The Haltech elite 2500 parallel install.
09-23-2015, 10:48 PM
#26
MBWorld Fanatic!
And please, do not be disillusioned by the people that will post that have no idea of the difference between a spark plug and an injector.
09-23-2015, 11:01 PM
#27
Member
Thread Starter
Ordered a Honeywell fuel pressure sensor yesterday as well. Im questioning whether to mount it pre injectors, or post. I really want to see this #8 cylinder lean problem is fact or fiction, but with data to explain more as to why.
Ill be ordering some micro quadlok connectors and dtm connectors tomorrow for tapping and rewiring things and try to find a reasonable location to mount the aftermarket ecu. (passenger side cowl cover is the easiest, but id like to put it closer to the factory ecu)
09-23-2015, 11:13 PM
#28
MBWorld Fanatic!
8 wide band o2 sensors (A friend of mine did this on a different car and found 40% differences in the 02 outputs under a "normal" tune), and you will be able to see what the tuned duty cycles come out to on those trouble cylinders.
I scoped an engine that died from 8 "going". It was tuned, but clearly there is no one that can add additional fuel to that one hole. It heats up the cylinder (Detonation? Just lean-burn heat? A mix of problems from hot coolant flow? - I don't know), and expands it enough start scoring the walls. Or the ring land breaks from hammering in det.
Let me know if there is anything I can do.
I scoped an engine that died from 8 "going". It was tuned, but clearly there is no one that can add additional fuel to that one hole. It heats up the cylinder (Detonation? Just lean-burn heat? A mix of problems from hot coolant flow? - I don't know), and expands it enough start scoring the walls. Or the ring land breaks from hammering in det.
Let me know if there is anything I can do.
09-23-2015, 11:24 PM
#29
Member
Thread Starter
8 wide band o2 sensors (A friend of mine did this on a different car and found 40% differences in the 02 outputs under a "normal" tune), and you will be able to see what the tuned duty cycles come out to on those trouble cylinders.
I scoped an engine that died from 8 "going". It was tuned, but clearly there is no one that can add additional fuel to that one hole. It heats up the cylinder (Detonation? Just lean-burn heat? A mix of problems from hot coolant flow? - I don't know), and expands it enough start scoring the walls. Or the ring land breaks from hammering in det.
Let me know if there is anything I can do.
I scoped an engine that died from 8 "going". It was tuned, but clearly there is no one that can add additional fuel to that one hole. It heats up the cylinder (Detonation? Just lean-burn heat? A mix of problems from hot coolant flow? - I don't know), and expands it enough start scoring the walls. Or the ring land breaks from hammering in det.
Let me know if there is anything I can do.
You could tune one side at a time with 4 widebands excluding the opposing bank while doing it (charge air typically isn't gonna be pulled from one side of the manifold all the way to the other side)
09-25-2015, 09:48 AM
#31
PLATINUM SPONSOR
Glad to see someone working on this. I spoke to Haltech a few years back and one of their shops had been working on making this work for our cars but kept hitting walls. Really need a great programmer with a ton of extra time. We all hope that is you
__________________
E63 Biturbo, UPD Cold Air induction kit, UPD performance crank pulley and UPD adjustable rear suspension with ride height adjustment.
CL55 UPD Cold Air Boost kit, UPD 3000 stall converter, UPD 77mm SC clutched pulley and beltwrap kit, Custom long tubes, UPD crank pulley , UPD suspension kit, UPD SC pulley, Aux. HE, Trunk tank w/rule 2000 pump, Mezeire pump, UPD 5pc idler set, Aluminum rotor hats.
www.ultimatepd.com
instagram @ultimate_pd
facebook.com/ultimatepd
E63 Biturbo, UPD Cold Air induction kit, UPD performance crank pulley and UPD adjustable rear suspension with ride height adjustment.
CL55 UPD Cold Air Boost kit, UPD 3000 stall converter, UPD 77mm SC clutched pulley and beltwrap kit, Custom long tubes, UPD crank pulley , UPD suspension kit, UPD SC pulley, Aux. HE, Trunk tank w/rule 2000 pump, Mezeire pump, UPD 5pc idler set, Aluminum rotor hats.
www.ultimatepd.com
instagram @ultimate_pd
facebook.com/ultimatepd
09-27-2015, 10:59 AM
#32
Best of luck with this...subscribed!
09-27-2015, 03:16 PM
#33
Member
Thread Starter
It does take a bit of headbanging against a wall at times before you'll make your way through problems. Being able to get and understand the data is the key part. Im not a huge mercedes guru by any means, but I am a pretty big electronics nerd and try to understand arduino code and programming. An eventual goal would be to get everything setup where I can decipher canbus channeling, but that'll be a long ways out.
09-27-2015, 09:19 PM
#34
MBWorld Fanatic!
I think you may also encounter some misinformation from this forum. So many here have simply repeated "Its just so complex" - but in reality, while there is indeed a canbus, I think understanding, once you know the communication going on at the canbus, you can utilize star to start turning things off, so they are not distracting.
An engine runs on Fuel and Air - and when you have both of those, you are doing quite well.
The ODB2 systems code would essentially be bypassed once you have the OEM ECU dummied on the sensors it needs to see to keep happy.
Assuming you can throw out the unimportant at first like the Radio and HVAC, and any interaction with the Suspension, then you essentially are dealing with systems revolving around the below, with both the need to control them, but also show the OEM ECU the parameters it is expecting.
Engine Speed
Vehicle Speed
TPS and DRW
IAT
Coolant Temp
Fuel Mixture (sensing for control of engine, but also parameters expected by ECU)
Boost pressure and Loop feedback
The purest idea of doing this, which I do not know whether it has been done, is instead of dummying sensor outputs the ECU is expecting - but rather to actually Sniff out the actual communication (CanBus or otherwise) that needs to happen to make everything work WITHOUT the ECU. Not even saying this is possible, or easy, but it would be purest.
Because if you flip the concept on its head, and everyone always says (on any newer platform) that the ECU is hard to trick - *Remove* the ECU. Then who is left to trick? Emulating the communication with the ECU's counterparts then becomes the issue. And who are these partners? The TCU, the ABS system, the brake-by-wire (which is a big big deal)
Anyway, just musing.
An engine runs on Fuel and Air - and when you have both of those, you are doing quite well.
The ODB2 systems code would essentially be bypassed once you have the OEM ECU dummied on the sensors it needs to see to keep happy.
Assuming you can throw out the unimportant at first like the Radio and HVAC, and any interaction with the Suspension, then you essentially are dealing with systems revolving around the below, with both the need to control them, but also show the OEM ECU the parameters it is expecting.
Engine Speed
Vehicle Speed
TPS and DRW
IAT
Coolant Temp
Fuel Mixture (sensing for control of engine, but also parameters expected by ECU)
Boost pressure and Loop feedback
The purest idea of doing this, which I do not know whether it has been done, is instead of dummying sensor outputs the ECU is expecting - but rather to actually Sniff out the actual communication (CanBus or otherwise) that needs to happen to make everything work WITHOUT the ECU. Not even saying this is possible, or easy, but it would be purest.
Because if you flip the concept on its head, and everyone always says (on any newer platform) that the ECU is hard to trick - *Remove* the ECU. Then who is left to trick? Emulating the communication with the ECU's counterparts then becomes the issue. And who are these partners? The TCU, the ABS system, the brake-by-wire (which is a big big deal)
Anyway, just musing.
09-27-2015, 11:22 PM
#35
Member
Thread Starter
I think you may also encounter some misinformation from this forum. So many here have simply repeated "Its just so complex" - but in reality, while there is indeed a canbus, I think understanding, once you know the communication going on at the canbus, you can utilize star to start turning things off, so they are not distracting.
An engine runs on Fuel and Air - and when you have both of those, you are doing quite well.
The ODB2 systems code would essentially be bypassed once you have the OEM ECU dummied on the sensors it needs to see to keep happy.
Assuming you can throw out the unimportant at first like the Radio and HVAC, and any interaction with the Suspension, then you essentially are dealing with systems revolving around the below, with both the need to control them, but also show the OEM ECU the parameters it is expecting.
Engine Speed
Vehicle Speed
TPS and DRW
IAT
Coolant Temp
Fuel Mixture (sensing for control of engine, but also parameters expected by ECU)
Boost pressure and Loop feedback
The purest idea of doing this, which I do not know whether it has been done, is instead of dummying sensor outputs the ECU is expecting - but rather to actually Sniff out the actual communication (CanBus or otherwise) that needs to happen to make everything work WITHOUT the ECU. Not even saying this is possible, or easy, but it would be purest.
Because if you flip the concept on its head, and everyone always says (on any newer platform) that the ECU is hard to trick - *Remove* the ECU. Then who is left to trick? Emulating the communication with the ECU's counterparts then becomes the issue. And who are these partners? The TCU, the ABS system, the brake-by-wire (which is a big big deal)
Anyway, just musing.
An engine runs on Fuel and Air - and when you have both of those, you are doing quite well.
The ODB2 systems code would essentially be bypassed once you have the OEM ECU dummied on the sensors it needs to see to keep happy.
Assuming you can throw out the unimportant at first like the Radio and HVAC, and any interaction with the Suspension, then you essentially are dealing with systems revolving around the below, with both the need to control them, but also show the OEM ECU the parameters it is expecting.
Engine Speed
Vehicle Speed
TPS and DRW
IAT
Coolant Temp
Fuel Mixture (sensing for control of engine, but also parameters expected by ECU)
Boost pressure and Loop feedback
The purest idea of doing this, which I do not know whether it has been done, is instead of dummying sensor outputs the ECU is expecting - but rather to actually Sniff out the actual communication (CanBus or otherwise) that needs to happen to make everything work WITHOUT the ECU. Not even saying this is possible, or easy, but it would be purest.
Because if you flip the concept on its head, and everyone always says (on any newer platform) that the ECU is hard to trick - *Remove* the ECU. Then who is left to trick? Emulating the communication with the ECU's counterparts then becomes the issue. And who are these partners? The TCU, the ABS system, the brake-by-wire (which is a big big deal)
Anyway, just musing.
yeah.
i'd love to reverse engineer the CANbus on the car once running the stand alone down the road. it becomes quite challenging because every input and output has a channel, it's simply a matter of defining the channels and their values, and going through the data to understand what's happening.
Emulating values isn't really a big deal, mainly because there's little need to do it, and secondly most sensors can be rescaled with an arduino module with ease. (analogreadwrite)
01-31-2016, 10:57 AM
#38
Member
Thread Starter
Nothing too exciting, i should be making some progress for mounting everything and wiring in the coming weeks of February.
Ive gotten everything except 2 widebands, and the extension harness for the oe ecu. Once I've made headway on the mounting bracket and wired power and ground through the cabin to the battery in the back, things should move rather quickly.
Ive gotten everything except 2 widebands, and the extension harness for the oe ecu. Once I've made headway on the mounting bracket and wired power and ground through the cabin to the battery in the back, things should move rather quickly.
03-15-2017, 12:20 PM
#40
MBWorld Fanatic!
Ok, it's been over a year since the last update. This thread went quiet when it was just getting interesting. Updates??
03-15-2017, 03:28 PM
#42
Member
Thread Starter
Yep, after taking a step deeper into the system, i doubted the capabilities of the ignition control as there's a torque control variable on the bus. Throwing timing during shifts is where i question things as timing is indeed retarded during shifting, so i started to dig deeper.
On my spare time im reverse engineering the bus, but time has been pretty limited recently.
Currently the only 2 options for a complete package are m150 developmental and bosch ms6.1 /ms6.2
12-19-2018, 01:30 AM
#47
Senior Member
I’ve put a Link Thunder in my car and had it running for well over a year now. On the engine side of things it’s all pretty straightforward but I didn’t bother with the canbus hacking and just swapped the auto gerbox for manual. I did want to get a SD/Syvecs kit but it has not been released until now. Shame really as any modern stand-alone is so much more advanced than the ME2.8...
12-19-2018, 08:30 AM
#48
Member
Thread Starter
I’ve put a Link Thunder in my car and had it running for well over a year now. On the engine side of things it’s all pretty straightforward but I didn’t bother with the canbus hacking and just swapped the auto gerbox for manual. I did want to get a SD/Syvecs kit but it has not been released until now. Shame really as any modern stand-alone is so much more advanced than the ME2.8...
Ill have can mapping strategies for link thunder expected in q1 2019.
The following users liked this post:
Alex L (12-19-2018)
