STR Prep - ECU and Tuning Discusson
09-21-2012, 06:16 PM
#622
Update:
Very happy to report that I've now been CEL free for close to 3 weeks while running my map that forces open loop early. The difference being I raised(with Brian Karwan's guidance) when open loop is forced to around 25%TPS instead of in the single digit percentage as previously set to, which was pretty close to idle.
The key to eliminating the codes for us was raising when we forced open loop in the RPM range, we can probably force open loop at a lower TPS percentage without producing a CEL but we were noticing a small annoying kick when it switched from closed to open loop when set between say 10% - 20% TPS but not when set pretty low, close to idle. I still want to play around with lowering it a bit more and seeing where this kick isn't distracting during an autox run but for now it stays at ~25% as I am SUPER happy not having to worry about a code being thrown at an event or when driving around town. Besides, during a run we're mostly above 25%TPS anyway, so it really doesn't matter much AND when I am driving it to work and just cruising I can pretty much cruise in closed loop
.
Huge thanks to Brian Karwan @ Karcept's Inc for putting in crazy long hours, testing and research into figuring this all out. The tune feels great and I no longer worry about a CEL popping up.
Now back to suspension tuning...
Very happy to report that I've now been CEL free for close to 3 weeks while running my map that forces open loop early. The difference being I raised(with Brian Karwan's guidance) when open loop is forced to around 25%TPS instead of in the single digit percentage as previously set to, which was pretty close to idle.
The key to eliminating the codes for us was raising when we forced open loop in the RPM range, we can probably force open loop at a lower TPS percentage without producing a CEL but we were noticing a small annoying kick when it switched from closed to open loop when set between say 10% - 20% TPS but not when set pretty low, close to idle. I still want to play around with lowering it a bit more and seeing where this kick isn't distracting during an autox run but for now it stays at ~25% as I am SUPER happy not having to worry about a code being thrown at an event or when driving around town. Besides, during a run we're mostly above 25%TPS anyway, so it really doesn't matter much AND when I am driving it to work and just cruising I can pretty much cruise in closed loop
. Huge thanks to Brian Karwan @ Karcept's Inc for putting in crazy long hours, testing and research into figuring this all out. The tune feels great and I no longer worry about a CEL popping up.
Now back to suspension tuning...
10-19-2012, 02:31 PM
#623
Join Date: Apr 2006
Location: Atlanta, GA
Posts: 464
Likes: 0
Received 0 Likes
on
0 Posts
OK throwing in the towel and needing some input. Was hoping to get this up for an autocross this weekend, but I can't get the car to start.
Car starts and runs fine with stock ECU directly plugged in.
I have an AP1-2002
have a used Emanage ultimate with ver2.23
I have the boomslang K+W (knock and water temp) which I had them "fix" to the map "tap" setup that is now shipped with.
I have tried a base tune I know is working good on another AP1 from a forum member. My emanage car setup settings are the same as most others posted. I have also tried adding a temp sensor under the CH settings. I even went to the extreme of seeing if the karcepts ap2 tune would turn over (haha I know!).
I have the same pin settings as karcepts and have tried to change pins 14 and 17 to see if any changes. none.
I am beginning to suspect the harness. I may check pin by pin, but I would hope those boomslang folks know what they are doing.
Any ideas?
Car starts and runs fine with stock ECU directly plugged in.
I have an AP1-2002
have a used Emanage ultimate with ver2.23
I have the boomslang K+W (knock and water temp) which I had them "fix" to the map "tap" setup that is now shipped with.
I have tried a base tune I know is working good on another AP1 from a forum member. My emanage car setup settings are the same as most others posted. I have also tried adding a temp sensor under the CH settings. I even went to the extreme of seeing if the karcepts ap2 tune would turn over (haha I know!).
I have the same pin settings as karcepts and have tried to change pins 14 and 17 to see if any changes. none.
I am beginning to suspect the harness. I may check pin by pin, but I would hope those boomslang folks know what they are doing.
Any ideas?
10-22-2012, 11:35 PM
#626
Former Sponsor
Join Date: Aug 2011
Posts: 76
Likes: 0
Received 0 Likes
on
0 Posts
(continued from post #580)
I don't know what to tell you AP1 guys, but keep plugging away! If you cannot get the car to even turn over, its definitely wiring or possibly a faulty EMU.
Well, as promised, I'm back to share a little more. As Rad had mentioned, we are finally CEL free when forcing the cars into open loop early (in order to tune part throttle fully). We cannot force the car into open loop as early as we had previously attempted (~5%TPS), as that was what was causing the CEL's; however, it actually works out near perfect to have our new closed/open transition around 25%TPS, as instead of having 2 tunes (one for fuel efficiency and one for race), the tune we ended up with satisfies both. We have a target 14.7AFR from 0-25%TPS, perfect for highway cruising; and then from 25%-100%TPS (with a lower AFR target and some aggressive ignition advance) the car is an absolute beast! I know the initial setup of the EMU can be frustrating, but after having sorted all those steps, I truly appreciate the power of the EMU and could not be any happier with it.
Attached is a speed density (MAP sensor driven) base map for an AP2v1 with a rev limit set to 8500 rpm where the close/open transition point is roughly 25%TPS. This file is ready to start tuning and making power with consistent operation and will be code free, assuming the following:
1. Chassis = 2004-2005 Honda S2000
2. Use of Greddy E-Manage Ultimate using software Version 2.23 and jumper settings specified
3. Use of Boomslang harness purchased with Knock Sensor & Intake Air Temp option + modified to MAP “tap”
4. Must also have the additional harness modification to a TPS "intercept" (shown at the bottom of this post)
5. Lastly, the "making power" part of my statement can only happen at part throttle with the use of a load holding dyno and using the live tuning capability of the EMU to determine optimum fuel and ignition adjustments for all load vs. rpm points
To briefly explain the Analog Output Setting a little more clearly and how it is used, below is how I have set the adjustments on this shared file:
On our s2k's, 0% throttle is around 0.25 volts and 100% throttle is around 4.50 volts. Every car will be slightly different, but you can connect your laptop live with the EMU with the ignition switch on (car not running) and then under the Throttle tab, you can play with the throttle and observe the corresponding voltage at every TPS point. As I mentioned previously, I discovered the OEM closed/open transition point was at 3.34volts or 73%TPS on my vehicle. You can hook up a scan tool while playing with the throttle to verify the closed/open transition point for yourself. Clearly, the TPS position solely defines whether the car is in closed or open loop.
Knowing this, we can modify the output to a fake TPS setting that forces the car into open loop. Note all the 3.40's in the output row. I rounded up a bit (from the 3.34 volts) to add a little fudge to guarantee you all should be in open loop. You still should verify this for yourself and may need to adjust the number slightly higher if necessary. Now take a look at the inputs and observe the highlighted 1.32. 1.32 volt TPS input on my vehicle corresponded to 25% TPS. So what we are saying at 25% throttle, trick the ECU into thinking we are actually around 75% throttle (3.4 volt output). We continue to trick the ECU by keeping 3.4 volt outputs all the way until the input numbers are greater than 3.4 volts, then the outputs can match as they normally would.
Looking again at the 1.32 volt input, if you desire to enter open loop sooner, then change this voltage value. For example, 1.10 volts on my car = 20% TPS. 1.52 volts = 30% TPS. You can play around with that value to adjust the transition point. Like Rad mentioned, we could actually feel a little kick when you slowly transition into open loop by this forced method. The higher the TPS transition, the less of a kick was felt. At 30%, the kick was non-existent. At 20%, I felt it and thought it was a little annoying. 25% seemed right where it was starting to fade.
Lastly, when actually dyno tuning, I will temporarily set the Analog Output Setting to the following during the session:
I highlighted the 3 outputs that are changed from the base map. So if following the lesson above, you will see that now I'm calling for open loop at 0.40 volts (~3% TPS). I'll simply do this to know with certainty I am in open loop all the time in order to be able to fill in all load vs. rpm points accurately without having to worry about the closed/open loop transition. CEL's may occur, but I always have a scan tool connected to measure ECT's, so its easy enough to clear codes or see if you have any pending that you can clear before they actually show up.
In the near future I should be playing with an AP1 on the dyno; so if that goes smoothly, I'll share a starting point like I have for the AP2v1's.
(continued on post #673)
I don't know what to tell you AP1 guys, but keep plugging away! If you cannot get the car to even turn over, its definitely wiring or possibly a faulty EMU.
Well, as promised, I'm back to share a little more. As Rad had mentioned, we are finally CEL free when forcing the cars into open loop early (in order to tune part throttle fully). We cannot force the car into open loop as early as we had previously attempted (~5%TPS), as that was what was causing the CEL's; however, it actually works out near perfect to have our new closed/open transition around 25%TPS, as instead of having 2 tunes (one for fuel efficiency and one for race), the tune we ended up with satisfies both. We have a target 14.7AFR from 0-25%TPS, perfect for highway cruising; and then from 25%-100%TPS (with a lower AFR target and some aggressive ignition advance) the car is an absolute beast! I know the initial setup of the EMU can be frustrating, but after having sorted all those steps, I truly appreciate the power of the EMU and could not be any happier with it.
Attached is a speed density (MAP sensor driven) base map for an AP2v1 with a rev limit set to 8500 rpm where the close/open transition point is roughly 25%TPS. This file is ready to start tuning and making power with consistent operation and will be code free, assuming the following:
1. Chassis = 2004-2005 Honda S2000
2. Use of Greddy E-Manage Ultimate using software Version 2.23 and jumper settings specified
3. Use of Boomslang harness purchased with Knock Sensor & Intake Air Temp option + modified to MAP “tap”
4. Must also have the additional harness modification to a TPS "intercept" (shown at the bottom of this post)
5. Lastly, the "making power" part of my statement can only happen at part throttle with the use of a load holding dyno and using the live tuning capability of the EMU to determine optimum fuel and ignition adjustments for all load vs. rpm points
To briefly explain the Analog Output Setting a little more clearly and how it is used, below is how I have set the adjustments on this shared file:
On our s2k's, 0% throttle is around 0.25 volts and 100% throttle is around 4.50 volts. Every car will be slightly different, but you can connect your laptop live with the EMU with the ignition switch on (car not running) and then under the Throttle tab, you can play with the throttle and observe the corresponding voltage at every TPS point. As I mentioned previously, I discovered the OEM closed/open transition point was at 3.34volts or 73%TPS on my vehicle. You can hook up a scan tool while playing with the throttle to verify the closed/open transition point for yourself. Clearly, the TPS position solely defines whether the car is in closed or open loop.
Knowing this, we can modify the output to a fake TPS setting that forces the car into open loop. Note all the 3.40's in the output row. I rounded up a bit (from the 3.34 volts) to add a little fudge to guarantee you all should be in open loop. You still should verify this for yourself and may need to adjust the number slightly higher if necessary. Now take a look at the inputs and observe the highlighted 1.32. 1.32 volt TPS input on my vehicle corresponded to 25% TPS. So what we are saying at 25% throttle, trick the ECU into thinking we are actually around 75% throttle (3.4 volt output). We continue to trick the ECU by keeping 3.4 volt outputs all the way until the input numbers are greater than 3.4 volts, then the outputs can match as they normally would.
Looking again at the 1.32 volt input, if you desire to enter open loop sooner, then change this voltage value. For example, 1.10 volts on my car = 20% TPS. 1.52 volts = 30% TPS. You can play around with that value to adjust the transition point. Like Rad mentioned, we could actually feel a little kick when you slowly transition into open loop by this forced method. The higher the TPS transition, the less of a kick was felt. At 30%, the kick was non-existent. At 20%, I felt it and thought it was a little annoying. 25% seemed right where it was starting to fade.
Lastly, when actually dyno tuning, I will temporarily set the Analog Output Setting to the following during the session:
I highlighted the 3 outputs that are changed from the base map. So if following the lesson above, you will see that now I'm calling for open loop at 0.40 volts (~3% TPS). I'll simply do this to know with certainty I am in open loop all the time in order to be able to fill in all load vs. rpm points accurately without having to worry about the closed/open loop transition. CEL's may occur, but I always have a scan tool connected to measure ECT's, so its easy enough to clear codes or see if you have any pending that you can clear before they actually show up.
In the near future I should be playing with an AP1 on the dyno; so if that goes smoothly, I'll share a starting point like I have for the AP2v1's.
(continued on post #673)
10-23-2012, 03:33 AM
#627
Join Date: Jul 2004
Location: Kentucky
Posts: 1,578
Likes: 0
Received 0 Likes
on
0 Posts
Guys, my AP1 has been solid, reliable, and code free with the EMU, so I kinda forgot about this thread. My bad.
While it has not been dyno'ed, it feels very strong for an AP1 with PTuning's AP1 tune and Mark May's re-wiring. I also must disclose that this car does not see a lot of street miles, so maybe it just hasn't seen enough miles for it to start misbehaving. Either way, once the autocross season comes to a close, I am happy to share my setup. If someone knowledgeable nearby wants to make a road trip to the Lexington, KY area to document my setup for others to replicate, count me in.
-Dave
While it has not been dyno'ed, it feels very strong for an AP1 with PTuning's AP1 tune and Mark May's re-wiring. I also must disclose that this car does not see a lot of street miles, so maybe it just hasn't seen enough miles for it to start misbehaving. Either way, once the autocross season comes to a close, I am happy to share my setup. If someone knowledgeable nearby wants to make a road trip to the Lexington, KY area to document my setup for others to replicate, count me in.
-Dave
10-23-2012, 04:34 AM
#628
I'd also like to chime in with a good report on the gemu option for an AP1.
I bought Noel's Ptuning modified harness and a random used emanage ultimate off fleabay. Ptuning provided a basic tune for IHE equipped 02 model via e-mail for a very fair price. I did have a problem loading the tune for two reasons.
1. The software wouldent work with windows 7
2. The gemu wanted a password to load the tune and wouldent accept the password Ptuning suggested.
I was lucky enough to know a local guy who as far as I can tell invented the internet and had little trouble making it work with one of his older laptops that still had windows xp (he did say that he could have made it work with windows 7 but it would be a pain in the ass). We got it working on Friday of last week and I drove to Dallas (5 hrs) for an autox on Sunday. The car does seem to have a mis after slowing from highway speeds (~ 80mph @4300rpm) for a while at partial throttle such as around town. The tune seems strong in the midrange with a smoothing of the vtec power jump and a possible gain in top end power as well. Afr's seem to be safe also. I plan to get the car on a dyno to have it checked closer and tweaked where and if needed.
I bought Noel's Ptuning modified harness and a random used emanage ultimate off fleabay. Ptuning provided a basic tune for IHE equipped 02 model via e-mail for a very fair price. I did have a problem loading the tune for two reasons.
1. The software wouldent work with windows 7
2. The gemu wanted a password to load the tune and wouldent accept the password Ptuning suggested.
I was lucky enough to know a local guy who as far as I can tell invented the internet and had little trouble making it work with one of his older laptops that still had windows xp (he did say that he could have made it work with windows 7 but it would be a pain in the ass). We got it working on Friday of last week and I drove to Dallas (5 hrs) for an autox on Sunday. The car does seem to have a mis after slowing from highway speeds (~ 80mph @4300rpm) for a while at partial throttle such as around town. The tune seems strong in the midrange with a smoothing of the vtec power jump and a possible gain in top end power as well. Afr's seem to be safe also. I plan to get the car on a dyno to have it checked closer and tweaked where and if needed.
10-24-2012, 09:32 AM
#630
Former Sponsor
Join Date: Aug 2011
Posts: 76
Likes: 0
Received 0 Likes
on
0 Posts
I can't speak for the AP1, but the AP2 knock sensor logging I have not found to be slow and would not believe so on an AP1. The main problem with the Honda knock sensors are that they are overly sensitive to engine noise/vibrations. So where it may show knock, it will not actually mean detonation. Log and un-tuned car and you will see knock counts all over the upper rpm range. With the modern cylinder head design and on 93 octane in NA STR trim, I have not observed knock to be much of a factor with dyno tuning. You can basically tune for maximum power at every rpm/load point. You'll get to a point where throwing in additional timing stops increasing power, so you just back the timing down to the minimum amount that still created that peak value. I have found the timing levels at this point are still well under the threshold of detonation.