Need help solving surging deceleration and idle

[Chewy1576]

Hey all,

I recently started over with my 383. I used stock AWSY (Originally AKWH) as a base, then updated for 90# injectors and disabled EGR. I have done one tuning run for the VE table and it probably needs a few more to get it right. It idles pretty smoothly at 700 RPM with IAC counts around 5 when warmed up. The issue I am seeing is that when I let off of the gas pedal and coast, I can feel the engine surging as it's decelerating with the transmission in gear. It's soft but noticeably in 4th and stronger in 3rd. I will go to neutral after 3rd until I come to a stop, then the RPM will drop suddenly, almost die, then come back to normal idle. It has only actually died once, but fired right back up.

I have read that a lean condition during deceleration will cause surging. I have tried changing some of the DFCO parameters to keep from pulling too much fuel or introducing more air but they haven't helped. This morning I disabled DFCO entirely but it still hasn't helped. Does anyone have any other ideas of things to look at to eliminate the surging and smooth out deceleration?

[buddrow]

The surge could be caused by too much spark advance as well as a lean condition. I'm guessing you are using a manual transmission according to the bins you noted. If not, then there is part of the problem.

Buddrow

[Chewy1576]

I guess it would help if I attach my bin and xdf huh? Yes, manual transmission. This morning I burned another chip real quick before leaving for work. I removed 3 degrees of timing and bumped the idle speed up from 700 to 800 RPM. I was using the spark table from the chip that came with the crate engine and it has about 6 degrees of timing added across the table. It seemed to run more smoothly, but I can still feel a little surging when decelerating in 3rd. I will put some more drive time on it today and see how it does.

[Roadknee]

Look in your datalog and see if the surge in engine speed coincides with the O2 sensor switching from rich to lean. Some engines with increased overlap just don't handle lean AFR's at low rpm during coasting or light throttle very well. You'll need to adjust the fuel trim parameters, which start with "O2" in your XDF, to reduce the amount of time the O2 sensor output dwells in the lean position.

[Chewy1576]

You mean these guys?

9019

I'm trying to look through some of the other more developed xdf's for comments on what these parameters do exactly. What about the "prop gain" parameters? Would those have an impact in this case?

Here is a log I took when I first created this bin. Only differences between then and now is the decrease in timing and increase in idle RPM I did this morning.

[Roadknee]

Yeah, those are it plus the tables. I've only tuned a few trucks with the 427 PCM and $OD. The parameters are similar to yours, but there are differences. I reached out for help on various message boards without luck. Tuners seem to hold this information pretty close. Maybe you'll have better luck than me.

The surge you're experiencing doesn't look too bad. I've worked with a pretty hot vortec headed 350 that would surge between 1200 and 1500 rpm when the throttle was cracked in Park. Deceleration under 1,800 rpm wasn't much better. You might try reducing the first three cells of the proportional gain table by 33% and blend the next three cells into the upper cells. The end result would look something like this:

9.17
11.78
18.33
25.37
33.99
43.78
56.29
56.29
56.29

You might also reduce the High O2 Limit Slow Filter (scalar) from 851 mv to 800 mv.

These revisions might make it better, worse, or have no affect. Other changes might be needed. I had to experiment with a lot of trial and error to deal with the surge in the 350.

[Chewy1576]

I will try those changes in the morning. I was also reading this article (http://www.dynamicefi.com/Tune_Intro2.php) about deceleration and it indicated the injector offset may be too low, especially for engines with different injectors or increased fuel pressure. My application has both, as I am running 80# BBC injectors at 18 psi. I modified the BPW for the larger injectors and injector bias according to the values from Dave W's TBI Calculation spreadsheet, but maybe the bias is too low? I also noticed that the battery voltage in the datalog cycles between 13.3 - 13.8 volts indicating the alternator could be on it's last leg. Increasing the bias means I will need to do more work with the VE table, but that needs to be done anyway. I'm also looking at other injector related parameters in a 7.4L manual transmission bin for comparison to see if I may have left anything out. I just have tomorrow to tinker before I leave town for the weekend, so I may not get to it for a few days.

[91dime]

Do you have a wideband o2 hooked up?

Not trying to hijack the thread but im also having similar issues but i have an automatic. I have the 350 injectors running at 20psi on a 5.7L so its a big change from stock and never had the issue before the fuel system overhaul, from what i can remember at least. Now the truck drives and idles fine but on deceleration it acts like DCFO is turned on although iv tried turning it off for testing. At any speed when i let off the throttle, i can feel the fuel shutting off and the wideband confirms that. It starts kicking to max lean, around 22 afr, then back to normal then back to max over and over until i apply throttle then its smooth until i let off again. From what i can see it happens no matter how fast im driving. Even shut off once while slowing down for a light.

[Chewy1576]

Unfortunately, no wideband O2, yet. I was looking at the Innovative Motorsports LC-2 to replace the stock O2 sensor and use the narrowband simulation output to feed the ECM and put in a gauge also, but that will be down the road.

The suggested O2 parameter changes and a few BPW/injector related tweaks have helped with the surging, but it's not completely gone. There are times where it will decelerate while in gear and transition to idle smoothly, and others where it will surge and jerk when getting to idle. My next change will be increasing the injector offset a little bit more, but that will have to wait until next week. I have attached my last data log from today. This one was for VE corrections.

[1BadAction]

Some of these newer OBD1 ECMs are just too sensitive for their own good, they severely jerk the IAC back and forth when they attempt to control the idle, and that is an excellent way to induce a surge in an engine thats not a soft piss weak compression smog motor.

After going around with this 100 times, what I ended up doing to control mine, was first off, ignore what most people say about setting the minimum air. It is a good start but usually ends up being too low to use as a final setting. Get the engine up to temperature and park the vehicle with it idling. rev the throttle and note how much it drops down below requested idle. It will most likely then recover too hard and start surging.

Shut the truck off, and turn the idle screw in 1/4 turn. Repeat what you did before and if it only drops about 50rpm before it recovers, you should be good, if its more, turn the screw in another 1/4 turn and try it. It may surge slightly, but that will be learned out after a little bit of driving. When this is close, you can now take it for a drive and start tweaking the settings. You might end up finding that a lot of settings don't even need to be changed. On mine, I did cut the gains back about 25% and turned the overspeed spark corrections OFF. If you are running a BBC throttle body, you'll also want to match the big block IAC position vs airflow percentages, if your ECM uses that particular table.

[91dime]

Unfortunately, no wideband O2, yet. I was looking at the Innovative Motorsports LC-2 to replace the stock O2 sensor and use the narrowband simulation output to feed the ECM and put in a gauge also, but that will be down the road.

I personally use the innovate mtx-l and i enjoy it. Its been a long time, but iv read before how some guys find it hard to tune as its not the same as the real thing. My setup is NB on pass side and WB on the driver side. I have used it for about a year and really enjoy it. You can actually use an output wire from the WB and connect it to an egr wire on the ecu and have WB datalogs along with the NB. I tried it the other day and its amazing the amount of data i get now. I just wish i knew what every parameter meant and what each one does and affects what. I also have the 7427

Originally Posted by HaulnA$$
.............As for data logging the WB O2, I am using the "Linear EGR Pintle Position" input at pin B16 of the PCM for the 0-5V analog 2
output of the LM-1. Obviously I am not running an EGR and I have code 32 disabled as well as all EGR enable terms. Bits
1, "LINEAR EGR/ 0 = EVRV EGR" and 5, "USE L4780 TBL FOR %EGR (Back pressure EGR)" at L400D (AFR Mode Wd 3) must be set, bit 2 does not matter and bit 3 must not be set or the input will not be read. I left bit 4 on but it did not seem to matter.
The tables at L4700, "DESIRED EGR Vs. RPM & LOAD, (VAC or MAP)" and L4780, "FUEL REDUCTION PCT WITH EGR ON Vs. RPM & VAC", must also zero'ed or the WB O2 input will affect the fueling. I set up the LM-1 and .ads file in TunerPro Rt so the AFR will be displayed properly from 10 to 20 AFR. See attached images.
It works great and the EGR routines do not affect fueling...........

Edit: to Chewy1576 (http://www.gearhead-efi.com/Fuel-Injection/member.php?6596-Chewy1576), are you running a NB o2 at the moment? I unplugged mine and took it for a drive and it drove like a dream and DCFO in fact is working properly, tested by coasting down a big hill in gear. After a few seconds the WB showed fuel shutting off for the duration of the hill and fuel returned to normal once leveling out. Did not surge or drop RPM at any point while off throttle like it does with the NB plugged in. In other words it drove like it was meant to drive.

If you unplug yours and experience similar results would that mean the o2 tables need more work for both of us? or could there be something else?

[Chewy1576]

I have adjusted the minimum air several times, and it wasn't until I got the IAC settings worked out to solve a high idle that I actually got the IAC counts to drop when turning the minimum air screw in. They were at about 5 with 700 RPM idle. I haven't touched it since.

Yes I am using a NB O2. I put in a new one in March. I can try driving around for a bit with it unplugged and see if it improves.

[91dime]

I have adjusted the minimum air several times, and it wasn't until I got the IAC settings worked out to solve a high idle that I actually got the IAC counts to drop when turning the minimum air screw in. They were at about 5 with 700 RPM idle. I haven't touched it since.

Yes I am using a NB O2. I put in a new one in March. I can try driving around for a bit with it unplugged and see if it improves.

Still driving around with mine unplugged and its been doing great all around, no noticeable RPM drops or close stalls when slowing down. Im seriously thinking of ditching the NB o2 since it seems to be more of a pain then anything else.

[Roadknee]

Still driving around with mine unplugged and its been doing great all around, no noticeable RPM drops or close stalls when slowing down. Im seriously thinking of ditching the NB o2 since it seems to be more of a pain then anything else.

The problem you're describing is not the norm for the 7427 ECM and TBI on a stock Vortec long block. The surging is caused by the O2 sensor swings on deceleration. I wonder if you have a very small exhaust leak (bad gasket, cracked weld in header or exhaust system, etc.) where fresh air is being pulled into the exhaust on deceleration only. The O2 sensor would see this as lean and the ECM would overcompensate until it swings rich, then overcompensate back to lean an on and on...

[Roadknee]

Chewy - you can also try reducing the Low O2 Limit Slow Filter. I think it's about 150 in your BIN. Try cutting it in half. This will increase the rate at which fueling changes are made to swing the O2 from lean to rich.

[91dime]

The problem you're describing is not the norm for the 7427 ECM and TBI on a stock Vortec long block. The surging is caused by the O2 sensor swings on deceleration. I wonder if you have a very small exhaust leak (bad gasket, cracked weld in header or exhaust system, etc.) where fresh air is being pulled into the exhaust on deceleration only. The O2 sensor would see this as lean and the ECM would overcompensate until it swings rich, then overcompensate back to lean an on and on...

You sir are good, I looked and sure enough a nut has worked its way off the collector on the side that the NB is on.

[Chewy1576]

On my way home from Yellowstone as I type. I will make that change and get a new data log asap. Thanks for the tip!

[91dime]

edit: double post, sorry

[Chewy1576]

I burned a new chip this morning and tested on the drive to work. Initial feel says that it is much improved, but my drive is barely long enough to get up to operating temp haha! I will do some more driving later to confirm.

[Roadknee]

chewy- what cylinder heads are on your 383?

[Chewy1576]

They are modified TBI heads. Here is a link to the crate engine specs - http://www.blueprintengines.com/index.php/products/bp-gm-crate-engines-landing/gm-383-truck-suv-main/item/gm-383-base

[Chewy1576]

I had some more time at lunch and after work to really test the decel. I could run it up to 3000 rpm in 3rd and it would decelerate smoothly all the way to idle. Even 2nd gear deceleration was smooth. I think we have it taken care of! I'm really happy with how it's running now. :rockon:

[Roadknee]

I'm happy it worked out and a bit surprised it was that easy. I've been fiddling with fuel trim parameters on the vortec headed 350 off and on for months and it's still not perfect. I looked at your engine specs and I'll bet that thing does run really good, particularly compared to a stock worn out 350.

[Chewy1576]

It's a night and day difference compared to the old 350 with 2 dead cylinders. It's even better now that the 80# injectors are dialed in. The engine was built to run with a stock 350 TBI setup, just swap and turn the key, but I had a feeling the old 55# injectors were being overloaded. Bigger injectors and a new EP381 fuel pump are enough to feed the beast for now. I didn't think about touching the O2 parameters because I was running a stock replacement sensor so thank you for pointing them out to me!

I will probably leave the engine alone for the rest of the year as I just found a donor truck to source an NV4500 and 4.10 gears to swap with my SM465 and 3.42's. It's going to be a busy summer! :happy:

[91dime]

Good to hear you got it running! I thought i had fixed my exhaust leak but its still acting up so i may leave the o2 out since it seems to run much better all around.

[Chewy1576]

The idle and deceleration surge is back! :( I changed out the SB throttle body for a BB unit with adjustable fuel pressure regulator. Small adapter harnesses were used to connect the TPS and IAC to the existing plugs. The live readings in the ECM look correct. Installed the Actron FP adapter inline with the supply line and the gauge reads 18 psi. I used a 7.4L Manual bin for IAC parameters and the idle will get down to the commanded 800 RPM, but it surges. i have attached a video of the madness. Any ideas?

https://drive.google.com/open?id=0B-yp1rbEKaEPNVM4T1pyUHZCNVU

EDIT: Jumpered Pins A and B in the plug to do an IAC reset, and noticed there was a Code 51 present. Changed the Mask ID in the bin to "AA" and burned another chip. The idle is much smoother.