Now that I seem to have my electrical problems resolved, I am turning my thoughts back to tuning. I have been playing with start up fuel coefficient, which is a multiplier of the base fuel map for start up, and have been getting somewhat inconsistent results. This table has various ECT break points and can be adjusted differently for the different ECT temps. Right now, the engine doesn't always start the same way. At least not as reliably as my Toyota. Some times it seems to rquire considerably more cranking before it fires, and sometimes it fires right off.

There are at least two temp related fueling coefficients that are multiplied times the base VE number in the fuel map. One is for steady state, after-start operation at various ECT temps. The other is a start up fuel value. I am not certain if the ECU multipies both in the start up formula, or uses just the base value in it;s start up calculation. The start up and steady state both have the same break points in the temp ranges. After the engine starts, there is a phase out of the start up value, and then the warm up coeffiecent is active.

So in thinking about this, I decided to watch the fuel map cell pointer as I cranked the engine. It does not always settle in the same cell for cranking. Sometimes it is in the 100KPA/200RPM cell (zero vacuum), and sometimes in the cell right above it which would be a very slight vacuum. The two values in this cell are not now the same. And further more, the values in the cells just to the right of them, which would be 750 RPM, are not the same.

The ECU program reads the actual value of the KPA and the RPM, not just what is in the cell which is highlighted at the moment. It then interpolates, or averages the neighboring cell values into the value it applies in the fueling formula. This depends on how close the actual cranking values are to those of the neighboring cell's KPA/RPM figures. Since all the values in the four cell group that would be affecting the starting fueling are indeed different values, then this might account for the varying results I am getting. It probably is hunting around as the actual KPA/RPM value drifts up and down during cranking. Especially if some oil gets into the cylinder walls and starts to build a little compression. This might account for some of the seeming change in its behavior after a few start up attempts.

So I am thinking I should be adjusting all four start up area fuel map cell values the same as I hunt for the best start up fueling, based on warm start up. THEN, go back and tune the cold start values. Without this approach, there is still randomness to the start-up fueling, since the cold start coefficient simply multiplies the value in the base map to determine the fueling for any given start up temp. It seems as if the engine would never operate in these cells once the engine is running, unless I let it get down to 750 RPM in high gear, and then floored the gas pdeal. I think they would call this user error.

I think I will try to get the steady state warm up coefficient dialed in first and then begin playing with the start up/ECT fuel cofficient
I will will give this tuning a try, and let you know how things go.

OH yeah, I am still waiting for someone with a slant EFI setup to look in their fuel map and see what VE value they have plugged in at start up RPM and KPA. Anyone? The four cells in question on my map run from 201 to 405. That will change soon.

And on to timing.

Dan, when you reported that Mopars research indicated 5 degrees advance was the best start up timing, was this for all temperatures? It seems to be true for warm start, but I was wondering if there is proven data on this, just to eliminate a ittle of the searching on my own. Forgive my asking if your earlier post was clear on this.

Sam

_________________

Is the fuel pressure behind the injectors fully pressurized before you try to start?

My only thought is that it bleeds down and you're not pressurized until after some cranks of the starter.

I had this problem with my toyota truck and rigged the pump to run and pressurized the system before starting (some injection systems do it automatically, and some don't)

_________________
Ed
64 Valiant 225 / 904 / 42:1 manual steering / 9" drum brakes

This system pressurizes the rail before hand, but it does cut off after 5 seconds if it does not start. I always wait for the pressure to reach 40 PSI before cranking. So, the pressure should be consistent in the various attempts to start it..I do have a fuel pressure gauge inside the cabin.

How did you rig your truck? Did you put a manual over-ride on the pump relay feed? I guess since this is a return system, it should not hurt anything to be able to put pressure on the fuel rail when you want to.

Sam

_________________

Sam, it wasn't that Chrysler found 5 degrees was the best for starting, it was that they found advancing the timing significantly beyond factory specs significantly worsened cold starting. Based on that, and the factory specs at the time the experimentation was done, I had recommended between 0 and 5 BTDC as the best setting for cold starting. As for warm or hot starting...I'd stand by the same recommendation. I'd probably go with 3 degrees total cranking advance at all temperatures.

_________________
一期一会
Too many people who were born on third base actually believe they've hit a triple.

An analog timer...............

The fuel pump has a test connector that grounds the fuel pump relay so you can test the pump without the engine on......

I got a large capacitor from an industrial liquidator store (local) for a dollar or two and hooked it up to the test connector. When you turn the ignition on, (but not the start position), the capacitor looks like a ground to the relay until it gets charged up. Lasts about 20 seconds (played around with various size until I got what I wanted). When I get into the truck, I turn the ignition on, then seat belts and mirrors (giving time to pressurize system) then start engine.

_________________
Ed
64 Valiant 225 / 904 / 42:1 manual steering / 9" drum brakes

This evening I went out and played with the tuning again. I did make all four start-up cells the same 201 VE percent, and the engine seems to start up much more reliably. I started it the first time tonight at 40 degrees ECT, and then let it warm up to about 160 ECT before I stopped playing with it. I tweaked the curve all the way down to a coefficent of 1 at 164, which means no modification to the base table. The four cells are 94, and 101 KPA at 200 and 750 RPM. I will play with the value in those cells more once I have the engine fully warmed up, and then go back and tweak the start up coefficient at the different ECT temps after I have the value in those cells ideal for a fully warmed up engine. IN the mean time, I think I am no longer concerned about it not starting, and leaving me stranded.

If it hadn't been raining cats and dogs here in Maryland, I would have taken it out for a cruise and tune session. Thanks again for following along, and offering info and tips.

Sam

_________________

Thanks for the update, Sam.

Also thanks for the timing discussion. My 68 Dart has around 10 deg initial and starts a bit tough in the cold. I need to get that MS2 in there and program the spark map...

Lou

_________________
Home of Slant6-powered fun machines since 1988

YOu are welcome Lou. Thanks to Dan here. Can you take a look sometime at what VE value you have in your 100KPA crank fuel map cell? I kind of think 201 is a bit lean. But am trying to sneak up on the perfect value.

The problem is, it gets inconvenient, and embassing if you kill the engine in traffic (argh!) and it won't fire right back up, so I am eager to get this worked out. This has only happened once, but that is enough incentive. I was waiting in a parking lot entrance to the highway, so this failure to start didn't inconvenience anyone else. But when you go to start it, and it tries and fails, too lean can sound just like too rich, and you feel uncertain about what to do to get the thing going in a pressure situation. That is the kind of thing that can make a woman lose all faith,and want to get rid of an automobile.

With the Accel ECU you can kick the trottle twice rapidly, and it squirts a bit of start up fuel in , just like a carb. But, I am reluctant to do that in a pressure, traffic situation. If it is not starting because it is flooded, extra fuel will make things worse. Too lean is better than too rich. If it doesn't start becasue it is too rich, you can't do anything about it. If it is too lean, you CAN kick the throttle for more fuel.

I have to say, this is the fun part of the process. Fixing the wiring eliminated the wierd, disfunctional stuff that was just confusing, depressing, and discouraging. The warm up miss seems to be gone for now. I am reluctant to say gone for good until I have driven it a bit more.

Sam

_________________

That sounds like a HOT restart is a issue. This is addressed seperately in the configuration menus. Cold running issues (dying after keyturn-Vroom!) is cold/ warm-up area afterstart settings are not tuned enough.

When I get a balky engine that doesn't want to keep going after cold initial fire (turn key-vroom, wait 30 seconds or so, kick throttle (for carb) but for injection things: putt putt putt splutters dies), is I hold the throttle cracked open till engine get hot enough via throttle (above 1,000 rpm but not faster than 1,800 rpm. Till I have time to tune this right. Weber is opening choke too QUICKLY or fast idle rpm too slow (ran out of screw).

Base timing does matter on particular engines than how fueling types are done (carb, TBI or port injection) for ease of startiablity/running ablity and does affect rest of the timing curves and most inportant at idle.

Computerized spark stuff does adjust the rpm at idle as engine responds to this DYNAMIC ignition timing adjustments much easier than trying to adjust the air idle valve which takes lot of change to show difference and responds so slow. YET computer does use both.

Cheers, Wizard

When it died in traffic the ECT was only about 123, so it was not warm enough for base start up tune. To tune the base start up, I will adjust the start up cell's VE number after it reaches at least 160 ECT.. Then I may have to go back and retune the warm up start fuel curve. I expect this to take awhile. You only get to start it a few times a day. Until this is done, I kind of want to have the lap top with me when I drive.

Sam

_________________

123 in F? That's in warm up phase of ramping up warmup process so engine needs more help with fuel and timing. around 32 F to 75 F is critical period for COLD engine where it needs more rpm and extra fuel and this two ramps down gradually to normal in reaction to the coolant temp.

All liquid cooled engines till around 150F or so especially with TBI and carb need much help with cold starts and warmup temps restart so need extra shot of fuel with timing at base setting always. Hot engine usually need no help as engine is very hot. Like I can start up a hot 2.2 without kicking choke on.

Port injection engines get this issues much easier due to less fuel wetting on the walls and more direct into the cylinders so it's less of a problem with warm up but cold starts still need first minute RICH running to keep engine alive at 2,000+rpm for that period to "blow the mixture down" into that shivering engine because the mainfold is too cold. Like the 132F, the manifold is not warmed enough and engine is bit almost getting there but need more RPM to carry this through. That why carb had fast idle cam steps For example: my carb 2.2 have appox 2,600rpm if cold, 2,200rpm afterstart (choke kicker kicked in) and 1,600rpm fast idle once base idle rpm is correctly set. Mine is dying because fast idle is too SLOW.

Coolant temp sensor location is CRITICAL. This is the key log.
The one at lower block is COLDER because of cooler coolant flowing from either radiators (heater and/or radiators) and gives false reading that in truth the engine is actually hotter by large margin. Fooling the computer badly and engine very ill-tempered. This is the REASON so many makes of all models and all kinds of engines had this sensor right near the thermostat behind the bypass hose so coolant is flowing pas the sensor for accurate reading.

A engine that is finally HOT enough is considered around 170F to 200F where you can have stock settings for normal running engine.

Remember IRON head lags in heat flow to the coolant so combustion chambers is running hotter than the coolant for awhile this can also can screw you trying to straighten this set up process. The aluminum heads is more closer together on coolant temp vs the chambers by several degrees.

Remember, thermal conduction thru different materials, Silver, copper and some other flows heat conduction rather well, while steel iron etc, flows rather poorly. That is why I have big trouble trying to solder in copper heatsinks tabs into the circuit boards where I have very easy time doing this same with steel heatsinks. Yes there are many cheapo electronics that use plate steel as heatsinks. Aluminum is good conductor of heat but not as good as copper but for what it it is, vastly better than steel/iron by far.

Cheers, Wizard

Thenks for the Temp Guidlines Wizard. I just realized as i was waking up in the New Year, that there are two more cells that efffect start up tuning, and those would be the cells with slight boost at 200 and 750 RPM.. Even though the engine certainly never sees boost during crank, those cells will have an influence on the number the fueling formula uses for start, by virtue of the interpolation process with neighboring cells. Maybe that is why the top NA cell is 101, and not 100, so as to minimize that affect. Anyway, I will set them the same. If these cells were influencing things before, then the start up fuel will be too lean when I make those cell numbers the same, since I will be taking a big number out of there which might have been having an effect on the start up fueling numbers.

I think I will crank it first before changing anything, and watch the milliseconds of fuel being delivered. Then I will change the cells and watch again. If fuel ms number drops, then I will raise the entire group of start up cells until it is delivering at least the same number as before. It is still too lean, but I am getting closer. I tried to start up in second gear yesterday, leaving the gas station, and of course killed it again. This time it started right back up. But it can still be better. There are simply too many things to pay attention to when out tuning with the laptop to be certain of what gear it is in.

This is the kind of stuff I haven't been able to get out of manuals and tech lines. I am sharing in the hope that following this process is helpfull to someone out there.

Sam

_________________

Another "hidden" reason the starts up usually revved up a bit more than usual even when hot for first few seconds were to force computer controlled ignition or mechanical distributor flyweights to advance the spark to keep engine alive. This is more needed when engine is really cold and warm up phase and this is controlled indirectly by the idle rpm via computer or choke mechanism.

Reason the computer controlled injection/ignition control knows exactly is going on and hence doesn't revs so high as carb or TBI did except when is especially COLD like -15C or below then computer commands extra high revs (1,800-2,000 RPM) or else the engine simply dies after first day of the startup.

The base timing like 5, 10 or 12 etc BTC (mechanically tweaking the distributor with timing gun or putting toothed wheel correctly relationship with crank or camshaft by specific degrees) is for the start up AND the basis of placing the ignition curve at right location. It is like placing the picture on wall at right spot.

Cheers, Wizard

This engine is really starting nicely now. I raised the base map start-up cell VE numbers 30%, from 201 to 265. This number will affect all the other modifiers, so there may still be some triming to do in the warm up tables. This is with 36lb injectors. The start -up injection open times are about double the injector's open time at idle in MS's for any given ECT. So if the car idles at 14.8 with 3.1 MS of fuel, then the start up injection time will be around 6 ms. I will check this more carefully later, when I am sure things are begining to take on their final tune specs, at least in the NA side of things.

I have also learned to understand Accel's port temp estimation program, and how to tweak it to keep the A/F ratio constant as the port temp goes up. This would be as the manifold gets heat soaked when you turn it off for awhile, or when it is sitting in traffic and the air is not moving over the manifold any longer. As the manifold gets warm, the mixture wants to lean out. There is a fueling adjustment coefficient which lets you raise or lower the percent at various estimated intake port temp (IPT) values. I simply watch the wide band A/F gauge as the port temp goes up, and adjust this coefficient to keep the A/F ratio the same. It works well. Eventually I will turn on the O2 feedback system and let it work, once I have this tuning worked out. The closer it is without the O2 sensor, the less compensation must be done when in feedback mode.

The ACcel Gen VII will let me tailor the timing at start up to advance after start, and then phase out, so that might be a good idea, based on what you wrote, Wizard.

I really think all the trouble I have been having for 18 months can be attributed to bad basic wiring, and a bad ignition switch in particular.

Sam

_________________

Sam
I know this really doesnt help. However, I think you are blessed with a greater knowledge of this programming than anyone else here. This has made you a fantastic asset for those of us who will follow.
Thanks for the continuing updates,
Frank

_________________
Scrapple: Because a mind is a terrible thing to waste.
73 Duster - Race Car
66 Dart Wagon - DD
178" FED
82 D150
All Slant powered