Warm restart and idle issues
02-19-2016, 07:24 PM
#1
Junior Member
Thread Starter
Join Date: Sep 2012
Location: Redmond, WA
Posts: 139
Total Cats: 2
Warm restart and idle issues
I'm starting a new thread as a continuation of my questions from https://www.miataturbo.net/megasquir...s-79819/page3/ because it appears my issues are moving in a different direction and I don't want to derail that thread.
To summarize, I previously wrote:
Then Braineack wrote back:
And I wrote...
I poked around a bit today and have some more to share. I'll put that in the next post.
To summarize, I previously wrote:
I'm having the same lean AFRs on hot restart that the rest of you are describing.
I can see the WB O2 come online and then the AFRs hang out around 17 until they slowly converge to the expected value around stoich.
When I started troubleshooting this, it seemed reasonable to guess that the problem was caused by ASE not running long enough. I set my ASR taper the same as DMakinson suggested and the rest of my settings were pretty close already. I'm still seeing the problem even with the extended ASE times though.
Is this something we just have to live with because of the heat soak on the IAT sensor? In my case, MAT doesn't look that high in the beginning so I wouldn't expect the MS to lean things out because it's excessively hot. Also, I'm using the GM sensor that Rev sent along with my MS3 and that seems to be the one people like.
I'm attaching a small log file in case it'll help. Many thanks in advance...
I can see the WB O2 come online and then the AFRs hang out around 17 until they slowly converge to the expected value around stoich.
When I started troubleshooting this, it seemed reasonable to guess that the problem was caused by ASE not running long enough. I set my ASR taper the same as DMakinson suggested and the rest of my settings were pretty close already. I'm still seeing the problem even with the extended ASE times though.
Is this something we just have to live with because of the heat soak on the IAT sensor? In my case, MAT doesn't look that high in the beginning so I wouldn't expect the MS to lean things out because it's excessively hot. Also, I'm using the GM sensor that Rev sent along with my MS3 and that seems to be the one people like.
I'm attaching a small log file in case it'll help. Many thanks in advance...
your AIT is not heatsoaked, and your engine is already warmed.
your fueling simplly does not change based on the engine demands.
As you idle and it drops from 17AFR to ~14AFR but the PW remains roughly the same -- in the low 2.0ms range. Youre idling at the same VE table % 57.7 throughout the log and you only have about 5% fuel enrichments from warmup, which as they reduce you actually richen.
So at 1,400-1,300RPM, after the car starts, you're asking for the exact same amount of fuel as 850RPM later in the log.
and why is your spark advance so high? is there any reason why you're idling at more spark advance than the car demands at WOT? Something is telling the MS to add 5° of spark above the already ridiculously high spark table number.
why does your log show 100%TPS during cranking?
why is your battery voltage so low?
I don't believe ASE is your issue. You dont really need it when your car is already warmed up. and it only should last 2-3 seconds.
You richen up as your RPM better matches the fuel that your injecting. So either lower your cranking idle table and targets so the idle duty starts running the car at 850rpm or tune your fuel map so you actually have the fuel required when idling 1200-1400RPM. and then fix your spark while youre at it before you melt your pistons.
your fueling simplly does not change based on the engine demands.
As you idle and it drops from 17AFR to ~14AFR but the PW remains roughly the same -- in the low 2.0ms range. Youre idling at the same VE table % 57.7 throughout the log and you only have about 5% fuel enrichments from warmup, which as they reduce you actually richen.
So at 1,400-1,300RPM, after the car starts, you're asking for the exact same amount of fuel as 850RPM later in the log.
and why is your spark advance so high? is there any reason why you're idling at more spark advance than the car demands at WOT? Something is telling the MS to add 5° of spark above the already ridiculously high spark table number.
why does your log show 100%TPS during cranking?
why is your battery voltage so low?
I don't believe ASE is your issue. You dont really need it when your car is already warmed up. and it only should last 2-3 seconds.
You richen up as your RPM better matches the fuel that your injecting. So either lower your cranking idle table and targets so the idle duty starts running the car at 850rpm or tune your fuel map so you actually have the fuel required when idling 1200-1400RPM. and then fix your spark while youre at it before you melt your pistons.
And I wrote...
Thanks, Brain. It looks like I've got some trouble in my tune that's goes beyond the fuel temp issues others are seeing. If we need to, I'll start a new thread to drill into my specific issues.
In the meantime, I can't comment on your PW and spark advance questions because I didn't tune those. I've sent an email to my tuner asking him to jump in with those answers. I can answer some of the questions though.
I started and stopped the engine a few times and in this particular case the engine cranked but didn't want to start. I hit the throttle to put the MS in flood clear mode and the engine started.
Not sure about this one. The battery is pretty new. Voltage may have been a little low because of the multiple starts. Voltage looks OK to me once the engine is running.
Glad to hear that. I was trying to match the settings suggested on this thread but the numbers sure seemed strange. Maybe we'll fix the other issues and then dial in the ASE.
I'll let my tuner, Nate, comment on this. Totally agree that I don't want the mixture to be as lean as we're seeing.
In the meantime, I can't comment on your PW and spark advance questions because I didn't tune those. I've sent an email to my tuner asking him to jump in with those answers. I can answer some of the questions though.
I started and stopped the engine a few times and in this particular case the engine cranked but didn't want to start. I hit the throttle to put the MS in flood clear mode and the engine started.
Not sure about this one. The battery is pretty new. Voltage may have been a little low because of the multiple starts. Voltage looks OK to me once the engine is running.
You richen up as your RPM better matches the fuel that your injecting. So either lower your cranking idle table and targets so the idle duty starts running the car at 850rpm or tune your fuel map so you actually have the fuel required when idling 1200-1400RPM. and then fix your spark while youre at it before you melt your pistons.
Reply
0
0
02-19-2016, 08:11 PM
#2
Junior Member
Thread Starter
Join Date: Sep 2012
Location: Redmond, WA
Posts: 139
Total Cats: 2
So I looked through my spark table first because of the feedback about the massive amount of advance. First, my tune is configured with a 6* offset, which I understand to mean that I should remove 6 from any of the timing numbers I see in the log and my spark isn't as advanced as it appears.
Also, the numbers in my ignition table are more advanced than in base maps I found online, but wouldn't I expect that if the table had been tuned on a dyno? My tuner told me previously that they played with the timing numbers until they faintly detected pre-ignition and then backed off from there. So how much advance is too much?
Next I worked on the mixture in the fuel map to verify it's correct at 1200-1400rpm as suggested. I checked AFRs in that area and hand tuned my Idle VE to be ~14.7.
Then I noticed in the previous log that the AFRs were way lean for a while after the EGO came online and then converged to ~14.7 after a few seconds. I poked around and found that EGO Delay After Start was set to about 30 seconds, which was way longer than it took for the O2 sensor to come online. I dropped it down to 5 seconds and specified elsewhere that any reading below 8.0 should be ignored, which covers me before the sensor comes online. I figured this would cause the MS to pay attention to the EGO input as soon as it was available. It looks like this has improved the situation.
I'm attaching the latest log file so you can take a look. Interestingly, the engine is good and hot now and it appears the injectors are handling it OK. If this all looks good to you then I've got another question to ask.
Also, the numbers in my ignition table are more advanced than in base maps I found online, but wouldn't I expect that if the table had been tuned on a dyno? My tuner told me previously that they played with the timing numbers until they faintly detected pre-ignition and then backed off from there. So how much advance is too much?
Next I worked on the mixture in the fuel map to verify it's correct at 1200-1400rpm as suggested. I checked AFRs in that area and hand tuned my Idle VE to be ~14.7.
Then I noticed in the previous log that the AFRs were way lean for a while after the EGO came online and then converged to ~14.7 after a few seconds. I poked around and found that EGO Delay After Start was set to about 30 seconds, which was way longer than it took for the O2 sensor to come online. I dropped it down to 5 seconds and specified elsewhere that any reading below 8.0 should be ignored, which covers me before the sensor comes online. I figured this would cause the MS to pay attention to the EGO input as soon as it was available. It looks like this has improved the situation.
I'm attaching the latest log file so you can take a look. Interestingly, the engine is good and hot now and it appears the injectors are handling it OK. If this all looks good to you then I've got another question to ask.
Reply
0
0
02-19-2016, 08:22 PM
#3
Cpt. Slow
iTrader: (25)
Join Date: Oct 2005
Location: Oregon City, OR
Posts: 14,189
Total Cats: 1,135
Incorrect. The 6* makes the log and timing tables correct for what your engine is actually getting.
Also your tuner is tuning WOT spark and we're discussing idle spark.
No offense but it's pretty clear you're new to MS and TS. Which is great! Everyone starts here. I'd read around as much as you can before making any assumptions or changes toyour setup.
Also your tuner is tuning WOT spark and we're discussing idle spark.
No offense but it's pretty clear you're new to MS and TS. Which is great! Everyone starts here. I'd read around as much as you can before making any assumptions or changes toyour setup.
Last edited by curly; 02-19-2016 at 08:37 PM.
Reply
0
0
02-19-2016, 08:27 PM
#4
Junior Member
Thread Starter
Join Date: Sep 2012
Location: Redmond, WA
Posts: 139
Total Cats: 2
Hmmm...so my timing really is way more advanced than in the base maps. I'll ask my tuner about that.
Still, shouldn't a dyno tune have dialed in the ignition properly? That's what I thought I was paying for.
I'm attaching my ignition table for your comments.
Still, shouldn't a dyno tune have dialed in the ignition properly? That's what I thought I was paying for.
I'm attaching my ignition table for your comments.
Reply
0
0
02-19-2016, 08:31 PM
#5
mkturbo.com
iTrader: (24)
Join Date: May 2006
Location: Charleston SC
Posts: 15,177
Total Cats: 1,681
Yes a good tuner would have tuned the spark table all over. From the looks of it, I don't think your tuner is very good.
Reply
0
0
02-19-2016, 08:40 PM
#6
Junior Member
Thread Starter
Join Date: Sep 2012
Location: Redmond, WA
Posts: 139
Total Cats: 2
Well, thanks for the honest feedback. That's the only way for me to know whether things are running as they should be. Can you point me to an ignition table that's closer to what I'm looking for?
Also, I figure I can't just put in new values in the table, right? Wouldn't I need to retune all kinds of stuff, including the VE table and WUE?
Also, I figure I can't just put in new values in the table, right? Wouldn't I need to retune all kinds of stuff, including the VE table and WUE?
Reply
0
0
02-19-2016, 08:42 PM
#7
Cpt. Slow
iTrader: (25)
Join Date: Oct 2005
Location: Oregon City, OR
Posts: 14,189
Total Cats: 1,135
quick test: go to the 3D view of your timing table. Your WOT cell (this is the 100kpa row with N/A engines), should have a similar shape as your dyno plot. So according to that timing table, you have a perfectly linear power curve, no loss in power until redline.
Most Miatas turbo or not don't do this. Unless you're supercharged.
Most Miatas turbo or not don't do this. Unless you're supercharged.
Reply
0
0
02-19-2016, 08:44 PM
#8
Cpt. Slow
iTrader: (25)
Join Date: Oct 2005
Location: Oregon City, OR
Posts: 14,189
Total Cats: 1,135
Yeah don't touch your tune. Especially if someone else tuned it. If there's any difference and it blows up, you'd be liable. They did a very basic power tune my guess so hopefully you didn't pay more than ~$350. that should be ~$100/hr for the dyno, take 1-2 hours, and a little extra for their time.
I'm not saying it's bad or unsafe, but with your experience I wouldn't change it much before we or someone local helps out a little.
I'm not saying it's bad or unsafe, but with your experience I wouldn't change it much before we or someone local helps out a little.
Reply
0
0
02-19-2016, 08:52 PM
#9
Junior Member
Join Date: Jan 2016
Location: Redmond, WA
Posts: 48
Total Cats: 0
hey guys, i work at the shop that tuned the car, im the one that installed the setup, and then started fixing issues in the tune after it initially left, im here to help so any and all feedback is appreciated, it will help me learn as well, im not the actual tuner at the shop but i also have a megasquirt miata (n/a) and really want to help get Ace car running properly and make him happy
Reply
0
0
02-19-2016, 08:58 PM
#10
Cpt. Slow
iTrader: (25)
Join Date: Oct 2005
Location: Oregon City, OR
Posts: 14,189
Total Cats: 1,135
Oops and here we are all bashing you.
The timing tables just Raise a few eyebrows, especially that jump from 33 to 19, and why is it at 33 at all? I thought DIY's MSM base map was high at 25. I personally do 20. The stock 10 seems a little low the more I tune idle.
The timing tables just Raise a few eyebrows, especially that jump from 33 to 19, and why is it at 33 at all? I thought DIY's MSM base map was high at 25. I personally do 20. The stock 10 seems a little low the more I tune idle.
Reply
0
0
02-19-2016, 08:58 PM
#11
Junior Member
Thread Starter
Join Date: Sep 2012
Location: Redmond, WA
Posts: 139
Total Cats: 2
Thanks for the pointer. Yeah, I wasn't planning to play with the timing table unless I got specific direction.
My tuner spent a lot more than 3 hours on the tune, especially since I know they've run through it twice. I can't explain what we're seeing at this point but I appreciate everybody's help in getting to the bottom of it.
I'm attaching a 3D view of my ignition table for your review.
My tuner spent a lot more than 3 hours on the tune, especially since I know they've run through it twice. I can't explain what we're seeing at this point but I appreciate everybody's help in getting to the bottom of it.
I'm attaching a 3D view of my ignition table for your review.
Reply
0
0
02-19-2016, 09:04 PM
#12
Cpt. Slow
iTrader: (25)
Join Date: Oct 2005
Location: Oregon City, OR
Posts: 14,189
Total Cats: 1,135
It's just a linear table, which is a bit of a novice move.
As you change timing, let's use advance for this example, you'll see some rpms gain power and others stay the same. Add 2* across the board and see only gains between 3 and 5000? 0-3 and 5-redline should probably revert back to the old timing numbers while 3-5 keeps the new timing.
Keep doing this over and over again and you'll get a fairly typical Miata dyno plot.
As you change timing, let's use advance for this example, you'll see some rpms gain power and others stay the same. Add 2* across the board and see only gains between 3 and 5000? 0-3 and 5-redline should probably revert back to the old timing numbers while 3-5 keeps the new timing.
Keep doing this over and over again and you'll get a fairly typical Miata dyno plot.
Reply
0
0
02-19-2016, 10:16 PM
#13
Junior Member
Join Date: Jan 2016
Location: Redmond, WA
Posts: 48
Total Cats: 0
timing is something im not 100% with yet so im not surprised i made a novice move on that one, was trying to stay safe and get MPG out of it as Ace was already happy with the power output of the car. didnt want to do anything crazy. so looking at his map what advice would you give me for adjustments that will help with his warm restart issues?
Reply
0
0
02-20-2016, 10:55 AM
#14
Boost Czar
iTrader: (62)
Join Date: May 2005
Location: Chantilly, VA
Posts: 79,493
Total Cats: 4,080
at 100kPa you have about ~13° more advance than the stock ECU.
in boost, you've never retarded past the OE spark timing. A good rule of thumb is 1° retard for every 1-2psi in boost...
just do the math here...
I would literally reduce your entire timing map by 10°. It would still be aggressive, but it would be within reason.
when I add in 10° to the timing map I use, your current is still more aggressive:
I'm seriously shocked your motor hasnt blown yet.
please just post your entire MSQ.
Last edited by Braineack; 02-20-2016 at 11:09 AM.
Reply
0
0
02-20-2016, 11:11 AM
#15
Cpt. Slow
iTrader: (25)
Join Date: Oct 2005
Location: Oregon City, OR
Posts: 14,189
Total Cats: 1,135
Ooh yeah i follow that rule of thumb too. If you set your timing tables with 2 psi deltas (14ish kpa) you can get your timing curve at waste gate safely and then have a good idea for max curve.
Last edited by curly; 02-20-2016 at 11:24 AM.
Reply
0
0
02-20-2016, 12:06 PM
#17
Junior Member
Thread Starter
Join Date: Sep 2012
Location: Redmond, WA
Posts: 139
Total Cats: 2
Thanks for the feedback, guys. Let's assume for the sake of argument that my tuner isn't trying to blow up my car and that we have a difference of understanding about what an ignition table should look like. I'm new to the Megasquirt so I'm clearly learning how it works. But I've got some background with the Link so I'm happy to share that I don't feel *totally* clueless with this stuff. As you can imagine, I feel stuck in the middle here and I'm looking for truth. I really appreciate everybody's help in getting there.
So I've gone ahead and uploaded my tune. I'm looking forward to your feedback.
Meanwhile, it preliminarily sounds like I should back off the timing across the board. Let's say we collectively decide that's the right thing to do as a first step. What other changes would we need to make to the tune? I figure I'd need to retune the VE table at least, right? How extensive would any other changes be?
Regarding shuiend's suggestion:
My understanding is that you can't just upgrade firmware on a Megasquirt like on an iPhone because the tune doesn't come over. You have to apply the new firmware version and then manually copy over the configuration. Is that correct? Would somebody please point me to a process for upgrading the firmware?
So I've gone ahead and uploaded my tune. I'm looking forward to your feedback.
Meanwhile, it preliminarily sounds like I should back off the timing across the board. Let's say we collectively decide that's the right thing to do as a first step. What other changes would we need to make to the tune? I figure I'd need to retune the VE table at least, right? How extensive would any other changes be?
Regarding shuiend's suggestion:
My understanding is that you can't just upgrade firmware on a Megasquirt like on an iPhone because the tune doesn't come over. You have to apply the new firmware version and then manually copy over the configuration. Is that correct? Would somebody please point me to a process for upgrading the firmware?
Reply
0
0
02-20-2016, 12:17 PM
#18
Cpt. Slow
iTrader: (25)
Join Date: Oct 2005
Location: Oregon City, OR
Posts: 14,189
Total Cats: 1,135
Nope it copies over with few errors. Back in the day it was stuff like "tune has 1.0, firmware has 1.00", but it's a little more complicated now. I think it works with just one round of I'm-freaking-out-bro warnings.
I'd load a base map timing table and double check sync, then retune for max power using a slightly different method than before.
I'd load a base map timing table and double check sync, then retune for max power using a slightly different method than before.
Reply
0
0
02-20-2016, 12:18 PM
#19
Boost Czar
iTrader: (62)
Join Date: May 2005
Location: Chantilly, VA
Posts: 79,493
Total Cats: 4,080
I dont like this:
you're basically adding 8°-5° of advance as your motor warms up. that's just odd, you should only be using this below say 60°F.
spark table 3 looks pretty good, one does not.
looks like they setup table switching and why you have fuel and spark tables 3 activated.
do you even use this?
what type of injectors? i see you have larger ones, but what brand?
I dont like that deadtime curve compared to what I've seen from different manufacturers. seems very conservative and could lead to underfueling.
ASE and warmup enrichments are very very low:
you can autotune warmup if youre lazy.
priming table is crazy long, ever notice how long your fuel pump runs when you put the key to ON?
your fueling algorithm includes AFR, I'm not the biggest fan of your AFR map
your VE table actually looks pretty standard, like it was never tuned for that AFR table correctly imho. goes with what i was saying before about not enough fuel in certain spots, that coupled with the ****-poor warmup and ase tables isnt helping.
you need to copy your spark 3 values into your spark 1 table PRONTO.
you're basically adding 8°-5° of advance as your motor warms up. that's just odd, you should only be using this below say 60°F.
spark table 3 looks pretty good, one does not.
looks like they setup table switching and why you have fuel and spark tables 3 activated.
do you even use this?
what type of injectors? i see you have larger ones, but what brand?
I dont like that deadtime curve compared to what I've seen from different manufacturers. seems very conservative and could lead to underfueling.
ASE and warmup enrichments are very very low:
you can autotune warmup if youre lazy.
priming table is crazy long, ever notice how long your fuel pump runs when you put the key to ON?
your fueling algorithm includes AFR, I'm not the biggest fan of your AFR map
your VE table actually looks pretty standard, like it was never tuned for that AFR table correctly imho. goes with what i was saying before about not enough fuel in certain spots, that coupled with the ****-poor warmup and ase tables isnt helping.
you need to copy your spark 3 values into your spark 1 table PRONTO.
Reply
0
0
02-20-2016, 12:35 PM
#20
Cpt. Slow
iTrader: (25)
Join Date: Oct 2005
Location: Oregon City, OR
Posts: 14,189
Total Cats: 1,135
Wow.
Like wtf?
I don't like cold Advance and wue at standard operating temps, but that's just me.
I also put that 210 cold advance to -10 starting at 205-207, so if it starts to overheat it reduces power a bunch. I do that for wue fuel too.
But that's besides the point. Braineack brings up some good questions.
Like wtf?
I don't like cold Advance and wue at standard operating temps, but that's just me.
I also put that 210 cold advance to -10 starting at 205-207, so if it starts to overheat it reduces power a bunch. I do that for wue fuel too.
But that's besides the point. Braineack brings up some good questions.
Reply
0
0
