cold start FTL
 

well, i THOUGHT i had his ironed out last season, but apparently i was lying to myself.

seems that below 55 or so i can't start my car... not a pop or a sputter, just cranks on the starter over and over again...

unfortunately, the battery in the laptop i've been tuning with is also junk, so i can't even mess with settings right now.

blaaaaaahhhhh...

now, if i'm not mistaken, the only window that should matter as far as getting the car to fire beyond cranking on the starter is the "cranking settings" window, right? after it starts but won't keep running, it's ASE, correct?

i need to either get a USB to serial adapter or a new battery for this laptop...

You need a $10 inverter from walmart so you can keep the laptop plugged in. Alot cheaper than a $30 usb-serial adapter.

i have one. unfortunately the miata battery doesn't seem to have enough juice to run the one i bought without it complaining unless the car's running...

Yep our electrical system is underpowered.

well, i'll get a laptop connected to it at some point in the next couple days, but i'm really after the question i posted.

am i correct with my assumptions about what i need to mess with to try to get this thing to start up?

also, anyone want to reassure me that a slight change on that dialog can mean the difference between doing nothing but cranking on the starter, and actually firing up? because my settings don't look like they're THAT far off from what people have posted in the past.

Cranking Pulsewidths to get it to start-- increase them in small increments at lower temps to get them to start. Don't mess with them at the temps where they've been working fine at or you'll have problems when it's warm-- it was fine there already.

Afterstart Enrichment (ASE) if it dies pretty much immediately after starting successfully.

Warmup Enrichment (WE) if it's too lean/rich while getting itself to operating temp.

ASE and WE work together for the first few seconds while ASE is engaged so keep an eye on how they can effect one another.

ok. sounds like what i was expecting, then. hopefully a few minor tweaks will let me start the damn thing when it's below 60 or so outside... lol

mspnp? if not, make sure your ignition mods are cherry. its probably fueling though.

it's not mspnp. and i also didn't build it. hopefully just tweaking the cranking settings will get me there, because i don't really know the first thing about the ignition mods... lol

ugh... i can't figure this out for the life of me. i've changed everything i can think of in the cranking settings window, and messed with the "more cranking stuff" window as well.. nothing.

i hopped in today and it started then sputter-died quickly thereafter, and then every time after that just cranked and cranked...

are there any other settings i should be playing with to try to get this to fire when it's 50F or less? that's not even that cold, but i just don't have a single problem with it when it's warm out...

so frustrating..

here's the thing. should i just be increasing the cells it's currently in (it's 50 deg CLT reading, so the 40 and 60 cell would apply), or should i increase everything from those down to the coldest?

also, how do i know if i've gone too far and am just flooding it?

hmm, after looking at the MSPNP 90-93 base map, my values are possibly off by a bit...

maybe i'll give those a shot...

it runs!

i ganked DIY's cranking settings from the 90-93 MSPNP without AFM, and it fired right up immediately with a 43 deg coolant temp and 60 deg manifold temp.

feel a little silly for not being able to figure it out, but at least i can drive the car now. :-)

Thanks DIY!

Great-- I figured it would, we got a chance to dial those in pretty well last winter in coolant temps down to around 40 degrees. I'm glad it worked out for you man!

me too. lol. been really bummin about that.

my question, though, is how the values in cells below the current temp might affect startup. for instance, if it's 43deg CLT, does having a higher or lower value in 0deg make any difference? my gut tells me no, but i thoguht i remembered something in the megamanual about MS using all the values to essentially draw out a curve. that said, i could be imagining that i read that.

oh well. i'll keep an eye on it if i get down below 40deg. i'll probably see a 20deg morning before i put her away for the winter.

took me about 15 minutes of fustin' around to get mine to crank this morning (40° coolant). how not fun. had to drive really fast to make up for lost time, which equaled things out, so my day has thus started as usual.

lol.

mine actually wouldn't start last night. probably just under 40deg...

i wonder what the calculation is that MS actually does using those tables, when you're between cells. anyone know?

like if you're at 43deg does it use the 40deg cell value alone? but if you're at 38 will it use the 20 deg cell?

i noticed that DIY's values are a little strange. i would think that the values should be higher as the temp goes down, but the DIY table is higher at 40 than at 20. Jerry just mentioned they messed with it down to about 40, which might explain this. i'll hopefully get a chance to fiddle with the <40 cells tonite and see if i can get something for 20deg.

also, i have noticed that i'll get a different start/no start outcome depending on manifold temp, even when coolant readings are the same. looks like MS takes ONLY the coolant reading into account when cranking, which probably explains why it started in the afternoon at 60deg manifold temp and 43 coolant, but not last night at 10:30 with probably closer to 40 - 45 deg manifold temp and roughly the same coolant temp as in the afternoon. (i didn't run the extension cord out to the laptop to check...)

you can have it average the two.

is that the "CLT only" value in the cranking settings dropdown?

i looked at the other value in that dropdown and i coulda sworn it said something about MAF values. so i thought it best not to bother (probably shoulda bothered) and left my settings the same as what DIY's got, which are CLT only, always on, twice, and priming table.

that said, i may try giving that a shot tonite when it's cold again, since again, i don't think that the coolant temp was much different on the two startup attempts. afternoon, it fired IMMEDIATELY, but at night with colder air, it just spun on the starter again.

just got get unlazy and mess with it for awhile at a time again.

still no dice. thoroughly confused this morning, since it started up INSTANTLY this morning with 38deg coolant temp and 43deg manifold temp. so once it got over the initial quick stumble and started idling, i shut it down quick before the temp rose more than a degree or two, and tried to fire it again. zilch. nothing.

messed with it for probably ten minutes and couldn't get it to do a damn thing. like i said, confused...

though, once i got into work i did remember that i may have cooked my ignitor when i was doing the install, burning MSQ files (file->open->burn) without unplugging the ignitor. at one point right after i finished the install i had no tach signal to the cluster, so i know it's at least been less than fully operation at one point. it's acted fine since, but i imagine that could POSSIBLY have something to do with my cold start issues?

i have a replacement and am going to swap it in when i get home from work tonite and cross my fingers. thoughts, anyone?

It's VERY easy to flood the engine when working with cranking enrichments, and in the scenario you presented (successful start, quick shut down to try again) adds to that because you likely have heavy afterstart enrichments kicking in immediately after that first cold start-- if the engine is under those heavy ASE when you shut down the cylinders may end up pretty wet and then not want to restart even though your cranking PW may be fairly close.

It's likely a combination of slightly less ASE, and maybe slightly less cranking PW needed at that temp. Also- check the priming pulse-- what's it sending in there? That's adding to it too.

A heavily flooded engine with wet plugs can be VERY hard to start-- you pretty much have to pull the plugs, let the plugs and the cylinders dry out, and then start over. Been there, done that a few times.

yeah, i have thought of that, actually. no idea how to tell if i'm giving TOO MUCH instead of too little.

mine started right up this morning, i basically divided my PWs in half.

Im sure i flooded it yesterday. I was lucky it caught at all. i guess all the backfires helped clean out the system :gay:

divided them in half, huh? yeah, i noticed the map i got from you was REALLY high at the colder temps when i compared it to the DIY settings. i figured longer on colder starts was better. lol.

see, my problem is when it's not starting, i don't get ANYTHING, no backfires, no hiccups, nothing.

i did take a look at the flood clear setting, and don't know if it's right or not. since i don't have a linear TPS, and that's listed as a "raw" value (which i assume means voltage?) where would that be if i want to be able to depress the gas pedal to crank it for a bit in flood clear mode, with no fuel?

mine are:

7.1
6.7
5.4
4.4
3.5
2.9
2.3
1.9
1.5
1.3

with 460cc injectors.

Without a linear TPS you can't use flood clear mode. And in my experience it's possible on some engines to flood them bad enough that this doesn't always 'clear' everything very well anyways. It helps blow out the cylinders some, but it won't dry the plugs and some plugs don't like to fire soaking wet very well.

On another note mentioned above, the trend should be longer cranking PW at colder temps. What you're trying to figure out is 'how long?'.

With the engine clear (not flooded) try cutting the CPW back 20% at that 40* point and see if it still fires reliably. If so then you probably had more fuel there than you needed and you can likely scale some of the surrounding CPW accordingly too (TEST!).

Make small adjustments, test between each change, and save off .msq files with descriptive names so you can go back to a 'known good' or 'known best so far' setup until you get it perfect for your setup...

yeah, mine are a LOT higher than your new ones, brain, since what i've had were damn similar to the ones you sent me in that msq. i think i'm around 8.2 at 40* and like 5, 4, 3 etc... below that.

another thought that crossed my mind was whether the injector size comes into play here. commonly i thought that ALL one should need to change is req_fuel, but these are millisecond pulsewidth values. bigger injectors, smaller values?

i thought the same, but that req_fuel doesnt come into play.

since my injector is 50% larger than stock and i was using DIYs vaules. I divided them in half. seems to work out perfectly. since my larger injector will squirt about the same fuel as the smaller one now in the same situation.

Read the mspnp manual, and you will find that it explicitly instructs you to modify cranking PW with injector change. Not only that, it shows you, mathematically, HOW to change them ;) Good stuff, that DIY!

how did i miss that... i need to start having my coffee BEFORE i try to start the car... though, without starting it, how do i get to Dunkin Donuts? ;-)

Remember that PW includes injector open/close time, so you'll get better results if done according to DIY instructions (essentially subtract out open/close time, change to account for injector change ratio, add back open/close time, voila!)

i'll second that, though. can't thank Jerry enough.

stock 1.6 blue tops are 203CC, right?

230cc I believe...

Well, I two was experiencing the no restart problem....but mine is turning out to be a heat soaked AIT sensor and bad clt-related air density corrections.

after about 3 secs of the engine being cut i begin to heat soak. by the time im about to start the car the sensor is reading 100*. my density correction values were screwing it up somehow, im not even sure the are good now. but once (if) it cranks over it's incredibly lean.

MY cold start seems perfect now, so it's on to troubleshooting this. What seemed to work was using DIYs density corrections. I let it sit for a few mintues, verified a 100* intake temp and it started right up and was rich for a few seconds before it quickly dropped down to 70*. Hopefully some sort of heat sheilding or wrap will help....

I never had a problem with IAT heat soak the whole time it was mounted in my plastic stock crossover tube. Now that I've got higher underhood temps with the turbo, and the IAT is mounted in a piece of steel, I get pretty bad IAT heat soak.

I wish there were startup-only IAT-based corrections.

-Mike

Isn't that what "Cranking Pulse Width Temps" (MAT Only) under Cranking Prime Table is for?

that's the 1.8 browntops 1.6 bluetops are 205cc :)

yeah, i dunno what the deal is now. even without heat soak it wont restart without tons of effort.

this morning i stopped by Rosenthal, intake temps of 50° coolant at 190°. stopped for about 3 minutes and came back out. Tried to crank it and no go, all temps about the same as when I left it. :(

That's what I had thought too originally, but I had based the 230cc off the FAQ... so I can do some digging, but which is it?

Here's the DIY Turbo FAQ entry on it:

Taken from https://www.miataturbo.net/diy-turbo-discussion-14/diy-faq-all-your-answers-one-big-post-4288/

those numbers have been verified from a number of sources.

now if jerry can only solve my issue.

have you noticed much of a difference with MAT only, CLT only, or averaged?

seems to me that if your intake temps are 50, and the coolant is 190, averaging them would put you in a spot where nobody's happy... i'm just supposing, though.

no its clt only, i think i want to try averaging them.

yeah, i was thinking about that the other day, when mentioned that i fired the car up with 43* CLT and 60* intake, but then later couldn't fire it with similar CLT and 40* intake.

i've read that FAQ a lot and never noticed it quoted different numbers than what I'm used to.... hrm

ah well always happy to be proved wrong :)

Don't know if I can magically solve anything-- but I'll surely try and help ;)

Firstly I don't know that i'd average IAT and CLT for this-- the engine temp is much more critical for this calculation. If you're averaging the two, and IAT is at 50 while CLT is at 190, you need a tiny little PW to start the car, but you're getting a much bigger one because the averaged in IAT reading in bringing the temp used to calculate cranking PW way down to a much lower temp, where a much higher PW is commanded, instantly soaking your plugs and flooding you out.

That would be my theory anyways...

Me too-- much better to be proved wrong, learn something from it and be better off the next time than hardheaded about it and dumb forever right? ;)

I had always used the 205cc value before on the 1.6 and 230cc on the 1.8, probably just forum misinformation which is all too common, and manufacturers really don't make this type of info the easiest to come by alot of the time. This is actually part (not all) of the reason why the AFR number used to set REQ_FUEL when scaling to different injector sizes on the MSPNP MM9093 and MM9495 isn't 14.7-- I had to account after the fact for the true injector sizes so it would scale properly for you guys. It won't change anything though as long as it's consistent.

i guess ill try lowering the PWs a tad more and turning off the priming pulses.

that seemed to have worked. i drove home and parked and waited untill 100* playing with my ait based corrections and PWs some more got it to start everytime from 80* to 108*

hopefully it will do the same tomorrow at 190* CLT and 50* intake.

So...what 1.6L injector size did you use to rescale the diyautotune cold start pw values?

Actual injector sizes will apply for this as it was tuned with the actual factory injectors in the car, so 230cc for the 90-93, and 265cc for the 94-95.

Glad to hear it looks like you've made some progress Scott--

Thanks very much! Exactly the info I was looking for.

they also use 13.7 as the afr multipler, however, iirc.

Anyone know why DIYs cranking/prime table counts down until it reaches 40F then it jumps up again and then continues counting down versus counting down the entire time?
This might just be an error of me not having MT setup to read their MSQ but I don't think that the values themselves would change because of that.

Jerry, what kind of AIT related air density have you used on turbo cars? Just curious if you have a table to share.

It never got below 40* in Atlanta, so they couldn't tune below that.

i ended up cutting the DIY cranking values down, by multiplying them by 75%, and so far so good. only tested in 50* temps so far, though. tomorrow morning i should get a solid 40* startup try, so i'll see how it works tomorrow morning. optimistic, though.

hahahhaha That's right, makes sense, i didn't think about that.

heh, i had the same thought as you, cj. then the "oh yeah... atlanta... 40deg..." thought hit me, too.