I still have not hit upon the correct start up combination. The fueling that worked with the short tip plugs, did not work at all well with the extended tip plugs. I think I was putting too much fuel in with the short tip plugs in an attempt to get it to start with the timing advanced too far. I reduced the timing to 6 BTDC at start up, and it worked OK until I changed the plugs to the stock plug depth and the weather got colder. So now it will not start when it is cold, and I have kind fo lost my way. I think the longer plugs are more prone to fouling out from getting wet. So houw do I proceed?

First question: What are the clues as to whether it is failing to start from too lean or too rich. When it has failed to start, it has not fired once. It acted like the ignition was dead, except the RPM count on the laptop was giving me a read, so I knew the pickup was working. When it did start it was after about 3 or 4 tries and extended cranking. I tried increasing the fuel, and decreasing the fuel, but was unclear what made it actually fire. I think it acted flooded when it did start, as it sounded loaded up, and slowly came up to speed as it gained stability.

When the engine is cranking, the pointer in the VE fuel map goes to the 100 KPA cell at the lowest RPM. What should that value be as proportion of normal operating VE? Does an engine need more fuel when cranking at 200RPM than it needs running at 750RPM. Should the VE value be more or less that idle VE?

How does a slant behave that is trying to start too lean? Does it fire and then fail to catch. Are there any half baked sputters, or is it the same deadness I am getting from the engine now? It does start fine when it is warm. I know when the accelrator pump linkage fell off my old Holley two barrel the car would not fire at all to start up, so a slant will not fire if there is no extra fuel for start. I am just trying to figure out a good strategy to get this dialed in. Thanks.

Sam

_________________

Offhand I can't think of a reason the long-tip plugs would be more prone to wetting out than the short-tip plugs. You may want to make an (unsuccessful) cold-start attempt, then immediately go remove a plug or two. If they are dripping with fuel, then obviously you've got too much. If they smell of fuel but there's no dampness, that's an indeterminate result. If they don't smell at all of fuel, then you've obviously got not enough (or none at all). My guess is that you have too much fuel rather than too little. I think if you had not enough fuel, you'd have the engine firing and "trying" to start. Your "slowly came up to speed as it gained stability" description is much more likely to result from excessive fuel than insufficient. I am not familiar with the units in which fuel is manipulated in your system, so I don't know whether to say "cut it in half" or "take out five units" or what. Can't answer your VE question, sorry.

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

I opened the Mega Manual for Mega Squirt, which is Megamanual.com, and went to the start up fuel calculation strategy and that was a help. I got an idea of what was a reasonable start up fuel requirement in milliseconds, and the change needed between cold and hot.

The important realization is that you must be within .5 MS for a really fine start up. That is not much of a difference. So I realized I was probably changing things by too large an increment when I did try dialing it in. I went out and fooled around with the start up fuel from 45 degrees ECT to 185 ECT, and I got really, really fine start ups there. I spent about an hour starting it, adjusting, and restarting it. I let it idle and warm up, and dialed in many ECT values between the two temps.

It starts so fast now that you could not hear the starter motor crank at all. I just kept playing with it, up and down, and looking for that effortless quick start once I heard what could be done. Now if I can get it to start when it gets cold again! I will keep working on it. I hope these tunings are stable.

Sam

_________________

Sounds like great progress! The bearish thing about cold-start tuning (whether it's with wrenches and screwdrivers or with a keyboard and screen) is that you really only get one or maybe two truly cold starts per session...then you have to wait until everything's stone cold again...

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

Do the same timing parameters hold true for a slant when really cold? Would a few extra degrees of timing help? It is set at 5 BTDC for all ECT temps now. I can vary the start up timing accross the temp range from -40 to 225 in roughly 20 dgree increments. Several weeks ago the timing start up term was set to be advanced an additional 20 dgrees when cold, and then I flat lined that as per your advice. Do you know about the cold,cold start behavior of a slant in regards to timing?
Sam

_________________

The usual strategy is to start out a bit too lean and gradually add fuel in 0.1-0.2 msec increments until it starts, at least with Megasquirt. This helps avoid flooding the engine.

_________________
"Mad Scientist" Matt Cramer
'66 Dart - turbocharged 225
My blog - Mad Scientist Matt's Lair

Not really, I think most of time it is just one base timing at all temps.

This is very critical that the BASE timing is set without the computer control intially, this is why is spelled out in FSM on older vehicles that uses computer-controlled spark using a distributor. On ones that do not use distributor, still have the intial advance DESIGNED into the crank or cam. For example, ford uses intial 10 degree BTC installed on the 36-1 tooth wheel on the crank for starting and limp-in timing for EDIS. That's where computer takes the base timing for start up timing because RPM is too low to give accurate dynamic timing during start.

Now you do understand why that engines needs rich mix in first few revolutions of start and slightly less richer for first few 10-30 seconds of run at 1,500-2,000rpm to help engine stay alive till there is some heat in the head materials so it won't quench the fire. This takes 10-30 even a minute (especially -20C and this is fortunely few. Some people work around it by gently pressing on throttle just in case if computer is insufficient to keep engine going at that extreme cold in that critical warm up phase. You'll know it: engine is missing and bucking if weather is that freezing cold like -25C.

Computer also advances the timings at same time once the computer sensed engine is above RPM setpoint (like 500rpm even bit higher) for cold start ups. This is set just above the starting rpm and not too high or it won't advance soon enough and engine strumbles.

Start up ratio mixture is usually just about same as WOT mixture ratio, slightly richer. IIRC. Richer mix is EASIER to ingite but too lean or extremely rich is hard to torch off in a cold engine.

Also there is warm and hot restart temps to consider about as well that comes later after you estalished the reliable cold start ups.

Lot of computer have code to perform "flood clear" mode by pressing down throttle competely (this tells computer to disable injectors) and turn engine over to clear flooding. Even modern cars do have this feature.

Cheers, Wizard

I really don't think you want much of any advance when cranking at any temperature, and I can't think of any reason why you'd want more advance at lower ambient temps.

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

Should it be set to zero when cranking cold? Also, in repeated tries going from lean to richer, is there a build up of fuel. Here is a dumb question. Why does flooding the engine keep it from starting? I mean, scientifically, and detailed, not just, "there's too much fuel"
Sam

_________________

I thought about that...you may want to give it a try. Consider: The whole point of spark advance is to account for engine speed. You want the biggest part of the bang to happen right at TDC, but there's some lag between firing the spark and the biggest part of the bang, so as engine speed increases you need to fire the spark progressively earlier to maintain the optimal timing of the biggest part of the bang. The colder the ambient temperature, the slower will be the cranking (battery's weaker, oil's thicker, etc.), so if you fire the spark 5° BTDC, the bang will be well underway already by the time TDC occurs, which means the piston on its way up will be fighting the bang. You may want to try 0° advance for very cold cranking and see what result you get.
Because any fuel will burn only within a specific range of ratio to air. Too little or too much fuel relative to the amount of air available...no combustion. Which really does boil directly down to "too much fuel".

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

So the mixture simply gets too rich to light off? I always thought the spark plugs got saturated,and were grounded out. Silly me.

Here is another thought. There is a feature with the ACcel DFI that allows a delay of fuel for a specified number of crank revolutions. This might minimize the tendency for it to flood. I say might, simply because it would crank air through for one or two revolutions and clear it. I don;t know if this is of any value or not here.
Another feature is that the start up timing can be phased in and phased back out at a pace I determine. This allows it to find the ideal timing as the engine cranks, in theory. What do you think of either of these ideas? You realize you are my grounding force here, and serve to keep me from getting too carried away with creative ideas that may or may not have any value. Thanks for serving as such.

One more thought. There is no flood clear mode in the ACcel DFI. Would it
work to run the ground leads fo the injectors through a rocker switch, or relay that I could cut off when I wanted to clear a flooded situation? Is there any hidden problem with this?
Sam

_________________

Yep. Remember, gasoline does not conduct electricity.
I don't see any value here, because there's no means by which the engine could be flooded and require clearing out at first crank, that wouldn't also flood the engine after a non-fuel revolution or two. If you've got too much fuel dialed in at cold startup, it's going to be too much whether you start injecting it immediately or after a revolution. But you might have a thought regarding air at cranking. Next time you're doing a dead-cold start, you may want to try holding your foot on the accelerator, about 1/4 of the way down, while you crank. Don't pump, just open the throttle a little. If this nets you an instant start, it tells you a few things: (1) you may have too much fuel during cranking, and (2) you might want to see if the IAC can be ordered to open during cranking, instead of just after engine lightoff.
Not sure I follow you here. Please elaborate?
I can't think of any reason why this would be a problem, but it would introduce a potential failure point...

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

There are three (actually more) parameters for start up timing, one specifies the number of degrees of advance, one brings this speciified amount of additional timing in over a number of revolutions, and the last phases it back out. This is added to the value in the timing map at 200 RPM and 100KPA, which I can set as well. So if I set the base timing at 0, and put in 6 degrees advance for starting, then it would start to crank at 0, raise the timing to 6 as it cranked, and then back it back out after a set number of crank rotations, determined by my calibrations..

Thanks for the tip on opening the throttle slightly to try and get it to start when cold. This sheds light on why they tell you not to touch the throttle on an OEM ECU. They have that bad boy set to fire with just the amount of air the IAS allows. I will try what you suggested. I was going to work on it tonight, but went to the gymn instead. Maybe tomorrow.

Sam

_________________

I still touch the throttle lightly on some balky FI cars if I need to get it fired up if battery is bit weak or it just splutters and barely fire, touching the throttle give engine a chance to spin faster if it catches. I had this on couple GM J-body cars (91 and 89), thankfully they are turned into soda cans now long ago.

Cheers, Wizard

I just re-read the manual for the start up ignition advance management. Here is a more detailed acount of how it works:

When you first crank, the timing is set at the crank off set number of degrees, which in my case is zero. Then it begins to add timing advance based on the base table plus the start up table, at a rate specified in it's own table which has 4 different temp ranges, (140-32-113-180).When the total specified amount is rached it begins to phase it back out towards the crank off set.

In my case, the phase in interval is currently set at zero accross the board. This means there is no phase in. The timing is just there from the first moment of cranking. The phase out interval is set at roughly 1/2 second. .25 degrees of advance are removed every 1/2 second until the timing reaches the crank offset specification. If I change the phase in interval to anything at all, then the timing will climb and then go back as the engine downis cranked.

The phase in increments are in milleseconds, and can be as large as 790. It seems as if this process should take just a few seconds at the longest, but you could spread it out over more than 12 seconds in the case of a zero to 6 timing phase in and back out.

I guess I will just go play with it. The other 'bear' as Dan says, is that you have to sit in a very cold car while you are fooling with the cool start tuning. But, I don;t see anyother way to do this. Any slant EFI tuners out there with some numbers, whose cars are starting right up when it is cold?
What size injectors are you running, what does your lowest RPM/highest KPA VE cell say, and what cold start coefficients are you looking at. maye how many ms your injectors are firing at start up. Just for the record, mine is firing at 2.7 ms when it is starting warm,with 6@ 36 lb injectors. Your numbers might help us all find a target calibration. I will report what I come down with when I find it.

Sam

_________________