Pellucid

This is the US 1.6 ECOTEC drive-by-wire engine with coil on plug ignition. I can't do the non-GM tool Learn procedure because the car will not idle at all.


Long story short, my friend's Aveo had the engine die due to overheating so I found her a 60,000mi used engine and installed it. It won't idle but runs fine if I can get the RPMs over 3000. At high revs the engine responds normally to the throttle but at lower than 3,000 it does not and giving it gas tends to kill it.

When you crank and start it revs to 2,500 then drops down to around 500, goes back to 2,000 and then back below 1,000 where it's missing and running rough then dies out. This happens consistently and if I can catch it and get it over 3k it responds to throttle normally consistently too. I have used both coil packs and both sets of plugs with no difference.

I've run it through three heat cycles now and it's throwing no active or pending DTC's. I'd try the non-GM scan tool learning procedure but it won't idle. I did try running it at 3,500 for 180 seconds and then shutting it off for 60 seconds but there was no difference in low speed operation.

I'm not really good with DBW throttles, don't like them and never had to troubleshoot one before but I thought maybe I had a bad throttle unit on the new engine based on readings so I hooked up the old throttle and got very similar numbers.

Is there a way to do diagnostics on the pedal assembly?

Here are the numbers I got with the designations from my HF scanner:

Engine Off Values, "new" throttle on engine

TP Closed - 27.2% WOT 39.6%
TP_R - 14.9% / 26.3%
TP_B - 28.6% / 39.6%
APP_D - 19.6% / 85.1%
APP_E - 9.8% / 42.4%
TAC_PCT 19.6% / 37.3%

"old" throttle connected to harness

TP Closed - 27.5% WOT 39.6%
TP_R - 14.5% / 26.3%
TP_B - 28.2% / 40.4%
APP_D - 19.6% / 85.1%
APP_E - 9.8% / 42.4%
TAC_PCT 19.2% / 37.3%

I had thought that maybe the new unit was bad and I'd see more matching values but since they're the almost the same with both units for now I'm assuming they are both good. The car drove to my house under it's own power with only a dying engine and I'm at a loss here.

Any help appreciated. I really need to get this thing out of my garage and do a timing belt on my own car before I've got two engines two replace.

cleveland63b

have you checked fuel trims? maf readings? fuel pressure? vacuum leaks?

Pellucid

I didn't note the specific values but everything was within the tolerances for the non-GM scan tool idle learn. RPM 450 - 4,000, vac greater than 5kpa, MAF greater than 2 grams/second and the voltage is good. All hoses and connections are good. The MAP sensor on the new engine got damaged so I swapped in the old one but I've run it with them both with no difference. The damage was to the connector so the terminals are exposed. No good for long term running but not an issue in a garage.

Wouldn't fuel pressure, MAF or vacuum leaks probably throw some sort of code after 3 heat cycles? (Cold to operating temp) Fuel trims I don't really know what I'm looking for. I can run it again and get those numbers for short and long if it would help. Fairly sure short term was in the low single digits and long term was around 8%

Thanks for your questions. I'll get the fuel trim numbers tonight or tomorrow. Anything else I can take note of? Do those TP sensor reading look right?

I checked the harness at the firewall again this evening to make sure the connectors were well seated and that checked out.

cleveland63b

i normally look at voltages when I'm looking at tps or app data but if there was no change between to two i'd think its ok as long as the wiring is good.


If you have to keep it throttled up to stay running the obd monitors may not be running for all the codes, fuel trims rule of thumb should be between -10% and 10% closer to 0% is better which once again aren't a good reading to go off of if it only runs with throttle. My first checks would be maf/unmetered air/vacuum leaks from the symptoms you're giving.


do you know anything about the used engine you got? how long did the vehicle sit? I would still do fuel pressure and volume just to eliminate it and get a fuel sample

Pellucid

Thanks for the thoughts and ideas.

I guess I need a better/pro level reader to read those voltages? I'm assuming both throttle assemblies are good because both the one the car drove in with and the one on the used engine give virtually identical readings.

I don't have a pressure gauge I can put on the schrader valve but I can rig one up or buy one. Again though I'm assuming fuel flow and pressure are okay because the car drove here and drove fine other than mixing oil and coolant from overheating too many times. As far as running and driving there were no issues pre-swap.


I didn't get to run it again tonight but I'll take note of the fuel trim tomorrow. I'm positive it was in that -10 +10 range though. My instinct tells me I really want to check the pedal somehow but I'm lost there. Maybe I'll pull the firewall connectors again tomorrow and make sure a pin didn't get bent somehow. I haven't even been able to find any kind of wiring diagram to start with. What I really want is a damn throttle cable!

I looked over the engine well before I dropped it in and everything external looked to be in great shape. I guess I'll drop a driver down there and make sure the manifold bolts are snug. It came with the intake on it, whole engine minus the starter and driven accessories but maybe someone started to loosen up the manifold. I guess I could see about pulling the injectors and fuel rail and swap those.

The intake boot is solid, MAF sensor should be good, clamps are all tight. Again it's the one it drove in with and there are no DTC's set or pending so I'm assuming it's a good MAF sensor. I don't remember the reading exactly but it was reasonable. I'll write that down next time.

It seems VERY sensitive on RPM which makes me feel like this is electrical and not a mechanical/vacuum type thing. Anything above 3,000 and the throttle reacts to the pedal 100% normally. Fine changes or punching it feels totally normal, anything below 3,000 and it wants to choke on giving it pedal or acts like it isn't seeing it. I'm trying to minimize any of that misfire running to protect the cat.

I can swap in the upstream O2 sensor from the old engine, I put that in last summer, but for a cold start idle I wouldn't think it would matter much.

Well I got it off ebay... I'm skeptical by nature but it seemed like a good buy. Seller with 99.9% positive on over 14k sales, good pics of the engine and car. The car had damage on the driver side body but that was it. 60k on the clock. I didn't pull the cam cover but looking through the filler I'd say the car probably had full synthetic it's whole life, almost no varnish. The plug bores were dry as a bone, no sign of any leaks anywhere and the plugs all looked even and good. I swapped over the brand new water pump the dealer just installed for her (for $800...) and the coolant passage behind the pump had no crud and showed traces of dex-cool.

So I don't know how long it sat but it's a big looking operation and they warehouse their engines. It was on a pallet on a shelf when they listed it and I'm guessing it didn't sit around their lot for long before they stripped the car down. Everything about the listing and the packing of the engine when I got it was top notch. .. I've got about twice as many caplugs as I used to. I don't know if you can post external links here but the listing it still up with the pics.

One thing that MAY be important, the engine came from an automatic and this car is a 5spd. I had to swap the engine harness over because there's a sensor/switch on the manual that isn't on the auto harness but everything else plugged right up. Both harnesses seem to be in good to great shape.

One of the first things I did was check to see if the auto engine listed a different part number for the throttle or coil pack than the manual but they were both the same and the physical PN on the coil packs are the same. Any info on things I could check on the pedal harness or out at the throttle body connector? Thanks again.

Pellucid

So here's a crazy idea...

I've got two throttle bodies, two intake boots and a helper. What if I leave the spare TB connected to the harness and have someone else crank the engine while I work the throttle?

If I can manipulate the installed throttle by hand and get it to run for even 10 seconds at a reasonable speed, 1000-1,500rpm or so, that's going to indicate that fuel and spark are where they should be, more or less. Right?

It'll throw a code for something for sure but it might run too. If I try this should I leave the MAF connected or is there a "limp" mode if it's disconnected that might work better. Something that assumes a reasonable airflow based on load and RPM.

Crazy idea I know.

Pellucid

I think I have PROBABLY found the problem but I can't find the solution yet.

I had only looked at the fuel trim numbers when it was running at high rpm. 3,500 or so. At that speed long term was pretty steady at -1 to 1, mostly around -.8 and short term was fluctuating in the negative range down to about -7.5 but at steady throttle mostly -1 to -2.

I finally got it to idle at a reasonable speed, though unsteady and popping once in a while, by just starting it and letting it try about 15 times in a row. Once it was idling I was able to look at the fuel trim numbers there and they were both about +25!!

That's when I noticed the AAT said -40. Now I'd think I would have noticed that before but I wasn't really looking for air temps so maybe I overlooked it.

It seems to make sense to me that even it it thinks it's very cold it might revert to more normal fueling above a certain RPM. I'm no expert but once you're turning 3,000 you probably don't need all that extra fuel to keep things going.

My problem now is that I can't physically locate the AAT sensor yet. I'm hoping I missed a plug, pulled a wire or maybe the AAT sensor just happened to die while the car was sitting.

cleveland63b

ambient air temp sensors shouldn't effect fuel trims, if you have +25% at idle then you have a vacuum leak somewhere, easiest way to find them is to get an alternative fuel (brake cleaner works) and with it idling watch your short term fuel trims while spraying areas that could potentially leak - intake gaskets, pcv hoses etc. and if the trims drop drastically while you're spraying you've found the leak.


that being said if your ambient air temp is reading -40 then theres definitely an issue, possibly that you missed plugging it in or a short/bad sensor

Pellucid

Thanks. I'll try propane or mapp and a hose, it's my normal go-to for vacuum leaks, not sure why I didn't think of it already. I like it because it's no muss, no fuss and if the worst happens and you get a fire you don't have carb/brake cleaner all over. Plus with a hose you can really zero in on a leak.

I think the AAT reading was a fluke. I'm pretty sure I would have see it when I was looking through the readings several different times the day before. I probably shut off the ignition and lost communication and forgot about it. Same thing happened today fuel trim values at one point, they were reading 0 while it was running and I'd had the ignition off.

IAT and AAT readings have been normal for the garage, 60F or so. I still can't find that AAT sensor though and I had the whole front of the car apart looking for it.

Pellucid

Well, I'm stuck.

I went over every vacuum line again, checked on the old engine to look for anything hidden under the manifold, vacuum checked the evap hardlines, took off the intake boot again and checked that again. Nothing found.

I played around until I got the car to idle again, sitting around +25 on both fuel trims and hunting and hunting from about 700-850rpm but running. I went all over it with mapp gas and a hose, then went around with another hose stuck in my ear. That's been good for isolating vacuum leaks in the past too. Finally I went around with a stick of incense for a smoke test. Other than around the alternator the smoke wasn't disturbed at all. I went around that whole area again with mapp gas and the 'ol listen'n hose. Nothing.

The one thing I noticed is that every now and then the short term trim will drop down into a normal range for a moment, then back up. It'll go negative but then back to 25 or so. And the long term trim tended downwards to about 21 over time. I still feel like it's electrical. Above 3k it's perfect, 2990 and it's a fat dog. I wish it would throw a code. I'm starting to think I'm going to have to talk one of the local shops into a house call with a better scanner to get this thing out of my garage.

Thanks for all the help so far.