LHU/LDK swap into Cobalt: idle help

Full disclosure: I've skimmed a few threads here, just enough to know that it would take a while before I felt comfortable actually doing anything to my ECU. Also, I do not have HPTuners yet and this question is helping me gauge whether I need it.

Anyways, I recently completed a swap of an LDK for my cracked LNF. Everything is stock including ECU. I'd prefer to keep it that way, but if it's not possible so be it. Once the swap was complete everything worked nicely, except I noticed that the idle was a little more rough. Probably about once every 15 start attempts the car fails to start. It'll crank a few times then stop. A couple of times I'll get a P2105 code (throttle body forced shut-off), but not always. I noticed these values from my generic OBDII scanner during idle: throttle position is 17%, load is 26%, spark advance is +5*. With AC on, 19%, 35% and +3* respectively. I never checked these numbers before swapping the engine so I can't compare them to what they were.

My next idea was to clean the MAF and TB. I'm not sure that the MAF would do much here since my issue is at idle.

I was told on css.net that the engine like 10-20* of spark advance, 15-20% load, and 4-6% throttle position. These drive-by-wire systems are pretty new to me, so I've got a few fundamental questions and a specific one. First, in the ECU's logic, I'm assuming that there are a few variables that are working in concert during idle. Obviously I'm commanding 0 throttle, so I'm wondering if there is something commanding the throttle to open up to 17-19% or if there's just something wrong there? I'll try cleaning/resetting tonight or tomorrow so that may answer that. I just thought of reading the throttle position without the engine running to see what that says.

The other specific question, assuming cleaning the tb wouldn't do anything, is if anyone is familiar with the stock LDK vs LNF ECU table values? I'm wondering if there is a difference in the low engine load areas that might be causing a rough, high-load idle. I understand that the compression is 9.3 vs. 9.2, so I'm wondering if that difference is enough that they would account for it in the tables.

Sorry for the convoluted post. Thanks for the help!

I will start by saying out of all the tunes I've done I have honestly never tuned an LDK surprisingly. Anyways....


As for the 15-18* of timing and throttle plate below 5% etc that is on a properly tuned LNF. I don't believe there is enough difference between the two motors to make a difference but I'll let someone else chime in on that.

The stock tune is far from an ideal tune and will idle that low but it also has torque limiters that don't allow for you to know it's not idling how it ideally should be.

Give more info on what all is new and what is not. Such as was the throttle body replaced or was it the original, etc.

I haven't done an LDK, either. But, couple things I would suggest. First, hptuners. Doing a swap without tuning is like playing pin the tail on the donkey. Similar to buying a carb and slapping it on a modded engine. No one can help beyond guesses without a log and a tune. That includes people with LDK experience (cssob?).

The LNF/LDK is not drive by wire. The throttle is a torque requested pedal and doesn't control anything directly including the tb.

If you can afford the engine, seriously consider the software to tune it. Otherwise you're only halfway there.

First, it's a new long block. Intake manifold is original, along with everything outwards of it. Fuel rail, exhaust Mani, turbo, all original.

To be honest, my motivation to keep the ecu untouched was to keep resale as high as possible. When my lnf cracked, I was quite upset and the first thought was to get it running, broken in, and sell it and use the money for a different project. I might not do that, but that was my motivation. Now, if the car doesn't run properly without a tune then I wouldn't have much choice. And you must admit, this forum would be a bit biased towards tuning;)

My biggest concern right now is the car failing to start. It actually happened twice today out of 5 attempts, and one was bad enough to throw a code. I'm not sure if a tune would fix that...

Sounds mechanical to me and I would look at all of the likely causes. What is a rough idle exactly? Lots of different options when you say rough so if you could explain a little better.

"Rough" is more vibes than normal. It's nothing like limp mode, but it's more than I'm used to or would like especially with stock engine mounts.

I did a few experiments last night and am leaning mechanical as well. The engine was off and I moved the gas pedal. When I'm not on the pedal, the throttle position is 21.2%. Full pedal is 83.1%. When I leave the pedal alone, after 10 seconds it goes to 34.9%. I then let it idle. I turned A/C on and idle got a bit rougher, the boost gage read 24inHg vac (it would come up to 20inHg on occasion), throttle position was 16-18%. The engine-off limit check makes me think that the TPS is reporting bad numbers, which may be fooling the computer and throwing off the timing tables. I'd like to find a "control" where someone can check their limits on a known good TB. Anyways, I know this forum isn't for mechanical troubleshooting but thanks for indulging me anyways. I came for the LHU/LDK vs LNF tuning question originally. If/when I get HPT I'll be back;)

In all honesty the numbers your throwing out sound close to right off of memory. I don't care much about what that part is doing but what is your rpms at and what are they doing? The butter smooth idle with the numbers you want take some experience tuning.

If I had to guess your idle is just high meaning rpms are probably 1000 or so. If that is the case or if they are higher I would guess when the throttle body was off someone moved the plate by hand. If that is the case you need to do a relearn. There is a few different ways to do that. Google will be your friend on that one. I don't recall the specific steps people did that had luck with it but I know you should find it pretty easy.

RPM's are 800. I've looked into TB relearn and such. Someone at CSS.net helped me with the throttle position numbers and theirs are the same. I'm putting the idle issue as a tuning issue and may address it soon. Not sure if the shutoff issue is related but I got a tip that I should wait longer for the fuel pump to prime. I'll see.

Now I'm even more curious. Can you send me a PM with your number. I'd like to contact you and get a video so I can actually see what is happening?

This is a perplexing issue. I, also, have never tuned an LDK so I defer on that, but I understand the differences to be very minimal.

First, I feel like it could still be the TPS, but you confided with someone that their numbers were similar and I have never firsthand checked mine, so I cannot speak to that.

Second, if you're running a stock calibration and having a no start issue, then I will say it's mechanical in nature. Stock calibration should start 99.9x out of 100. I have never had an OE no-start issue that was tune related, pretty sure GM made these calibrations to start in all conditions.

Third, the fuel pump prime issue would point to a mechanical problem, in my opinion. I jump in the car, depress the clutch, and turn the key all the way, always starts.

Fourth, does the car run normally at cruise and WOT?

Now, what would the mechanical issue be? I can only speculate. Seems like you're sucking a lot of vacuum (though my Kappa doesn't have a vacuum gauge yet, so I might be off, but I thought normal was 20inhg). I would propose that maybe the IACV is bad/going bad. Without more details I can't say much more...

I'll PM Mike.

I'm also thinking mechanical. I have a "new" TB/TPS on the way, but I'm starting to lean towards the alternator now. I bought a used one when I did my swap because I broke the dang stud for the cable on mine. I tested it at Autozone, but I think their tester was only checking voltage and not spinning resistance. When I tried starting this morning it went into limp mode, but I also heard a quick squeal of the belt. Now I'm thinking that the alternator is binding at start and resisting too much at idle, causing a rough idle. The conditions above where my vac pressure stayed at 24inHg with A/C was after only idling to warm it up; no driving. It seems like if I drive for a bit then it has a better time staying at the appropriate 20inHg, and I'm thinking it's because the alternator is loosening. This is just a hypothesis. Unfortunately it's a pretty expensive one to experiment with.

I agree that if it's a startup thing it's likely mechanical. I don't know much about tunes, but I doubt there's much to mess with in startup, even cross-platform (LHU/LDK vs. LNF). If it were just idle then it'd be different and I'd be shopping for hpt.

Fuel pump prime isn't it. After work yesterday I let it prime and it started into limp mode. So that was easy to rule out.

The car does run normally at WOT and cruise.

So that's why I'm thinking mechanical as well and looking at the alternator. I'm also thinking alternator because I'm trying to figure out what's different from before the swap. The alternator is different, A/C compressor is not. The alternator was used, and it had a good amount of corrosion on the surface. I thought I covered the bases by testing it first, but I am less confident about the tester's ability to catch this sort of thing. The engine acts like it's struggling for air at idle (high % throttle, low spark advance, bogging with A/C) so that's what was leading me to TPS. I also checked the airbox for an obstruction. BTW, I understand that there is no IACV on these. And yeah, 20inHg is normal. That seems to be pretty standard across manufacturers.

I'm appreciate of the help. I'm not expecting much in this forum regarding mechanical stuff since it's the HPT forum so I will try not to bog it down much here. I'll obviously answer questions especially since you're trying to help:) And who knows, this may help you or a customer if they come in with a similar situation!

Another thing. I noticed when my turbo failed that the ECU learned a lot of habits that couldn't be unlearned (turbo choked out the engine with too much backpressure). This is also related to other people having some corrupted ECU files which was fixed with a write entire. Not saying it can't be something mechanical (loose connector, ground, bad sensor, etc.). I would try a write entire first just to be sure it isn't an ECU issue. You're leaving it stock still, but the write entire will undo any previously learned habits.

I got another P0606 (PCM malfunction) code yesterday along with a throttle actuator malfunction. Is there anyone in the PHX area that can hook me up with a cheap write entire?

Check all your sensors and wires first to make sure nothing got pinched?

The dealer is going to replace the ecm under the emissions warranty. Hopefully that's the answer

Thought I'd update. This is a little embarrassing tho. The new ecu didn't change anything. I noticed the voltage was low (<12v) at the same time my vacuum was high, so I thought alternator. Turns out it was a loose ground. Smh. The idle numbers are still basically what I quoted earlier, but I don't think there's anything "wrong." Depending on what my plans are with the car it may become a project to get the idle nicer and open it up a bit, but there are some things at home that govern whether I'd keep the car much longer. Thanks for the help!