I've pretty much always had this issue, just never looked into much yet..On a hot re-start, the motor runs lean for maybe one minute, then runs fine. The wideband shows super lean during this time as well, and motor will stuble and run like crap. Over at the MS website, people are all talking about heat soak of the IAT, but the intake temps are damn near identical at the restart. Like i can shut the car off, then 10 seconds later restart it, and it will run lean until a minute, when the ego correction starts adding fuel. Anyone else have this issue? Its done this with every code version i've tried to date.

Hot afterstart enrichment?

Originally posted by ner947
Hot afterstart enrichment?

Is there such a thing? Also, the ego gauge on MT shows that its adding like 10% more fuel once it goes into closed loop after the hot restart...rather than it being only +/- 5 % when going from a cold soak start to fully warm..wierd..

I have the same problem. I had the ASE explained to me pretty good the other day on the msfi forums. If you look at a datalog while starting the ASE only kicks in for a few seconds. After that is goes into warmup. I dont think the ASE has to do with this problem. What wideband are you using? Sometimes I think my innovate comunicates to slow with my msns. Just a thought.

Originally posted by selmerguy2003
I have the same problem. I had the ASE explained to me pretty good the other day on the msfi forums. If you look at a datalog while starting the ASE only kicks in for a few seconds. After that is goes into warmup. I dont think the ASE has to do with this problem. What wideband are you using? Sometimes I think my innovate comunicates to slow with my msns. Just a thought.

I'm using the AEM wideband..Its not that because the MS isnt seeing the signal from it right after warm startup, until like a minute, then the ego starts correcting by adding fuel, and the motor smoothes out...

Why is your EGO correction adding fuel at idle? Maybe you simply have VE values that are too low.

Matt- maybe you could send me a copy of your MSQ file and I can compare it to a lot of MSQs that I have, that I KNOW work great. Maybe I can find something there.

I have my ego helping at idle alittle too. It was the easiest way to smooth out the idle. For me on the ve table there was a big difference between 53 and 54. Im not talking about the ve table under constants but the ve table under tuning. When I changed the constance ve table to get the number to change on the tunning ve the afr at idle went from like 13.6:1 to 15:1. So I used the ego to help alittle. Thats what it is for right?

Originally posted by dougie2
Why is your EGO correction adding fuel at idle? Maybe you simply have VE values that are too low. Matt- maybe you could send me a copy of your MSQ file and I can compare it to a lot of MSQs that I have, that I KNOW work great. Maybe I can find something there.

Well let me explain in more detail i guess. If i start it up from a cold soak, it warms up fine, and once it reaches my set temp of 160 degrees, i have ego running closed loop at idle. The stock ECU runs closed loop idle at stoich, so i should be able to do the same with the MS.. Well the problem is if i do a hot restart, it runs really lean, with lean misfires, and it runs pretty bad..(maybe 16-18.0:1 A/f) if i didnt have ego correction at idle, it would always run lean...it runs fine after the ego kicks in to richen up the mixture. This is why i kinda have to run ego closed loop at idle.

Its just strange that a cold start to fully warm, it runs great, adn my closed loop idle correction is maybe damn near perfect, at +/- 5% on the ego gauge. Even today at a car meet, i let the car sit for maybe an hour or so, so it had to go into warmup enrich, coolant gauge was just a little bit above the full cold mark. Well the same thing happened, it ran lean on the re start, and even LEAN during the warmup period, UNTIL ego correction kicked on to richen it up to normal A/F's. When this happens, the ego correction isnt at +/- 5% anymore like it is on a cold soak start, its more around 108-114%! Its almost like the baro correction is kicking in for some reason, and the entire map is much leaner(baro correction is turned OFF) All my A/F's, fully hot from a cold soak start, are PERFECT! lol i have no clue whats going on..Its done this with every code version i've tried. I'm currently using the 024s13c firmware, and it seems really good so far besides this one issue. I just dont get it.. I've searched a lot on msefi boards, and seems like many are having this issue as well..

Doug, i'll get an msq to you to see if you see anything odd...

Originally posted by XtremeRS
I'm currently using the 024s13c firmware

I've though about downgrading to that. The new alpha versions have been quite a bit more unstable than that one.

Unless it's only in the alpha versions, the "hot" afterstart settings are under afterstart enrichments, the table goes from -40 to +160

My ase table only goes to 130. How can I change this?

Update to the newest alpha code. I have 029k running on the bench (car is down) with MT 756.

The other cool thing about the newer alpha releases is that they support real-time barometric correction - you need a second MAP sensor identical to the one hooked up to the manifold, though. It works on the bench but I haven't tried it on the car yet. I might end up putting a "large" (0.1-1.0 uF) capacitor across the second MAP sensor output to average the output more. It seems to fluctuate around 99-100 or 100-101 kPa very quickly at times, where as actual barometric pressure would take on the order of a minute or more to change that much depending on weather conditions, vertical speed etc... It would be nice if there was a code change to accept a different type of MAP sensor, like a 1 BAR or something which would have better accuracy and precision.

As far as the barometer reading Matt, everytime you turn the key on, the MS take a barometer reading. You cant stop that. To my knowledge the barometeric correction makes the MS periodically take a barometer reading. This would help someoen who lives in ahilly region.

For me the hot start lean issue only lasts for a short duration, it isnt constant. I do notice after the car has been driven for a while, it continues to lean out. From a cold start my idle AFR in 13's. Once its good and hot and driven for a while it drops to the 14.8.

I havent had much chance to really go after that issue yet but i feel confident the issue is in the VE table.

Terry

Originally posted by Star Turbo Talon
As far as the barometer reading Matt, everytime you turn the key on, the MS take a barometer reading.

I remember reading somewhere that it takes the reading once when you first power up the unit and it keeps that value until you restart it again. Makes sense to me, because if it took the reading perodically how would it know when is a good time to take the reading? The RPM's would have to be 0. I can't prove it though, I'm too lazy to search for the link ;)

It only takes baro readings during resets, and power on, not when its running..

Originally posted by XtremeRS
It only takes baro readings during resets, and power on, not when its running..

True. I was just reading up in the manual. It states however that you could run a separate baro meter and have the ms check it periodically.

It also said that the ego correction starts after 30 seconds of the car being on and that it will +/- fuel after every 8 ignition events. That's probably why it takes a minute for it to stabalize. Hope that helps.

edit: reading a little further into it, can you change the 'Ignition events per step' to a lesser number to help cure the problem faster? Instead of 8, could you make it 4 or something?

Originally posted by WIDECLIPSE

Originally posted by XtremeRS It only takes baro readings during resets, and power on, not when its running..

True. I was just reading up in the manual. It states however that you could run a separate baro meter and have the ms check it periodically. It also said that the ego correction starts after 30 seconds of the car being on and that it will +/- fuel after every 8 ignition events. That's probably why it takes a minute for it to stabalize. Hope that helps. edit: reading a little further into it, can you change the 'Ignition events per step' to a lesser number to help cure the problem faster? Instead of 8, could you make it 4 or something?

Yep, you can change the ignition events per step..go to "fuel tables1+3"
at the top of the MT screen, and then go to "exhaust gas settings". You will see the step size box there. I run 16 igntion events per step, it seems to do the best around there. You run too many events like 8 per step, it can cause the ego correction to be too erratic. Too little events per step like 36 for example,and it will react too slow. Unless you are in a base tune, use 1% step size.

Originally posted by WIDECLIPSE
edit: reading a little further into it, can you change the 'Ignition events per step' to a lesser number to help cure the problem faster? Instead of 8, could you make it 4 or something?

There's a limit on how low you can pratically go with that number. I've found that with my setup anything much less than 10 or so is unstable. You don't want to go too high either or else the time between updates at idle will get rather large. I just checked my MSQ the number I am currently using is 25.

I may have stumbled onto something that will help this problem. I have a little time this weekend to mess with my ms somemore and I found something intresting. I was told that the asd relay was a god way to power my ms. It works but I have always had a choppy idle. For shits and grins I powered it off the the battery today and the car idles much better. Also I dont have the hot start issue. I have started and stopped my car 5 times with out having the lean idle. A much more intresting thing is that my afr changes between the two power sources. It runs much leaner off the asd relay. I would guess that the power source is not enought to control the ms and injectors.

Anybody else try this? I have be playing all day and man the car idles better then ever. I hooked up my ms to a toggle switch with power direct from the battery. No more choppy idle and none of that lean hot start either. This is the first time I have been able to get the map down to a consistant 36kpa at idle too. I love this stuff.

Yes, this is a known issue. Using the ASD relay will cause the problems you were having as well as ignition misfires and other quirks. Never use the ASD relay for MS power. Use a simple 12V 30-amp SPST automotive relay to power your MegaSquirt.

I'm using my ignition wire from the ignition switch to power a relay that turns the MS on and off...the power to the relay and to the MS comes from the batter, a redtop Optima. I also use a noise filter cap on the power supply line. I played around with it more today, and its still doing it. From an overnight cold soak, it warms up to op temp, and it runs great, ego correction is +/- 5%. Shut the car off for a few minutes, and it runs really lean until o2 correction kicks in, then the ego correction gauge is reading +8 to +15%! i dont get it..I even swaped in a new processor, same thing. Other than that, its running great :shrug

Well i think i've figured this out for the most part. Other MS users have had lean start issues, and i've figured out that the air density correction algorithm was not adding enough fuel for rising intake air temps. It was pretty clear just by watching my intake temps rise, and then watching the o2 correction add in more fuel as air got hotter, esp when going above 95 degrees, and really bad when in the low 100's.

The new alpha test code i'm using is 029o, which has air density correction table in it. YOu can choose either coolant correction, or intake air temp correction. It seemed to work quite well so far, and i'm only really using the correction at idle and slightly above, which is when intake air temps can get hotter with the slow air flow, and a large intercooler. When intake temps were around 150 on a hot restart, i needed approx 30% more fuel to keep the o2 correction close to 100%.

So there it is, if anyone has this issue, just try out the new alpha code and see what happens. The car idled the best so far with this code version as well. Very smooth.

One more issue that i figured out was the hunting idle, and roughness, was my PCV valve was sometimes letting air by, and sometimes not, which was messing with the idle speed, and fluctuating the MAP signal, which was messing with the smoothness of the idle. So i just took that out, capped off the intake port, and ran a line from the fitting on the valve cover to my catch can. It's working great so far.