Engine Cuts Out At Low Load

[Gollum]

Long story so I won't go there, but my engine was running great before I swapped my head (from P90 to P90). The new head is from a 81 so it's got a slightly different cam, but not by a ton. Both heads were solid lifter heads. I'm using the same intake and exhaust manifolds. Same ECU. Same AFM. Same TPS. Same Turbo. Same Exhaust.

 

But, in the shuffle of heads I had to change the coolant temp sensor because the old one didn't want to come out without falling apart on me. Then wiring the engine back up the hacked up TPS wiring finally gave up the ghost, so I got a new plug from a NA model and wired it in (I'm pretty sure it's correct now).

 

But, I can't seem to get it running right. It'll start ok (no cold start system left, I just rev it a few times on a cold morning and then it'll idle), but at all temp levels, it won't keep rev's at very light throttle settings. It gets really rough and wants to die. If I snap the throttle open quickly the motor revs nicely, and under accel the motor runs nice. But at soon as I start to hold a speed it starts it's sputtering. I've noticed that at those low throttle positions if I open the AFM a bit to give it some more fuel the revs pick right up and are stable. But I've loosened the AFM and that didn't help at all. I've actually moved the AFM a lot both directions to figure it out but it's still just not happening.

 

The only thing I can figure, is that maybe the AFM worked on the old head, but is now just way off for the new one for some reason. Or maybe the new coolant temp sensor is the wrong part, but I tired the one from my 81' and it didn't make a difference. Or maybe my fuel was bad, but I've burned through that and filled it up, but this is the first tank so it's a bit mixed. Or maybe my plugs are going bad? I pulled them and put in some older NA plugs to see if there was a difference and there wasn't any. The other thing that I should probably check, is the FPR, but I don't understand how that could make it run lean at high vacuum... Grounding problem somehow?

 

 

So what am I overlooking guys? I've done my best to change one thing at a time, but the problem seem to be extremely static.

 

The car's drivable, and pulls really good once boost kicks in, but that transitional area just sucks.

[stprasinz]

did you place the maf closer to the turbo inlet? air leaks on the intake? timing different....

Edited February 2, 2011 by stprasinz

[mario_82_ZXT]

The TPS on the turbo cars is only used at idle. 2 pins have continuity and that's it. Cruise is closed loop, anything above like 3.5k or so is open loop.

 

I'm thinking bad wiring or bad CHTS. Check the TPS pins for continuity at idle (only) and then check the harness itself.

 

Or go megasquirt

[Gollum]

did you place the maf closer to the turbo inlet? air leaks on the intake? timing different....

 

 

The AFM had been right up against the turbo since I got the car, ran fine like that but I knew it was too close. In trying to troubleshoot I've now moved it just behind the radiator core support. I checked for leaks on the intake and actually had one, but fixing it didn't change anything. Timing hasn't been checked with a light, but I know it's close just by ear, which is how I've timed all my L6 motors for the last 6 years.

 

The TPS on the turbo cars is only used at idle. 2 pins have continuity and that's it. Cruise is closed loop, anything above like 3.5k or so is open loop.

 

I'm thinking bad wiring or bad CHTS. Check the TPS pins for continuity at idle (only) and then check the harness itself.

 

Or go megasquirt

 

I'll double check the TPS wiring again. Maybe the ECU thinks the TPS is closed 24/7? That could be bad maybe.

 

I'm pretty sure the CHTS is ok, since the one from my other car did the same thing and that car is running fine.

[stravi757]

IS the car running rich or leaning out when this happens? I would suspect the AFM if rich. If its leaning out, maybe you missed a rather large vacuum hose when you reinstalled your cylinder head.

[rayaapp2]

I cannot remember whether or not the ECCS system on those years used all three or just 2 out of 3. 2 out of three sounds right, but I do not know why. Ive always calibrated the throttle switch using all three pins. Its the only way you know its really setup in the correct position. In any case you should be able to cut the TPS wires off or cross them all together and still not end up with your situation. The TPS is fine trim on the mapping. Compounded issues with TPS out of wack can do what your describing. There is an ancient post about TPS map values, but I cant remember which board it was on, who was involved in the thread, what it was called, or when I saw it. If you want to search through all of TonyD's posts on Hybridz and Zcar you might find it.

 

Cylinder Head Temp Sensor as Mario pointed out may be something to check just to be sure. Average infrared on the brass base and cylinder head around base is what I do and graph the resistance at cold and full operating temp with a few plots in between and you will should be able to compare that to the graphs in the FSM(My 82 Maxima FSM has them so the ZX manual should as well) or online somewhere. Check it at the ECU if possible. Where did you get the cylinder head temp sensor? None of the Kragen or Autozone replacement I have checked(And I checked a lot of them) were even close to whats suppose to be in these cars. I have checked both replacement parts for the CHTS and CTS(thermostat ones) and have had no luck. I have had better luck sourcing the sensors out of wrecking yards and cleaning them up meticulously then re-using them. Im not sure if they are available through the dealer, but my gut says yes and expensive. As far as this sensor goes its a major one. It could cause your woes but not without help.

 

Double(or triple?) check for vacuum leaks. Perhaps there is one hiding somewhere around the manifold or a crack in the middle of a rubber hose?

 

I would stress the phase "Compounded Issues" with these old systems and the fact that you pulled it all apart disturbing its natural state of being. Stuff Im sure you know Nathan.

 

Now that you've played with the AFM thats definitely something you should re-set somehow. Get it back to as stock as possible. Otherwise all your going to do is mask other problems. Unless the AFM has worn out the carbon traces somehow during your head swap, I doubt it is the issue. If it ran well before the head swap it should run the same unless there was some major problem in the old head you didnt mention and the AFM was tuned for that. I have seen AFM's carbon traces wear down and cause similar issues as the AFM cross that point of the carbon trace... Cruise and part throttle being an area that the arm sweeps across the majority of the time during the cars use, it tends to wear that area more than the rest.

I dont remember anything about testing the VAFM. Again FSM? See what you can find.

 

Have you preformed wiggle tests? Wiggle your electrical and vacuum connections and make sure you dont hear, smell, or see any changes.

 

Forgive me if Im stating the obvious. Im almost positive you know most of this. I have a feeling youve entered that fuzzy state of thought that occurs when youve been staring at something for to long.

I know I havent given you much to go on, but hopefully that helps you eliminate some stuff that might have been given you "fuzzy" thoughts.

[Gollum]

All good advice, and stuff I've thought of as you suspected.

 

I think it's going to have to come down to actually testing components via FSM standards. I'll test the CHTS tonight on the stove with a cooking thermometer. I'm not sure how I'll test the AFM, but the traces might very well be the main issue here, as it's never had a cover in the length that I've owned it and my engine bay is QUITE oily. Maybe I should pull that too and check for continuity on the traces.

[rayaapp2]

All good advice, and stuff I've thought of as you suspected.

 

I think it's going to have to come down to actually testing components via FSM standards. I'll test the CHTS tonight on the stove with a cooking thermometer. I'm not sure how I'll test the AFM, but the traces might very well be the main issue here, as it's never had a cover in the length that I've owned it and my engine bay is QUITE oily. Maybe I should pull that too and check for continuity on the traces.

 

 

The Carbon traces are a resistance board. Continuity will not tell you much of anything. You will need specific Ohm values to check for at pins.

Because these resistance values affect current that is what the ECU reads as current draw through those circuits. You affect the resistance with contaminants or wear and the reading goes out the window for the ECU.

 

 

GL

[stravi757]

All good advice, and stuff I've thought of as you suspected.

 

I think it's going to have to come down to actually testing components via FSM standards. I'll test the CHTS tonight on the stove with a cooking thermometer. I'm not sure how I'll test the AFM, but the traces might very well be the main issue here, as it's never had a cover in the length that I've owned it and my engine bay is QUITE oily. Maybe I should pull that too and check for continuity on the traces.

 

Your welcome to use one of my AFM's to see if thats the problem, I think I have like three or four laying around. I think a I have couple chts also.

[Gollum]

Are any of those turbo AFM's? I believe the turbo AFM's were a bit different, at least I'm certain there's minor differences, not sure how it affects the tune. Maybe we can work out a trade or something.

[rayaapp2]

Turbo VAFM's are bigger and have a different circuit board than NA. Im sure everything related to tune including spring, couterweight, etc are all different as well between the 2.

You can swap them back and forth, but Ive never been able to get a NA to tune for a Turbo.

[stravi757]

Are any of those turbo AFM's? I believe the turbo AFM's were a bit different, at least I'm certain there's minor differences, not sure how it affects the tune. Maybe we can work out a trade or something.

 

Yes,they're all turbo AFM's.

 

I Reread your problem and I dont think it has anything to do with the AFM. If the AFM went bad your engine should be running rich. And I doubt the CHTS sensor you bought is causing these problems. If your fuel pressure is right then it still sounds to me like a big vacuum leak(especially since you said when you press the AFM flapper back it runs better)

Edited February 4, 2011 by stravi757

[Gollum]

Finally did some more troubleshooting today. Checked the CHTS and it's maybe a little low in resistance at low temp, but at operating temp it's right in the middle of the graph in the FSM. So it seems this part should do the job well enough.

 

I cleaned the AFM traces with alcohol and when I was running it again I was noticing it couldn't be the AFM because when I move the AFM open it very consistently richens up and runs smoother, not erratically as though the traces are bad/cutting out.

 

BUT!!! I noticed that if I pull the vacuum line from the FPR that NOTHING CHANGES.... Idle doesn't change at all, full throttle doesn't change (though it shouldn't) and light throttle doesn't change a bit...

 

Bad FPR? I'm gonna change out the whole fuel rail for one I have sitting around with another set of turbo injectors and different FPR on it. I'll see if that changes anything. Maybe I've been tuned for a bad fuel pressure all this time I've had the car? Or could it have gone bad in the last year sitting around my garage?... Strange question to have to ask, if you ask me.

[rayaapp2]

Finally did some more troubleshooting today. Checked the CHTS and it's maybe a little low in resistance at low temp, but at operating temp it's right in the middle of the graph in the FSM. So it seems this part should do the job well enough.

 

I cleaned the AFM traces with alcohol and when I was running it again I was noticing it couldn't be the AFM because when I move the AFM open it very consistently richens up and runs smoother, not erratically as though the traces are bad/cutting out.

 

BUT!!! I noticed that if I pull the vacuum line from the FPR that NOTHING CHANGES.... Idle doesn't change at all, full throttle doesn't change (though it shouldn't) and light throttle doesn't change a bit...

 

Bad FPR? I'm gonna change out the whole fuel rail for one I have sitting around with another set of turbo injectors and different FPR on it. I'll see if that changes anything. Maybe I've been tuned for a bad fuel pressure all this time I've had the car? Or could it have gone bad in the last year sitting around my garage?... Strange question to have to ask, if you ask me.

 

Im going with hidden vacuum leak still.

Typically when the FPR goes out its the diaphragm. When the diaphragm goes out fuel comes out the vacuum hose as well. To proper diagnose FPR issues you need a fuel pressure gauge. I suggest a trip to harbor freight for one of those if your not so equipped. The change in pressure is going to richen up the mixture enough to keep the injectors supplied at WOT... you may not notice a big change at idle unless you have a 5 gas or wideband on the exhaust. Ive pulled them off before when they were known to be good and not seen a change in idle or anywhere else. Its just going to richen up the lower mixture enough to notice with the correct equipment. I would suggest against a fuel rail/injector swap. Dont waste your time unless you KNOW for sure the one you have is bad by diagnosis. Your already fully aware of the problems with swapping those injectors back and forth. There is a better chance that you end up creating another problem before solving the current one. Save yourself the headache in my opinion. Worth the $15 for a fuel pressure gauge than another headache. The FPR by itself would not cause your issues. The FPR as stated above will cause a rich condition at idle and part throttle conditions not lean as indicated by your AFM door opening(assuming AFM setting was still stock)

 

Now that you know you CHTS is good, retest it at the ECU pins. Make sure the connector and harness arent adding in more resistance. At least if it isnt the CHTS you will know at that point and cross it off the list of diag like a flow chart from the FSM. I have seen the CHTS cause this issue before in Nissan and BMW... Both fail rich though.

 

Personally Id invest in some sort of gas analyzer or wideband Nathan. You mess around with these engines and cars enough that it should be something in your arsenal. I run a Innovate wide band in one of my cars. Soon as I have some extra car cash Im going to get a second Innovate and set it up to be portable to use on multiple cars so I can tune my other cars at home. Knowing what your AFR was doing at the point of the stutter will greatly help in your diagnosis. Its also great for tuning those cars in and if your still running a ZX will help you to a point with your smog testing. The LC-1 runs $140 with no gauge. With a gauge it runs $160 or so. You can use any gauge with it though so if you find a cheap one with a digital readout you will be good to go. I dont think the LC-1 datalogs though. I have a separate datalogger setup with my LC-1. So you would have to be keeping an eye on it while trying to get it to mess up.

 

As far as the vacuum leak goes, bust out a can of brake clean and start squirting spots of the intake where there is any rubber hose or gasket.

I doubt this is it, but have you checked the oil cap seal, PCV system, and valve cover breather? That pesky PCV hose under the manifold always caught me on my worst days. Having a leak there with the system setup as stock can cause a good vacuum leak as that is all "metered" air in there. Opening the oil cap at idle should cause the engine to stumble and possibly stall. Insta-vacuum leak!

[stprasinz]

I didn't even think to mention the pcv system metering the air.. And I was trying to find the change. (all things working fine, change some parts and now he has this.... )And I've had problems with the meter being to close to the turbo and the inlet not having enough pipe before it... as for the tps, 2 settings one supplies the 5 volt reference to the signal of closed throttle, and on auto's it sends reference to full throttle. Its like 2 switches nothing more.

 

if you pull the hose from the fpr the idle should jump (vacum leak from the hose being disconnected. just a thought As for the wide band they're pricey and in that case, he could buy a complete ecu and harness from a z31 and the sensors and be done worrying about the older analog ecu... I'd try one of those spare afm too....

[Gollum]

I'm planning on going megasquirt soon enough, probably next winter. And a LM1 is part of my budget for it. But I can't really justify it right now where I'm at with the car.

 

I got it running a bit better today, to where it only hiccups every now and then if I'm cruising bellow 2k.

 

I measured the CHTS at the ECU and it's showing about 10% high, which should be slightly rich, but it's still close to FSM specs. I'm about 95% sure it's not a PCV related issue, and I'm about as sure that I don't have a vacuum leak. The PCV wasn't even hooked up when I got the car 2 years ago, but now I've got the valve cover going to pre-AFM (so if anything it'd be causing it to go rich) but the lower block opening isn't hooked up to anything, and the PCV valve has been removed and has been replaced with a plug. Nearly everything on the intake has been plugged. The only think left on it is the FPR and boost gauge. And just for good practice I checked it today and couldn't find any leaks.

 

All that being said, I think I've concluded that my compounding issues have come down to poor electical signals and me futzing around with the AFM too much. I think my grounds are in poor shape, and that's lead me to having the AFM way out of adjustment. I'm going to work on some grounding solutions tomorrow and then see how it runs and adjust the AFM for how it runs at that point. If the car runs alight but the AFM seems extremely loose then I'll have to keep looking into it, but I have a feeling that after it ran so well today it's going to come down to making sure all my electrical connections are 100% perfect. It wasn't pulling 100% clean WOT from low RPM, but it's certainly workable from here.

 

In conclusion I think something that's mucked this thing up from the beginning was that the car didn't run well at it's first start with the new head for many reasons, and I'd been adjusting the AFM trying to figure out if it was lean or rich in certain conditions, and I've adjusted it so much now I'd almost forgotten where it should be in good working order.

 

More to follow soon enough.

[Gollum]

Well the weather was decent today. Cold, but dry at least. So I finally got out there and changed the fuel rail, took about 45 minutes or so taking my time. And surprise surprise, the engine is running a ton better. It coughed a little bit as the injectors were getting used to being used again, but after a short engine warm up it was revving better than it has yet. So I took it for a drive and WOW. Pulls strong all the way to redline again. Yay!

 

When I pulled the vacuum line off the FPR before it didn't spit fuel, but the engine was running. This time the engine had been off a good couple of hours and the FPR spit fuel. Strange eh?

 

Existing problems still: Engine is backfiring under in gear decel, and I still can't get the motor to easily rev up heel-toeing between shifts. This might be TPS wiring, but I also noticed that off the line the engine wanted to die again, probably due to the AFM having been tweeked to make the car drivable. So I've put the AFM back a bit so it's still a little loose, but closer to where it should be I think. We'll see how that drives tomorrow.

 

Overall that seems to have been the BIG problem that was making other issues completely masked. Now I should be able to troubleshoot the other issues better.

[Gollum]

So with my own annual BBQ/meet coming up, the pressure is on to get this thing truly streetable again and sort out the last bugs.

 

So the engine was running better pulling air from atmosphere rather than post AFM so I reconfigured all my intake... again... and it was still running rough, but better. I figured it was just off on mixture a taste, so I wired in a potentiometer to the CHTS, but on my way to doing so, I found a lovely 1k resistor already in series.... Lovely.

 

I'm not that upset with the guy who did this, but rather just upset that I've had to go through this car with such a fine toothed comb and overlooked this. This is an obvious mod, and the only thing that hid it was the fact that there was electrical tape EVERYWHERE, therefor the electrical tape on the CHTS wiring went totally unnoticed.

 

So I get the resistor removed, and pot installed and it's running slightly better still, at idle at least. But it was still coughing a bit here and there, and I decided to go back to an old diagnostic I'd done a million times thus far, which was checking the injectors to make sure they were opening and that all cylinders were running well. Well I get to cylinder 5 and pulling the plug doesn't do much... So I plug it in, unplug... nothing. Put a screwdriver on it and listened in and sure enough no clicking.

 

Switched it out today and tada, it idles like a champ. In fact for the first time since I had this engine torn apart I was able to lower my idle adjustment screw back to where it should be. I've normally had the idle adjustment wide open in order to get it to idle at 800. Now it was idling at a solid 1400 and didn't consider dying when coming off throttle.

 

Only..... (of course, there always has to be SOMETHING wrong right?)

 

Though it finally drives around at parking lot speeds great for the first time since the build, it now doesn't like good doses of throttle.

 

My hunch is that it's related to the CHTS. I'm going to get a new plug for the CHTS so I can get enough length to rewire it from scratch and see how it runs without any added resistance. Just to double check I'm going to drive it around tomorrow while it's cold to see how it drives. With the ultra high resistance at cold temps being a little off here and there shouldn't through it off too much. If it's driving a lot better at cold temps then that means it's certainly the CHTS causing it to run rich or lean once it's warmed up.

 

The saga continues.

[BlueStag]

.... he could buy a complete ecu and harness from a z31 and the sensors and be done worrying about the older analog ecu... I'd try one of those spare afm too....

 

What's that? The ECU and harness of a 3.0 turbo can be laid on a 2.8 turbo? To advantage?

 

I gotta go search for that......

[Gollum]

What's that? The ECU and harness of a 3.0 turbo can be laid on a 2.8 turbo? To advantage?

 

I gotta go search for that......

 

It takes a little bit of work, but it's well worth the reward. Most of the conversion work is in converting over the Z31 CAS into the 280ZXT Dizzy.

 

The biggest upside is that the Z31 can have a programmable eprom installed and then you can run an open source tuning software to recalibrate. Even out the door though the Z31 tune works quite well on the L28ET.

 

The only reason I haven't gone that route is that local yards want way too much money for a wiring harness+ecu, and local part outs normally have no engine and harness. That combined with how much money I'd spend getting a new erpom installed you might as well just go megasquirt for just a little bit of money. A good chunk of a MS2 budget for me is stuff I'd still be buying with a Z31 ecu setup. Overall from what I've seen a Z31 ECU would only save me about $150, and I'd be loosing some easy of use and tune-ability.