When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Car dying at stop lights and stop signs and exhaust question
Hey everyone - a bit of a catch all post for a couple questions I have.
My car has been street tuned a few times and then the tune has been reviewed by another reputable tuner; however, I have an issue with the engine dying as I clutch in to come up to the stop lights. It is very intermittent and mostly only shows up after the car is nice and hot. I've had the remote tuner look and another tuner evaluate my tune and both of them said it's your lightened flywheel that's the problem go back to a stock one. I really dont want to drop the trans again, so I thought I would ask if anyone else has experience with tuning on a lightened flywheel and if this is just something I have to live with or if I do need to spend several hours going back to a stock weight flywheel.
I had a custom 3" exhaust installed after I put on my turbo stuff. Its a little loud but mostly I have an issue with what is like a resonant vibration/drone. I asked the exhaust guy to see what he can do about it. He suggested adding back a highflow cat which I'm having him do. I'm curious to anyone's thoughts on the logic of, 'even if I put something on the car which makes it less efficient (i.e. a cat), I can turn up the boost to counteract that because the real limiting factor on a stock block setup is the rods at roughly 250lb-ft. So if for instance I am making that 250 at 10psi, then I add an exhaust restriction of cat and it drops it to 240lb-ft, if I turn it up to 11psi, its really same same? Hopefully that question makes sense. It logics in my head but I thought I'd ask guys who have been doing this a lot longer than I have.
The method I just used for setting/tuning my idle after replacing a bad idle air control valve is as follows (I don’t know if this is the 1000% correct way, but it works and is logical to me). It is also worth mentioning that I have a Skunk2 throttle body and stock flywheel.
1. Start the car, let it get up to temp (oil and water temp both just to be safe), raise the idle using the control screw on the throttle body or the throttle stop screw to the point to where you can unplug the IAC valve and have the car not die. Don’t have any accessories going if possible. No headlights, radio, etc.
2. At this point adjust the idle until your RPMs are at the lowest you should ever want them to be. Maybe even 50-100 RPM or so under to account for possible alternator/cooling fan load. For me I stopped at around 900rpms.
3. Reconnect your idle air control valve, and set up the initial values target table or whatever it is called for closed loop idle. Have a point at your target rpm for fully warm idle, another point for your fully cold idle, and then a few points in between (for me I target 1050rpm warm idle because it feels smoother with my motor mounts/slightly better throttle response and 1200 cold). Add enough duty cycle to where it actually overshoots your target rpm slightly by around 100 rpm’s give or take.
4. Set up your closed loop idle settings to correct that small overshoot over time. I can’t recall my exact settings off of the top of my head, but I want to say I just through in random PID values and it worked sufficiently well. Something like 0-50-8 iirc was good enough for me to not bother with it anymore. It could be faster but I am just happy my IAC is fixed for the time being.
If done correctly you’ll get a small overshoot when returning to idle before closed loop control activates and narrows in on the true target rpm. This is dependent on having the other closed loop activation settings configured correctly. I’m also on a MS2PNP unit so your set up may vary.
I can’t comment too much on keeping the stock block alive, I’ve got a bad history with them, but your logic is sound to me for the most part. Only thing I might point out is that while torque does bend rods, I believe the underlying culprit is more peak cylinder pressure.
For example if you were somehow able to make the same torque two different ways (on average throughout the cranks rotation, not instantaneous since that would be impossible) but one method did it with a sudden spike of pressure vs one that has a lower peak pressure over a longer duration, the longer duration option would be safer. How this plays in with the lower VE that will result from adding a catalytic converter, that’s where I’m not sure. Also 250ftlbs of torque at the wheels is very much in the flirting with danger territory. Me personally, I wouldn’t stray much above 220ftlb at the wheels if I wanted to last without a dyno, 240lbft at most depending on the accuracy of the dyno. Not saying it can’t be done, but again personally, I’m not testing my luck after blowing up 2 stock engines.
The method I just used for setting/tuning my idle after replacing a bad idle air control valve is as follows (I don’t know if this is the 1000% correct way, but it works and is logical to me). It is also worth mentioning that I have a Skunk2 throttle body and stock flywheel.
1. Start the car, let it get up to temp (oil and water temp both just to be safe), raise the idle using the control screw on the throttle body or the throttle stop screw to the point to where you can unplug the IAC valve and have the car not die. Don’t have any accessories going if possible. No headlights, radio, etc.
2. At this point adjust the idle until your RPMs are at the lowest you should ever want them to be. Maybe even 50-100 RPM or so under to account for possible alternator/cooling fan load. For me I stopped at around 900rpms.
3. Reconnect your idle air control valve, and set up the initial values target table or whatever it is called for closed loop idle. Have a point at your target rpm for fully warm idle, another point for your fully cold idle, and then a few points in between (for me I target 1050rpm warm idle because it feels smoother with my motor mounts/slightly better throttle response and 1200 cold). Add enough duty cycle to where it actually overshoots your target rpm slightly by around 100 rpm’s give or take.
4. Set up your closed loop idle settings to correct that small overshoot over time. I can’t recall my exact settings off of the top of my head, but I want to say I just through in random PID values and it worked sufficiently well. Something like 0-50-8 iirc was good enough for me to not bother with it anymore. It could be faster but I am just happy my IAC is fixed for the time being.
If done correctly you’ll get a small overshoot when returning to idle before closed loop control activates and narrows in on the true target rpm. This is dependent on having the other closed loop activation settings configured correctly. I’m also on a MS2PNP unit so your set up may vary.
I can’t comment too much on keeping the stock block alive, I’ve got a bad history with them, but your logic is sound to me for the most part. Only thing I might point out is that while torque does bend rods, I believe the underlying culprit is more peak cylinder pressure.
For example if you were somehow able to make the same torque two different ways (on average throughout the cranks rotation, not instantaneous since that would be impossible) but one method did it with a sudden spike of pressure vs one that has a lower peak pressure over a longer duration, the longer duration option would be safer. How this plays in with the lower VE that will result from adding a catalytic converter, that’s where I’m not sure. Also 250ftlbs of torque at the wheels is very much in the flirting with danger territory. Me personally, I wouldn’t stray much above 220ftlb at the wheels if I wanted to last without a dyno, 240lbft at most depending on the accuracy of the dyno. Not saying it can’t be done, but again personally, I’m not testing my luck after blowing up 2 stock engines.
I guess that is true, if it is peak cylinder pressure that determines bent rods and you need more cylinder to return to the same tq value, then you are still technically running closer to the limit/failure point by doing what i did. This is why I asked! Thanks. I'll see about the idle stuff, but I should clarify, its not that it won't idle, its that as the RPMS fall, it simply will dump all the way to 0. I could just let it idle for hours without issue.
The air bypass should be set so that the ecu can adjust the idle down if it is high. I would set the static idle to 150-200rpm below the normal warm idle. The idle initial values for the hotter coolant temps needs to be increased and maybe the dashpot adder. Increase the initial values first and do not use the values in the pic. Increase your values by a few tenths at a time.
Slightly off-topic, I never understood what the vertical axis was on the CL Initial Target was for. At some point, mine was set to all of the same values and the description on the left axis didn't make sense so I never bothered with it. I feel like such an idiot now for missing that. Now I get to fix my slightly lowered initial idle when first warming up
Also OP, I thought about this on my way home, but check and make sure your over-run fuel cut upper and lower thresholds aren't set too low. I'm not sure what the general consensus is, but I have my upper limit set to 2200 RPM and the lower limit set to 2000 (my understanding is that my values might be slightly higher than necessary). If these are set too low (ie 1200 UL and 1000 LL), the engine might not have time to catch itself, if that makes sense.
Edit: We understand that the car idles fine once stable and only dies when the engine's RPM is decreasing. The issue is likely that the engine isn't getting enough air in time. Raising the minimum threshold by messing with the throttle screw raises the minimum amount of air the engine will be able to intake. Raising the initial duty in the target table gives the car more air when it first starts to attempt to idle before the closed loop settings come into play to correct any error. The only other thing that hasn't already been mentioned is the engine states which only apply to MS3 if activated, if I understand correctly. I have only ever tuned MS2-equipped miatas, so I am not of any use there.
Also, if you could send a tune file or datalog we would likely have a lot more to work with here rather than just speculating. I'd be happy to take a look for you.
In addition to the advice already given, notice in LeoNA’s settings that valve close delay is set to 0. This is very helpful for a lightweight flywheel. Without seeing your current tune my initial guesses would be that your initial table is too low, or that your valve is set to open at too low an RPM. Setting the delay to 0 (keeping the valve open all the time) would be my first test.
07-10-2024, 10:59 AM
Car dying at stop lights and stop signs and exhaust question
0
