2011 CTS-V (70KPA idle) rough / rich idle issues

Hi all,

Trying to help a buddy remotely diagnose an idle issue with a recent build on a 2011 CTS-V.

Specs:
Stock LSA bottom
Injector Dynamics 850 injectors
Texas Speed ported heads
Stage 3 Texas Speed Blower Cam
10% over drive crank pulley
2.55 blower pulley

I referenced the injector data from this post https://forum.hptuners.com/showthrea...l=1#post330042 and from the Injector Dynamic website.

Issue:
Issues with startup and idle. LTFT trims quickly go to -17% on both banks once it goes into closed loop fueling. The MAP is reading 70KPA when it starts up and is idling (reads 99KPA key on not running). As a test we put it into MAF only to bypass the Virtual VE table and took 15% of fuel out of the MAF table and it still idles reading at -17% rich.

Questions:
Is this a potential vacuum leak?
I feel the 70KPA at idle is high, even with a 231/239, .640"/.629? cam
I plan on visiting soon and getting wide band readings for him to try and isolate further

Appreciate any insight on where to point him next, etc. I?ve attached the tune file and log to show the issue.

prob not actual issue but the offset v inj temp table u may have to move the table up a little so the 0.0078 shift all up so that is in the 65c cell and rest above like they are see if it helps (unless tip temp is in the 45c cell running temps then divide the table by 0.6 see if that helps but should be setup right for the new injectors), in the eoit ect table change the 110 to 90 should help with the camshaft a little, makeup pulse minimum change to stock or 0.25 i dont like lowering that too much, u are always using the ve/vve as the maf isnt fast enough for quick transients so it will look up the vve when it needs to so u have to make that right also, also do not even try to go into PE it may not end well needs to be setup PE EQ needs work and throttle cant be 100% get the WB on there

you are missing short pulse adder for ID850s

Not to hijack thread but will increasing idle min airflow in park and neutral lean out idle at all? I had this same problem and reduced neg trims some through short pulse and min airflow mods, but I can’t tell which one effected it or if both did. Still need to do a little more...

Well first, is your bypass actuator connected correctly? The blower can still compress the vacuum behind the throttle, so if the bypass is closed and you're at idle, it will compress that 35kpa into whatever your pressure ratio is, likely somewhere right around 70kpa.

Quote:

---

Originally Posted by ttz06vette

Not to hijack thread but will increasing idle min airflow in park and neutral lean out idle at all? I had this same problem and reduced neg trims some through short pulse and min airflow mods, but I can’t tell which one effected it or if both did. Still need to do a little more...

---

it shouldnt as ur not increasing actual airflow ur increasing the minimum allowed like a boundary but if u increase it too high it will still try to reach commanded idle rpm so will remove timing to do so, short pulse would affect it as its an adder but u shouldnt adjust injector data to change fueling (unless data is wrong) but offsets can be adjusted as they are against shifting factors temp/pressure, so if say injector tip temp offset is wrong it can be adjusted to help, maf, ve/vve for fueling all other tables make those accurate

Quote:

---

Originally Posted by 07GTS

it shouldnt as ur not increasing actual airflow ur increasing the minimum allowed like a boundary but if u increase it too high it will still try to reach commanded idle rpm so will remove timing to do so, short pulse would affect it as its an adder but u shouldnt adjust injector data to change fueling (unless data is wrong) but offsets can be adjusted as they are against shifting factors temp/pressure, so if say injector tip temp offset is wrong it can be adjusted to help, maf, ve/vve for fueling all other tables make those accurate

---

Makes sense. Thanks...I’m close.

Quote:

---

Originally Posted by 07GTS

... u are always using the ve/vve as the maf isnt fast enough for quick transients so it will look up the vve when it needs to so u have to make that right also, also do not even try to go into PE it may not end well needs to be setup PE EQ needs work and throttle cant be 100% get the WB on there

---

1. It was my understanding that by setting the High RPM Disable and Re-enable to something where it would never kick in would effectively bypass the VVE and only use the MAF? Is this not true, or only applicable to certain ECU's?

2. We haven't touched the PE EQ and we set the throttle to 100% to avoid skewing any fuel collecting at part throttle, etc. This would be adjusted before doing any WOT.

Quote:

---

Originally Posted by jsllc

you are missing short pulse adder for ID850s

---

I just double checked I have the data from the excel sheet inputed in the short pulse adder table. Are you seeing another table that's missing data? I'm not seeing one based on the data they provide.

Quote:

---

Originally Posted by smokeshow

Well first, is your bypass actuator connected correctly? The blower can still compress the vacuum behind the throttle, so if the bypass is closed and you're at idle, it will compress that 35kpa into whatever your pressure ratio is, likely somewhere right around 70kpa.

---

I'll have him double check and will check when I go visit. Would you say 70kpa could be normal on some setups? It just seems insanely high based on the setup (but I'm not 100% privy to blower cars).

1. NOT TRUE. This ECM always uses DYNAMIC. NO EXCEPTION. What makes up dynamic is a mystery to all of us. We all have theories. There is no doubt the there are options that bias it towards MAF or VVE or if patched VE but even in a failed MAF state dynamic does not follow VVE or VE perfectly. There are other modifiers.

2. correct filtering during logging takes care of skew.

3. interesting, I have specs from a number of years ago and the numbers are slightly different. Do not know why.

As for the rest. No way this takes 20g/sec base idle air then goes down to 14. An E67 has three map sensors. BARO, After TB and After SC as well as manifold vacuum. Log all 4. The car in the PIC and listed below idles with 35-39kpa of vacuum.

Quote:

---

Originally Posted by jsllc

1. NOT TRUE. This ECM always uses DYNAMIC. NO EXCEPTION. What makes up dynamic is a mystery to all of us. We all have theories. There is no doubt the there are options that bias it towards MAF or VVE or if patched VE but even in a failed MAF state dynamic does not follow VVE or VE perfectly. There are other modifiers.

2. correct filtering during logging takes care of skew.

3. interesting, I have specs from a number of years ago and the numbers are slightly different. Do not know why.

As for the rest. No way this takes 20g/sec base idle air then goes down to 14. An E67 has three map sensors. BARO, After TB and After SC as well as manifold vacuum. Log all 4. The car in the PIC and listed below idles with 35-39kpa of vacuum.

---

Dynamic airflow isn't a mystery lol. Chris has explained it in detail on this forum before. And if you really want the info, the algorithm is patented. It is publicly available on google patents. I get a ton of information about automotive software that way.

If all you do is disable the MAF, sure it won't follow VE/VVE because it still has an active kalman filter. But you can disable that and remove transient airflow altogether... Lots of people do. Any deviations you see otherwise are either temperature bias issues or data reporting latency affecting what you are reading vs what you expect in your cal file. Remember, CAN is slow...by the time all of the info is broadcast/polled for at whatever rate/polling frequency those might be and finally reaches your PC, some of it is out of date. Not to mention physical delays of the hardware. Can't always totally trust it.

How the code is written and math behind it is not on Google. Written code for a living 30 years now well versed in how it works. As for what is displayed. It is only out sync if you choose to make it so. Can bus delays do not affect how the data is iterpolated inside the ecm. Sync issues happen between devices not within. We don't even have all of the parameters exposed. I had over 100 added to hpt 2 years ago for the e40, e38 and e67. Some of those are used in the predictive coefficients for airflow. That math is also not documented nor has anyone reverse engineered it. Just like injector tip temp. Took me 6 months to find and document that one so hpt would add it. You need to do more research. Happy Googling.

Quote:

---

Originally Posted by jsllc

How the code is written and math behind it is not on Google. Written code for a living 30 years now well versed in how it works. As for what is displayed. It is only out sync if you choose to make it so. Can bus delays do not affect how the data is iterpolated inside the ecm. Sync issues happen between devices not within. We don't even have all of the parameters exposed. I had over 100 added to hpt 2 years ago for the e40, e38 and e67. Some of those are used in the predictive coefficients for airflow. That math is also not documented nor has anyone reverse engineered it. Just like injector tip temp. Took me 6 months to find and document that one so hpt would add it. You need to do more research. Happy Googling.

---

It is literally all right here... https://patents.google.com/patent/US20050060084A1/en

I know it is indeed that because I was a software engineer at GM and have been through damn near all of the E67 engine code. Can lead a horse to water...

https://i.ibb.co/SnD6JRx/Capture.png

Quote:

---

Originally Posted by jsllc

How the code is written and math behind it is not on Google. Written code for a living 30 years now well versed in how it works. As for what is displayed. It is only out sync if you choose to make it so. Can bus delays do not affect how the data is iterpolated inside the ecm. Sync issues happen between devices not within. We don't even have all of the parameters exposed. I had over 100 added to hpt 2 years ago for the e40, e38 and e67. Some of those are used in the predictive coefficients for airflow. That math is also not documented nor has anyone reverse engineered it. Just like injector tip temp. Took me 6 months to find and document that one so hpt would add it. You need to do more research. Happy Googling.

---

Sounds like these would have come in handy.
Attachment 101073

Quote:

---

Originally Posted by smokeshow

It is literally all right here... https://patents.google.com/patent/US20050060084A1/en

I know it is indeed that because I was a software engineer at GM and have been through damn near all of the E67 engine code. Can lead a horse to water...

https://i.ibb.co/SnD6JRx/Capture.png

---

You know nothing, Jake from State Farm.

This concludes my yearly login to the HPT forum. Lol.

So now what is the value of weighted???
How is it weighted???
More signals? and if so what are they?
What does schedule estimator correction mean? How is it done or the math behind it.

Useless and if the engineer you claim then you know that. Just enough to fool someone into believing. In the world of software this is known as a block diagram. No details as to the actual workings. Not much different than a road map with only major cites and interstates.

Without the details and math there is no way to do this. Exactly my point thanks for making it.

As useful as this is for tuning.
1. buy some software
2. change some stuff
3. apply changes
4. check changes
5. if changes weren't good go to step 2.
6. done

But if you are who you say. Share all of the info.

Quote:

---

Originally Posted by smokeshow

It is literally all right here... https://patents.google.com/patent/US20050060084A1/en

I know it is indeed that because I was a software engineer at GM and have been through damn near all of the E67 engine code. Can lead a horse to water...

https://i.ibb.co/SnD6JRx/Capture.png

---

So which of these are false????


How the code is written and math behind it is not on Google? (Block diagram is not code)

Can bus delays do not affect how the data is interpolated inside the ecm?
Sync issues happen between devices not within?
We don't even have all of the parameters exposed?
Predictive airflow math is also not documented nor has anyone reverse engineered it?

I mean... he?s not lying about who he is and his background. Can confirm. Lol.

smokeshow,

So if the d**k measuring contest is over I am all ears and ready to see the math and logic. I am ready to learn

1. Prediction Coefficients - We only see values of meaningless numbers : Base, Current, Old, Corrected, MAP, MAP old 1, MAP old 2, TPS, TPS old 1, TPS old2. How are these calculated? How are they used and modifiers. Please be specific and provide the math.

2. Missing parameters - After documenting and finding over 100 others I know for certain there are many we do not have. Please document and describe them. It would be very helpful to offsets in the ECM so we can use user defined parameters.

3. Dynamic Air - There is biasing between MAF and VVE. What is the the math and logic behind that?

4. Virtual TQ - We are missing more here. Please explain the math and logic used to determine the values.

5. Body Control - This module modifies many of the above. Please explain how that is done and under what circumstances. How is this calculated into the ECM modifiers.

This is a great start for those of us who have had to reverse engineer and have not had the opportunity to see all of the code.

You do this for us and you will get a big thank you from everyone.

Quote:

---

Originally Posted by jsllc

smokeshow,

So if the d**k measuring contest is over I am all ears and ready to see the math and logic. I am ready to learn

1. Prediction Coefficients - We only see values of meaningless numbers : Base, Current, Old, Corrected, MAP, MAP old 1, MAP old 2, TPS, TPS old 1, TPS old2. How are these calculated? How are they used and modifiers. Please be specific and provide the math.

2. Missing parameters - After documenting and finding over 100 others I know for certain there are many we do not have. Please document and describe them. It would be very helpful to offsets in the ECM so we can use user defined parameters.

3. Dynamic Air - There is biasing between MAF and VVE. What is the the math and logic behind that?

4. Virtual TQ - We are missing more here. Please explain the math and logic used to determine the values.

5. Body Control - This module modifies many of the above. Please explain how that is done and under what circumstances. How is this calculated into the ECM modifiers.

This is a great start for those of us who have had to reverse engineer and have not had the opportunity to see all of the code.

You do this for us and you will get a big thank you from everyone.

---

If you wish I will build a site where you can document all of this for everyone to share. Take me about a week to get the posting mechanism and limited forum coded. Then we can ask questions and you can be specific with answers. I know twenty to thirty advanced tuners that would love to have all of this information. Just tell me when.

...and a bunch of other engineers like me who would love to truly understand some of the hidden stuff, even though we just tune for ourselves and a few friends.

Of course we all know enough to make it work (and work well) at this point, thanks to guys like Dave and Scott and others. That said, it would be cool to better understand the whole thing.

For example smokeshow, I read in the other thread about DFCO... I certainly didn't know the real reason for that. I always thought fuel savings sounded a little suspect, because seriously- how much could you really save while coasting.

I can tell you it is exhausting answering questions, so if he doesn’t want to, I get it. I bother him, but I also buy him stuff like 91mm turbos to get answers. Lol.

Absolutely. Info usually isn't free, and when you are an expert in something, it certainly gets old explaining it over and over again, especially if you don't really have the time, desire or monetary incentive to do it. :)

I'm just glad you and others are willing to help dudes like me shorten my learning curve at times.