Intermittent Rough Running Diagnosis

[tony-baroni]

I have a 350 chevy small block, TBI motor. In a G-30 van, 1995 model. The engine is a crate GM truck motor, with summit AL heads, roller cam (stock truck cam), rebuilt TBI, upgraded fuel pump, new distributor and all the tune up stuf when it was built. Maybe 20k on the rebuild. It is running a programmed chip from harris. Lots of new stuff on this one, it's been running fairly well for a while, but I recently started having trouble.


The trouble started in Southern California, after filling up at Costco, I had a lot of stumbles on the highway. Took it easy, switched to Chevron gas. It got better, kept getting smoother after multiple tanks of new gas, as we drove it back to Northern California.

Then another road trip, to Oregon border, broke down with a power steering pump, took some time to get that fixed, and drove back home. It had the minor stumbles throughout the trip, but nothing to write home about.
Then it sat parked for 3 weeks or so.

The next drive, was a 75 mile drive, and it ran like crap the first hour, really shuddering stumbles, and it had no power under WOT. It even struggled to raise the RPms and reach highway speeds on uphill sections. Then after about an hour it ran normally. Drove a bit around for a day or two, then drove the 75 miles back home. Intermittent mild stumbles coming back home. It sat another 3 weeks or so.

Yesterday I drove it locally, and it was really bad, shaking the van again with the stumbles. Short trips, but it was worrisome.

Today I got my ALDL cable out, and PC loaded with TunerPro again, and monitored it while it was running, and some short 1 mile drives. It was back to the mild stumbles.

I was able cause some bad shudders by holding down the gas while parked in drive with my foot on the brake. I recorded that in TunerPro.

I did see that the system is adding knock retard during that shuddering time.

Could someone take a look at the file and let me know if they see any possible items to dig into? I am new to this, and am not sure where to look first. And I don't know if the numbers are what they should be anyway! :-0

[sturgillbd]

You didn't get your datalog to attach to your post. Have you checked fuel pressure? Have you inspected your distributor cap and rotor? There are lots of things that can cause the shuddering and loss of power. Post your datalog and a copy of your bin if you have a way to read the chip.

[tony-baroni]

Thanks Sturgillbd

I did a visual inspection of the wires, and cap (without taking it off). Everything looked good. Then inspected fuel injectors with the key on, and then idling, and then just a little gas, it seemed fine.

Then I checked the fuel pressure, and it was too low. Figured a bad regulator, took that out, and found the adjustable set screw that holds the regulator in place was fully backed out. It must have backed itself out from vibration.

I'll have to get on my PC to upload a log and bin....now I'm focused on figuring out what I can put on those threads to prevent that from backing out again, that won't fall apart in the fuel and clog up the injectors, any ideas?



You didn't get your datalog to attach to your post. Have you checked fuel pressure? Have you inspected your distributor cap and rotor? There are lots of things that can cause the shuddering and loss of power. Post your datalog and a copy of your bin if you have a way to read the chip.

[sturgillbd]

If its a stock style regulator, the threads shouldn't be exposed to fuel. You could use some light duty loctite on the threads of the adjustment screw.

[tony-baroni]

Thanks for pointing that out, I have since learned more about this unit. It's been sent back to the guy who sold it to me, he adjusted it and confirmed the adjustable fuel pressure regulator was good.
I have set the FP to 15psi, and put the stock memcal back into it. That hasn't helped at all, I have to check timing next.
What do you think about the state of the distributor cap and rotor? This is with about 10k miles on it.
14718
14719

[tony-baroni]

Okay here's another update. It's idling high, around 1900. The TPS voltage with throttle closed is 0.6V. The TBI is whistling very loudly, a piercing whistle, when the throttle is closed. I was able to make a log file, starting it from nearly cold, getting it to closed loop mode was possible while idling because of the high idle. Then drove it, but tried driving it without giving it any gas really. Then turned around and drove it a bit with around 15% throttle. The driving starts around 6m mark on the log.

I hope the file upload worked properly. The bin is a custom one, based on BNKM, and I am using a definition file for $31.

How can I get the idle down to where it should be?
Is it normal for the knock retard to stay so consistently around 6 deg?


Other items, I checked manifold vacuum and it's in the green around 21 in Mg on my gage with this high idle. When I put it in gear, the idle drops, and then it shows in the red zone, indicating late ignition timing, only around 15 in Hg Vac

[tony-baroni]

One more update. I went out again and made a 2nd log, with more throttle, and got onto the freeway briefly. it runs like crap. Tons of knock.

I noticed a few things when I looked at the parameter list, instead of the dashboard.
Knock counts are off the charts, getting up to 50k and reseting every few seconds.

When I come to a stop, the spark advance shows 23 thousand and change, that doesn't seem right.

The check engine light came on at a stop light while the vehicle was stopped. I haven't yet figured out how to see which code was generated yet.

I'm all ears to feedback and what to try next!

Thanks all!

[tony-baroni]

Found the error codes.
35 - IAC Failure
54 - Fuel Pump Relay Failure

[sturgillbd]

Just replayed your log file. Your bin file appears to be $E6 instead of $0D. That is why you are seeing the constant knock retard. As for your high Idle, you either have the throttle stop screw open too far or you have a vacuum leak. You are getting an IAC failure because the IAC is closing completely and the pcm cannot control the idle. There are some odd things going on with the spark advance. The beginning of the log looks normal. As the log progresses, I see the timing drop from ~20 degrees down to ~5 degrees with it still idling in park. At another point, I see the spark advance pegged at 50 degrees. The PCM uses timing to help control idle speed but this seems extreme. Adjust the minimum air adjustment screw to where you have approximately 10 IAC counts at hot idle in park. You will have to power the key off and wait a few seconds for the pcm to power down so the pcm can relearn the closed throttle position. You will know when the PCM is powered down if you leave the data dash up when you key off. Tuner pro will start flashing DA Not connected at the bottom of the window. It will reconnect as soon as you key on again. It would be nice if you could read the chip and post a copy of your bin. The relay failure is a signal back into the PCM from the fuel pump relay on pin B12 of the red PCM connector. Make sure you have battery voltage on this pin with the engine running. Clear your PCM codes before adjusting the minimum air screw on the throttle body if you don't have a vacuum leak.

Thanks,
Brian

[tony-baroni]

Thanks for checking it!

I have been looking for a vacuum leak, but haven't found any. All the hoses are connected from what I can tell.

I've uploaded my bin. I thought it was a $31. The vehicle is a 1995 G30, with a 5.7L and 4L80E. Engine isn't stock...

I will look into IAC...

[sturgillbd]

I knew it wasnt $0D . Thats one reason I asked. Ill look through and see if I can find anything in the datalog that stands out. To log knock retard on some of the masks, you have to make a bin change so it shows up in the datastream. I'll see what it takes on yours in a bit.

[tony-baroni]

Thanks Brian!

I'm not sure where to go next.
Here's what I did:
1. cleared codes
2. Started engine, pulled vacuum lines one by one of the TBI to see what effect it would have. Some of them changed the idle slightly, and then I put my finger over the port on the TBI, and it would act just like as if the hose was in place. Other hoses, did not change the throttle when I unplugged them, and left them open.
The big difference came when I unplugged the vacuum hose that comes from the rear of the TBI, and plugs into a sensor (MAP?) beside the TBI. When I unplug that, the idle drops considerably, from the sound of it, near a normal level.
3. I didn't understand your IAC advice. The screw I think you're talking about is the one that sets the position of the throttle blades when the throttle is closed, right? On my engine, it is already backed out all the way. I backed it out about a 1/4 turn before the screw pulls away off the throttle bracket, because the blades are already touching the sides of the bores in the TBI.

How do I set the IAC now?

I wonder if there might be a break in my base gasket under the TBI....I've traced all the vacuum lines I can find twice now, and they all appear plugged in where they should be.

[sturgillbd]

Check your base gasket on the throttle body. If no vacuum leaks, Adjust the minimum air screw to close the throttle blades for approximately 10 IAC counts at hot idle. Clear PCM codes. You have constant knock retard. You could possibly have a bad knock sensor or bad filter board on the memcal. Do you hear any spark knock/ping? You can simulate a knock sensor on this pcm with a 3.9k resistor. It thinks a knock sensor is there but gets no knock feedback. You have extremely low timing everywhere in the datalog and that is why it is lacking in power. I also see the timing all over the place with a constant throttle/load. Something is making the PCM pull timing. As always, make sure all of the engine and PCM grounds are in good shape. Make sure plug wires aren't inducing noise back into the PCM through the engine harness. Get your fault codes cleared up and not returning then datalog again posting results here.

[sturgillbd]

If base gasket is good, you could have an intake gasket leak. Make sure the pcv valve is free and if it has vacuum power brakes, make sure you don't have a leak there. It appears you have a vacuum leak somewhere if you have backed the minimum air screw all the way out with no change.

[mecanicus]

this helped when I had to set my IAC :

[tony-baroni]

Terrific guide, thank you mecanius!

[tony-baroni]

Finding the root cause continues to elude me...let me share what I've done in a couple checks this evening and last.

Replaced the TBI base gasket, torqued to specs 12 ft lbs.
Verified the hose to the MAP holds vacuum
Verified the hose and vacuum canister hold pressure (though they slowly leak, not enough to make this idle so high)
Verified the PCV valve and hose don't have any cracks, and is in place
Verified the hose from the TBI to the EGR control unit is in place.
Verified the hose from the EGR to the EGR control unit holds vacuum

Found that the charcoal canister hose, that runs from a port on the front of the TBI, does NOT hold vacuum. I do not know if it is supposed to!

Then I tried setting the IAC.
Warmed engine up and in closed loop mode.
Turned key off, jumped the 1 & 2 terminals on the ALDL port, put key to ON for ten seconds. I have read the IAC should be buzzing at this point, but there was no noise. Unplugged IAC. Key off, jumper out. Plugged in PC, and started vehicle, IAC showed zero, and idle was still 1900. Is it possible my IAC is not working and holding it wide open? Any other checks I should try?

Pressed on the underside of the EGR valve, which dropped the idle while I held the valve open. So that doesn't appear to be a leak...

I also tried using Wd40 spray on the ports and intake gaskets but it didn't seem to have any effect. I then tried spraying it directly into the TBI, and it didn't have any effect there either. The fan is so strong at 2000 rpms that it's just dispersing the wd40 too quickly.

[stew86MCSS396]

That's because that procedure to adjust the minimum air doesn't work for PCMs. Middle of post #9 mentions how to adjust it.

[tony-baroni]

Good morning all. Here's a recent log file, after I swapped the memcal. When running with this one, I'm not seeing the massive number of knocks. I didn't drive it, just was idling.

IAC shows zero at idle, which I think means it's fully closed based on the fact the idle speed is so high.
BLM and INT both are below 128, which I believe means the engine is running rich and the PCM is trying to pull fuel out. BLM goes as low as 96, while INT is rising back up towards 128.

This is a bit counter intuitive. I've been assuming that the idle is high due to a vacuum leak, which would indicate a lean condition, and the system would be adding fuel. The PCM should know it's supposed to be at idle since the TPS shows a zero or near zero throttle position. What do you all think about that?

Re: IAC
I don't follow the steps in post 9 to set IAC. It says to adjust the idle screw until you see a certain count, but that's already backed off all the way. So opening the throttle blades further would increase the engine speed further, and keep the IAC at zero, right?
It also talks about turning off the key for 10s, which I do every time I stop troubleshooting. So according to my interpretation, my IAC should be set. I'm just uncertain....and trying to confirm everything.


I'm going to have to take a break as I'm headed out of town tomorrow AM. I might be able to read the stock chip and post it here tonight as I ordered a part from Moates to allow me to do that.

[tony-baroni]

Tonight I did my best to confirm no vacuum leak. Sprayed starting fluid everywhere...no indication of any leak.

Did the battery off reset of IAC.

I confirmed it appears my vehicle doesn't have a solonoid valve on the EVAP canister, so that acts lke a constant vacuum leak....but I confirmed with a short plugged off vacuum line that it had no effect anyways on the idle.

I don't know what to check next, so unless the group has an idea it's going to have to go to my mechanic..what a bummer! I hope he can use a scan tool to confrm the IAC is responding. Perhaps it's stuck wide open and that's causing the problem...

[stew86MCSS396]

Watching one of your logs^^^10-13-19 your IAC almost never ventures from 0. Depending on what's programmed into the PCM typically 145, Key On Engine Off, then 145 is what you should see and you should be able to see this in TPRT. IDK...going back and reading your 60 posts...I'd be suspect of the SPR bored out TBI you have. Was that ever changed to perhaps a stock unit? It appears that this engine should run on a stock tune based on the information that you provided.:confused:

[tony-baroni]

I sent in my TBI to SPR as a core, and he transferred the throttle brackets over, so I haven't tried a stock one on this engine. I do not have a spare. I agree that I didn't need to perform that upgrade....perhaps I will try a stock burn chip. I was hoping to get it back to running as well as possible with the current chip and setup, before changing things further.

[tony-baroni]

I burned a chip with the stock bin, and validated it with Moates burn2. Chip is from Moates.

Installed it, and the thing started up. Set the base timing back to zero.
Went through the IAC reset procedure.

The service engine light is now flashing on/off just over once per second. TP5 shows an error at the same rate in the bottom status bar. It also doesn't show many of the sensor values, and shows every error code has been set.

I removed the chip and adaptor, reset it to make sure it was seated properly, tried again, same result.

I will have to do some more searching to see what might be causing this one...

[stew86MCSS396]

Last time I've seen anything like this, cable used for emulating was dirty. Blew it out with compressed air problem solved, but I think maybe an offset issue. With the wrong offset you can verify the chip all day long and it'll report back good but won't work in the 'puter. http://www.gearhead-efi.com/Fuel-Injection/showthread.php?1477-Reading-Stock-GM-Chips-Burn-1-Burn-2-AutoProm-TunerPro-Flash-and-Burn
Also you're probably running off limp mode hence no sensor readings in TPRT. For some reason with PCMs, after resolving whatever problem it is, I could never get the CEL to go away without having to disconnect the battery.

[tony-baroni]

Good morning! Thanks Stew for the tip, this community has a wealth of knowledge. I'd read that page at some point in the past, but didn't remember it when I was burning a chip. The offsets were correct though, since my size bin was the default Moates used in his SW.

This led me on a full search for Moates and chip burning posts. Using the moates guide, and realized I needed to use erase chip and verify blank steps before programming. Since the chip had been verified, this was a bit surprising, but that appears to be the only thing that changed which allowed it to work. Or there is something else happening that I haven't found yet.

I now have the stock chip burned and the vehicle running, with the codes cleared. It still has the problematic high idle and shaking. However, the INT and BLM values with the stock chip at idle and part throttle are much closer to 128. Knock counts are lower.

[vilefly]

Has anyone checked if the IAC valve is even moving in the real world? It should seat itself and open back up once upon KOEO conditions. Another problem is mismatched pintle styles that do not seal properly against the throttle body idle passage. They all swap. Try plugging off the inlet port to the IAC, and see if the rpm drops.....use a rubbermaid glove or something, whatever works. If no change, then remove and plug off all non-critical vacuum ports, remove the pcv valve from the valve cover, start engine, and block off the pcv vent on the other valve cover. Now block off the pcv valve grommet in the remaining valve cover with your hand, and see if the crankcase still has vacuum. If it does, then you have an intake gasket problem.

The only other thing is setting the tps to exactly .50v at closed throttle, and have the timing set to stock timing at idle (on the chip, too), but I suspect this has already been done.

[Kilroy]

Well read your whole thread and one thing comes to mind working on these vehicles, IGNITION MODULE. It will screw up your timing, intermittently cut off ignition signal to PCM. You will be chasing your tail blaming everything. The PCM rarely codes for the module. Also the trigger wheel , a hole burnt through ignition rotor causing spark to ground out through distributor shaft. But your files always show timing all over the place, IGNTION MODULE main cause for that. Also don't buy cheapo buy AC DELCO . Brand new I've seen same problem, good for a day or two same problem so you think module not the problem and its the cheapo failing. Hope this helps.

[tony-baroni]

I have an update, I've been hoping to review a scan before posting but my cable stopped working so I have another diagnosis to do!

I brought it to my mechanic, they confirmed the IAC was bad. SPR performance service has been great during this process, and they sent me a replacement valve, as well as tuning up the TBI.

The mechanic installed the IAC and confirmed that fixed it.

When I picked it up, it still had a miss, and as it warmed up and I got onto the freeway it got a lot worse. So much for paying them to help with diagnosis!

I discovered a youtuber, ScannerDanner, who teaches auto diagnosis. I watched a bunch of his videos, and worked though his basic evaluations for a miss. I diagnosed a bad spark plug wire on cylinder #5, which was also the spark plug that looked the worst. Now I know that's another sign I should pay closer attention to.

Fixing that had a really big impact on the performance. That was last Sunday, and I drove it 60 miles that day. I still felt a little something going on, but it's driveable now. I also am using the stock chip program, the shifting is much smoother. I'm curious to see how my mileage and my acceleration performance will compare.

Once I fix my cable I'll take some more logs and ask the group here what they think about the timing.

I will also look around to see what I can find for a test procedure for the ignition module. I'm disappointed in the AC delco performance spark plug wires, they don't grab onto the end of the spark plug very well. They have a lifetime guarantee, I plan to visit the dealer today with the one bad wire and see what they do for me.

[vilefly]

Ignition modules either have the 5v square wave HEI reference signal fail, or Fail to fire the coil while still output an HEI ref signal. 1% will have the Bypass circuit mess up and set a code for it.


The test for the pickup coil signal gets interesting, and can have random symptoms like rough idle, randomly dies, stumbles on acceleration, etc. The quickest test is to drive it once in a normal condition, then disconnect the "set timing" connector with the tan/blk wire, and drive it again to see if any symptoms disappear. If the idle quality does not change, either.....then the problem is not the pickup coil or a cracked magnet in the distributor reluctor. If it does change, replace the distributor or just the pickup coil if you are a gambler.