hector

Brain, you may be right on the fuel which is why I suggested to go back to running batch at cranking. This puts the fuel as it was during the first video. Yet with less fuel and different spark settings in the fourth video it sounded better on cranking.

So maybe the rough starting has to do with ignition?

The tables he has are very similar to mine.

cranking pulse %

-19.5* F 326%
8.5 321
30.3 276
51.9 251
80 226
107 201
131 176
152 151
175 140
198 132

His is the same except at 174* and 180* (his table ends) which are 125%, and 100%.

And priming pulse

-19.5* F 6.0ms
8.5 5.6
30.3 5.2
51.9 4.8
80 4.4
107 4.0
131 3.6
152 3.2
175 2.7
198 2.1

His numbers are almost the same as his temp goes up by 20* from 10* to 180* unlike my arbitrary scaled up numbers.

However, during our "cold" starts which are ~90* at this time of year, we have the same numbers.

My car cranks over a bit before starting but I don't mind it one bit. Its similar to OEM in length to start but smooth once it does start. I tried uploading a video of it but I suck at internet and MT.net didn't let me from my phone and uploading to you tube was also shot down. In any case, it started as usual even after two weeks of not being used with an old 10lb battery and the air on.

hector

Don't know if this worked but this is the "cold" start.

Edit: Thanks Andy, that did work. I was trying the you tube link next to the earth link but it delivered that white page for some reason.

Mech5700

Above the message text box, click the little pic of the earth with a chain link under it, paste the url. Then post.

Mech5700

I see. It cranks longer but fires right up when it catches.

Went back to batch fire, kept 15* cranking advance, upped the Cranking pw% (180* is now 200%. I tried increments of 25% from 100 on up and saw some improvement at higher percentages), Cranking RPM @ 350, and used your recommended ignition settings. My battery is crap (about 11.5v every time I try to start after it sits for a little) so I am going to pick up a new one and retest. So far it is better, but not yet where I want it. Will update ya later bud, thanks.

Mech5700

Had an idea, wanted to spray a shot of starting fluid in the throttle body and see what happens. See if I'm looking at too much or too little fuel. Well, nothing really happened. Crank, no start at all. Put the throttle down to the floor and cleared it out, popped and kicked a few times so I know there was too much fuel in there. Scaled the cranking PW% back down to where it was before and finally got it started again.

A couple other things I tried:
-raised up the injector dead time voltage % curve
-saw crank rpm was about 205-210 when I was in clear flood mode, so I set the cranking RPM to 215
-took the default crank pw% and scaled it down so 90* is 150% and 180* is 100%
-set crank advance to 10*

After a bunch of test starts and small changes listed above, I am finally seeing some improvement. But it obviously likes to start better when its warmed up.

The true test will be tomorrow morning before work.

Braineack

iTrader: (62)

put the deadtime back to where it was when i shipped the unit.
cranking rpm needs to be way higher than 215. it needs to be 350-400rpm.

all you need to tune is the cranking PW curve.

Mech5700

The dead time is still set to .8ms, I only adjusted the curve to be higher.

With the cranking PW%, can it be less than 100% at 180*? I swear it still feels like it's getting too much fuel...

Braineack

iTrader: (62)

if its getting too much fuel itll never catch and then backfire eventually and smell like fuel.

Mech5700

Moar loggage.

Is "timing err" something to worry about? It goes wacky when trying to start.

I also see "lost sync count" happens when the rpm drops and flatlines during cranking. Is that just because the crank stops moving for a split second?

Made some adjustments to lower cranking PW%. I was surprised to see that I could start the engine @ 200*F with 50% cranking PW%...

Mech5700

Well, no luck today. Tried to start with the tune posted above and couldn't even get it it catch. Tried for almost ten minutes, tried adjusting PW% back to defaults, nothing. Was so bad I almost killed the battery (the one i bought yesterday) and had to take the truck to work....

richyvrlimited

iTrader: (1)

Why so much advance when cranking? I use 2deg of advance.

Cranking doesn't exactly spin the motor fast, so you need less advance. 15deg is way too much IMO.

Set cranking advance back to something low and then concentrate on the fuel side of things to fine tune.

Braineack

iTrader: (62)

no, it's nothing you can even use.

Like in startup 4 when it pauses during cranking? How often does cranking like that happen, because that's not good. sounds like a weak battery plus too much fuel.


yes you need to be careful with amount of cranking you're putting on the battery, you can kill one fast.

When cranking and trying different settings, you need to flood clear more often so leftover fuel isn't messing up what you end up with.

The cranking pulsewidth %, iirc, is just a multiplier on the Req_fuel value. I find the best way of tuning the cranking pulsewidth is starting with the the table at 100% across the board and tuning for a specific temperature on a cold engine. It's easier to start with too little fuel and add fuel until there's the correct amount to catch. If you're starting with too much, it'll just lead to headaches and the car may never start and you can drain a battery fast.

If you can start a cold engine with the cranking PW% at 50% then it suggests the req_fuel is too high to start with or the fuel map during cranking is too rich (this is an area that normal isn't touched while driving).

You may find that fooling with the timing advance and other things may help, but the main focus should be on getting the correct fueling during cranking at a particular temperature, you pretty much only have one chance a day to tune cold-starts, a warm motor should be starting without much effort/tuning.

Something else to try may be increasing the start-up delay time on your alternator control settings to make sure the alternator isn't trying to charge while you are cranking.

Mech5700

Richy, I had it set to 0* for the last week or so and it would always eventually start in the morning. I set it to 15* yesterday, but it was when the engine was hot. It's obviously too much when cold. Will set back to 0 and try again tonight after work.

Brain, I will ignore timing err.

And the crank flatline has been happening since I've installed the MS... it's never wanted to start correctly. And the log from last night is with the new battery.

Req fuel was set by you and I verified it with the calculation for my injectors. 11.6 I believe it is.

I will try the other suggestions tonight after work, I will have to charge up the battery a bit and give it a shot.

Mech5700

Brain, you mention the fuel map too rich in the cranking area. Does it use the the VE table values while cranking? I was under the impression that only req fuel and cranking pw% was used to calculate fuel delivery while cranking?

Braineack

iTrader: (62)

unplug your MAP sensor, try to crank, and ask me that again

req_fuel is nothing but a multiplier applied to the fueling algorithm--alone it means little to nothing--the cranking pw% increases/decreasing that multipler. But yes to answer the question, the VE table is calculated during cranking.

Your cranking settings are pretty much out-of-the-box defaults. I unfortuentaly havent had a local NB to tune myself to get better settings. They are typically pretty close, my own setup didn't deviate much from them. I'll see if I can reach out and get a better set of startup settings for you from other NB guys.

Mech5700

Please do, I would greatly appreciate it.

I didn't even think about changing the VE table at that area for cranking...

Mech5700

Update for the evening:

Put back to stock, ECU, MAF, everything, to make sure everything was OK because it WOULD NOT start no matter what. I even pulled the valve cover off to verify that I didn't skip timing... It was literally fighting me that hard. Luckily it started fine, of course. I did note that the plugs were BLACK. BLACK AS BLACK CAN BE. It is now clearly obvious I am fighting a rich start problem. Let it warm up so I can start it with the MS. After more fidling, reverting back to the base tune I got with it, adjusting the Crank PW% to flat 100% I finally got the damn thing to start.

I let it run to warm up and tried again, still a struggle but it started. Went to the gas station, topped off, and tried to start again. Fought me of course. I then started whacking the VE table down on the lowest RPM column. It used to be high 60s-70s, and it actually acted like it wanted to start down by 40s-50s. Got it runnin so I hit the road.

Did a 34 mile round trip (about the same route as I do daily for work) and filled up again after. I got decent bump in mpg! I normally hang around 22-23 if I am really light on it. This time I was at 27.6! Granted, I didn't go all the way to work, maybe additional 2 miles on the streets, but hey, I'll take it!

-crank advance is -2*
-alternator does not seem to be attempting to charge too early from what I see on the log
-crank RPM 375
-default req_fuel, injector dead time, and curve
-crank pw% is flat at 100% accross

Faith has been somewhat restored, I know I'm close now... I came home and did one more log for ya to see if there is anything else you guys notice.

If you look close at the cranking RPM pattern, that dip to 0 is what I need to figure out. "Status1" drops, "crank" and "run" on the log records go from "Y" to "N", "lost sync" hits, "timing err" hits, it's like it loses it's mind for a split second before it catches and runs...

I shaved my head so I can't pull out any more hair.

And also made a poem about Brain during my 6am almost late to work no start rage:

Zaphod

I can't see that working..

Mech5700

See brain's post #32. It's a starting point.

What are your pw% values?

Braineack

iTrader: (62)

I'll have a new tune for you to try this afternoon.