls smoothly when in a "sweet spot" as far as loading, so if I rev it up to 2500-3000 rpm before dropping the clutch, it does OK. Or if I floor it when he engine is already at a fairly high rpm, such as 4K or above, it doesn't stumble too much. And the problem is MUCH worse after the car has been driven awhile, like 20+ minutes on a hot day. The first 20 minutes of so of driving is basically perfect, no matter how much I lug or load down the engine.



Here's what I've done so far:



- Checked engine flash codes (came up perfect)

- Tested throttle position sensor, and installed one Raj loaned me. No impact, which gives me confidence the flash code test was accurate

- Replaced DME relay

- Cleaned throttle body with chemical flush sucked through the brake booster line

- Checked spark plugs (all perfect)

- Checked distributor cap and rotor (again, perfect)

- Plug wires are just over a year old

- Ran two tankfuls of dessicant to remove any water that may have been in gas

- Had injectors cleaned and blueprinted by Marren. Amazingly, this didn't help at all, though the car runs better than before when in its "sweet spot."

- Replaced fuel filter.



The fact that the injector clean didn't help has me thinking that it's not fuel related, so I don't think checking fuel pressure or flow rate would be very fruitful. The engine runs very smoothly, particularly since I got the injectors back, when not under a condition that causes stumbling, so I don't thing there is an issue with engine timing. I talked to a mechanic today who suggested putting the stock DME chip back in (I have a Racer-X chip), which doesn't sound unreasonable, but I hear it's a real pain to get to, so I don't want to do it unless there's a reasonable probability that it will have an impact. And I view the time driven (temperature) factor as a major clue, so the fact that the chip resides in the air-conditioned interior seems to argue against that as a probable cause.



What would you guys try next? Would repeating the flash code test after running the car hard more likely identify a problem? I'm planning to take the car to my first DE next weekend, and this stumble will really take a lot of the fun out of it, so I'm starting to get a little desperate. Thanks!



Andy N

Partial Post: Please Login or Register to read the full post.

[Greimann]

Typically, in any engine, hesitation can be cured by advancing the timing. Considering that the 968 has non-adjustable computer controlled timing, the only other way to bump the timing is to feed it more fuel or a higher octane too keep the knock sensors happy. When they are happy, they don't tell the computer to dial the timing back.



Try 5 gallons of 100 octane race gas. If you can't find that, then go to a paint store and get a gallon of toluene (sometimes labeled toluol) and dose the tank. This blend will bump the octane 3 or 4 numbers and tell you whether your hesitation is octane, and therefore ignition timing related.



Otherwise, try changing the O2 sensor. They need replacement at about 70K to 80k miles and directly affect the fuel mixture.

Partial Post: Please Login or Register to read the full post.

[rustech]

I wonder if the Mass Airflow Sensor could cause these symptoms...but yet it's not consistent. The car runs perfect at certain RPMs. The O2 sensor also sounds like a reasonable probability, but I wouldn't think the symptoms you are experiencing would be so severe.

Partial Post: Please Login or Register to read the full post.

[Cloud9...68]

Thanks for the inputs. The fact that the problem comes on only after 20+ minutes of driving, particularly on a hot day, has me questioning the ignition timing theory, though I appreciate all ideas, so I'm wondering if I'm missing something. Is there a reason you would think timing retardation would come on after 20+ minutes of driving? Thanks.

Partial Post: Please Login or Register to read the full post.

[Greimann]

[quote name='Cloud9...68' date='Jun 19 2005, 05:53 AM']....  Is there a reason you would think timing retardation would come on after 20+ minutes of driving?

[right][post="6065"]<{POST_SNAPBACK}>[/post][/right][/quote]

The engine has had a chance to build up to full temperature and that is when the potential for knock is the best. The combustion chamber is good and toasty by then and knock is more likely.

Partial Post: Please Login or Register to read the full post.

[Cloud9...68]

[quote name='Greimann' date='Jun 18 2005, 07:49 PM']...go to a paint store and get a gallon of toluene (sometimes labeled toluol) and dose the tank.  This blend will bump the octane 3 or 4 numbers and tell you whether your hesitation is octane, and therefore ignition timing related.

[right][post="6052"]<{POST_SNAPBACK}>[/post][/right][/quote]



So you're saying to mix one gallon of toluene with a full tank of gas? Thanks

Partial Post: Please Login or Register to read the full post.

[Greimann]

[quote name='Cloud9...68' date='Jun 19 2005, 09:02 AM']So you're saying to mix one gallon of toluene with a full tank of gas?  Thanks

[right][post="6071"]<{POST_SNAPBACK}>[/post][/right][/quote]

Right. Toluene is rated at about 118 octane, so it will give you a nice bump. You may notice a reddish deposit on the spark plugs if you check them, but it is not harmful if you use in moderation. <img src="/forum/images/smilies/968/smile.gif" class="smilie" alt="" />

Partial Post: Please Login or Register to read the full post.

[Cloud9...68]

OK, on my way to Home Depot (my favorite source for Posrche performance products <img src="/forum/images/smilies/968/laugh.gif" class="smilie" alt="" /> ) for a gallon of toluene. Thanks for the tip; it will be a good way to rule in or out a major category of potential problem.

Partial Post: Please Login or Register to read the full post.

[Cloud9...68]

Well, I added the toluene (two quarts to half a tank of gas), drove it around a good hour, and saw no improvement in the stumble. The car seems to like the octane boost, so, similar to the injector clean/blueprint, the cars runs better when it's not at a condition that causes the stumble, but the higher octane made no fundamental improvement to the stumble itself.



I have a neighbor who has not one, but two spare Bosch coils that fit the 968 (he has a 911 and a 944), so that's the next thing I will try. Wish me luck, as this has dragged on just about long enough <img src="/forum/images/smilies/968/mad.gif" class="smilie" alt="" />

Partial Post: Please Login or Register to read the full post.

[Cloud9...68]

Now I can add one more thing to the list of things I've tried that haven't worked: the coil. I really had high hopes for this one, but installing my neighbor's spare Bosch coil really didn't significantly improve the stumble under load, though similarly to the injector clean, the car actually seems to run better under conditions where it doesn't tend to stumble.



The fact that boosting the octane by adding toluene to the tank didn't help exonerates the knock sensors. This, combined with my confirmation that the throttle position sensor wasnt't the culprit gives me further confidence that the flash code test was accurate when it came up clean.



So, what's left to check? I suppose I'll try putting the stock chip back in (I have a Racer-X), though I cringe at the thought of bending the tabs on the DME box again. Plus I hear getting the box out is a real pain. And there's a possibility that my voltage regulator is acting up (see my post on this), though most likely it's just the gauge, because it's reading a solid 13.5V using a VOM.



Since it doesn't tend to stumble at the high-rpm conditions I'll be running in the DE this weekend, I guess it will be OK to go ahead and participate (and I'll have a built-in excuse for being slow <img src="/forum/images/smilies/968/biggrin.gif" class="smilie" alt="" /> ). But it sure would be nice to get this problem fixed. Any other ideas anyone? Thanks.

Partial Post: Please Login or Register to read the full post.

[NeedPorscheSpeed]

Since you have already ruled out the TPS, I am leaning towards the Oxygen Sensor. You car seems to run fine while until the O2 sensor is heated up. Once it comes into play, it seems to be sending the wrong signals to your DME, thus causing your car to run too lean or too rich. Do you have a strong gasoline smell coming from your exhaust? If so, more than likely, it is your O2 sensor. Good luck. <img src="/forum/images/smilies/968/smile.gif" class="smilie" alt="" />

Partial Post: Please Login or Register to read the full post.

[Vedder]

It might be the blue sensor, it tells the DME to richen on lean the mixture, it might be leanning or richen it out, that plus a weared out O2 sensor can do that, i used to saw that in the VW G60's , check the resistance at cold , and then at operating temp, should give rough 1000 ohms at cold and 150 at operating temp.

Dont forget that these sensors must use somewhere the same ground point for them.

So even if the resistance at the sensor is correct, it might be reaching the DME, curropted information.

my 2 cents

Partial Post: Please Login or Register to read the full post.

[Cloud9...68]

[quote name='NeedPorscheSpeed' date='Jun 19 2005, 11:20 PM']Since you have already ruled out the TPS, I am leaning towards the Oxygen Sensor.  You car seems to run fine while until the O2 sensor is heated up.  Once it comes into play, it seems to be sending the wrong signals to your DME, thus causing your car to run too lean or too rich.  Do you have a strong gasoline smell coming from your exhaust?  If so, more than likely, it is your O2 sensor.  Good luck.  <img src="/forum/images/smilies/968/smile.gif" class="smilie" alt="" />

[right][post="6105"]<{POST_SNAPBACK}>[/post][/right][/quote]



My only hesitation with it being the O2 sensor is that it didn't show up in the flash codes, nor is the check engine light coming on. Also, I thought from reading previous posts on 968.net, I thought a bad O2 sensor is more likely to affect the way the car runs while it is cold (though I don't understand why). It's cheap enough that I might go ahead and replace it anyway (the car has 87K miles, so I suppose it's due). Thanks for the input.

Partial Post: Please Login or Register to read the full post.

[Cloud9...68]

Oh, yeah, I was going to ask, is it possible that the aftermarket chip is impacting the function of the flash codes, giving me a false positive? Again, I don't want to remove the chip unless I think there's a pretty godd chance it's the culprit, but I'm running low on options. Thanks.

Partial Post: Please Login or Register to read the full post.

[Cloud9...68]

[quote name='Cloud9...68' date='Jun 20 2005, 04:50 AM']...giving me a false positive? [right][post="6111"]<{POST_SNAPBACK}>[/post][/right][/quote]



I meant a false negative.

Partial Post: Please Login or Register to read the full post.

[Greimann]

[quote name='Cloud9...68' date='Jun 20 2005, 03:50 AM']Oh, yeah, I was going to ask, is it possible that the aftermarket chip is impacting the function of the flash codes, giving me a false positive?  Again, I don't want to remove the chip unless I think there's a pretty good chance it's the culprit, but I'm running low on options.  Thanks.

[right][post="6111"]<{POST_SNAPBACK}>[/post][/right][/quote]

The chip does not affect the flash codes. I would save that as a last resort. If anything it should help correct a stumble because chips add fuel and timing.



O2 Sensors can get "lazy" before they fail alltogether. In the "lazy" state, they won't necessairly throw a code.

Partial Post: Please Login or Register to read the full post.

[NeedPorscheSpeed]

[quote name='Cloud9...68' date='Jun 20 2005, 05:49 AM']Also, I thought from reading previous posts on 968.net, I thought a bad O2 sensor is more likely to affect the way the car runs while it is cold (though I don't understand why).[right][post="6110"]<{POST_SNAPBACK}>[/post][/right][/quote]



Hmm... interesting. My understanding is the exact opposite: the O2 sensor takes time to "heat" up, and until that time, the engine operates in an "open loop" mode, strictly using its pre-defined maps. Once the O2 sensor is ready for operation, the computer shifts into "closed loop" mode, responding according to the environmental inputs that are sent to it via the O2 sensor. One way to think of it is that "open loop" mode is static, and "closed loop" is dynamic. At least, this is my understanding... <img src="/forum/images/smilies/968/biggrin.gif" class="smilie" alt="" /> Good luck. <img src="/forum/images/smilies/968/smile.gif" class="smilie" alt="" />

Partial Post: Please Login or Register to read the full post.

[flash]

ok - a couple of things to look at



good injectors do not mean good flow to them - a volume check is the only way to do insure that - even correct pressure does not necessarily mean good flow - i've had this very problem on another car - i had plenty of pressure, and the jets were perfect, but there wasn't enough volume under load - consequently the bowls would empty and it would fall on its face - the fact that it only happens when warm though rather makes this less likely



say it only happens after it's warm - here's a possibility - i've had this happen too - you could have a slight lift occurring at the head, or a crack in it - it would only show up when things get hot - have you done a compression check when hot? even better, a leak down test - is there any coolant loss or corruption in the oil? (by the way, just because there isn't either of these, it does not mean the head gasket is not leaking)

Partial Post: Please Login or Register to read the full post.

[mitch968]

Couldn't hurt to check the Mass Air Sensor. I shot some carb cleaner on mine as I saw suggested somewhere. It seemed to make the engine a bit more responsive.



If you don't know how old the O2 sensor is, I would change that too. Mine wasn't flashing the O2 code when I replaced it, but I did notice a difference. In particular, it was more responsive at low rpms. You can get the universal connector o2 sensor for about 50 bucks, and it's an easy install.

Partial Post: Please Login or Register to read the full post.

[Cloud9...68]

OK, I'm getting several inputs to replace the O2 sensor. I have a few questions about that:



1) Given that the O2 sensor is in the exhaust stream, I would think it would heat up very quickly. And yet the car runs great for about 20 minutes.



2) The car runs fine as long as the engine is not "loaded". For example, if it is already at an rpm where it's producing a lot of power, like 4000+, flooring it from there gives pretty smooth, strong acceleration. Likewise, revving the engine in neutral produces not a hint of a hiccup. At the other extreme, flooring it in a high gear at 2000 rpm produces a stumble-fest. It just bucks a hesitates, making minimal forward progress. I would lose a race to a 20-year-old Yugo pulling a yacht uphill in this mode. A quick downshift cures the problem. Would a bad O2 sensor explain this?



As far as Flash's head gasket/cracked head (Gasp!) theory. Same question - why would it run so strong from an rpm where it's less relatively unloaded? Though a compression test might be a good precaution, in case it is something more serious like this.



I'm not trying to shoot down anybody's ideas. I very grateful to everybody who's trying to help. I just want to make sure I understand your thought process in case I'm missing something.



Just to rule it out, I think I will try putting the stock chip back in. It may be a long shot, but it's relatively simple.

Partial Post: Please Login or Register to read the full post.