Projection tune help needed

vettestr · #1 (**permalink**) 04-08-2009, 08:32 AM

Hey Gang,
I have an older Pro-jection 4DI system. I am half a$$ed decent about tunning this but am a bit lost or keep overlooking the same thing.

The only symptom is what acts like a lean condition stumble only at light cruz. If at say 2500 RPM and just maintaining that speed the cam is happy and engine is smooth, no ignition issues found but a light hesitation starts, can not see RPM loss but can just feel it. The 02 sensor readings are 0.70ish at this point ??? Any throttle increase or decrease stops the very light stumble. I have fattened up the range but no help.

What am I missing here gang?
Your help and ideas are appreciated.
THANKS Jeff C

Gaz64 · #2 (**permalink**) 04-09-2009, 04:03 AM

Jeff,

I played with a 4di a few years ago, was a headache at the start, learnt a lot, LOL.

Here's a helpful tuning aid for 4di:

http://autochart.com/

Works much better than the Holley software.
Heaps easier to adjust your maps and setup parameters.

I might be able to send you a copy if you want to try it.

vettestr · #3 (**permalink**) 04-09-2009, 08:18 AM

Gary,
Thanks for the suggestion. I have and use their software with grate success most of the time. Problem before me is the (verified function) 02 sensor is not indicating a lean condition. I fattened her up anyway without fixing the stumble. All other areas look right but I am missing something and can no longer see the forrest for all the dam trees in my way??
Jeff c

olddog · #4 (**permalink**) 04-09-2009, 05:43 PM

Could you have a bad spot in the throttle position sensor?

vettestr · #5 (**permalink**) 04-09-2009, 07:52 PM

Dog,
I will pull it off and check it. The Program Gary mentioned below allows you to record all functions as you drive. So will look at the TPS output on a run also. Jeff C

Gaz64 · #6 (**permalink**) 04-10-2009, 01:20 AM

Jeff,

Can you look at what settings you have under ECU Programming/Engine/Oxygen Sensor Control Setup?

There are target parameters versus tps (1/100th volt) and "Throttle position to go open loop".
At steady throttle cruise, the oxygen compensation could be pulling fuel out to the point of a light throttle surge.

You can also see this on the AChart data logger.

Naturally don't ask the logger for everything.

I used to look at coolant, air, rpm, tps, O2, O2 compensation, map, accel volume and time, injection volume and time.

vettestr · #7 (**permalink**) 04-10-2009, 08:27 AM

Gary,
I must be missing your point, can you tell me in a different way? It is all blending together, what am I missing here? It is not entering open loop at cruise as verified by logger. It is correcting/removing fuel but to reasonable 02 readings and have increased target points multiple times.

I am entering closed loop at 99 degrees and remain closed until until close to full throttle (maybe 7/8's of full range). So engine is closed loop during or when the surge happens, confirmed by data log. In closed loop it uses the map parameters but adjusts to feedback from the 02 sensor so always delivers desired A/F ratio. On cruise and closed loop it is definitely correcting or pulling a little fuel to reach desired offset but final 02 reading shows a fat mixture. I have increased that offset twice without resolve.

Thanks for taking time to explain your points. Jeff C

Gaz64 · #8 (**permalink**) 04-10-2009, 03:12 PM

Jeff,

I know I'm no expert on the 4di, but between the pair of us, we share some experience, so I'll throw some ideas at you and see what you think.

I'd reduce the Throttle Position to go Open Loop value from 7/8 throttle to zero as a test so it never goes closed loop if you feel the fuel map is correct, or near correct

I'd guess you know the system fairly well since you're using the AChart program.

Can I ask how you have scaled the rpm, map, tps scales?

Can you give some details on your engine combo - capacity, compression, cam, intake manifold, injector lbs/hr etc?

vettestr · #9 (**permalink**) 04-10-2009, 07:12 PM

The fuel map is rich across its scale by a factor of 15 above stoichiometric. What are you looking for by going to open loop?

The engine is an alum. Dart gen 6, Pro One CNC'd 325 heads at 540 inches and 12.4 compression.

Gaz64 · #10 (**permalink**) 04-10-2009, 07:55 PM

Quote:

Originally Posted by vettestr

The fuel map is rich across its scale by a factor of 15 above stoichiometric. What are you looking for by going to open loop?

The engine is an alum. Dart gen 6, Pro One CNC'd 325 heads at 540 inches and 12.4 compression.

Jeff,

I'm thinking the closed loop correction is your surge, so by leaving in open loop just as a test, and only if your fueling is at near stoich at cruise, go for a drive and see if the surge is gone.

This all assumes your A/F ratio is correct in the map.

Maybe your O2 sensor is "lazy".

Your engine is most certainly a bigger combo than I expected, bet it goes hard.

fkemmerer · #11 (**permalink**) 04-10-2009, 09:31 PM

Quote:

Originally Posted by vettestr

Hey Gang,
I have an older Pro-jection 4DI system. I am half a$$ed decent about tunning this but am a bit lost or keep overlooking the same thing.

The only symptom is what acts like a lean condition stumble only at light cruz. If at say 2500 RPM and just maintaining that speed the cam is happy and engine is smooth, no ignition issues found but a light hesitation starts, can not see RPM loss but can just feel it. The 02 sensor readings are 0.70ish at this point ??? Any throttle increase or decrease stops the very light stumble. I have fattened up the range but no help.

What am I missing here gang?
Your help and ideas are appreciated.
THANKS Jeff C

Jeff,

Does your EFI system have a data recording capability? If so, I'd suggest you record the following when you make the problem happen: Target and Actual A/F Ratios, ignition timing, acceleration enrichment (both TPS and MAP-based, injector pulse width, MAP Senor (or load), VE %, and RPM. If you post this data, we can probably figure out what is going on. Is the RPM and load point where you are having the problem between VE table entries? If you are splitting adjacent cells, are there big differences between the cells?

BTW, this is a pretty common problem. I've had basically the exact same experience with 2 EFI controlled engines I've built. It should be easy to tune out.

- Fred

vettestr · #12 (**permalink**) 04-11-2009, 07:06 AM

Gary, Rain is not a common problem here in the Phoenix valley but is now raining cats n dogs. I will post results of changes and run when I can but maybe a couple of days delay. This car on wet pavement or puddles can get scary.

Fred, I have several runs recorded but think file type will not allow them to be opened by most folks. They are DOS based files and either Holley or Achart Pro files. I will email direct or I follow directions well but am otherwise clueless how to display.

fkemmerer · #13 (**permalink**) 04-11-2009, 07:22 AM

Quote:

Originally Posted by vettestr

Gary, Rain is not a common problem here in the Phoenix valley but is now raining cats n dogs. I will post results of changes and run when I can but maybe a couple of days delay. This car on wet pavement or puddles can get scary.

Fred, I have several runs recorded but think file type will not allow them to be opened by most folks. They are DOS based files and either Holley or Achart Pro files. I will email direct or I follow directions well but am otherwise clueless how to display.

Hi Jeff,

Can you print the results of the runs to a .pdf and email or post?

- Fred

vettestr · #14 (**permalink**) 04-11-2009, 07:59 AM

Quote:

Originally Posted by fkemmerer

Hi Jeff,

Can you print the results of the runs to a .pdf and email or post?

- Fred

I can try a few things. The resolution of any 1 page with say even just 4 chosen parameters would be of questionable value. The program displays a numeric value for all intersection points on the graph. That is a snapshot of only that second of maybe a 15 min. test ride??

Because the injection program needs DOS I am running all this on what must be a WWII relic with Windows 95 as the OS. I save all these to a floppy but that computer is the only one I have that still has a floppy.

fkemmerer · #15 (**permalink**) 04-13-2009, 02:50 PM

Quote:

Originally Posted by vettestr

I can try a few things. The resolution of any 1 page with say even just 4 chosen parameters would be of questionable value. The program displays a numeric value for all intersection points on the graph. That is a snapshot of only that second of maybe a 15 min. test ride??

Because the injection program needs DOS I am running all this on what must be a WWII relic with Windows 95 as the OS. I save all these to a floppy but that computer is the only one I have that still has a floppy.

Jeff,

The things to look at would be:

1) What does the AFR reading do at the time of the stumble?
2) Is the system providing any transient fuel (enrichment) related to the change in throttle setting that causes the stumble
3) Does the fuel or timming program make a bit move in the VE%, Timming, and/or target AFR before, during, and after the stumble

- Fred