Page 1 of 2 12 LastLast
Results 1 to 20 of 33

Thread: Intermittent Hard Starts (LS Swapped Mazda RX8, 2004 GTO Donor)

  1. #1
    Tuner in Training
    Join Date
    Jul 2019
    Posts
    19

    Intermittent Hard Starts (LS Swapped Mazda RX8, 2004 GTO Donor)

    Whats up HPTuners community!

    To be honest I have done a shit ton of research on this issue, but have yet to find a similar scenario to my current problem. To give some background I have a 2006 Mazda RX8 that I put a LS1/T56 into from a 2004 Pontiac GTO. The car runs great, however I have been fighting this intermittent hard start issue. Sometimes you hop in it fires right up and your good to go, other times it take a few times and extended crank time to get it up and running. Once it starts though it idles fine, and drives without any noticeable issues. Hasn't thrown any codes either. Initially for the first several months (not exaggerating, I have been fighting this since Jan and am at my wits end) I assumed it was hardware related so I went through the following:

    -Replaced all coil packs (thinking one may be intermittently not working)
    -Rebuilt Fuel injectors (replaced the o-rings and filters thinking 1 may be leaking by, and verified before and after that they were not leaking when pressurized to 58psi)
    -Plugs are brand new (thinking maybe 1 is fouled or something)
    -Plug Wires are new (thinking maybe 1 is cut/compromised causing issues)
    -Replaced Crank sensor (cheap and simple, and lined up with a potential cause of hard starts)
    -Replaced Cam Sensor (again cheap and simple, and lined up with a potential cause of hard start)
    -Replaced Fuel pressure regulator (was feeling good about this one, since the current FPR was indeed bad, but now i get 58PSI on prime and through cranking, and start problem still exists)

    At this point, there really isn't more left to replace/check from a hardware standpoint, not to mention it would be odd to have a hardware problem, and it not affect the car actually running/driving and only it starting (in my opinion). This leads me to believe that the intermittent hard start problem I have been chasing may be tune related. Some background on the tune.....I swallowed my pride, and took it somewhere to be tuned so that it could be done right (and paid for it too....jesus) thinking that the hard start problem was tune related, and I was told (yep shes good to go now) but that was bullshit, and it still has the exact same intermittent hard start problem it did prior to the tune.

    I was hoping with all the data/background I have provided, and the logs I have attached that you could take a look and see if anything jumps out at you in the tune or the logs that I can try to finally resolve this issue. I have looked at them a thousand times and don't know where to start (no pun intended) to try and solve an issue like this.

    To further clarify, the engine is a stock LS1, pulled right out of a 2004 GTO. Has a custom CAI, headers to a 3 in exhaust, and the 5 wire MAF (with the IAT built in), but that is it. No cam, or aftermarket internals, ect.

    As for the files...."True_Street_Tune" is the dyno tune I paid for, "TST_Hard_Start" is the True street tune, but i played with the fuel pressure delay thinking that it was telling it to fuel prior to the fuel pressure being built yet (didnt help). "Start" log is one of the times when I just hop in, it starts right up no issues. "Hard_Start" log is another time i hop in, and it takes a few times to get it fired up.

    I know this is a big post, and alot of info, but I wanted to be respectful of everyone's times, and try to provide as much detail, and data as I could. Thanks in advance for any help/guidance.
    Attached Files Attached Files

  2. #2
    Tuner in Training
    Join Date
    Jul 2019
    Posts
    19
    I really have 3 options for starting problems right? You have the cranking VE table on the air side, the cranking fuel table on the fuel side, and the startup/cranking spark table on the spark side. Engine needs air, fuel, and spark to fire off.

    I've seen that people just adjust the cranking fuel table down 10% at the temperature they are having problems at and viola its fixed, I guess I will start there and see what happens. If no change, then set it back, and move to spark, rinse and repeat for air as well.

  3. #3
    I had a similar problem. I replaced the fuel pump and that fixed it. I think the check valve on the fuel pump was failing. Sometimes priming the key 3 times helped, but eventually i had 100% long cranks.

  4. #4
    Tuner in Training
    Join Date
    Jul 2019
    Posts
    19
    Quote Originally Posted by Just Chuck View Post
    I had a similar problem. I replaced the fuel pump and that fixed it. I think the check valve on the fuel pump was failing. Sometimes priming the key 3 times helped, but eventually i had 100% long cranks.
    If the pump was failing or bad then I would think I would see issues with Fuel pressure, but its rock solid at 58psi on startup everytime.

  5. #5
    Its not the pump itself failing. There is a check valve that holds the pressure even when the its not running. So if its leaking after you key on, the pump runs for 2 seconds then shut off then If the pressure bleeds off too quickly, the pump has re prime itself while you are cranking.

    I guess a good test would be to put a fuel pressure gauge and see what it does after you key on ( don't crank it). If you don't hold 58psi after the pump primes, that is a problem.

  6. #6
    Tuner in Training
    Join Date
    Jul 2019
    Posts
    19
    Quote Originally Posted by Just Chuck View Post
    Its not the pump itself failing. There is a check valve that holds the pressure even when the its not running. So if its leaking after you key on, the pump runs for 2 seconds then shut off then If the pressure bleeds off too quickly, the pump has re prime itself while you are cranking.

    I guess a good test would be to put a fuel pressure gauge and see what it does after you key on ( don't crank it). If you don't hold 58psi after the pump primes, that is a problem.
    Sorry maybe I wasn't clear enough in my original post, but this has been checked a few times to say the least. My initial thought was this was a hardware issue, not a tuning issue so the fuel pressure has been verified. I recorded videos of a close up on the gauge on my Aeromotive FPR and it primes to 58 and holds before, during and after cranking.

  7. #7
    Advanced Tuner
    Join Date
    Jun 2015
    Posts
    253
    I would change your stoich afr under Fuel then General to 14.7 if you are running pump gas. You should also log injector pulse width for both banks and short term idle trims

  8. #8
    Tuner in Training
    Join Date
    Jul 2019
    Posts
    19
    Quote Originally Posted by ryan_axberg View Post
    I would change your stoich afr under Fuel then General to 14.7 if you are running pump gas. You should also log injector pulse width for both banks and short term idle trims
    I am running normal 93 pump gas yes. Any idea why the shop that tuned it would tune it for ~14.0 like it shows instead of targeting 14.7? I added injector PW for both banks and short term idle to my layout on scanner software now.
    Will make a few tweaks as suggested, and re-log to see what she does. Thanks for the input.

  9. #9
    Tuner in Training
    Join Date
    Jul 2019
    Posts
    19
    Quote Originally Posted by ryan_axberg View Post
    I would change your stoich afr under Fuel then General to 14.7 if you are running pump gas. You should also log injector pulse width for both banks and short term idle trims
    I am willing to try anything so I will try this, this weekend and see what happens, but I thought about it, and if I do this essentially I will be starting over right? Meaning all the VE/MAF tuning that was done at the previous AFR will be dead, and I will have to get it re-tuned for an AFR of 14.7 now, right?

  10. #10
    Tuner in Training
    Join Date
    Jul 2019
    Posts
    19
    Ok, so I have an update and what I think is good news.....I have been just racking my brain against this problem for a while now so I decided to take another step back, and start checking things I wouldnt normally think about that I did during the swap. One being the purge solenoid valve. I deleted this, and I know it was tuned out, but when I went to physically look at in on the car I realized that I plugged the intake manifold port where it normally plugs in, but then back at the fuel tank on the charcoal canister instead of also capping/plugging that I just put a little filter on the outlet. I sat there and stared at it then it dawned on me....this is essentially a little vacuum leak. It can allow un-metered air into the fuel system causing a "lean" condition. Since this was like that when it was tuned (not knowing until now)...I'm wondering if that is why a richer AFR was targeted (14:1) as opposed to (14.7:1). Would that make sense? If so then maybe it is worth changing the AFR to 14.7 and seeing what happens?

  11. #11
    Senior Tuner kingtal0n's Avatar
    Join Date
    Feb 2017
    Location
    miami
    Posts
    1,799
    I used to have a hard start when the car was hot in the sun. Turned out the fuel lines would empty and need a complete re-prime. Just like somebody up there said.

    If you have a fuel pressure gauge inside the car to watch while you crank, then fine rule it out. But if your fuel pressure gauge is in the engine bay like most, and you are just assuming its not a fuel pressure issue... you might be surprised.

    To fix this issue I put a little fuel pump prime button in my car on a 555 timer so it runs for 3.5~ seconds to prime the system so I dont have to key on and off three times when the car is hot to get the fuel pressure back. I also have a couple fuel cut-off switches that are easy to access from the front and rear of the vehicle in case it ever rolls. Just be careful with anything fuel related that you decide to compensate for with electronics.

    Besides all that. The other thing that made a huge difference was IAC park position. Turns out the factory has it set to reduce airflow when IAT is high, which can make it easy to flood the engine.
    I now have mine set to open even more when IAT goes up, so I can keep roughly the same fuel pulse and not worry about having to fine tune the cranking fuel down to reasonable amount for hot engine starts.

  12. #12
    Tuner in Training
    Join Date
    Jul 2019
    Posts
    19
    Quote Originally Posted by kingtal0n View Post
    I used to have a hard start when the car was hot in the sun. Turned out the fuel lines would empty and need a complete re-prime. Just like somebody up there said.

    If you have a fuel pressure gauge inside the car to watch while you crank, then fine rule it out. But if your fuel pressure gauge is in the engine bay like most, and you are just assuming its not a fuel pressure issue... you might be surprised.

    To fix this issue I put a little fuel pump prime button in my car on a 555 timer so it runs for 3.5~ seconds to prime the system so I dont have to key on and off three times when the car is hot to get the fuel pressure back. I also have a couple fuel cut-off switches that are easy to access from the front and rear of the vehicle in case it ever rolls. Just be careful with anything fuel related that you decide to compensate for with electronics.

    Besides all that. The other thing that made a huge difference was IAC park position. Turns out the factory has it set to reduce airflow when IAT is high, which can make it easy to flood the engine.
    I now have mine set to open even more when IAT goes up, so I can keep roughly the same fuel pulse and not worry about having to fine tune the cranking fuel down to reasonable amount for hot engine starts.
    I understand this is a very common fail-mode for the symptoms I am seeing, and I tested this by setting up a go-pro that is fixed on the fuel pressure gauge while I am cranking. I have reviewed 20-30 videos of both good starts and hard starts, all of them show the same thing. Key on, prime, ~58 PSI and it stays through cranking. I agree never say never, but from the data I have everything on the fuel pressure side seems to be consistent, and correct.

    The second point you bring up is interesting. I looked at both the stock tune and the one I currently am running from the tuner, and yes, the IAC park position does go down as you get higher in ECT. This made me think that it may be starving for air when trying to crank. Especially since I have a custom CAI that is physically longer then most if I had to guess. I started researching this more, and it seems 2.0-2.25 g/s is what a lot of cars like in the "startup airflow initial" table.

    I have made that change, as well as adding 2 degrees across the board to my cranking spark, and the car fired right up at 150 degrees ECT. The real test will be tomorrow. I will start it cold, then go up to ~185 or max temp for me, then try it at 185 and see what happens.

  13. #13
    Tuner in Training
    Join Date
    Jul 2019
    Posts
    19
    Still not getting anywhere regardless of the changes I've made. Below is the "word vomit" that I have been documenting to try and see if something jumps out that is common with the good or the bad starts, but as you can see there are good starts and bad starts with all conditions being the same.
    Testing

    5/21/20

    ? Reduced Cranking FA table by 10% on ECT temps up to 212, and down to 12 (made it leaner)

    o Had 2 back to back good cold starts

    o Hot start was bad at fedex (~185-190 degrees)

    ? Changed timing (+) to 10 from 140 degrees ECT to 284 degrees ECT on cranking spark table

    o Hot Start bad (~180 degrees)

    ? Changed all values from 0-200 RPM to 0.80 (+) on cranking VE table (made it richer)

    o Hot start good (~169 degrees)

    o Cool start bad (~117 degrees)

    5/22/20

    ? Removed all previous changes, and went back to TST tune

    o Then reduced cranking ve by 20% for 0-800rpm on 80-100 Kpa cells (made it leaner)

    o Also reduced primary ve by 20% for 400/800 on 85-105 Kpa cells (made it leaner)

    o Finally added 2 degrees of timing to the cranking spark table

    ? Would not start cold at all.

    5/26/20

    ? Removed all previous changes and went back to TST tune

    o Reduced Crank FA table by 10% for ECTs of 14-86 degrees (made it leaner)

    o Increased Crank FA table by 18% for ECTs of 122-284 (made it richer)

    o Finally added 2 degrees of timing to the cranking spark

    ? Cold start was bad (75 degrees), but forgot to hook PCV back up

    ? Hooked PCV back up and it fired right up (100 degrees)

    ? Next day it fired right up again cold (75 degrees)

    ? Then I plugged the charcoal canister where before it was open to atmosphere (through a filter)

    o Car would not start at all ~cold/cool (too lean now?)

    ? Then I reverted back to TST tune (richer)

    o I was able to get it to fire at ~90 degrees and 165 degrees, but the 90 degree start took 3 tries

    5/27/20

    ? Changed nothing, and cold start was good, hot start was bad.

    o Added 2 degrees to the cranking spark

    ? 150 degree start was bad (38 lb/h startup airflow)

    o Changed everything from 104 degrees up on IAC park position table to 20 g/s

    ? 185 degree start was good (25 lb/h startup airflow)

    ? Next hot start ~1.5 hours later was bad (25 lb/h startup airflow)

    ? Changed IAC park position back so now I am running the TST base tune, with a 2 sec FP delay, and 2 degrees more cranking spark.

    ? Changed all values in the startup airflow initial table to 2 g/s

    o 150 degree start was good (48 lb/h startup airflow)

    5/28/20

    ? Car would not start at cold temp (82 degrees)

    o Set 68 and 104 degree cells back to 1.0 g/s from 2.0 g/s

    o Car started, but 2 took two cranks

    o Car started at 185, but also took two cranks

    o Car started at 194 (immediate restart)

    ***notes***

    ? ECT, Idle Startup Airflow, Start Outcome


    ? 75, 50, bad (started but had to crank twice)

    ? 75, 50, good

    ? 75, 50, good

    ? 82, 58, real bad (no start)

    ? 82, 50, bad (started but had to crank twice)


    ? 163, 32, good (3 min wait)

    ? 151, 38, bad (2 hr wait)

    ? 151, 48, good (2 hr wait)


    ? 189, 25, bad (10 min wait)

    ? 194, 24, good (only 5 min wait)

    ? 187, 41, bad (10 min wait)

    ? 198, 40, good (only 2 min wait)

    I assume its the engineer in me that doesn't allow me to just throw money at a problem, and let the data drive my decision, but at this point I cant help but to keep thinking about 1 thing. My injectors....I have pulled them a number of times out of the intake, but still in the rail, and pressurized the line from priming the fuel pump a few times, and have never seen anything leak out of any of them. I figured this would rule out a "leaky" injector causing this issue, but on so many threads I have seem something similar to this, and that is what it ended up being the problem. I guess my question is, has anyone done this same test I explained, not got the injectors to leak, then replaced them and the problem went away? If so I assume that means that they only leak at certain times (i.e. when hot)? Has anyone see this, where they wont leak "cold", but once they are warmed up they leak when shutoff? Another thing I keep thinking about it when I bought the motor I don't know how long it sat prior to me getting it, then it sat for a few years before the swap was done. Will these stock 28# bosch injectors just go bad from extended times of non-use?
    Attached Files Attached Files

  14. #14
    Tuner in Training
    Join Date
    Jul 2019
    Posts
    19
    Don't think its the problem, but i have decided to pull the trigger on the injectors. We will see if the issue goes away or remains.

  15. #15
    Advanced Tuner
    Join Date
    Apr 2019
    Location
    Parts store
    Posts
    476
    What mods if any have been made?

  16. #16
    Tuner in Training
    Join Date
    Jul 2019
    Posts
    19
    Quote Originally Posted by Hondaeater View Post
    What mods if any have been made?
    None its a stock 2004 LS1 other than longtube headers and exhaust, and a custom CAI.

    I got the injectors in the other day, just hadn't had time to actually test it yet, but my fingers are crossed.

  17. #17
    Advanced Tuner
    Join Date
    Apr 2019
    Location
    Parts store
    Posts
    476
    If it was stock you really shouldn't need to change all that stuff? Headers and a CAI would require calibration of VE and MAF tables.
    Hope you get it figured out. Have fun

  18. #18
    Senior Tuner Lakegoat's Avatar
    Join Date
    Jul 2012
    Location
    Florida
    Posts
    1,424
    Your tuner used 14.1 as stoich because all pump gas has corn juice and is approx. 14.1.
    2000 Camaro SS 2015 L83 port injected, Whipple 3.0, 4L80E, 8.8 Ford
    2013 Silverado 5.3, 6L80k 8.8

  19. #19
    Advanced Tuner
    Join Date
    Jun 2015
    Posts
    253
    I think that would mess him up though because even if the real stoich is 14.1 if he has a little ethanol mixed with the gas his gauge wouldnt read that as stoich. Look up Jay's Tech Tip #35 on youtube. He says that widebands are set up thinking that most people are running pump gas and even if you use e85 which has a stoich of 9.7 the gauge would show 14.7 when it runnin stoich. Unless you have a gauge that reads in lambda.

  20. #20
    Tuner in Training
    Join Date
    Jul 2019
    Posts
    19
    Quote Originally Posted by Hondaeater View Post
    If it was stock you really shouldn't need to change all that stuff? Headers and a CAI would require calibration of VE and MAF tables.
    Hope you get it figured out. Have fun
    That?s what I was thinking....why would so much have to change. Shit just doesn?t make sense.


    Quote Originally Posted by Lakegoat View Post
    Your tuner used 14.1 as stoich because all pump gas has corn juice and is approx. 14.1.
    I have read this other places after I posted it, but was thinking I may try changing it just to see what happens. If anything.

    Quote Originally Posted by ryan_axberg View Post
    I think that would mess him up though because even if the real stoich is 14.1 if he has a little ethanol mixed with the gas his gauge wouldnt read that as stoich. Look up Jay's Tech Tip #35 on youtube. He says that widebands are set up thinking that most people are running pump gas and even if you use e85 which has a stoich of 9.7 the gauge would show 14.7 when it runnin stoich. Unless you have a gauge that reads in lambda.
    I dropped it off to get tuned. I know it was dyno tuned, but have no clue if he hooked up a wideband to tune it or not. I?ll definitely look up the video you are talking about tonight and see if it has anything helpful in it.

    In other news...the new injectors are in, car leak checked good at 58psi so I started it up, and it fired right up no leaks which is nice. That being said I could tell it felt different when it started up. Like it was a little bogged down, then fired up. I didn?t want to read to much into it, so after it got to temp I shut it off and let it get back down to ?cold?.

    The next day I went out to fire it up, and it would not start. My assumption was IF my old injectors were bad/clogged/leaking and these new ones are working like they should then maybe I really am getting too much fuel now. I did the flood clear trick and it still didn?t start but it was noticeably different when I attempted to start it, so that re-enforced my theory.

    I cut 10% cranking fuel across the board and it fired right up no issues. Drove it around for 20 mins, then shut it off and let it sit for ~10 mins to try and replicate the tests I ran prior to the injectors being replaced. Went to fire it up at temp (~194) and it didn?t fire, but fired right up the second time.

    I put the tube file back to baseline, and only cut 10% cranking fuel at temps below 120, and will re-run the same test tomorrow and see where I?m at. I?m hoping that the injectors were in-deed the problem, and now I just have to go back through the steps I already tried with the bad injectors but will get different results. We will see, don?t know what else to try at this point.