Zeroing in. Any helpful ideas?

[Sam Powell]

I took the Dart out today, and made great strides in the tuning. It feels pretty strong now once it is warmed up. Got to data logging, and found out some interesting things. It is far from perfect, but it does not fall on its face any longer when it goes into boost. I feel like I can even out the rough spots in both the timing and VE tables.
I have a few questions that someone here may be able to answer. The drivability is very poor when it is cold, and that problem continues up until it is really warmed up. About one block from the house, it wants to die, and is so far out of calibration that I cannot tell if it is dying rich or dying lean. I get back-firing through the exhaust, so maybe it is rich, and trying to put out the flame. After a minute or two of struggling to keep it running, it comes around and settles down, but after a few heart attacks.
Anybody out there got ideas for dealing with this?

1. Is there a strategy for getting this right?
2. Can I use data logging for this too?
3. What variables should I be logging?
4. I do not have the MST hooked up. Does anybody here know what difference this makes? How does it come into the picture?
5. Even with Mega Squirt, everything went lean, and still does with Accel after I drove the car for an hour or so and got things really warm. This means if I do my tuning after it is REALLY warm, it will be rich for the first 30 minutes the next day. If I do the tuning when it is half hour into driving, then it will get lean after another half hour.

What changes after an hour of driving with these slants? It does not appear that the coolant gets any warmer. Does the combustion chamber or exhaust get really hot and change things? Should I be tuning on the really warmed up condition, and then adjust the warmup VE modifier somehow? Do you think the solid lifters are changing the valve lift a slight amount? Would this effect the VE enough to be seen in the resulting mixture? If so, I will have to come up with an average fuel map that is just a little lean at full warm, and a little rich for the first half hour.
SamAgain, I appreciate any advice here. I am making progress.

_________________
A REAL professional is someone who knows when to hit it with a bigger hammer.

[Pierre]

Does that computer use a Manifold Air Temp sensor? I have trouble with warm idle speed, and I *think* it may be related to the MAT sensor reaching 120+ deg when warmed.

Whats an MST?

Cold start and idle are one of the toughest things to get straight. You will likely need to run rich when cold but not so much you get exhaust pops. You will have to go through quite a few (never ending? teehee) true cold starts to get it close. iac motor/solenoid opening, cold cranking, cold idle, and cold acceleration enrichments all have to be nearly perfect to get modern car cold driveability.

[Sam Powell]

MST means Manifold Surface Temperature. It is just hanging there, and I need to fab a bung that bolts to the manifold somewhere. They do not give you this piece of hardware, nor a picture of what it should look like. I don't know if air is supposed to flow past it, or just sit in contact with the metal of the manifold. The MST is turned off in the config file. Once it is mounted, I will turn it back on. I am pretty sure they mean intake manifold. I will look that up again.

So you think the exhaust popping is too rich?
Sam

_________________
A REAL professional is someone who knows when to hit it with a bigger hammer.

[GunPilot]

Sam, you might want to bite the bullet and install a wideband O2 sensor. I did that a couple months ago on my MS EFI'd Ramcharger, and the tuning went from mysterious to simple. No more guessing.

The Innovate LC-1 costs about $200, and comes with software. Hook it up to your computer and read your air/fuel ratio in real time.

Here's some stuff I wrote about it over on RCC: CLICK HERE

Just got back from Moab, wheeling that truck and it was flawless the whole time.

_________________
It's a Slant thing. Even I don't understand.

1974 Duster, EFI /6 soon to be turbo...

Get that Monkay! Get that nasty thing!!

[Pierre]

You should be able to tell that easily enough with your narrow band Sam - whats it showing? I'm not 100% certain what popping is but too rich with gas combusting in exhaust seems legit. Reading the plugs should show you this too. I know if you put your hand at end of pipe and the stream isn't steady but puffy, thats a lean misfire or valve issue. Don't know about popping.

And I defiantly agree about the wideband LC-1. That will tell you how much rich or how much lean making tuning easier, whereas narrow band will just say rich or lean.

The LC-1 setup is pretty trick too, fully independent programable dual channel - so you can set it for 0-5v on output #1 to mate with your computer whereas output #2 can be set to 0-1v to mimic narrow band and make it compatible with narrow band gauges.

[Sam Powell]

I do have a wide band gauge, which works most of the time, and maybe it works all the time, and truly weird things are happening. The wide band has datalogging capability, but I don't know how you record this in conjunction with the other sensor feedbacks. Do these stand-alone wide band gauges have the ability to read right along with the datalogging of the other sensors? Do you open two windows, and tile them, or can they somehow make a graph of the selected ECM inputs that includes the wideband O2 sensors feedback?

The thing is, when things get hairy in the cold start situation, the wide band gauge really seems to go insane. It swings quite lean, and quite high as if something is trying to correct, and it overshoots.

Accel lets you chose from a wide array of ECM inputs and outputs to data log, but you can only do five at a time. So far I think I have not hit on the best choices, because I will look at the graph and say to myself, "now why did it do that?". For instance, on the cold start data log I thought I chose Engine Coolant Temp, or EGT, but when I got back the graph it had two Map sensor values, one is KPA and one is PSI. So that was kind of a bust. I haven't yet included the VE table value, which admittedly I can determine by looking at a printed copy of the table and cross reference the RPM and KPA value, but the less processing for clerical stuff, probably the better, since I am trying to grasp a concept here, and the more available brain power, probably the better.

I hate to take up so much space here on the slant forum with EFI talk, but you guys seem to have very useful feedback, and there seems to be a genuine interest here. I also hate to look so dumb, but then if it walks like a duck, etc. I am really open here about my ignorance. No BS coverup job is in the works.

Once the basics are done, there is some really sophisticated stuff Accel will do which I don't understand. I will get to that and share it in time. For instance, there is some kind of correction factor for each individual cylinder to correct for lean and rich. You have to program in the correction coefficient in the system setup. I will get that figured out in time. I have been told for years here on the forum that cylinders number 1 and 6 burn lean, so this will be something that can correct for that. I am a long way from that at this point though. Everyone says get your basic VE table set up well first before playing with variables and correction factors. Thanks again for any thoughts and advice.

In thinking back over this engine's entire history, it actually exhibited the same characteristics when it was carbureted. I am pretty sure it got to running leaner after an hour of driving. So what would cause this? If I can answer that question, then maybe I can figure out how to program a correction for it. I will watch the engine coolant temperature more closely in the future.

Sam

_________________
A REAL professional is someone who knows when to hit it with a bigger hammer.

[Charrlie_S]

[dakight]

I believe that cylinders #1 and #6 run lean because the runners are longer and some of the fuel condenses and settles out between the carburetor and the valve. This should not be an issue with EFI if you're injecting at the valve port.

If the car is running lean after an hour of running I'm wondering if it's related to the underhood temperature. Do you have a sensor for the temp of the incoming air charge? Do you have an open element air filter or some kind of cold air induction system? The air under the hood can get very toasty and it will take longer to reach its ultimate operating temp than the engine itself. Modern EFI engines all have sensors for intake air temp and I'm wondering if that isn't what you're calling the MST. On my Dakota it was located in either the #2 or #4 intake runner and I moved it to the side of the air hat to isolate it better from the engine heat. I picked up some low end throttle response when I did that. Either way it was in contact with the intake air stream. Seems to me that you will have trouble getting it properly modulated until you have some way to measure and compensate for intake air temperature.

_________________
David Kight
'62 Valiant Signet, White
'62 Valiant Signet, Red
'98 Dodge Dakota
'05 Jeep Liberty

Growing older is unavoidable but growing up is strictly optional.

[Dart270]

Intake air temp (Manifold absolute temp or MAT) is essential and may be why you are having lean/rich troubles on warmup. FWIW, I have that running on my MSEFI setup and I don't like the correction factor very much (the dependence of the fuel metered vs. MAT reading) and need to change it by modifying a little code.

I assume you are using warmup enrichments with the DFI? We had those setup on MS before so you could look back for those values.

DFI should also allow MAT correction factor adjustment somehow.

Best of luck and keep on it!

Lou

_________________
Balancing out the driver's weight for more than 50 yrs

[Sam Powell]

The MAT sensor is in the intake stream right before the throttle body. It is after the turbo and the intercooler. The coolant temp correction values are almost analog in nature. It is represented by a red line that horizontally extends across the screen with temp values on the horizontal, and correction coefficient on the vertical axis. You can click and drag, or step by step incrementally any part of the line up and down. The range is from -40 to 225. A vertical blue line represents the current temp, and all major sensor values are displayed digitally in the lower left hand corner of the screen. The MAT reading is not one of those however.

You have to go to a separate engine analysis screen for the many detailed feedbacks and outputs that are happening. If this lean condition, after extended running, correlates to coolant temp I should be able to get it evened out on the coolant temp correction scale. If it is related to something else, like manifold temperature, then maybe it can still be figured it out, but I don't know enough yet to do it.

Thanks for the kind words Charlie. I feel like I don't even know enough to ask intelligent questions. I guess if you ask the question correctly, then maybe the answer is encoded in the question. So guys, help me ask the right question here.

Sam

_________________
A REAL professional is someone who knows when to hit it with a bigger hammer.

[LUCKY13]

Sam


I believe the MST will help this problem out alot. It is intended to just read the Intake surface when you do mount it & not be touching the internal air of the intake( just the intake metal itself). My guess is that after so much heat gets built up in the engine & compartment that the intake is getting hot and affecting how the fuel vaporizes.


Injector Coeffiecant settings is for trimming the fuel for each cylinder separatly. This should NOT be ajusted until you are completely tuned. Then through CARFULL Spark Plug Readings you can determain the state of each cylinder and if any are lean or rich this can be used to balance them with each other. At this piont some slight retuning may be needed after they are balanced.


If I undestand right you have about a 30 to 45 minute window that things are steady ( after warmup and before the change(going lean) when it has been driven for a hour) I would tune in this window and then use the MST table to correct the change that accures when this piont is reached. That is if the coolant temp is not changing when it reaches this piont of going lean after a hours driving. It may need triming from both maps. (coolant correction & MST maps)


The cold start tuning will be the hardest to achive. The fuel & timing will be needed to achive this. If it was so extremely rich to cause this stubling backfiring problem I would assume you would have some black smoke ( but with the turbo this may not be so). So I am going to guess that lean is the problem you are having. But to be getting backfire out of the exhaust sounds like a rich condision. So this is a bit confussing. SO maybe it is getting to much timing plus being lean ( you will just have to expeirament) because it could be the other way around. The start up fuel maps & the Ignition Startup Term map is where you will be able to correct this. You will have to make the changes & then data log the warming up stage to see the results and it should help you get it right. It will need to be rich I would assume for cold running and tapper off as it warms.


Jess

[Sam Powell]

Thanks all. Jesse, is the start up term map one of those I will need the Sport Key to get into. How does this map work? What does it control. I do not have the sport key yet, but will eventually get it.

_________________
A REAL professional is someone who knows when to hit it with a bigger hammer.

[sandy in BC]

It occured to me today while packing drywall (building new house) that your inlet air temperature is going to be somewhat unique. The combination of turbo and intercooler is going to produce an inlet air temp cycle very different from a naturally aspirated setup. Im used to 747 ECUs that dont use IAT.....and I chip accordingly.....but that is naturally aspirated and a wide open engine department with open element aircleaner. Another application I have uses IAT and has a temperature controlled air cleaner....I chip accordingly.

I think there is some potential that your inlet air on warmup is very cold....(intercooler ....no heat....cold engine ....no boost). After an hour of driving the intercooler is warmer...boost is on ....engine coolant is warmer. In short the engine has much more heat imbeded all around. The warmer the intercooler is the more the IAT climbs.

I would make sure I had a way to measure IAT that was reflective of my cars situation.
I would datalog IAT at all cycles.

_________________
Yeah....Im the one who destroyed this rare, vintage automobile.....


http://www.arrowtipi.com/My%20Webs/index%20Valiant.htm

[Sam Powell]

Good idea Sandy. I was wondering about putting together a digital, plug in tuning set of gauges tht showed more stuff than the Calmap displays or can be datalogued at once. It would just be a set of inexpensive volt meters with appropriate labels on them. You dont; need EGT or IAT all the time, but for tuning it would be nice.
Sam

_________________
A REAL professional is someone who knows when to hit it with a bigger hammer.

[LUCKY13]

Sam, Ignition Start Up Term basicaly adds or subtracs timing from the base timing when in the startup mode. The DFI determens when to applie this corrected timing from the coolant temp.


The map for this is in Igintion - Startup - Ignition Startup Term, it is one of the maps that looks like it is almost anilog.


The MAT as Sandy meantioned will come into play on alot of the tuning, but I dont think it effects anything until it is out of the startup mode. It will be good for triming the fuel on a 1/4 mile pass because as you learn what to data log you will see when you get to making 1/4 passes that as you get to the end of a 1/4 pass the air will get hotter from the compression of the turbo. The amount of heat that it gains will be effected by how good your intercooler works and how effeicant your turbo is.

The more datalogging you do the tunning will become easier to understand in what it needs ( maybe not at first but with practice).


Jess