You can find 165's rather common on ebay, but you take the risk of getting a bad one.

http://www.rosesandwrenches.com/ecms/ for a "gauranteed" ECM (if the ECM they send is bad, they'll replace it for you).

Alternately you can go to most autoparts stores and replace your ECM (with core) for around $100, and you know it will be good, or at least returnable if it isn't.

Start there. 1227165s are notorious for going bad. I'm on my 3rd. When they go bad, there's lots of strange-ness, including surging (I've experienced it), high idle (car will not idle below 1500, even though the commanded idle is normal - I've experienced this), TPS and other 5V reference sensors maxing out on the low end (I've experienced this - couldn't figure out why my TPS was only going to 2.5v).

Replace the ECM first, or even better, replace it with a buddy's if it is available and see if the problem goes away.

The next thing to suspect is the MAF.

To check for vacuum leaks, use some carb cleaner or other such combustable fluid/gas (some use propane) and spray around your intake. If the RPMs surge, you've found your leak. A mechanics stethescope is also helpful here.

Disabling the code thresholds will not keep the problem that is causing the code to go away. Don't bother.

Find the source of the codes (again, the ECM could be bad and throwing false codes due to failing sensor A/D electronics internally).

Don't give up. You'll get it, and you'll feel better than you've ever felt when you do.

M

 

I concur.

It's time to go to another ECM, if for no other reason than to rule it out as a source of all these problems.

 

im still inclined to belive a bad ground could be cuasing all of these problems. but i am also inclided to agree it could also be a bad ECM. i have replaced a few 165's but not lots of them.

 

I'll fourth the bad ecm. the 1120 idle RPM is a good clue. Plus all the other wierd stuff. I had a 165 go bad too, but I think I tortured it a little doing an ECU splitter that allowed me to run between two ECUs (but there were a few mistakes in the beginning, oops).

 

I went to the junkyard today to look for an ECM... it wasn't until I was halfway home that I realized the one I found in a Blazer cross-referenced. Oh well...

Besides, it was probably better that I passed on that -- 208,000 miles on a very "beat" vehicle doesn't usually say too much about the condition of the parts.

I talked to my friend, who works in a GM parts department, and he was able to find me a brand new unit for $81.00. All I have to do is give them my old one (gladly!). It's part number 88999194, ECM 16198259, which is the updated version of the '165.

Throughout all of this, I just wanted to extend a "thank you" to everyone who's taken the time to offer their assistance. I hope that this whole saga will be over with a new ECM... let's all pray together! I should be receiving the computer by next Tuesday, so I'll post some updates when I get everything installed.

 

I looked at this thread last night and like some others listed I am experiencing similar problems. Reading over the comments made, has given me lots of ideas to try. Blue86 I hope your problem gets solved but I wanted to thank you for posting as I and many others have problably learned alot from it.

 

** Edit

I read the first few posts and my comment was negated.. nevermind.

Deffinitely an interesting read though.

 

lol, I started a thead similar to this but never had time to check it out. Finally got to reading this whole thread and I think we're on the same problem which makes me think it's the ecm that's bad.
Check out my post about idle issues and the BLM's going to extreams, nothing we do has helped.

 

Thanks for all the support, guys. I'm glad to see that there's some consistency in diagnosing such an inconsistent problem! It would be great if the new ECM fixes things.

JPrevost, it certainly seems like we're experiencing similar issues. Now that you mention it, I remember seeing a commanded AFR at idle upwards of 15.xx for brief moments, and I could never figure out why it was doing that. The commanded AFR in open loop often was (unusually) at 14.71, where it should have been a bit lower.

 

I feel I am having a very similar, hard to find problem too. Its a miss on cylinder #2.

I just can't make pretty graphs like you can. At one point in time, I USED to be good at excel, but man, my graphs look nothing like yours. Its pretty hard to do any comparison with my weak graphs

 

The open loop commanded AFR is based on LV8 and CTS. As the MAF flow changes (as seen by the ECM), the LV8 changes which in turn changes the commanded AFR.

RBob.

 

Time out - what is CTS? Been following this interesting thread as I too have the $6E MAF set up and I'm getting my tuning equipment just as soon as Moates.net can get it here. Can't wait to get started!

Now back to the thread

 

Coolant Temperature Sensor.

 

New ecm did nothing! We're still having a surging problem. Next step we're going to completely disable the IAC physically and in the code. I can't figure this thing out only that when it's cold it's the worst, even warm restarts... anytime it's in open loop! What's really annoying is that 15.6 commanded AFR which I have a feeling is causing the problem. Closed loop and it's puuurefect, until the BLMs start to move. As if it's a vac leak. We'll install the wideband and verify the real deal then monitor the manifold pressure. If that doesn't work I'm going 7730 and never touching another 165... been too much of a headache getting these masks to be streetable with a far from stock motor.

 

I am having similar problems... Open loop is terrible. Cloud of fuel out the tail pipes and doesn't want to idle. Question about CTS. If your dash guage is close to what your TunerPro monitoing is saying, is it safe to say the CTS is good. My guage does have a 20 degree difference but the guage probably isn't perfect either.

 

Blue86Iroc,

Suppose I told you I have been ******* the exact same symptoms as you on my Xfire. I have TFS heads with a
CompCams Hyd Roller with similar specs to the LPE 219 cam with TPIS headers. I'm using a 7747.
Motor surges at idle, sometimes shutting off. When I come to a stop, motor will idle for a moment then go into surge fits. Will yours also surge when you put your foot on accelerator with no load and raise the idle speed?
As RBob and JPrevost can tell you, I have been all over the motor with new parts, re-doing the gaskets etc. First of all, set the FP so your commanded AFR is what you get at WOT. You'll obviously need to adjust the BPC accordingly. Then tune part throttle. I can tell you the jobwas made much easier by using a WB O2. In this case, I locked the BLMs and INT at 128 so motor ran off VE tables and MAP. I got the AFR to be in the mid 14's throughout the rpm range into the 5200/50kPa cells. In tuning the idle area, I had removed fuel in the VE runoff area.
Here's a couple of things to try:

1. Disable C/L Enlean if your ECM uses it.
2. Increase the VE in the 20-30kPa levels for the 400-1600rpm
range (Or equivalent for MAF systems)

I believe the issue is that induction system runs dry in decel. Once it turns back on, it gets caught up in a seesaw between low and high rpms where the discrepancy in fuel delivered is significant. The IAC's may be reacting too quickly causing the A/F mixture to go from lean to rich drastically and back again.

 

jackballs529, the engine CTS and the gauge sending unit are two different items. The gauge sending unit is nowhere near accurate, and neither is the gauge itself -- my temperature gauge reads on the 220º line when the CTS indicates 150º. Always trust the CTS over the gauge.

Dominic, thank you for the suggestions. I basically did to the MAF tables what you did to your VE table (read on...). I'm planning on getting a WB oxygen sensor in the future -- should make tuning a lot easier, especially at WOT.

JPrevost, take a look at what I've found out. Might help you out...

My new ECM came in yesterday and I had some time to tinker. When I removed the old ECM, I noticed that it was (interestingly) a remanufactured unit, just like the new one I bought. I looked for a build date, but I couldn't find one. Hmm...

I first used my inital ARAP chip (v0.01), containing only timing changes and adjustments needed for pre-'89 vehicles (i.e. fan flag). The engine still surged at idle and would not arrive at the target idle speed of 800 RPM. BLMs were still 160 in cell 0, and the resetting of the INT was still linked to the RPM fluctuations. It threw Code 33, too. I was using my OEM MAF sensor at this time.

Replacing the OEM MAF with my brand new aftermarket MAF, open loop idle was much more unstable. The motor went into a high idle (~1500 RPM) right before shutdown, just like tests from the other day. No codes at idle, but a '33 about ten seconds into driving.

Next, I tried a stock ABWT $32B bin, the only change being a slightly higher Idle RPM vs. Coolant Temperature table. The engine ran horribly, surged like crazy, had no power, and that stupid high idle came back again. Code 33.

Finally, I re-replaced the aftermarket MAF with the OEM MAF. I also loaded up "v0.08" of my ARAP bin, which is identical to v0.01, except the timing at idle has been raised to about 22º and the first few entries of MAF Table #1 have been increased to account for a supposed lean condition in BLM cell 0. When I first started the car, the engine speed shot up to about 1500 RPM and stayed there. IAC steps were 0 (closed). Confused, I restarted the engine and watched the ALDL output -- IAC started at 160 then very quickly went to 0. The idle remained at 1500 RPM again. Still a little intrigued, I left the engine running and adjusted the minimum air screw until there were about 40 steps at 800 RPM. I let it run like this for about a minute or two. There was no surging. None at all. Let me say it again:

The engine did not surge at idle.

The BLM value in cell 0 was 116, not 160. This means that adjusting the MAF table DID do something, unlike before -- the bin that I used this time was the exact one that I tried before, with the old computer, when I noticed no change in anything.

This car has never, ever felt as smooth as it does right now. The RPMs stay in line perfectly at idle, during braking, and even when I lift my foot off the accelerator while coasting. It actually drives like a normal vehicle!

Also, no codes were thrown. I believe this is due to the fact that the minimum air was, for some reason, set too high and allowed the MAF to read enough air to trip the code -- but before the data logger could react. That's probably why I didn't see anything in my logs that made sense in regard to a Code 33 error. How the minimum air got so high, I don't really know. I set it about a dozen times beforehand.

I played with the idle speed a bit and found 750 RPM to be satisfactory. I might try lowering this a tad when I straighten out the rest of the idle fuel mixture.

In essence, I surmise that the super-lean condition in cell 0 was the root cause of the RPM drop (I don't want to say "surging," because that implies "stall saver," which WAS working correctly -- the rapid decrease of engine speed was the real problem). Note that the RPM drop occurred under two cases: 1) INT resetting itself, and 2) entering BLM cell 0.

In the first case, I still believe that the INT was playing a large role in fuel correction. It's designed to periodically reset, which removes the correction factor, but it SHOULD be made up for by advancing (or retarding) the BLM. When the BLM was maxed out, as it was, it can't go any further -- the INT resets as normal, without compensating using the BLM, and the RPM drops. A possible way to test this theory would have been to raise the BLM upper limit and see what happens.

In the second case, BLM cell 0 was usually entered when the accelerator was released while at speed. This motion would cause an RPM drop. Entering a maxed-out cell, the ECM has no idea how much fuel to add (or remove). It's possible that the engine speed dropped because the correction factor just wasn't there. Since the INT is reset to 128 when a new cell is entered, INT compensation couldn't be factored in -- the stall saver took over and added fuel as necessary to bring the RPM up to where it should be.

Now, do I believe that the ECM was also at fault? Absolutely. The bin that richened the mixture at idle, via the MAF tables, had no effect with the old ECM. My BLM values don't "stick" anymore, either.

Also, I feel that the aftermarket MAF is somewhat to blame. I'm not surprised about aftermarket not equalling OEM, but I am surprised that my old, beat, "screens removed then replaced when I learned it wasn't a good idea" OEM MAF still works like it should.

Based upon experimental data, tepid analysis, and the gracious help of you guys, I've come up with a postulate:

A bad ECM, in the company of a super-lean condition at idle, can cause problems.

Its respective corollary:

A lean condition in an idle BLM cell can, and probably will, cause engine surging due to repeated, rapid drops in rotational speed.

And finally, a theory:

Aftermarket MAFs suck.

I've put about thirty miles on my Camaro since everything started "going well," and there hasn't been one bit of a hiccup as far as driving is concerned. There is a slight issue during starting, but I want to try a few things before I post anything about that.

Needless to say, I'm about as giddy as can be .

 

Excellent write-up. Thanks for keeping it going and keeping us informed.

Like I said in an earlier post, its a good day when you figure out a problem like this!

 

Thanks, Mangus!

You are SO right!

 

That's what we are going to do Friday evening! Except I was going to go so far as to closed the IAC, unplug it, and disable it all together in the code. Then remove the idle over/under speed SA tables and see if we could a decent idle. Now we'll just adjust the idle speed and see where that leaves us, then increase fuel at idle... which I had done but with no change before so we'll see how it works out for us and I'll keep you updated.

 

BlueIroc,

So if I understand correctly your saying that the drop in rpm sends the BLMs into a different cell not normally associated with idle. That cell is very lean and starts to send the motor into a lean/rich tizzy. Have I summarized that correctly? It seems to match what I have observed. I've attached a WB scan that illustrates whats happening. The RPMs drop from a steady 850's into the 400 range. This sets up a spike in the AFR with a lean surge and then over again. Is the solution for me to re-map the low/med/high BLM mapping so I don't cross into another BLM cell when at idle?

 

JPrevost, I hope your next round of tests produces some results... good luck! Yes, definitely let us know what you find.

Dominic, those are some interesting plots -- definitely looks like something similar to what I was seeing with my car. Your summarization of the theory is a little off though... my initial explanation was probably a bit scattered, so I apologize for that. This is what seems to be happening based upon my observations:

Case I: Assume that you're sitting at idle with a BLM value of 160 in cell 0. The ECM is adding fuel to compensate, but the limit of compensation with respect to the BLM is maxed out. Now, the INT is still trying to update the pegged BLM, so it continues to increase in value, all the while adding fuel (since the INT is also a fuel correction factor). At a specified interval, the INT tries to update the BLM, fails to do so, and resets itself for a new sample of data. The previous correction via INT is lost because it has been reset to the nominal value of 128. Fuel is lost, RPM drops, and the stall saver raises the engine speed to the target idle. After a few more seconds, the process repeats (surging). Ultimately, the ECM knows that it needs to add fuel, but it can't "remember" the exact amount of fuel to add because the maxed-out BLM prevents it from doing so.

Case II: Assume that you're driving at 45 mph with the accelerator depressed 20%. Under these conditions, you might be in BLM cell 6. You want to coast, so you let your foot off the gas pedal. Because the RPM and load has changed, you enter BLM cell 0 (which is the same cell used during idle). The ECM is going to correct the mixture with what it already has learned -- the BLM value. If cell 0 is "super lean" (i.e. BLM = 160), this might not be enough fuel, because the INT may have been partially responsible for fuel correction (as in Case I). The RPM drops because there isn't enough fuel being added, and the stall saver once again prevents the engine from quitting. The INT can't really help (from a correction standpoint) upon cell 0 entrance because it gets reset to 128, and just doesn't update fast enough to do anything. It's very similar to Case I.

As I've mentioned, these conclusions that I draw are based on the analysis of my vehicle's problems. I can't validate the thought process of the ECM during Case I and Case II, but my observations indicate that they seem correct.

With regard to your question, if your sensors are working properly (MAF/MAP, specifically), and the ECM can calculate load and receive an accurate RPM reading, I don't see how you could be crossing into "wrong" BLM cells. I would investigate the mixture status in cell 0 (and cell 1, to be safe) and ensure that it's where it should be (I don't have a WB, so I'd just use BLMs). It appears that cell 0 is fairly important, since it's used during idle and coasting -- precisely the times when surging seems to be most prevalent.

 

blue86iroc

I have idle problems as well. My car starts and shoots a cloud of fuel out of the tailpipes. Then it runs really rough and if I hit the gas peddle and let off, it will stall. Once into closed loop it straightens out a bit and I can get my BLMs in range but the car lopes and feels like a mis. I to have changed my throttle screw to a longer one which gives me the idle I want. But as you say, I think that gives to much air as I get the code 33. The car also stalls when put under load, like shifting to reverse. I tried a new ECM for about 5 minutes and it didn't change a thing. I didn't have time to mess around and get it into closed loop. I just assumed if the car started and did the same thing that the new ECM didn't do anything. (was I wrong about that?) I have put lots of new parts in it hoping to get lucky. (New Aftermarket MAF, TPS, IAC, Relays, EGR, wires etc.) Just did a new rebuild last summer with all the goodies in my signature. Do you have an idea were I should start? I know that having the idle screw turned that far out is masking a problem.

 

jackballs529,

When I was trying to get my idle set right, I too bought a longer minimum air screw -- I thought that the factory screw wasn't allowing the throttle blades to open far enough for a correct idle. In my case, however, I was wrong about that. The stock screw is definitely up to the task; I mean, we're only talking about a few hundred RPMs here. By using a longer screw, the throttle was open too far and it wouldn't settle on an idle speed (but I have no idea why it worked some of the time). Your best bet would be to reinstall the factory screw (and spring), set it for where the car will idle with the IAC connected, and monitor your IAC steps in closed loop at idle. Turn the minimum air screw slowly until you achieve the desired steps -- I was happy somewhere between 30 and 40. If the ECM and IAC are working properly, your idle speed should not change unless you max out the IAC in either direction. The RPMs might fluctutate once or twice, but they'll stabilize. I like this method better than the other way that requires disconnection of the IAC.

If the original ECM isn't completely fried, you might not see a difference when connecting a new one up -- that's just from my experience with electronics and computers in general. The ECM is a complex device, and not all of it is going to break at the same time. I know that my old ECM would sometimes fail to report data, i.e. updating the BLMs. It also failed to perform certain calibrations, i.e. adding fuel via the MAF tables. I noticed basically no difference when I hooked up the new ECM -- but once I started logging data and making adjustments, it was like night and day. The thing actually worked!

 

BlueIroc,

This is a great thread because I can't count how many times I have read about cam swaps and people ending up with surging idles. The history behind mine goes something like this.
I originally had a CompCams Hyd Roller that was a 264/270 adv
with 112LSA, 212/218 @ .050 and .488/.495" using 1.5 rockers
Ran fine. Idled great. I was using 10.5psi FP which is normal for a TBI, but it was running lean at WOT. I felt I wasn't getting the most out of the TFS 23D heads so I ran a static compression check one day finding that it was over 200psi in all cylinders. To me that suggested overlap wasn't long enough causing some pumping losses with piston working against the A/F mixture in the chamber. So I sent my airflow specs to Motors and Machines and they came back with a cam very close to a single pattern, lower LSA, and greater overlap. Specs on the current cam are 269/269, 218/218, .495/.495 at 108LSA. So if you do the math, you find that the overlap went from 43d to 53d at adv and from -9d to +2d at .050. Currently using 1.6 rockers so it bumps the "effective" overlap a bit more. I tuned the motor just fine at the same 10.5psi FP. Ran great but still lean at WOT. We put a WB on it and decided to pump up the FP to 16.5psi for WOT which obviously made everything else rich, but idled fine. Still a bit lean at WOT. In the course of tuning PT, the surging began at idle particularly when the BLM/INT was unlocked. I too noticed that the stumbling idle happened when the INT reset. It dropped like a rock at idle and then would reset creating the stumble.
I've been at this for the past 10 months playing with just about every PID/IAC/Idle/VE/SA control I could find not to mention replacing parts. Tuning using a ZT-2 WB (the 160 baud ALDL in a 7747 is worthless) the rest of the VE table is in the mid to high 14's. No probs except for a slight stumble on trailing throttle decel. Now using 20psi for FP. What you find is that at idle the PW
is pretty short using this much pressure. One of the problems of using TBI is not having a real adequate selection of Injector sizes to do the job.
The following seems to have yielded the best results so far:

1. Lowered SA in the 4-600rpm cells
2. Bumped the VE in the Low kPa region of the VE table
3. Slowed down the IAC action at Low VSS
4. Disabled the C/L Decel Enleanment

I still get a surge when I first take the car out and then come to my first stop. Then the idle settles down. It will be much smoother on subsequent stops. BTW, I'm running O/L idle and I am still able to get the surging so not sure whether the problem's in the VE table. I've also gotten it even having disabled the Stall Saver.
I bring up the overlap increase because I believe it may be part of the reason for the idle seesaw. When in decel, I have seen kPa values that have been very low. Sometimes down to 8kPA and very definitely in the teens. Using a VERY uneducated guess,
I am theorizing that the high vacuum conditions in the plenum during decel effectively sucks out whatever fuel is there. I realize this may not be the case with "dry" manifolds like TPI. But if you're familiar with a Xfire manifold you know that there's ALWAYS fuel at the bottom of the plenum. And the high vacuum/low kPa sends the VE table into cells not normally used. If they were lean before, it didn't matter since the cal never entered that area of the VE table but now it does. In addition, with the higher FP being used the PW is very small at idle while it was nominal with lower FP. So accurate fuel metering suffers at such low PWs.
Food for thought and comment.

 

blue86iroc

You talk about Cell 0 and cell 1. I use TunerProRT, How do I look at those cells and what am I looking for?

Thanks,
Jason

 

Here's the boundaries on a 7747. As you can see they can be changed along with the tolerance for RPM and kPa.

 

I'm a MAF guy.

 

Stock ARAP boundries are based on RPM and air flow (g/sec).
RPM bounds; 1000, 1200, 2000
MAF bounds; 13, 21, 34

 

Interesting stuff to ponder over, Dominic. Since MAP sensors work via engine vacuum, I could understand how pulling varying amounts of vacuum with a high-duration cam could cause problems. That, and the fact that TBI is a wet injection system could complicate things. I've never worked with MAP air metering before, but maybe some other people have seen similar issues. With MAF, I can see just what MAF table I'm in by looking at the amount of air that's currently being reported -- is there a way to find out what portion of the VE table you're using at a given instance? If so, maybe you could adjust those sections of the table... assuming that they aren't needed under different conditions.

Hmm, maybe your idle suffers at higher fuel pressures not solely due to the small injector pulse widths. By increasing the FP, the conical spray of the injectors change... the stream at the center becomes more concentrated. It might be possible that the fuel isn't atomizing properly and is pooling up inside the manifold. What if you just run a few psi above factory and see what happens?

Jason, what JPrevost said. Depending on the status of the engine, you'll be in a certain BLM cell. At idle, when your engine speed is less than 1000 RPM and you're flowing less than 13 grams of air per second, you'll be in cell 0. The overall plan is to get your mixture correct in every cell that's used (BLM = 128). From what I've heard, MAP systems can become much more accurate here; I'd be happy between 126-130 with my MAF system. You'll need to change the MAF tables to adjust the fuel in each cell. If you want to change any value at the boundaries, you have to change the respective table scalar. There is plenty of literature about the MAF tables on this site, so be sure you understand how the tables work before you adjust anything. This thread should help you out a bit.

 

Well one things is clear, I hate these old MAF's compled with the 165 ecm. Just a small change in the MAF table in the 8-12gps range would swing the BLMs by 10. I was going from 132 to 122 and finally got a 126 and called it quits.
I also when through the WHOLE thing just making changes I felt would work because taking notes was getting really time consuming... the result was a perfect idle but I have no idea what change (or the callaboration of all the changes) fixed it.
One thing I did to get the IAC to come down was move the TPS voltage at closed throttle from .73 down to .63, that helped when the "Calc BPW (Don't use Table)" was set as in the ARAP, but only the ARAP uses this table and the idle was still too high (from that darn table). So I unset that and the idle pured near perfect. Then I was finally able to adjust the timing watching Diacom on the old laptop and we got things spot on. His transmission is SO violent though that it'll sometimes stall the engine when put into gear, like it's scary violent. We tried shifting from park to reverse at 1100rpm and I thought my butt left the seat from torquing the chassis so much... so that's something we'll have to fight with . I also disabled EGR by raising the MAT threshold, that might have done it. I don't know and I don't care at this point... I'm just fed up with the $6E code. I'm sticking to $8D and SD from here on out. There's something to be said about the responsiveness of the SD because when I tuned a relatively stock 95 Lt1 and put it in SD mode, modified the tables and the thing felt great! Much better than running on the MAF and yes, I spend a day and then some tuning the with the MAF but it just felt kind delayed... might just have been the AE or some other code, all I know is closed loop (and open) was "krisper" with SD making me even more biased .

 

JPrevost, sounds like 6E hasn't been treating you well. That's too bad, I must be one of the lucky ones. I have a radically cammed 1967 327 which now has a Holley Stealth Ram on it, and I have had very little to change to get the car to run and drive as well as if it came from the factory set up like this. Screams to 6500 rpms and idles with its characteristic heavy muscle car lope, but perfectly steady idle speed.

 

I bet it would treat me better if I had more control over the parts selection. Working with a bosch o2 sensor makes me want to scream and I have a feeling that is part of the problem if not all of it. I also think EGR was a factor, I don't know if he got the correct EGR valve. Also, this motor liked having the extra choke AFR... but I can not for the life of me figure out why his commanded AFRs were going LEAN (er than stoich). It was trying to idle with a 15.2-15.6 on a warm restart but all of the apparent "%" enrichment open loop values were positive values... so it's something I can't figure out, maybe a hidden table in the code of the ARAP.

 

i think alot of the variance come from the crappy Bosch maf not the code itself. ive seen dramatic swings in calibration from MAf to MAf with the bosch units. they are just junk.

 

Oh, and the fact that this one was de-screened probably didn't help any.
Now I'm probably stepping over that "line" but what makes more sense; Keeping the 165 and adding an LT1/LS1 maf or going to a 730 and doing some very minimal leg work? I think option 2 is rather inexpensive and GM did make more HP with it stock than the MAF (code maybe).

 

thats was what i like to call a compounding factor.

 

JPrevost, I think I'm starting to agree with you about SD -- it seems like my car is acting goofy again (and just when I thought I had things fixed!).

I took the car out the other day. I was looking forward to driving it, since things seemed to be working correctly. Had a real hard time starting it up. The RPMs shot up to about 2000, resulting in Code 33 again. That doesn't worry me too much, because I think I can solve it by lowering the "initial IAC steps" constant so the engine doesn't receive an influx of air upon startup.

I tried to let it idle for a few minutes before I started driving, but the thing quit on me about five times. No surging or anything, it was just like someone flipped the switch to "off." Kinda weird. Smelled very rich.

So, the time comes for me to take it for a drive. Remember how I had the BLMs around 116 in cell 0? Yeah, well, now they're back up to 160. The mixture looks leaner everywhere else, too, but it's this damn cell 0 that's still screwed up.

The car still drove alright, but you could tell that something just wasn't right. It didn't surge at all, but I noticed a minor (~200) RPM drop when I let off the accelerator at speed. Also, I barely even used cell 0... usually the ECM was in cell 1, which was equally lean.

I compared my logs and found virtually no difference between them. The only thing that's distantly related is the MAF airflow. My readings are slightly lower on the problematic run than before. I wonder if my GM MAF is finally giving up the fight.

It would be nice to get out and do some additional testing, but I'll have to wait until the snow melts .

 

blueiroc,

Its good to know I'm not alone on this. Not that I wish for this kind of a hassle on anybody, but this tuning issue seems to be prevalent when swapping cams. Be good to nail it down once and for all. I bumped the timing in the High kPa/Low RPm range and that seemed to help some, but I think I will turn my attention to the INT Delay and the Prop Gains. I know I read RBob's paper on PID controls and that with a more lively cam, you could turn these down, but mine seems to like it a bit higher. I'll keep you posted on my results today.

 

blueiroc,

Here's an update from this weekend's chip burning. I had some time so I re-read for the 50th time, Rbob's articles on IAC and
PID controls. Burned a chip having made some mods based on his recommendations. I also remembered someone having used the PID controls from the 4.3 van bin on a 7747. I figured the 6cyl would have greater idle variation issues than a stock V8. So I used that for a guide as well. When I took the car out last night I had a case of idle stumble when I stopped for the first time after getting under way. After that it seemed the idle calmed down which suggests the IAC reset routine(s). These are the changes that I made. Maybe they can help you as well. Be interesting to get a second person trying these:

1. Increased Idle Prop Term Duration 2F8
2. Put another 5% fuel in VE tables run off
at 20-35kPa and 400-2000rpm
3. Increased(slightly) the delays for small and
large error IAC retract/extend 5C7->5CA
4. Increased DFCO Stall Steps 5D3
5. Increased (lowered) IAC Low VSS extend delay 5E7
6. Increased (lowered) IAC Low VSS delay coeff 5E9
7. Lowered Stall Saver Decay Coeff 5EB
8. Increased throttle follower in D 5F1
9. Lowered MAP Transient Coeff 0F
10.Increased (slightly) the Prop Gain Flow v Air Flow
at bottom of the table for 0 and 16 gms/sec 4C7
11. Increase O2 Prop gain at idle 2F9

This isn't the final answer but the motor was much better behaved. Another related ( I think) symptom is trying to keep a steady raised idle at no load. The motor seesaws when I try to hold say 1200rpm. However, it can hold a steady high idle in warmup. If I had to take a guess at what was the one param
that made a difference, it would be the TF%. I mistakenly had thought that the TF% for Park at 5F0 was controlling idle. I have a manual trans so the TF% in D at 5F1 is what needed a change.

If you choose to make these changes do them in smaller increments otherwise you may go too far the other way.

 

Good observations, Dominic. Some of those adjustments aren't available to me, as I'm using MAF, but I'll look into the other ones (I just gotta find some time; I've been so busy lately!). Did you make all of those changes at once, or is that list in the order that you made the changes?

Just a quick update from my end: We skirted the rain this afternoon, which allowed me to try out my car again. I put the aftermarket MAF back in, making no other changes, and started the engine. As suspected, it ran horribly... barely idled, smelled rich, etc. Putting the OEM MAF back in its place let it run smoothly, just like before.

However, the surging returned. So, I reloaded my base ARAP chip, with no adjustments to the MAF tables or the timing. It ran great, idled great, felt wonderful, and... guess what? BLMs at idle were at 132, NOT 160 like it always was with this chip. Don't ask me why, because I have no clue.

The only thing that I can possibly surmise is the following: My OEM MAF might still be dying, but it was offering "skewed" readings before, which required the extra fuel to be added (re: my post from a week ago). However, the proper compensations were being made, and everything seemed like it was operating normally. I think I might be in a "sweet spot" right now, but again, I'm not sure. I don't even know if this logic is correct (I haven't studied the logs enough to be sure). The only thing that I can do at this point is just drive the car and see what happens.

 

blue86,

I made those changes in one burn. If you'll note, they deal primarily with the IAC coefficients and delays. In fact, I burned another chip that took these same changes and went an additional increment on them. Idles better than previous so it seems to be moving in the right direction. As I said, I think the main contributor to my problem may be not enough Throttle Follower. Did you try to hold a steady high idle under no load?
If you did, is it steady or does it surge like there's a vacuum leak?
Heve you thought of doing a "compare" between the first ARAP bin in the one that seems to work?
The description you give of the later edition ARAP sounds like the EGR is involved somehow.

 

I'm very much a noob, but I'm going to throw in my 2 cents anyway. There's a few things that make me very curious.

1. You've tuned, and tuned, and re-tuned without much drastic change to your idle quality. Kinda seems like putting a band-aid on a more severe problem.

2. You've checked for vacuum leaks time and time again.

3. You've replaced most if not all of your fuel delivery and air measuring electronics.

4. You've checked ecm and sensor grounds with no results.

5. Your graphs and charts seem to indicate that your engine is randomly cutting out, espeically at idle or under deceleration, or no load.

I'm thinking the random drop out may be caused by an ignition problem, more specifically a weak spark. If the major ignition parts are new, then I would check to ensure that your engine block is grounded. Even remove the ground strap, clean it, re-install it, or attatch another one.

Maybe I'm way out to lunch on this one, but it's just one more thing to check.

-HD

 

blue86,

Here's another thread that closely follows this one.
https://www.thirdgen.org/techbb2/sho...hreadid=293497


I have made the changes to timing suggested by JPrevost in this thread, and son-of-a-gun if he wasn't right. Too much timing. Is there anything in the base ARAP chip that lowers it in the idle area? I dropped 4d across the 4-600rpm range and 2d in the 800-1000rpm range. Major difference in the idle and the trailing throttle decel stumble I was experiencing. I also dropped 4d in the 30kPa rundown area.

 

Promise you guys aren't going to make fun of me...

My car no longer surges. No changes to the EPROM were necessary to fix this, just about an hour with a multimeter and a piece of wire before the light bulb finally went off (figuratively, of course ).

I decided it would be a good idea to test MAF burn-off functionality. I wasn't getting any codes, but something told me to check it. So, after examining a schematic, I made the necessary jumps between connectors and seemingly closed the burn-off circuit. However, the little wire in my OEM MAF was no glowing orange like it was supposed to. Curious, I measured the voltage at the appropriate pins in the MAF connector -- it looked good; no wiring problems here.

I initially suspected the MAF itself... maybe the burn-off circuitry was fried or something. I connected a lead from the MAF output signal pin to my DMM and measured the voltage with the ignition on. It was something like 0.66 V, which seemed a bit high for just sitting there with no airflow (it was probably dirty since burn-off wasn't working). Blowing through the MAF raised the voltage, imprecisely confirming that the sensor was doing something.

I tested burn-off on the OEM MAF once again, this time using my wire lead into the DMM to ensure that I was getting power when the sensor was hooked up. Sure enough, the MAF wire turned bright orange (and a nice puff of smoke even came out, haha). I fiddled with this for a few minutes, intermittently able to reproduce burn-off.

That's when I looked at the MAF wire harness -- something about those contacts just didn't look right. I bought another connector from a junkyard for comparison. The contacts in the old piece were bent back, preventing reliable contact with the MAF sensor. I spliced in the new connector and it's been fine for the last 200 miles. No codes, either. If I just would have listened to funstick in the beginning!

Now, I'm not sure how my interpretations throughout the past few weeks are going to change in light of this. I mean, you can't make general conclusions if your entire process is skewed. There's no way to tell when the MAF sensor was making contact or when it wasn't, so I'm sure most of my past logic is going to be incorrect. Hmm, I'll have to think about this when it's not one o'clock in the morning...

 

Don't I wish mine is that simple.

 

Well, my story isn't over yet... I was just too tired to finish writing this morning. I still have an idle-related issue (but it's not that bad).

I currently have the idle set to 750 RPM. Any time that I give the throttle a quick jab at idle, in Park, the motor almost stalls out as the RPMs fall. They seem to hit about 500-600 RPM, and then the ECM senses "something" and brings the engine speed up again, surging maybe once or twice. After that, it idles fine. I have no problems driving whatsoever. It becomes more of an issue if the idle speed is set lower (650 and 700 RPM), and less existent if it's higher (even 775 RPM is worlds different than 750 RPM).

The reason I said that the ECM senses "something" is because I don't think it's the stall saver -- I set the thresholds according to another thread to effectively remove the stall-saving feature. I noticed no changes in anything.

If the ECM could "ease into" the idle speed after a quick rev, I wouldn't be having this problem. Not sure if it can even do something like that.

I looked at my logs and the EGR is enabled right before the RPMs fall, shortly after I let off the gas. I still need to read up on EGR operation, but it might have something to do with it. Maybe also in the way the cam duration (208º/221º) affects the engine under a no-throttle condition, but as many have said, the ZZ4 cam isn't that radical.

Just some thoughts of mine; I'm going to try some experiments and see what I can find.

Dominic, you mentioned the EGR in one of your previous posts. Do you have any enlightening info?

 

You have just described my symptoms to a "T". I disabled the stall saver and no change. In fact, my EGR channel is blocked off and diabled in the bin. If you blip the motor, the RPMs will drop and then bounce up but then the engine feels like it goes into a stall but recovers. It may surge once or twice and then idle. Now I am doing this in Open Loop idle. If I enable C/L Idle, it will just continue to surge. In cruise, the motor is beautiful. AFR is in the high 14's. My EGR is diabled. I would recommend you do this as a matter of course when you tune anyway since the EGR will mess with the VE/SA table settings. Try to isolate those first. Get them right, then enable EGR if you need to.
I have had better success when slowing down the Low VSS IAC action and reducing the Initial Advance at idle and below as JPrevost suggested. I am also going to richen up the idle area somewhat.

1. If you are sitting at 0mph, and raise the idle speed with the accelerator, will your motor hold a higher idle speed or will it surge up and down as if there was a vacuum leak?

2. What is the BLM like when it goes down to that 5-600rpm range? My WB shows a lean condition at that point.

3. Do you notice a slight stumble when you are in a trailing throttle decel, not a completely closed throttle?

My cam is a 269/269 @ .006 and 218/218 @ .050 at a 108 LSA.
I have had to bump my FP for the TBI units to 20 psi to get enough fuel for WOT. Motor didn't do this with this same cam when I was running 10.5 psi.

Quote:

I looked at my logs and the EGR is enabled right before the RPMs fall, shortly after I let off the gas. I still need to read up on EGR operation, but it might have something to do with it. Maybe also in the way the cam duration (208º/221º) affects the engine under a no-throttle condition, but as many have said, the ZZ4 cam isn't that radical.

Just some thoughts of mine; I'm going to try some experiments and see what I can find.

Dominic, you mentioned the EGR in one of your previous posts. Do you have any enlightening info?

We need to keep at this till its solved.

 

As for the EGR, you need to take it off and test it. You can use a hand pressure tester and fill it with pressure and see if it holds. It is a potential vacuum leak if it doesn't. You also need to check the vacuum selenoid. I don't know how to do that. I have asked but no one really replied. Having idle issues of my own...

 

jackballs, thanks for the tip. I checked the EGR about a year ago when the motor was apart, and everything seemed good. I haven't tried anything with the solenoid, though.

Dominic, I must have misread one of your threads a while back -- I didn't realize that your surging was occurring when the throttle was opened and closed fast... same stuff I'm seeing.

My car does the "rev-drop-recover" thing in both closed and open loop idle. Holding the accelerator at a certain RPM and easing out of it slowly, there is no RPM drop. If I let go quickly, I can reproduce the problem. In either case, I have no surging whatsoever... only one little spike after the RPMs initially drop, and then it's fine. Trailing throttle deceleration doesn't even do this; no problems whatsoever. If I never thought to rev up the engine in Park, I would have never even noticed this, that's how well the car is running.

BLMs at 500-600 RPM are around 118-122. I still need to tweak the MAF tables a bit to raise this, but I have a feeling that the ECM might be lying to me -- I read somewhere that having a large-duration cam will sometimes show inaccurate BLM numbers in idle cells. I guess that's where having a WB would help.

Also, like JPrevost, I'm seeing more occurrences of a 15+ AFR ratio in open loop idle. I wonder if it's because my BLMs are so low, it's trying to lean things out (since the ECM calculates proper fuel in OL by commanding an AFR and using learned BLM values).

I have my idle timing set to about 24º. I've also tried the factory spark table, which is around 15º, and the problem is still there... no better and no worse.

I'm beginning to wonder if this isn't just a characteristic of our cams. I don't really have enough experience with valve timing to make a legitimate comment, though.

 

blue86,

One thing I did recently was to reset the IAC Steps at Idle. When I checked, it was at 25-30, so I increased the throttle plate opening to drop the IAC count. When I started to do this the idle surge started then it would subside. Then I would lower it some more, and the surge came back. So causing the IAC to move or change was inducing the surgeuntil the IAC could settle it again. Very similar to what you were describing. As I said, the cam didn't do this when I was running a leaner mixture. It idled just fine. This started when I began increasing the FP for WOT. Too short a PW at idle may have something to do with it.