Sponsored

Help with warm start issue - Whipple

OP
OP
khsonic03

khsonic03

Well-Known Member
Joined
Mar 5, 2019
Threads
9
Messages
171
Reaction score
113
Location
Jacksonville, FL
Vehicle(s)
2017 Mustang GT PP / 2016 F150 Lariat / 2018 Jeep Rubicon
This might sound stupid but you’re using an aftermarket IAT sensor right? Eg. Not the one supplied by Whipple.
Without throwing anyone under a bus, one of those ”big two” tuned a mates car and failed to notice that he was still using the Whipple sensor. Your issues sound very much like what he was seeing.
Yeah I'm using the VMP IAT sensor kit that goes into the back of the Whipple (post intercooler).

I can run it by the tuner and make sure they are accounting for it. Thanks for the heads up.
Sponsored

 

Jackson1320

Banned
Banned
Banned
Joined
Sep 26, 2019
Threads
26
Messages
3,057
Reaction score
1,220
Location
California
First Name
Jaxson
Vehicle(s)
2015 mustang gt
Smoke tested the car today. Found one leak at the TB set screw (see video: ). Got excited and thought I had figured it out. Sealed that set screw with some tape and no change. There are absolutely no other leaks. Pumped a ton of smoke in there.

LTFT's are 0 during startup and idle.

Did a step test log and fuel trims are within 2-3% most of the time. Its just startup and idle that is a disaster.

New MAF is due in tomorrow. Also going to try swapping the TB circuit board with the factory one. Will probably have to recalibrate the TPS setting again. That will eliminate MAF and TPS sensor as an issue.

This is getting old. Lol
I know your long term fuel trims are at zero on start up because you are on open loop but when it switches over to close loop where are they at. When you’re driving they have to = +/- five together. Start up is open loop so it is something in the open loop Part of the tune Or one of the few sensors that are used during open loop or a failed part
 
OP
OP
khsonic03

khsonic03

Well-Known Member
Joined
Mar 5, 2019
Threads
9
Messages
171
Reaction score
113
Location
Jacksonville, FL
Vehicle(s)
2017 Mustang GT PP / 2016 F150 Lariat / 2018 Jeep Rubicon
I know your long term fuel trims are at zero on start up because you are on open loop but when it switches over to close loop where are they at. When you’re driving they have to = +/- five together. Start up is open loop so it is something in the open loop Part of the tune Or one of the few sensors that are used during open loop or a failed part
Yes, I was watching the LTFT during open loop and then when it went into closed loop. Showed 0 the whole time.

MAF shows up tomorrow. Will see what happens then go from there.
 

ugstang17

Well-Known Member
Joined
Oct 26, 2018
Threads
9
Messages
1,329
Reaction score
694
Location
unknown
Vehicle(s)
none
A vacuum/boost or looking at your vacuum reading in the gauges section of your instrument panel will be needed. If you do not have a vac/boost gauge get one. An FI setup needs one.

Appears to be a lean start in the log. Been a while since I logged and looked at one so the acronyms are a bit hazy. What is your vacuum reading at idle on cold start as compared to hot start? About 17" - 19" of Hg is average on a stock cam setup at idle. If this is closer to zero when idling warm vs cold or even all the time then you may have a vacuum leak after warmup. The AFR is lean at startup and while it settles when the O2's come on line this is only because the STFT's are correcting massively for it. WHat I do not see (this tuner who I know does not monitor LTFT for some reason when they tune...the other one does) is LTFT's. If you could see LTFT's you would likely see that they are also high and trying to correct as well.....assuming a vacuum leak is present.

Datalog and increase throttle and observe STFTs at 2000, 3000, 4000RPM. Note if as your vacuum becomes more positive if the STFT's get smaller.

If you are using an ngauge set it up to monitor LTFT's on the gauge settings. If using an SCT device created a datalog cfg file to monitor LTFT1 LTFT2 STFT1 STFT2 LAMBS1 LAMBS2 AFRbank1, AFRbank2, THROTTLE POSITION, and RPM. Allow the car to idle and monitor at startup. save that datalog and then create another with the car still running and perform a ramped throttle up through about 4000 to 4500 RPM in 500 step increments. Hold each position steady for abotu 10 seconds.

If there is a vacuum LTFT will begin above 1.00 on both banks (if the leak is common to both) and will then drop closer to 1.00 (0 on your ngauge LTFT reading on the display but 1.00 in the log) as you increase in throttle position. When you let off the throttle, STFT and AFR will rise and then start to settle back to normal as LTFT begins to increase again. Now all you have to do is find the vacuum leak.


My guess is that when you disconnected your battery adaptive learning was lost (LTFT data) and now the car is relearning again. Until you can verify you have no vacuum leak you cannot expect the tuner to resolve this issue as it is a mechanical issue and not a tuning issue...assuming it is vacuum related. The LTFT's and vacuum gauge will tell you more.

In my case adaptive learning is disabled. So LTFT always reads zero. That is what Lund chose to do (or they simply forgot to turn it back on before completing the tuning process but the car runs fine). I suspect LTFT is enabled (adaptive learning active) in your case because of whaty happened after you disconnected your battery. Good luck.
 
OP
OP
khsonic03

khsonic03

Well-Known Member
Joined
Mar 5, 2019
Threads
9
Messages
171
Reaction score
113
Location
Jacksonville, FL
Vehicle(s)
2017 Mustang GT PP / 2016 F150 Lariat / 2018 Jeep Rubicon
A vacuum/boost or looking at your vacuum reading in the gauges section of your instrument panel will be needed. If you do not have a vac/boost gauge get one. An FI setup needs one.

Appears to be a lean start in the log. Been a while since I logged and looked at one so the acronyms are a bit hazy. What is your vacuum reading at idle on cold start as compared to hot start? About 17" - 19" of Hg is average on a stock cam setup at idle. If this is closer to zero when idling warm vs cold or even all the time then you may have a vacuum leak after warmup. The AFR is lean at startup and while it settles when the O2's come on line this is only because the STFT's are correcting massively for it. WHat I do not see (this tuner who I know does not monitor LTFT for some reason when they tune...the other one does) is LTFT's. If you could see LTFT's you would likely see that they are also high and trying to correct as well.....assuming a vacuum leak is present.

Datalog and increase throttle and observe STFTs at 2000, 3000, 4000RPM. Note if as your vacuum becomes more positive if the STFT's get smaller.

If you are using an ngauge set it up to monitor LTFT's on the gauge settings. If using an SCT device created a datalog cfg file to monitor LTFT1 LTFT2 STFT1 STFT2 LAMBS1 LAMBS2 AFRbank1, AFRbank2, THROTTLE POSITION, and RPM. Allow the car to idle and monitor at startup. save that datalog and then create another with the car still running and perform a ramped throttle up through about 4000 to 4500 RPM in 500 step increments. Hold each position steady for abotu 10 seconds.

If there is a vacuum LTFT will begin above 1.00 on both banks (if the leak is common to both) and will then drop closer to 1.00 (0 on your ngauge LTFT reading on the display but 1.00 in the log) as you increase in throttle position. When you let off the throttle, STFT and AFR will rise and then start to settle back to normal as LTFT begins to increase again. Now all you have to do is find the vacuum leak.


My guess is that when you disconnected your battery adaptive learning was lost (LTFT data) and now the car is relearning again. Until you can verify you have no vacuum leak you cannot expect the tuner to resolve this issue as it is a mechanical issue and not a tuning issue...assuming it is vacuum related. The LTFT's and vacuum gauge will tell you more.

In my case adaptive learning is disabled. So LTFT always reads zero. That is what Lund chose to do (or they simply forgot to turn it back on before completing the tuning process but the car runs fine). I suspect LTFT is enabled (adaptive learning active) in your case because of whaty happened after you disconnected your battery. Good luck.
Thanks for the reply and suggestions. I will take some more logs with LTFT in them and try the tests you mentioned above. I did note today that my LTFT were 0 on the ngauge during startup and idle (closed loop).

My vacuum at idle is a little over 21hg. I will compare cold to warm.

I did a smoke test today and found a small leak on the TB set screw. Sealed it and didn't appear to change the STFT at all. There are no other vacuum leaks.

I am still not totally convinced the TB settings aren't playing a part in this. I am within whipple spec for TP2 voltage, but wondering if something is not right with the TPS.
 

Sponsored

Jackson1320

Banned
Banned
Banned
Joined
Sep 26, 2019
Threads
26
Messages
3,057
Reaction score
1,220
Location
California
First Name
Jaxson
Vehicle(s)
2015 mustang gt
Thanks for the reply and suggestions. I will take some more logs with LTFT in them and try the tests you mentioned above. I did note today that my LTFT were 0 on the ngauge during startup and idle (closed loop).

My vacuum at idle is a little over 21hg. I will compare cold to warm.

I did a smoke test today and found a small leak on the TB set screw. Sealed it and didn't appear to change the STFT at all. There are no other vacuum leaks.

I am still not totally convinced the TB settings aren't playing a part in this. I am within whipple spec for TP2 voltage, but wondering if something is not right with the TPS.
I’m sure your Ltft’s are turned off. I have never seen a car at zero and not move at all.
 

Jackson1320

Banned
Banned
Banned
Joined
Sep 26, 2019
Threads
26
Messages
3,057
Reaction score
1,220
Location
California
First Name
Jaxson
Vehicle(s)
2015 mustang gt
Also your battery is considered discharged at 12.3 and charged at 12.6 without surface charge. Check it in the morning after sitting all night. It should be 12.6 or 12.5 no lower. Yours shows 12.0 that’s very low. Ether you used a lot of juice before you started logging or your battery is on its way out
 
OP
OP
khsonic03

khsonic03

Well-Known Member
Joined
Mar 5, 2019
Threads
9
Messages
171
Reaction score
113
Location
Jacksonville, FL
Vehicle(s)
2017 Mustang GT PP / 2016 F150 Lariat / 2018 Jeep Rubicon
Also your battery is considered discharged at 12.3 and charged at 12.6 without surface charge. Check it in the morning after sitting all night. It should be 12.6 or 12.5 no lower. Yours shows 12.0 that’s very low. Ether you used a lot of juice before you started logging or your battery is on its way out
Just checked it this morning - 12.55. I was doing a lot of starting and stopping when I took that log.
 

ugstang17

Well-Known Member
Joined
Oct 26, 2018
Threads
9
Messages
1,329
Reaction score
694
Location
unknown
Vehicle(s)
none
Not an expert, but it may be possible that since it does appear that they have adaptive learning turned off (LTFT's read zero) that they made table corrections to the tune while residual data in the PCM from LTFT's was still present - if that is possible. If this is possible and my theory is correct this would cause you to not have problems until a complete memory clear was performed which occurreed when you disconnected that battery which cauased a complete reset. So you may now need to send new datalogs back to Rob and explain what has happened so he could confirm this and then recorrect what he changed earlier.....if they choose to tune without adaptive learning active. Otherwise he needs to turn LTFT back on so that things will settle out with the current changes he made.

That is my opinion. I would think that a complete mem clear would occur at PCM load when a new tune is uploaded. If not then this may be the source of your issue, as you have verified that the vacuum is good at idle. I seriously doubt with 21" Hg at idle you will have a leak after warmup...but it could happen. Hot vacuum leaks usually tend to occur after the compressor on PD setups on the intakes or around the o-rings on the FI's.

I earlier referred to LAMBA1, 2. I meant LAMBSE1 LAMBSE2 (something like that) which is the target value the tune is telling the PCM to target and maintain which will be seen in AFRbank 1 nad AFR bank2 respectively. THe STFT is derived b y the difference between those two values IIRC. When LTFT is active it will change at a given point is STFT stays off by more than 7-9% (or about) for an extended period of time so that DTFT will then return within tolerance. IF LTFT goes to high that is when CEL lights start happening.

Sorry to ramble. I am thinking out load and trying to recall and refresh my thought process at the same time. I studied this several years ago when I was tuning a 14 GT PD setup with Alex (now at Lund) when he was still at VMP. It was an excellent experience and I wish I could do the same again with him at Lund but they are much more email oriented with customers on tuning rather than working over the phone....which worked very well because I could correct him if he was not understanding what I was conveying if my terminology was wrong or off. Plus I was able to learn from him and verify some of my theory basics I learned from reading a book he reccommended on tuning basics.

Good luck.
 
OP
OP
khsonic03

khsonic03

Well-Known Member
Joined
Mar 5, 2019
Threads
9
Messages
171
Reaction score
113
Location
Jacksonville, FL
Vehicle(s)
2017 Mustang GT PP / 2016 F150 Lariat / 2018 Jeep Rubicon
Not an expert, but it may be possible that since it does appear that they have adaptive learning turned off (LTFT's read zero) that they made table corrections to the tune while residual data in the PCM from LTFT's was still present - if that is possible. If this is possible and my theory is correct this would cause you to not have problems until a complete memory clear was performed which occurreed when you disconnected that battery which cauased a complete reset. So you may now need to send new datalogs back to Rob and explain what has happened so he could confirm this and then recorrect what he changed earlier.....if they choose to tune without adaptive learning active. Otherwise he needs to turn LTFT back on so that things will settle out with the current changes he made.

That is my opinion. I would think that a complete mem clear would occur at PCM load when a new tune is uploaded. If not then this may be the source of your issue, as you have verified that the vacuum is good at idle. I seriously doubt with 21" Hg at idle you will have a leak after warmup...but it could happen. Hot vacuum leaks usually tend to occur after the compressor on PD setups on the intakes or around the o-rings on the FI's.

I earlier referred to LAMBA1, 2. I meant LAMBSE1 LAMBSE2 (something like that) which is the target value the tune is telling the PCM to target and maintain which will be seen in AFRbank 1 nad AFR bank2 respectively. THe STFT is derived b y the difference between those two values IIRC. When LTFT is active it will change at a given point is STFT stays off by more than 7-9% (or about) for an extended period of time so that DTFT will then return within tolerance. IF LTFT goes to high that is when CEL lights start happening.

Sorry to ramble. I am thinking out load and trying to recall and refresh my thought process at the same time. I studied this several years ago when I was tuning a 14 GT PD setup with Alex (now at Lund) when he was still at VMP. It was an excellent experience and I wish I could do the same again with him at Lund but they are much more email oriented with customers on tuning rather than working over the phone....which worked very well because I could correct him if he was not understanding what I was conveying if my terminology was wrong or off. Plus I was able to learn from him and verify some of my theory basics I learned from reading a book he reccommended on tuning basics.

Good luck.
Thanks man. Appreciate the thought you are putting into this. I tried messing with the TP2 voltage more on the throttle body. No luck there. Also got a new tune from the tuner and that made no difference either.

I did pay attention to the vacuum and it was around 21.7.

In regard to your first theory - are you saying from the get-go he may have tuned off residual data in the PCM? The startup issue has been a problem since day 1. Even when I'm going back and looking at some of my original logs, the STFT's were always way off at startup and idle. I feel like the car adapted some over time and that's why the problem started to fade. Not sure I fully understand how though, since LTFTs seem to be turned off. I'm sure there's something I'm not understanding there.

If I had a vacuum leak from an injector o-ring or an intake manifold gasket, I'm assuming I'd see a difference between bank 1 and bank 2 since it would most likely impact one cylinder. However, both of my banks are fairly consistent, so if it were a vacuum leak, I'd think it'd be on the top end somewhere. I've thoroughly smoke tested the engine and there is definitely no vacuum leak. So thorough, that I found that tiny leak from the Whipple TB set screw (which Whipple didn't even believe was possible when I told them).

Here's where I start getting out of my league - is it possible that the fuel tables are not scaled properly and that's why at startup and idle I don't have enough fuel, but in higher RPMs, everything is fine? It truly seems like I'm not getting enough fuel at startup and idle (or I'm getting too much air).

Lastly, does anyone thing this could be a crank relearn issue? I still haven't done one yet, as I haven't touched the clutch/flywheel. I've done OPG, but if my timing was off, I'd know it. It only takes one tooth to throw a misfire code.

Still have more tests to do, just haven't had a ton of time yet.
 

Sponsored

ugstang17

Well-Known Member
Joined
Oct 26, 2018
Threads
9
Messages
1,329
Reaction score
694
Location
unknown
Vehicle(s)
none
Do the crank relearn just to get it off the table. I highly doubt there is an issue but at least by doing that you will rule out the people who think that is the end-all-be-all answer to every problem on a Coyote. Usually crank relearn related issues generate generate misfire issues. You aren't experiencing this. But again to be certain and rule it out...do it. Then you can move on. ITs a simple troubleshooting process of eliminating distracters.

In response to your question - Yes. And again I am theorizing - (would like a pro tech or tuner to chime in here in case I am in left field and correct me if I am wrong) - that if its possible residual data may have been in the PCM that the tuner was unknowingly tuning to, when you disconnected the battery and reset everything that data is now gone and won't be back because LTFT is not active so the STFT's will now be different. Again it is theoretical but holds water based on the fact that you state the idle problem worsened after the battery disconnect period.

My tune has LTFT shut off. At start up my STFT's will reach 17-19 adding fuel and then settle out as the idle settles and the O2's come on line. I am running a 2300 TVS setup with a twin 67mm TB. After warmup my STFT's at idle are anywhere from 0-7. Cruising they are 0-2.

Your car has no vacuum leak. We've answered that. And yes I agree that AFR bank 1 or bank 2 would be extremely leaner than the other if there was a leak at an FI o-ring causing a lean condition. So you can assure yourself and the tuner that the integrity of the engine is solid in that aspect. So you are back to either tuning or a very touchy TB.

I won't speculate on tables. I started to and then edited them. plenty can come into play on the idle process. But of the tables I would be least inclined to think that fueling tables would be off. Assuming you are running what Whipple provided or something better like ID 1050X's which are the cadillac brand and have some of the the best data and flow matching in the industry fuel table s would be pretty far down the list.

I suggest again sending the logs back to the tuner(s) review them. Explain what has taken place. Then ask...don't request....but ask if increasing hte RPM to 850 at idle or even 900 would help gain better control of this loop with the larger TB. Lund standardly tunes to 850ish idle on all S550 tunes in FI applications according to John and Alex. I am one ofhte few who have a Lund tune that actually idles at 750-800 and it took my tuner (who is a Lund distributor) and John Jr days to get the idle stable in all conditions including AC on/off to not surge or die at idle. That said as I mentioned above the STFT's are not 0-2. THey can peek up to 11 or so depending on air temps and so forth but they don't hang there. They usually average about 7-9 at idle after warmup but are much higher at cold start. THere can even be a little surge at cold start up to about 1500 RPM but it settles out after 2-3 minutes to about 900 and then drops to 750-800 at normal operating temps.

Your only other option would be to consider a smaller TB if you seek street manners better than what you have. I am certain I am about to be roasted here by 132mm TB users in the Whipple world and I am certain that there are those who have no issues with their setups, but not all people are as equally picky about cold starts and idle as others.

One last thing I just thought about. If oil has migrated into the intercooler and/or intake plenum area and has puddled up this can cause issues as well. Without a good catch can system this can easily happen will also make the idle tuning process more difficult. So you may wish to put the catch cans back on AND do an inspection even if it means pulling the head unit to inspect that intake plenum area and IC unit from oil migration. Then get that catch can back on. THe passenger side is notorious for sucking in oil even on modified N/A setups I have come to learn.

Best of luck to you. Sorry I have nothing else to contribute.
 
Last edited:

beetle6986

Well-Known Member
Joined
Apr 1, 2020
Threads
1
Messages
322
Reaction score
197
Location
Harrisburg
Vehicle(s)
16 Mustang GT, 04 Cobra, 2008 BMW 335
I noticed my car doing the exact same thing on warm start. Starts up, then idle drops to around 500 for a second (Almost like it stumbles) and then it settles. The only recent changes within the last few hundred miles is an aftermarket tune and 3 ounces of boostane in the tank I only ran one tank with boostane. I have a different tuner than you. My car is stock other than the phase 2 and Borla exhaust.
I don’t recall the car doing that before, but i’ve only put 1,000 miles on it. So i’m not 100% sure
 

kbreeze72

Well-Known Member
Joined
Oct 23, 2021
Threads
15
Messages
138
Reaction score
117
Location
Eastern Long Island, NY
Vehicle(s)
2018 F150 5.0 & 2021 C8 Z51
Did you ever resolve this mystery? I've read this far, I need a happy ending! lol.
 

19BULLITTwhipple

Well-Known Member
Joined
Mar 23, 2021
Threads
24
Messages
190
Reaction score
81
Location
California
First Name
Ramzi
Vehicle(s)
Bullitt 2019
I feel like someone should have mentioned that the boost-a-pump might not have a check valve. Which makes warm starts an issue. I had this problem on my Whipple until I got a fuel system. Chances are the issue is mechanical.

Screen Shot 2022-01-23 at 6.00.43 AM.png
 
OP
OP
khsonic03

khsonic03

Well-Known Member
Joined
Mar 5, 2019
Threads
9
Messages
171
Reaction score
113
Location
Jacksonville, FL
Vehicle(s)
2017 Mustang GT PP / 2016 F150 Lariat / 2018 Jeep Rubicon
Did you ever resolve this mystery? I've read this far, I need a happy ending! lol.
Sort of lol.

The engine was smoke tested multiple times and the only leakage was around the Whipple TB set screw. Got a new TB and there was no change.

The tuner made a few adjustments that finally seemed to settle it out. I would say 85% there.

I did read in another thread where intake alterations can affect idle/startup due to MAF turbulence. I have a heavily altered intake opening due to my aftermarket bumper. I intend on trying to block it off and see if that helps get me the rest of the way there, but haven't had a chance to play around with it yet.

I'll post an update when I finally get around to it, but as of now, the car is not experiencing the issues to the same degree that it was when I posted this, so not as much of a priority at this point.
Sponsored

 
 




Top