Fuel injector cut off and cut on

[redcoupe86]

My understanding is that when you release the throttle and it closes all the way, the injectors cut off and then they cut back on at some lower RPM. What tells the injectors to cut back on, what mechanism? The DME or another sensor? The issue I’m having is when I get past 4000 RPM and I let off the throttle the fuel injectors cut off, but they don’t cut back on in time to keep the engine running. I’m noticing this happens even if I’m not under hard boost, for instance, if I rev slowly to 4000 RPM in first gear and then let go of the throttle it still dies.

However, if I do the same thing under 4000 RPM say that 3000 RPM the engine has enough time for the injectors to kick back on.

[944 Ecology]

Check your TPS. The ECU looks at the RPM and the TPS and figures out how much injection pulse is needed.

However, this could be a symptom of the idle speed being set too low and the ICV being sticky or bad. Check out the BMW wiki for an engine with similar characteristics...

https://www.e30zone.net/e30wiki/index.php?title=Motronic_1.0

[Monkey Wrench]

interesting link. id also check for vacuum leaks first as any leak like a bad hose connection due to an old hose may affect things a lot , especially at idle. a smoke test may help find them. a leak in the intake manifold or any of the associated hoses might throw off the idle air fuel mixture so I'd look into that and then after assuring myself it's all ok, then go on to the other stuff mentioned. Otherwise it's a bit like painting yourself into a corner by adjusting air fuel mixture when you also have a way that air can enter the engine without being measured and causing imbalance that way.

[walfreyydo]

Quick and dirty way to check your TPS is to make sure it "clicks" when the throttle plate is opened. It should do it almost immediately upon throttle opening. The TPS on these (NA) cars sends a binary open/closed signal to the DME computer. On a turbo, the TPS is variable (I believe it sends a linear 0-5V signal). Testing procedures for both types are on Clarks garage.

However, the TPS may, or may not be, the cause of your problems. A lot of possible causes here. Vacuum leaks are always a good one to check for. Fuel pressure also another thing to immediately look at.

We also do not know the year/model 944 in question which is really important.

Last but not least, I would like to introduce to you and recommend reading through Clarks Garage
https://www.clarks-garage.com/shop-manual/ts-01.htm

[redcoupe86]

Thanks for your input, I’ll reply to some of the above. I’ve already checked the TPS and I know that it’s working. I verified with all the tests on Clark’s garage - and it’s also a new unit has less than 2000 miles on it. Im familiar with Clark’s garage. I’ve also done the smoke test to check for vacuum leaks on three different occasions and I have found no leaks anywhere, the only smoke I get comes out of the intake itself. II’m trying to sort out is what is actually telling the injectors to cut back on once the RPM falls when you shut the throttle? Is the turning back on of the injectors controlled by the DME knowing what rpm the engine is at and then cutting the injectors on, or is it some other method? Is the ICV somehow connected to this?

[walfreyydo]

How do you know your issue is related to "the injectors to cutting back"? Do you have any data or information to verify this? This sounds like an assumption you are making and may be unrelated to your actual issue.

It could be air metering issue (AFM), ICV, diverter valve, or something else. Yes when the throttle plate closes some air must still bypass the closed throttle plate through the ICV. Very easy to test and clean (once you get it off). ICV issues would also cause rough/poor idling.

Also, is this a turbo? You didnt specify your year/model 944. Do you have a wideband? How is your idle normally? Stable? I wonder if the stalling off throttle is more related to an air fuel ratio imbalance (overly rich/lean) and not the "injectors cutting back" - this is where a wideband would be helpful to see what your AFR's are doing when it does stall.

[redcoupe86]

It is an 86 turbo turbo.

I have tested the ICV and the TPS. Both test good. I’m reasoning if those are good then what is next on the list to check? It could be the wiring to or from these components is bad. I’m also watching the RPM drop to zero with no indication that the engine is running. While I don’t know for sure that the injectors are cutting out, that’s why I’m here trying to verify that fact one way or the other. If I determine the injectors are getting the signal to cut back on but the engine continues to die then I’ll need to do further investigation of course.

[Monkey Wrench]

just a remote possibility bu the engine in a car sort of rocks as you go on and off the gas, due to torque. if a wire happens to touch the block or maybe the exhaust its possible to get an intermittent issue that is triggered by the engine actually moving , maybe a weak speed or TDC sensor wire or similar that has seen better days? a plug wire worn through and touching? any wire that is attached to the body and the engine needs to flex and flex means stress on a wire, plus the oily conditions. intermittents like that can be hard to troubleshoot as they hide when you are looking and may contact for just a second and cause an issue. cleaning connectors, grounds etc can't hurt. you might swap plug wires as a test if you happen to have an old set on hand.

I know in my Volvo there is a ground to block on the fuel rail wiring and if I lift that ground the car will die. Porsche likely has a similar ground connection around the fuel rail and a engine ground strap. others might be aware of it's exact location. on my Volvo the bolts that hold on the fuel rail on, double as that ground point.

also refering to my Volvo there is a black relay on the innder fender in the engine bay, connected to , I believe the injectors but maybe the fuel pump, think it's intention is to cut fuel in a crash. If I pull that relay it instantly kills the engine. There could be a parallel to it in the Porsche world, I assume.

a trick Ive done is to try to connect a timing light to plug 1 and tape the button down, now you cnan drive or at least run the car and watch the pulsations of the actual plug wires You might be able to tell if the spark is interrupted even for a moment if you can see it flashing in real time. maybe one of those LED things that plugs between the injector and it's wire can help monitor the pulsations in real time too. (noid lamp)

a simple trick is just put a 12 V test lamp across the primary coil, as that coil sees its pulsations (to create the speak) so will the test lamp. then you can sort of see the coil flashing or flickering and look for interruptions in ignition pulse to the coil. If you find anything you can delve deeper into why the spark is interrupted.

[walfreyydo]

To be clear, the phenomenon you are trying to reference is called "over run fuel cut", which is common in standalone systems to cut fuel on decel. I am unsure if Motronic also has this feature, odds are it does.

However, this should not cause the car to shutoff. I know this because I have my megasquirt setup for overun fuel cut, to cut fuel on decel, and my wideband also verifies this going max lean (22.4 AFR or whatever). The car decelerates as normal and does not stall, which is how its supposed to work. So regardless, even if the injectors are cutting fuel, this isnt your problem...

This is why I am asking if you have a wideband. It would tell you if injectors are actually cutting off (full lean), or if perhaps something else is going on, like overly rich. IMO every turbo car should have a wideband O2 and AFR gauge, regardless, but thats besides the point.

Theres so many possibilities here, and the added complexity of the KLR and any potential issues with that. I think the first thing I would do is monitor fuel pressure when under the conditions leading to the stall, see what happens.

Also, I think you already answered this, but are you running the stock AFM or have you converted to MAF? Have you checked the AFM and done the voltage tests in Clarks Garage? Repositioned the pickup arms? That would be another thing to check.

[redcoupe86]

I have the Vitesse V-MAF+ system. In fact, this all started happening once I install the system, but after talking with John (Vitesse), he indicated that when the diverter valve dumps a bunch of air into the inlet tube, that when I’m experiencing is common on MAF systems with a diverter valve so close to sensor. However, he also says that should only happen went under acceleration. I’m also experiencing a cut off if I just leave the car in neutral and rev the engine past 4000. That’s why I’m still trying to see can answer to this.

Well, I have tested the ICV, but let’s assume for the moment that I tested something wrong and the ICV is not working correctly. Would what I’m experiencing - engine shut off under when I rev over 4000 - be related to the ICV was failing?

[redcoupe86]

BTW I have the MoMonitor that allows you live data monitoring, and based on John’s review the TPS is operating correctly.

To answer your earlier question I do not have an AFR gauge. Thigh I should.

[walfreyydo]

Originally Posted by redcoupe86

I have the Vitesse V-MAF+ system. In fact, this all started happening once I install the system, but after talking with John (Vitesse), he indicated that when the diverter valve dumps a bunch of air into the inlet tube, that when I’m experiencing is common on MAF systems with a diverter valve so close to sensor. However, he also says that should only happen went under acceleration. I’m also experiencing a cut off if I just leave the car in neutral and rev the engine past 4000. That’s why I’m still trying to see can answer to this.

Well, I have tested the ICV, but let’s assume for the moment that I tested something wrong and the ICV is not working correctly. Would what I’m experiencing - engine shut off under when I rev over 4000 - be related to the ICV was failing?

This is actually really great information. On a MAF based system the blow off valve MUST cycle back into the intake AFTER the MAF and before the turbo compressor. I assume this has been done and is correct but I mention because many folks confuse blow off valve with a diverter valve, and try to rig up their system to vent to atmosphere, which with an air metered system (MAF system) is a big no-no. Thats because the air coming in is metered, and if some of the air is vented to atmosphere, the computer doesnt know that and your system will run overly rich. So just want to put that out there in case there is any confusion - the "blow off valve" must divert back into the intake post MAF. A true blow off valve can only be used in a speed density or alpha-n system, which the stock Motronic is not.

A blow off valve only operates when you let your foot off the accelerator. Exactly the time when you seem to be having issues no? Thats because boost is trapped between the closed throttle plate and the compressor, yet its already been metered, so the pressure needs to vent back into the intake, before the turbo and after the MAF (so it doesnt get double counted). Are you sure the diverter/bypass/blow off valve is re-routing back into the intake correctly, after the MAF sensor? This is also why a vacuum or boost leak can cause lots of issues as well, especially while under boost, because that pressurized, metered air can escape through loose/cracked vacuum lines or boost pipes/connections.

Although if you are revving in neutral its probably not building nearly as much boost as under load and thus, this may not be your problem. To test the ICV its rather simple. Remove and hook up to +12V. At +12V the valve should snap fully open. If its sluggish, slow or does not open/close fully its likely gunked up. Letting it sit in some brake cleaner and then sprayed/cleaned out with another round should do the trick. However, if your ICV were the issue, you should also have problems with the car idling correctly, which it sounds like you dont have that issue.

Another thing you can do to check for boost/vacuum leaks is to hook up an aftermarket boost gauge (or a temporary vacuum gauge works too). The advantage of this over the stock gauge is that it will show vacuum as well as boost. Based on this thread, normal vacuum at idle is between 16-18 in hg. You can also see what your peak boost is (as with the OEM gauge) and you can use this to compare your peak boost vs what it should be at to determine if you have any boost leaks. If you should be at say, 14psi, but only hitting 10psi, thats a good indication of a leak (or an issue with the wastegate).


Here is an illustration (boost pipe lengths not to scale):

[redcoupe86]

So yes the system is designed such that the BOV is after the MAF. it is working correctly, it’s new for one and I can clearly hear it doing its job when I let off the accelerator.

Please note: the V-MAF+ system works great and does all it says it will do. Very happy with it. I’m just not sure why my RPM doesn’t stabilize quickly enough to stop the engine from dying when the RPM goes over 4000 - to be clear this happens with the engine under load, every time, or sometimes if the engine is just in neutral with no load and thus no boost.

Could this happen because of a bad ISV?

[walfreyydo]

As an aside, how much boost do you build when revving to 4,000 rpm in neutral? Can you verify your plumbing is correct per the above diagram?

I think John had good feedback- you dont want your diverter valve plumbing in directly next to the MAF, it could blow pressurized air backwards on the sensor causing all sorts of problems. Youd want to be sure its plumbed at least 6" behind the MAF, and not plumbed between the MAF and air filter, but between the MAF and the turbo compressor. Ill assume based on your reply that you have it setup correctly (was writing this while you posted)

Again, a bad ISV/ICV would also show up with a poor idle or inability to idle. However, if its badly gunked up, it may not be able to actuate quickly enough to recover after a pull to 4k. Im not sure thats the issue, but its worth checking. I addressed how I test the ICV in my earlier post.

Question. Does the car drop below normal idle and THEN die, or does it just die immediately as soon as you lift off the gas. Like, youre doing a pull, let off the gas - does the car decellerate at all and drop below normal idle and sputter and die, or does it literally shut down as soon as your foot is released from the gas pedal and doesnt even have a chance for RPMs to drop? To me it sounds like it immediately dies as soon as the throttle plate closes and diverter valve actuates correct?

This is a head scratcher for sure. Have you tried posting in the Turbo forum? This is the NA forum and although I have owned other turbo cars in the past (so I understand the overall system) Im not overly familiar with the 944 turbos. There is also a KLR system which manages boost and prevents overboost. This may be something to look into.

[redcoupe86]

Originally Posted by walfreyydo

As an aside, how much boost do you build when revving to 4,000 rpm in neutral? Can you verify your plumbing is correct per the above diagram?

I think John had good feedback- you dont want your diverter valve plumbing in directly next to the MAF, it could blow pressurized air backwards on the sensor causing all sorts of problems. Youd want to be sure its plumbed at least 6" behind the MAF, and not plumbed between the MAF and air filter, but between the MAF and the turbo compressor. Ill assume based on your reply that you have it setup correctly (was writing this while you posted)

Again, a bad ISV/ICV would also show up with a poor idle or inability to idle. However, if its badly gunked up, it may not be able to actuate quickly enough to recover after a pull to 4k. Im not sure thats the issue, but its worth checking. I addressed how I test the ICV in my earlier post.

Question. Does the car drop below normal idle and THEN die, or does it just die immediately as soon as you lift off the gas. Like, youre doing a pull, let off the gas - does the car decellerate at all and drop below normal idle and sputter and die, or does it literally shut down as soon as your foot is released from the gas pedal and doesnt even have a chance for RPMs to drop? To me it sounds like it immediately dies as soon as the throttle plate closes and diverter valve actuates correct?

This is a head scratcher for sure. Have you tried posting in the Turbo forum? This is the NA forum and although I have owned other turbo cars in the past (so I understand the overall system) Im not overly familiar with the 944 turbos. There is also a KLR system which manages boost and prevents overboost. This may be something to look into.

I am plumbed correctly, with 6” behind MAF. I dont build any boost to speak of in neutral revving to 4k, but build anywhere from 10-15 lbs of boost under load depending on what gear I’m in.

When I let off the gas, as long as the car is in gear, the RPM fall accordingly, and as long as I push in the clutch when coming to a stop below 4k the car engine catches and idle is maintained. But if I run the car past 4k and push in the clutch (so NOT in a gear), the RPM just falls all the way to 0 and doesn’t appear to try and catch idle at all and dies immediately. It’s like the key was just turned off after revving to 4k.

On a side note I checked to see if there was 5v between ports B and C on the test port with engine started per the test manual and there is, but I’m not sure what this proves? Does it mean there is a signal from the KLR?

I haven’t posted on the Turbo forum yet since there appears much fewer people on that one.

[walfreyydo]

I wonder if the KLR is doing something once you hit peak boost. Would you say you are still building boost when you rev it past 4K? Said differently, when you stop the car before hitting 4K, has it already reached peak boost, or is peak boost reached after 4k? Does the car pull all the way to redline if you wanted to?

(See Tom M'Guinn's comment here: https://rennlist.com/forums/944-turbo-and-turbo-s-forum/679005-951-turbo-klr-problems-implications.html)

Sorry I dont have much knowledge about the test port, my car doesnt have one.

[redcoupe86]

I eke out a bit more boost if I go to redline. But in general most of the boost comes in around 4k but otherwise the car runs great, it’s only an issue if I go past 4k and push in the clutch and let the RPM fall.