LT5 ECM Tuning thread

[XfireZ51]

Jeff,

Maybe we should start a fund and see if we could pay someone who has the knowledge to create a def file for use by members of this registry.

[-=Jeff=-]

Originally Posted by XfireZ51

Jeff,

Maybe we should start a fund and see if we could pay someone who has the knowledge to create a def file for use by members of this registry.

Maybe...

[kenthompson]

Let me know when you want the money!

[XfireZ51]

Jeff and I will keep you posted.

[Tyler Townsley]

Originally Posted by XfireZ51

Tyler,

The spark or SA tables are just that. Its 3D tables of the commanded
spark advance (SA). Based on the rpm and kPa at any given time, the ECM will command the stated timing. That's the simple explanation. However, there are several ancillary tables that are also used to modify the final SA depending on the parameters that are met (i.e. Idle, warmup, Decel, TPS, CTS etc.)With just what Ron has described I'm not convinced that the SA tables are the solution for Ron. Not sure that there is a solution for Ron. Getting some scantool and WB data would help in trying to narrow it down. But with big cams and more overlap, there's a tendency to overscavenge the chamber exposing the NB O2 to more air.
That causes the ECM to think the motor is running lean which in turn causes the ECM to try to "richen" the mixture which then increases the unburned HC. Lowering the O2 voltage swing points at lower airflow tells the ECM that the motor is not running lean. That's why the WB could be reading stoich but the BLMs show lean.
That's why I'm interested in the R/L O2 voltage tables. They're in the calibration just haven't been created in the .xdf or .tdf.
Make sense?

On another list there are people using advanced timing in the lower ranges to solve idle problems associated with cam overlap, that is what I was talking about. With a good datalog to see specific cells and a good wideband checked against the timing retard you might get the mixture to pass. You would have to go open loop all the time in the chip and customize it just to pass. If you went closed loop there is no way the calibration is going to let you get it right as it will try adjusting to 14.7 at all points. It may be that the motor needs cells at a different mixture range and spark point so it can interpolate between cells to get what he needs. IE you are starting the iginition early to get better combustion, since it is at lower speed should be able to monitor/adjust for any spark knock.

Todd comments?

Tyler

[Tyler Townsley]

Originally Posted by XfireZ51

Jeff,

Maybe we should start a fund and see if we could pay someone who has the knowledge to create a def file for use by members of this registry.

You can take the Tunercat .tdf file and create a .xdf file. The early tunercat files had a problem in the VE tables that Todd found and we contacted them and they corrected it. So you need to make sure you have the later def files. I am not sure what you need beyond what is in the .tdf files for tuning as they are pretty thorough. I have the complete 90 commented GM calibration printout and it is over 1 in thick and 250 pages. Since it is a 90 the calibration has only sketches of where they were going in the egr area so it is only good to 92.

Todd has spent literaly 100s of hours learning what he knows, he and I am somewhat leary of handing out information before we know that the reciepent has a good understanding of just what is involved. I can tell you from bitter experience you can mess up big time if you do not have a good grasp on what is happening in a calibration.

The biggest problem is as you have described, a good way to log what is going on and be able to review the logs watching what happens as you make a change. My car has been off the road for 5 years so I have not had a platform to take advantage of the new software which is why I have not had much to contribute.

I believe Todd uses a FJO wideband and datalog setup, but I also know whatever he used it was the hours on the road that was most important. There are very few who really want to play with stuff and yes it is confusing to a newcomer but thats why you can count on one hand those who have the skill and knowledge to get results.

Tyler

[XfireZ51]

If you went closed loop there is no way the calibration is going to let you get it right as it will try adjusting to 14.7 at all points. It may be that the motor needs cells at a different mixture range and spark point so it can interpolate between cells to get what he needs. IE you are starting the iginition early to get better combustion, since it is at lower speed should be able to monitor/adjust for any spark knock.

SA is likely not the issue. The condition you describe is precisely why you'd like to modify the O2 Window Terms. These will define the O2 sensor swing points that the ECM will try keeping the mixture at. And it can be varied on the basis of airflow. For those using cams with larger overlap, this helps tame idle surging and rich idle conditions. PID controls are helpful here too. These don't appear in the .tdfs and .xdfs generally available. There's nothing "hard-wired" in the cal for 14.7 being stoich. For those of us using gas with 10% ethanol, we benefit from changing stoich to 14.3:1 for greater tuning accuracy.

Originally Posted by Tyler Townsley

You can take the Tunercat .tdf file and create a .xdf file. The early tunercat files had a problem in the VE tables that Todd found and we contacted them and they corrected it. So you need to make sure you have the later def files. I am not sure what you need beyond what is in the .tdf files for tuning as they are pretty thorough.

The .tdf is pretty basic. It lacks PID controls, IAC controls, O2 sensor etc. Quite a few missing parameters allow someone to really fine tune the motor particularly if significant mods have been made. For example, I found that swapping in an AL FW benefitted from changes to the VE tables but also to IAC and Proportional Gains.

I have the complete 90 commented GM calibration printout and it is over 1 in thick and 250 pages. Since it is a 90 the calibration has only sketches of where they were going in the egr area so it is only good to 92.

Todd has spent literaly 100s of hours learning what he knows, he and I am somewhat leary of handing out information before we know that the reciepent has a good understanding of just what is involved. I can tell you from bitter experience you can mess up big time if you do not have a good grasp on what is happening in a calibration.

That's true to an extent. But you can't learn about what various controls do unless you can play with them and see what the effect is. Some of us can't disassemble code, but we certainly can utilize the parameters, once revealed, to tune the calibration to our liking. What would you like to know about someone like myself that would make you feel comfortable that we know what we are doing and therefore share the commented hack? With the majority of ZR-1s being 90s, this would be very useful for the community.

The biggest problem is as you have described, a good way to log what is going on and be able to review the logs watching what happens as you make a change. My car has been off the road for 5 years so I have not had a platform to take advantage of the new software which is why I have not had much to contribute.

I believe Todd uses a FJO wideband and datalog setup, but I also know whatever he used it was the hours on the road that was most important. There are very few who really want to play with stuff and yes it is confusing to a newcomer but thats why you can count on one hand those who have the skill and knowledge to get results.

Tyler

As =Jeff= can attest, I've burned over 1000 bins using either proms or Ostrich/TunerPro RT combo. I use a ZT-2 WB for WOT tuning. Datalogging with EASE and http://www.dynamicefi.com/. I think you'll find that there is quite a large community of "tuners" out there pretty adept at making all kinds of engine combos run well. I need to give a great deal of credit for to people like Bruce Plecan ( generally acknowledged as the father of GM ECM tuning) and Rob Rauscher (RBob moderator on 3rd Gen) for freely sharing what they learned.
I don't think tuning an LT-5 would be any more difficult than some of the stuff you see on 3rd Gen. The fundamentals are the same, and as Todd has stated, much of the calibration is based off of the $8D code.

[tpepmeie]

Originally Posted by XfireZ51

Some of us can't disassemble code, .

Dominic,
Tell you what, you guys give it a shot and post up here, or in a new thread, when you get stuck. I assure you if you get stuck along the way I will offer help. I couldn't disassemble code either when I started...heck, I have no training at all in this field.

Which calibration are you working with? AYBK? How far have you gotten into it?

Todd

[XfireZ51]

Todd,

I appreciate the offer of help. I have not started disassembling code. I have both the .tdf and .xdf editors. I have ported a .tdf over to an .ecu and .xdf file. (Pretty simple since both John and Mark built a function to do just that). I have previously created and modified .xdf items. I prefer working with TP but have used both as well as modifying calibrations directly in hex. I have modified .ads so as to output aldl items that were not supported in the scantool I was using. I've been able to introduce new output to the ALDL datastream (i.e. WB O2). Basically, if I have a hack, I can create the .xdf. What other tools do I need?
If you want to take people like myself, Shawn ZR-1 and =Jeff= through the steps, I'd do it as a learning experience. Not sure how that changes anything regarding my understanding and use of the parameters I'd like access to. But I'm willing to keep an open mind.

[Tyler Townsley]

Originally Posted by XfireZ51

Todd,

I appreciate the offer of help. I have not started disassembling code. I have both the .tdf and .xdf editors. I have ported a .tdf over to an .ecu and .xdf file. (Pretty simple since both John and Mark built a function to do just that). I have previously created and modified .xdf items. I prefer working with TP but have used both as well as modifying calibrations directly in hex. I have modified .ads so as to output aldl items that were not supported in the scantool I was using. I've been able to introduce new output to the ALDL datastream (i.e. WB O2). Basically, if I have a hack, I can create the .xdf. What other tools do I need?
If you want to take people like myself, Shawn ZR-1 and =Jeff= through the steps, I'd do it as a learning experience. Not sure how that changes anything regarding my understanding and use of the parameters I'd like access to. But I'm willing to keep an open mind.

I will offer some tips and questions to make sure you understand just what is being changes. The one that bit me first was something called.
'Secondary inj on fuel PW scaler.' Its in TC under a different heading, I used it on Rons car when Jim ran out of ideas, and TP, anyone know what it does and just how you can use it. I know Todd does but it is not to obvious and can have a dramatic difference.

Tyler