AIT/TPS Issues w/Tephra V7
Oct 3, 2011, 10:00 AM
#1
Evolving Member
Thread Starter
iTrader: (14)
Join Date: Aug 2007
Location: New York
Posts: 389
Likes: 0
Received 0 Likes
on
0 Posts
AIT/TPS Issues w/Tephra V7
OK, I'd like to start off with talking about a little bit of the history of the car before diving in.
(this way, it could be possible that someone with more experience/knowledge than myself with doing things such as this... could possibly spot/point out something I may have done wrong)
I have an Evo9. My car was put on an AEM EMS V1 for a long while.
My current mod configuration is in my signature. Everything is in there.
Brief recap of relevant mods: Tephra V7 on evo9 ecu, GM AIT, AEM 5Bar, greddy 3 port electronic boost control solenoid, Speed Density.
During the time of AEM, relevant modifications to the ECU include:
• Wiring in a 3 port greddy electronic boost control solenoid into the factory 2 wires that the original boost control solenoid was wired into. This way, AEM was able to control boost, and I didn't need to run a long vacuum line into the car or to a reachable manual boost controller under the hood.
• Wiring a switchable Ground for switch 7 (AEM switch 7) for alternating boost/maps (nitrous map enable/disable which can be used as an alternate fuel/timing map)
• Wired in the LC1 innovate wide-band o2 sensor into the AEM so that it picks up the o2 reading and can be tuned accordingly.
• Wired in a ground so that when the engine light goes on, a mounted bright white LED goes on as well (used as a shift light with AEM... you can set it so that after a certain RPM, it grounds a wire and lights up the engine light... which in turn, lit up my LED)
• Wired in a GM (AEM) AIT sensor (installed in UICP) into 2 wires that were part of the MAF harness. I found that information here. it worked. I didn't have to run wires down into the ECU... and after I spliced into those wires and T'd them, my AEM read the AIT perfectly.
So I sold the AEM and picked up a Stock ECU and got the Tephra V7 mod.
• I unwired the LC1 Innovate wideband sensor and re-wired the wire that I originally disconnected because I was unsure if this would affect the stock ecu wire-up.
• I unwired the switchable ground (switch 7 on aem) and reconnected the factory ecu wire for the same reason above.
• I left my LED wire in... which works regardless of ECU... as long as the engine light is on, the LED will go on (which is quite annoying at the moment as I have a TPS engine code, and the LED is always shining in my face... I covered it with tape)
• Left the boost solenoid wiring alone, as it is wired in to factory spec.
Now this is where the issue is... I looked up a post by mrfred (that was later incorporated into Tephra V7 if i remember right) that you are supposed to wire the AIT a certain way... it has two wires, ground one, and the other goes into pin 96 for the Evo9 ECU which is a fuel temp wire (used only for emissions which was patched to pass... so it didn't affect anything, only added functionality) So is the way I have it wired incorrect? The wires that I tapped into from the MAF harness... are those some sort of actual AIT reading wires? I can't imagine that they are fuel temp wires. Maybe I should un-do the sensor wiring, and install it as mrfred documented here.
Does having this AIT wired incorrectly affect the drivability of the car on speed density? I can see this being a possibility since the MAF harness is pulling in information from that GM AIT sensor... but I'm no ECU guru... maybe that harness is disabled or something for Speed Density??
I'll be re-wiring the sensor tonight.
Side note (very interesting):
When I went with AEM, I then went with a Magnus V5 Cast manifold (quite awesome by the way, especially since you can thread the AEM 5 bar right into it! No vacuum lines!) I obviously removed the factory intake manifold. Upon doing this, you remove ground wires. Keith at Turbo Trix Racing gave me some very insightful and interesting information... He said when he did that to his car, and eliminated his grounds on his stock ECU... he ended up melting a portion of his engine harness. Apparently many of the sensors around the intake manifold look for ground, and need a good ground to function correctly. For whatever reason... the AEM EMS doesn't need as much ground and doesn't have that issue. So I was driving around with the Stock ecu, and no intake manifold grounds with the Magnus V5 after I sold the AEM...
Now check this out: At that time... I had THREE engine codes... P0121 (TPS/Pedal Position Sensor A CKT Range/Perf)... it showed up as 2... one pending for whatever reason. Also P0335 (Crankshaft Position Sensor A Circuit Malfunction) And lastly P0500 (Veh Speed Sensor A Malfunction) ALL very necessary codes for a car to function. So I went over to Pep Boys, bought a couple fish-eye crimps for 10mm bolts... some 8ga black insulated wire loom... and made a ground kit for my Magnus V5. I grounded the head; one end near the cam angle sensor.. I had a spare 10mm thread there to use, and the other end over to the strut tower brace center area where there is a 10mm side threaded whole. I also grounded my intake manifold there by connecting a second ground cable from that brace area to the throttle cable bolt assembly on the intake manifold.
Here's what's funny... two of the 3 engine codes went away! I now only have a TPS code. Keith gave me another TPS to try out... I positioned it so that WOT read 100% in evo-scan... drove around... had no codes... but the car wouldn't boost past WG pressure. Stuck at 15psi. (car was tuned for 27-30). Then I decided to reflash the ECU with the same flash just for $hits and giggles... and the car boosted again... BUT the TPS came back on? Either way, I was happy that the 2 other codes are gone, but the TPS is interesting. I hope no circuitry is permanently damaged in my harness or anything.
The real PITA is that everytime I set off to drive the car from a 0mph situation... as soon as I lift off the clutch at anything below 2000-2200rpms... It goes completely lean... and takes up to around 2000-2200rpm's for it to normalize and have response/fuel. It seems load related, and not rpm. Even at cruise (when above that rpm) when I tip into throttle, there is a very hard hesitation that puts the car on it's face before it spits fuel/richens up the mixture for response.
More interestingly... when tuned with more fuel in those areas and the car get's reflashed, it drives fine. Then after about 5 minutes of driving... the lean issue comes back. ??????????
So there you have it. I don't know if the AIT being wired in wrong would cause that... but I'd absolutely LOVE (and pay) for anyone that can resolve this issue so that I can get it out of the way to finally get tuned!
-MR
UPDATE:
Got the AIT to work. I wired it like in the thread by mrfred. (Pin 96 for Evo9 Ecu).
LC1 works, had to flash MUT12 to 8409 according to mrfreds post here
Swapped AEM 5Bar for Omni 4Bar, rescaled according to this thread by... you guessed it... mrfred!
I multi-meter tested my TPS and set it so that at idle, the signal wire reads between 0.4 and 0.5 volts thanks to ezz-e (I traded map sensors with this man... he had an aem with an omni 4bar, and i had a stock ecu with an aem 5bar, so it made perfect sense
And guess what... No TPS code, no crank angle position sensor code, no speed sensor code... and stumble from dead stop issue is GONE. Now time to get a proper tune!!!!!
(this way, it could be possible that someone with more experience/knowledge than myself with doing things such as this... could possibly spot/point out something I may have done wrong)
I have an Evo9. My car was put on an AEM EMS V1 for a long while.
My current mod configuration is in my signature. Everything is in there.
Brief recap of relevant mods: Tephra V7 on evo9 ecu, GM AIT, AEM 5Bar, greddy 3 port electronic boost control solenoid, Speed Density.
During the time of AEM, relevant modifications to the ECU include:
• Wiring in a 3 port greddy electronic boost control solenoid into the factory 2 wires that the original boost control solenoid was wired into. This way, AEM was able to control boost, and I didn't need to run a long vacuum line into the car or to a reachable manual boost controller under the hood.
• Wiring a switchable Ground for switch 7 (AEM switch 7) for alternating boost/maps (nitrous map enable/disable which can be used as an alternate fuel/timing map)
• Wired in the LC1 innovate wide-band o2 sensor into the AEM so that it picks up the o2 reading and can be tuned accordingly.
• Wired in a ground so that when the engine light goes on, a mounted bright white LED goes on as well (used as a shift light with AEM... you can set it so that after a certain RPM, it grounds a wire and lights up the engine light... which in turn, lit up my LED)
• Wired in a GM (AEM) AIT sensor (installed in UICP) into 2 wires that were part of the MAF harness. I found that information here. it worked. I didn't have to run wires down into the ECU... and after I spliced into those wires and T'd them, my AEM read the AIT perfectly.
So I sold the AEM and picked up a Stock ECU and got the Tephra V7 mod.
• I unwired the LC1 Innovate wideband sensor and re-wired the wire that I originally disconnected because I was unsure if this would affect the stock ecu wire-up.
• I unwired the switchable ground (switch 7 on aem) and reconnected the factory ecu wire for the same reason above.
• I left my LED wire in... which works regardless of ECU... as long as the engine light is on, the LED will go on (which is quite annoying at the moment as I have a TPS engine code, and the LED is always shining in my face... I covered it with tape)
• Left the boost solenoid wiring alone, as it is wired in to factory spec.
Now this is where the issue is... I looked up a post by mrfred (that was later incorporated into Tephra V7 if i remember right) that you are supposed to wire the AIT a certain way... it has two wires, ground one, and the other goes into pin 96 for the Evo9 ECU which is a fuel temp wire (used only for emissions which was patched to pass... so it didn't affect anything, only added functionality) So is the way I have it wired incorrect? The wires that I tapped into from the MAF harness... are those some sort of actual AIT reading wires? I can't imagine that they are fuel temp wires. Maybe I should un-do the sensor wiring, and install it as mrfred documented here.
Does having this AIT wired incorrectly affect the drivability of the car on speed density? I can see this being a possibility since the MAF harness is pulling in information from that GM AIT sensor... but I'm no ECU guru... maybe that harness is disabled or something for Speed Density??
I'll be re-wiring the sensor tonight.
Side note (very interesting):
When I went with AEM, I then went with a Magnus V5 Cast manifold (quite awesome by the way, especially since you can thread the AEM 5 bar right into it! No vacuum lines!) I obviously removed the factory intake manifold. Upon doing this, you remove ground wires. Keith at Turbo Trix Racing gave me some very insightful and interesting information... He said when he did that to his car, and eliminated his grounds on his stock ECU... he ended up melting a portion of his engine harness. Apparently many of the sensors around the intake manifold look for ground, and need a good ground to function correctly. For whatever reason... the AEM EMS doesn't need as much ground and doesn't have that issue. So I was driving around with the Stock ecu, and no intake manifold grounds with the Magnus V5 after I sold the AEM...
Now check this out: At that time... I had THREE engine codes... P0121 (TPS/Pedal Position Sensor A CKT Range/Perf)... it showed up as 2... one pending for whatever reason. Also P0335 (Crankshaft Position Sensor A Circuit Malfunction) And lastly P0500 (Veh Speed Sensor A Malfunction) ALL very necessary codes for a car to function. So I went over to Pep Boys, bought a couple fish-eye crimps for 10mm bolts... some 8ga black insulated wire loom... and made a ground kit for my Magnus V5. I grounded the head; one end near the cam angle sensor.. I had a spare 10mm thread there to use, and the other end over to the strut tower brace center area where there is a 10mm side threaded whole. I also grounded my intake manifold there by connecting a second ground cable from that brace area to the throttle cable bolt assembly on the intake manifold.
Here's what's funny... two of the 3 engine codes went away! I now only have a TPS code. Keith gave me another TPS to try out... I positioned it so that WOT read 100% in evo-scan... drove around... had no codes... but the car wouldn't boost past WG pressure. Stuck at 15psi. (car was tuned for 27-30). Then I decided to reflash the ECU with the same flash just for $hits and giggles... and the car boosted again... BUT the TPS came back on? Either way, I was happy that the 2 other codes are gone, but the TPS is interesting. I hope no circuitry is permanently damaged in my harness or anything.
The real PITA is that everytime I set off to drive the car from a 0mph situation... as soon as I lift off the clutch at anything below 2000-2200rpms... It goes completely lean... and takes up to around 2000-2200rpm's for it to normalize and have response/fuel. It seems load related, and not rpm. Even at cruise (when above that rpm) when I tip into throttle, there is a very hard hesitation that puts the car on it's face before it spits fuel/richens up the mixture for response.
More interestingly... when tuned with more fuel in those areas and the car get's reflashed, it drives fine. Then after about 5 minutes of driving... the lean issue comes back. ??????????
So there you have it. I don't know if the AIT being wired in wrong would cause that... but I'd absolutely LOVE (and pay) for anyone that can resolve this issue so that I can get it out of the way to finally get tuned!
-MR
UPDATE:
Got the AIT to work. I wired it like in the thread by mrfred. (Pin 96 for Evo9 Ecu).
LC1 works, had to flash MUT12 to 8409 according to mrfreds post here
Swapped AEM 5Bar for Omni 4Bar, rescaled according to this thread by... you guessed it... mrfred!
I multi-meter tested my TPS and set it so that at idle, the signal wire reads between 0.4 and 0.5 volts thanks to ezz-e (I traded map sensors with this man... he had an aem with an omni 4bar, and i had a stock ecu with an aem 5bar, so it made perfect sense
And guess what... No TPS code, no crank angle position sensor code, no speed sensor code... and stumble from dead stop issue is GONE. Now time to get a proper tune!!!!!
Last edited by MagicManRed; Oct 16, 2011 at 09:40 PM. Reason: UPDATES!!
Oct 3, 2011, 10:05 AM
#2
Evolving Member
Thread Starter
iTrader: (14)
Join Date: Aug 2007
Location: New York
Posts: 389
Likes: 0
Received 0 Likes
on
0 Posts
I'd like to add that I am currently trading my AEM 5Bar for an Omni 4Bar as many people seem to say that it works better with the stock ECU. The trade makes sense since the person I'm trading with has AEM, and I have a Stock ECU
Oct 7, 2011, 11:54 AM
#6
wire the ecu stuff back to stock.
use the fuel temp as iat in the rom
install the 4bar map and calibrate the sd properly for it
along with setting proper map sensor settings in the rom
the tps with the boost thing sounds like a fluke
what you're experiencing is lean spikes and it must be fixed
by either map ve table / open loop / maf table / rpm ve or all
of the above
also mess with stutter fixes. lots of ways to tune sd..
if i remember right your car has no front o2 sensor.. so you must force in open loop
and tune that way. not that hard.. last car i did on SD the 790hp 2L with hta86 was all open loop.
no one is responding because people are lazy to read that big *** story you wrote.
use the fuel temp as iat in the rom
install the 4bar map and calibrate the sd properly for it
along with setting proper map sensor settings in the rom
the tps with the boost thing sounds like a fluke
what you're experiencing is lean spikes and it must be fixed
by either map ve table / open loop / maf table / rpm ve or all
of the above
also mess with stutter fixes. lots of ways to tune sd..
if i remember right your car has no front o2 sensor.. so you must force in open loop
and tune that way. not that hard.. last car i did on SD the 790hp 2L with hta86 was all open loop.
no one is responding because people are lazy to read that big *** story you wrote.
Oct 7, 2011, 12:04 PM
#7
Evolving Member
Thread Starter
iTrader: (14)
Join Date: Aug 2007
Location: New York
Posts: 389
Likes: 0
Received 0 Likes
on
0 Posts
Got the AIT to work. I wired it like in that thread. so that is A-OK.
Also got the LC1 to work. Had to flash MUT12 with a different code. Logs afr now without that serial cable.
I tried the jitter/stutter fixes. Didn't help. Returned values back to stock.
My AEM 5 bar seems to be working fine. As you said, it is probably just the SD VE, etc tables that need adjusting.
I disabled the P0500 code via the periphery tables, (speed sensor) and it seems the TPS went away with it. I adjusted my TPS so that it reads 99% WOT and 9.8% No throttle.
Did some WOT runs, and car isn't throwing codes at the moment.
The lean issue is still there. I guess it isn't TPS related, just tune related.
Thanks for your reply. I'd love to have you tune the car, but you only do remote... and I don't play that :P
And I agree with you... I'm sure most people aren't reading the whole thing... then again, those people, aren't the types of people I want replies from anyway
Also got the LC1 to work. Had to flash MUT12 with a different code. Logs afr now without that serial cable.
I tried the jitter/stutter fixes. Didn't help. Returned values back to stock.
My AEM 5 bar seems to be working fine. As you said, it is probably just the SD VE, etc tables that need adjusting.
I disabled the P0500 code via the periphery tables, (speed sensor) and it seems the TPS went away with it. I adjusted my TPS so that it reads 99% WOT and 9.8% No throttle.
Did some WOT runs, and car isn't throwing codes at the moment.
The lean issue is still there. I guess it isn't TPS related, just tune related.
Thanks for your reply. I'd love to have you tune the car, but you only do remote... and I don't play that :P
And I agree with you... I'm sure most people aren't reading the whole thing... then again, those people, aren't the types of people I want replies from anyway
Trending Topics
Oct 9, 2011, 09:20 AM
#8
Evolving Member
Thread Starter
iTrader: (14)
Join Date: Aug 2007
Location: New York
Posts: 389
Likes: 0
Received 0 Likes
on
0 Posts
Never mind. TPS code still shows up.
Thanks to scheides, I am changing the math in evo scan 2.8 to read PSI with the AEM 5 Bar.
This way I can start sorting out some issues.
Thanks to scheides, I am changing the math in evo scan 2.8 to read PSI with the AEM 5 Bar.
This way I can start sorting out some issues.
Oct 9, 2011, 08:51 PM
#10
Evolving Member
Thread Starter
iTrader: (14)
Join Date: Aug 2007
Location: New York
Posts: 389
Likes: 0
Received 0 Likes
on
0 Posts
V5 has speed density as well?
I'm fairly new to Tephra... but I'm very computer savvy and mechanically inclined. Just trying to get as much info as I can to sort out these small kinks. The car ran fine with the AEM EMS V1 (other than some weird throttle delay it had that I couldn't stand, as well as it taking forever to turn the motor over) so I got rid of it when I read up on Tephra V7 and it's features such as Valet mode, switchable maps, and knock light. (sucks there isn't a shift light option for it yet)
Anywho, looking forward to hearing back from you and the details of your journey from V5 to V7.
-MR
Oct 9, 2011, 09:16 PM
#11
Man you are a novel-writer! 2 PMs with you and over char limit, LOL.
So. V5 was not SD-enabled and neither was v7, the original tephra roms were patched with SD code. The v7 ones had a bunch of common hacks added to them because most of them were so vanilla you would not have a v7 car w/o all the other stuff anyways.
Ideas- make absolutely sure you are wiring everything to the correct ecu pins. The diagrams are easy enough, but when you have the harness in front of you and you are upside down in the car and looking at the pins on the plugs instead of the ecu itself it can be confusing. Sounds like you have MAT and wideband logging working so you're probably good, but still, triple check your work there.
I mentionedthis in a PM, but make sure you are logging the correct requestID for various items. If not, your output numbers in evos an will not make sense.
Also, try using some values others have posted (mine for example) for your VE tables in the 60-120 load area. Don't think that jus because you have a different MAP sensor that yours will be different...if it is scaled correctly then you should be fine.
You say you are computer savvy, but you can't find the data.XML file? Dig harder man!. I'm not sure where it is but it's gotta be there. If not, just use the GUI like I sent you via PM.
Lastly, make sure to use the scientific process through everything you do. If one setting does not work, document it, tweak, try again, and note the differences. The best wayto troubleshoot is to try a bunch of iterations, log and take notes on them all, then step back and look at the data altogether, so you might get a big picture on wha is going on. Change only one thing at a time. Too Manu variables will just cause yourself a CF and judging by all of the questions you have it sounds like you might have just that.
Keep trying!
So. V5 was not SD-enabled and neither was v7, the original tephra roms were patched with SD code. The v7 ones had a bunch of common hacks added to them because most of them were so vanilla you would not have a v7 car w/o all the other stuff anyways.
Ideas- make absolutely sure you are wiring everything to the correct ecu pins. The diagrams are easy enough, but when you have the harness in front of you and you are upside down in the car and looking at the pins on the plugs instead of the ecu itself it can be confusing. Sounds like you have MAT and wideband logging working so you're probably good, but still, triple check your work there.
I mentionedthis in a PM, but make sure you are logging the correct requestID for various items. If not, your output numbers in evos an will not make sense.
Also, try using some values others have posted (mine for example) for your VE tables in the 60-120 load area. Don't think that jus because you have a different MAP sensor that yours will be different...if it is scaled correctly then you should be fine.
You say you are computer savvy, but you can't find the data.XML file? Dig harder man!
. I'm not sure where it is but it's gotta be there. If not, just use the GUI like I sent you via PM.Lastly, make sure to use the scientific process through everything you do. If one setting does not work, document it, tweak, try again, and note the differences. The best wayto troubleshoot is to try a bunch of iterations, log and take notes on them all, then step back and look at the data altogether, so you might get a big picture on wha is going on. Change only one thing at a time. Too Manu variables will just cause yourself a CF and judging by all of the questions you have it sounds like you might have just that.
Keep trying!
Oct 10, 2011, 07:10 AM
#13
Evolving Member
Thread Starter
iTrader: (14)
Join Date: Aug 2007
Location: New York
Posts: 389
Likes: 0
Received 0 Likes
on
0 Posts
Triple checked all wiring. even pulled out Map sensor to check that the wires were wired in right.
AIT is wired in right. LC1 is wired in right. Misc Engine light ground wired in right.
No Data file for whatever reason. EvoScan 2.8.0005
Running 93 Octane. No E85. Sold my ID 2200's.
25-28psi on 10.5:1 compression.
Boost GM-5Bar MAP
Request: 38
Eval: 0.29411765*x-14.3
Units: psig
Doesnt read right
I keep getting a TPS P0121 code.
My throttle percentage at fully floored WOT reads 99% and change.
When I'm not on the throttle... reads just over 9%.
I think my TPS issue has a great deal to do with the code, and the lean issues. I'm guessing AEM didn't need very particulars with the TPS, which is why it seemed to have worked better than this.
I will post up a log in a moment along with my rom if you're feeling frisky... Maybe your eyes will see something mine don't
AIT is wired in right. LC1 is wired in right. Misc Engine light ground wired in right.
No Data file for whatever reason. EvoScan 2.8.0005
Running 93 Octane. No E85. Sold my ID 2200's.
25-28psi on 10.5:1 compression.
Boost GM-5Bar MAP
Request: 38
Eval: 0.29411765*x-14.3
Units: psig
Doesnt read right
I keep getting a TPS P0121 code.
My throttle percentage at fully floored WOT reads 99% and change.
When I'm not on the throttle... reads just over 9%.
I think my TPS issue has a great deal to do with the code, and the lean issues. I'm guessing AEM didn't need very particulars with the TPS, which is why it seemed to have worked better than this.
I will post up a log in a moment along with my rom if you're feeling frisky... Maybe your eyes will see something mine don't
Last edited by MagicManRed; Oct 10, 2011 at 07:30 AM. Reason: Added info