If I flip the toggle switch to start the fuel pump, and crank the starter for a few seconds before turning the ECU switch on, it fires immediately when the ECU is energized. This morning it would not start. And I could tell from the smell that it was flooded. So I turned the ECU off to clear the cylinders. I flipped the ECU toggle back up without stopping the cranking, and it fired immediately. BOOM. No almost start. No coughing, and sputtering. I tried this many times later in the dayat different air temps, and different ECT temperatures. It lights up like a bomb the instant the ECU is energized every time.

One key here is that the Accel ECU has no clear flood mode as mega squirt does. I was considering putting a momentary contact switch to kill the injectors, but was not sure how to wire this. I don;t want to do something that will hurt the ECU.

So it is pretty clear that I have the start up enrichment map set too rich for cool weather. It is still cool here in DC. And, I could fiddle with this, and maybe get it to start better in a more conventional manner. But I don;t really see anything wrong with using the fuel pump-crank, ECU sequence to get it to start when it is cold. One advantage is, the oil pressure builds as it is cranking. I can watch the pressure rise and not turn on the ECU until the pressure is up if I wish.

Any thoughts on this observation? Any ideas about what else might be going on, if anything? Do you see any down side to just keeping it as it is, and using this special technique to get it started? I know it is odd, but this entire car is pretty odd now.

Sam

_________________

The only downside I can think of is that to me (maybe or maybe not to you) it's aggravating when something isn't quite right, even if I have a workaround for it.

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

Dan, I do feel the same way, and would not stop thinking about it, and maybe playing with the tuning to see if I could get it "right". And, I am absolutely open to any and all suggestions for "fixes" here. But I now feel like I could drive this car anywhere once again. I was really afraid to get very far from home if there was any likelyhood that the weather was going to get too cool later in the day. More experience with it will build more confidence.

I can also see that maybe this could be rigged to create an ad hoc, accidental theft proof device. Maybe I am under estimating the brilliance of thieves, but it seems if it does not start right up, they are likely to just leave. If I made the start up enrichment too rich across the board, only those in the know would be able to start it up. Maybe this is silly. But you know the old saying about making lemonade if you get lemons.

Sam

_________________

H'mm. Well, leaving aside the anti-theft idea, IIRC you had good results taking a bunch of timing advance out of the crank/startup regime...why not try pullling out fuel, too?

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

That makes sense, of course. But the real question is, why does it behave differently to crank first, and then flip the ECU on? It really does not make sense. I think it is more complicated than just too much fuel. Or at least, more complex. Why would it fire right up when cranked first? In both cases, spark and fuel start , at least in theory. simultaneously. Or at least should in both cases. I have always felt there is something not quite reliable about the interface between the ECU and the distributor. It is like it does not really read and fire the MSD until the distributor has spun a few times. This is all speculation.

But, the problems starting might be, maybe, lack of spark on the first couple of cranks, and not too much fuel. This would flood it. And then when spark does come, it is flooded. I don;t know. LIke I said, this is just random ruminations here. The thing is, I have played with those start enrichment table forever, and could get it to start with one setting just fine, and then the next time I tried to start it, it wouldn't. I never had this problem with the megasquirt. Never. It always started just fine. And seemed to be not too terribly fussy about the start up enrichment. So I can't believe the mixture is all that terribly critical.

Sam

_________________

Hey Dan, if you are right, and someone else is wrong, you don;t have to convince them or your postition. Time will take care of that. In this case it was only about 8 hours. This morning I went out to test my new procedure in 38 degree weather, and the car would not start, and smelled of gas. . No maner of toggle wiggling would get the car to start. So, I marched back inside, got the lap top, and took some gas out of the start up map.

It did start. So, you are right for sure. The thing is, when it did start, it seems as if the flooded condition maybe complicates knowing if the current enrichment for 38 degrees is really correct. Maybe if the cylinder is completely dry upon cranking it will now not be rich enough. This is hard.

And, maybe there is is still some wierd intermitant ignition problem. Although I suspect that is not likely, it is possible. It fails to start properly only when the car has set for awhile, and is cold. That hardly seems like the kind of thing that would effect ignition. Every time I have thought to look at it, the RPM count has been working on the lap top during crank, which means the distributor is being read by the ECU.

I told myself I was not going to tackle this problem until I got the garage roof fixed, (Sunday) and got the new tank and fuel pump in. I just couldn;t help myself. Thanks in advance for any helpful hints or thoughts here. I know you have all dug into this some in the past. I am still optimitic that I will get this worked out.

Sam

_________________

You've reminded me of when I used to carry a 3/8" and a 7/16" end wrench in the glovebox of my '65 for progressive incremental choke adjustments so when the car would get "cranky" after sitting and getting cold in a snowstorm.
The hardest part is when thinking things through, and stumbling upon an idea, and convincing oneself that's gotta be the right problem/solution because it all makes sense and fits...then thinking about it some more and stumbling upon another idea, and convincing oneself that this one's gotta be right, because it all makes sense and fits...then going out to the garage and finding out that neither problem/solution is the right one.
I have to wonder if there's some kind of "prime shot" programmed into this ECU that puts a prime charge of fuel into the engine by firing all the injectors if it's been longer than nnn minutes since last engine-on and the ambient temperature is below ttt degrees. If so, and there's too much fuel programmed in at crank, that could cause flooding...
Our garage fell in (literally) about a month ago. Nothing of consequence in it at the time. It's been scraped off and hauled away. City bylaws are very picky and particular about what can/can't/must be built, and the results of building a compliant garage would be essentially useless. Let's hear it for life in the big city, governed by regulations written by people who just like to control others' lives.

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

Actually correct way to clear a flooded engine is press gas pedal all the way down (WOT) and crank for 2-3 seconds then release and attempt to start normally. This works on carbed engines and injected engines as well.

Even your computer *should* have this feature as well.
Megasquirt models have them also.

Sounds like injectors are still squirting where you are supposed not to have when engine is not turning. That shows me you have power noise issues. Important to have equal gauge (12ga or 10ga) for 12V and ground for the computer, 12V switched via relay seperately via computer's control (preferred) or by switch.

Cheers, Wizard

Correction, works on some injected engines. Some systems do have a flood-clear mode, and others do not. It is by no means universal.

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

According to Accel's tech line this ecu has no clear flood mode. But, such a thing is mentioned in a commercial publication that supports their earlier Gen VI ECU. So, either they eliminated this feature, or the tech rep did not know what he was talking about. It is not mentioned in their POS manual.

Dan you are are correct about why this process is so frustrating. But what makes it so hard is that there are so many layers between and you and the old air-fuel-spark formula. That is why theories are generated, tried and then marked as failed. It is very much a mental game and not a physical game. And the physical games are the ones I am good at.

You can't just look in and see if fuel is there. And checking for spark is not as simple as pulling a plug wire. And checking timing is not a timing light process any more. You have layers of sensors, and software, and relays that may or may not be working between you and the product itself. I have been told that car companies spend millions of dollars calibrating their ECU's for production use. I can believe it.

The other thing that makes it hard, is I can't tell for sure why it did start when it starts. It really seems like their is still some hidden, un-thought of random bug in this system that is eluding me.

The thing is, once it starts one time, from then on it behaves absolutely normally until it sits for a couple of hours. After that, it won;t start; then I can fiddle with the start up fuel coefficient and get it to start, but I am begining to doubt if this really had much to do with why it did start when it did. Then the next time I go to start it, it won't.

And, by the way, the toggle that will somethimes get it to start by flipping it off and then back on is not the ECU, I was wrong about that. It is the general ignition voltage. It is the circuits under the hood that were controlled by the ignition switch before, and are now run by a relay. So, my logic is now turned on its head. I was not cutting off fuel, and turning it back on. I was shutting off the ignition circuit, which has the MSD box on it, and turning it back on. Now, if I could smell gas while cranking-and-no start condition, you would conclude it is flooded right? But turning off the ignition would or should make it more flooded. And yet it fired up after a few cranks of NO IGNITION and still fuel. So I am dumb founded. This should flood it more, not clear it of fuel. Is it possible to smell fuel in the tail pipe when it is not flooded?

Is it possible the relay that is supposed to be turning on the ignition is unreliable? Why would this be temperature sensative? Maybe there is just no spark at all. But once it starts, it runs fine after that. I need a very logical sequence of checks here that can eliminate one variable at a time from contention. But there is the concern that what ever is failing here is not failing all the time. So even a good check may not be good for good.

By the way, there is a pre-start fuel coefficient that I think may effect things as you suggested. But the manual is not at all clear about whether this is during crank, or immediately after firing that it effects the fuel. It is correlated to ECT. And, I took this curve way down last evening when it would not start, and for the time being, it started better. But later, it did not want to start right up. The universe is really playing games with me.

I am just this("es) far from selling this car and buying a Miata. To make matters worse, my wife and kids are begining to give me a hard time about the polution this car generates. And, they are right. It is pretty stinky. I took on this EFI project to keep my mind young. I wanted the challenge of the digital world. And, I wanted to learn about modern engine management systems. But the bottom line is, I feel older than ever, and again very discouraged. This car is like Miss America with a chastity belt.

Sunday the garage gets fixed. There are no relays or software in the process.

Sam

_________________

I wonder if your MSD box is maybe a bit hard to wake up in the morning.
Certainly, if that fuel never got burnt!
No, they're not. Not if you look at the big picture. Manufacturing a vehicle is a hugely resource- and energy-intensive process that generates vastly more pollution than even an old car generates in its entire lifespan. Discontinuing your use of an old car and buying a new one because its exhaust is cleaner does not reduce pollution at all. It increases it. And unless you're planning on crushing the Dart so nobody can ever drive it again, it'll still be on the road, kicking out what your wife and kids consider excessive pollution (I'm sure their hearts are in the right place, but their facts are deficient). Still, even so, if you want to reduce pollution, put a catalytic converter on your car. You've got a programmable EFI system which, I imagine, would be more than happy to look at an O2 sensor and keep the mixture stoichiometric for you.

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

Can an MSD box have an intermitant kind of function like that? I assumed that being digital, and solid state it was either good, or dead. This thing is at least 13 years old. Does anybody have as diagnostic process to recommend for checking out the MSD box? I would prefer not to just throw parts at it. Although I certainly have done some of that in the past for sure.

Does anyone out there have a history with an MSD box behaving in a cranky manner? That is, some times OK, and somtimes not.

I tried turning the ignition toggle EARLY this evening, and letting it "cook" for a minute before cranking, and it fired right up. But, It was warm. I may not know the answer here until it gets cool again. I don;t know what the weather forcast is for the next ten days, but it is about time for the DC weather to get warm and stay warm. If it gets cool over night, I may get a chance to try it again.

Sam

_________________

By the way Dan, I really appreciated your sharing the fact that it creates more polution to make a car than that car creates in its waning years of driving. I shared that with my wife, and she said that fact gives her a new perspective on the old car. She has always been a fan of it as my hobby, but did have that concern of late. Your comments have helped.

Lou has shared this with me in the past, but this time it was timed just right to help heal some family friction. Thanks.

Sam

_________________

Glad to help. They are entirely correct that what leaves the tailpipe of your Dart is considerably dirtier than what leaves the tailpipe of anything made in the last twenty years or so, but that's only a small part of the picture. Nevertheless, your Dart's exhaust is going to be considerably cleaner than that from a carbureted Dart, especially one capable of the same performance level as yours. And the fact — uncomfortable though it may be to an egalitarian — is that the vanishingly tiny number of old/high-polluting cars on the road are "subsidised" by the overwhelmingly large number of new/low-polluting cars on the road. It's been pretty well demonstrated that there are so few pre-1981 cars in any kind of regular use that even if all of them were magically made to disappear tomorrow, the effect on air quality would be unmeasurable. Think about the ratios involved: One old Dart for every...what...ten thousand 2005-2008 models on the road? Twenty thousand? A hundred thousand? Let's say the ratio is one oldie for every ten thousand '05-'08 models, even though that guess is probably very low. Let's assume the '05-'08 models are all emitting at the 33 %ile of the range allowed in the mid-'90s by Federal certification protocols, so those numbers are pegged at 0.0825 g/mile HC, 0.132 g/mi NOx, and 1.12 g/mi CO.

Let's assume the oldie is a completely stock, typical 1963 model emitting 8.25 g/mi HC, 2 g/mi NOx, and 11 g/mi CO. That is exhaust between 10x and 100x dirtier than the late models.

Now let's calculate the averages

HC:
[10,000 (0.0825) + 8.25] ÷ 10,001 = 0.0833, which is less than a 1% increase over the fleet of '05-'08 cars without the '63 model factored in.

NOx:
[10,000 (0.132) + 2] ÷ 10,001 = 0.13218, which is just over one tenth of one percent increase over the fleet of '05-'08 cars alone.

CO:
[10,000 (1.12) + 11] ÷ 10,001 = 1.121, which is eight one-hundredths of one percent increase over the fleet of late-model cars alone.

Take a look at the stackup of conservative assumptions in this calculation and the tiny differences involved, and it becomes plain that the dirty exhaust from all the old cars remaining on the road is so trivial as to get lost in the noise and measurement.

Would I want all the cars on the road to be '60s models? No, we'd all be choking to death. It very nearly happened in big cities back in the '60s and '70s when there were many fewer, much dirtier cars on the road. Now we've got many more, much cleaner cars on the road. Whether there are too many cars on the road altogether is a different question, but one needn't feel guilty about driving an old car based on exhaust emissions.

(And that's before we get into the fact that the old car runs cleaner today than it did in the past due to cleaner gasolines, better ignition systems and engine oils, lower tire rolling resistance, etc.)

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

I know this is drifting from the topic, but it seems to be the item of the moment.

Your car can be extremely clean running if you already have a working fuel injection system.
Maybe right now, you are busy working out the kinks in the startup routine, but you can always keep in mind the more lofty goals.

I am still young in the game with upgrading to EFI (like lurking here and stalking megasquirt auctions on ebay) but I've looked at a few of these new car engine management systems. The big features that play into clean burning are not all in reach for a slant six (I'm guessing combustion chamber design is less than optimal) but with efi can come 02 sensors and computer enhanced timing control and even effective use of catalytic exhaust systems. Those features might get you into the later part of last century as far as pollution standards. Even if your specific experience/setup doesn't make your car run clean as a new one, you are helping other owners to improve their options for classic car updates.

I have read many of your posts and gained a great deal of insight from them. Your effort will be multiplied many times over as others build upon it. Thanks for taking the time to share it and keep up the good work. (You too Dan)

Paul

_________________
1980 Aspen 225 super six