Page 3 of 5 FirstFirst 12345 LastLast
Results 41 to 60 of 83

Thread: NN tuning help

  1. #41
    Advanced Tuner
    Join Date
    Jul 2015
    Posts
    925
    HellcatANNSTFTlog.jpg

    Yes you copy it and multiply it into the the table in the tune, but you are only going to change the InjPW in both tables.

    In the pic, at .019 fuel mass the STFT is -2.3(roughly), so to achieve 0% you would reduce the InjPW by -2.3 to achieve the target commanded FA at that given fuel mass (This is just for example, I'd be happy with that)

    Does that make sense? It's no different than tuning a MAF or VE.
    Attached Images Attached Images
    Last edited by Jay@HAP; 03-23-2017 at 07:29 AM.

  2. #42
    Yeah makes sense. Sorry i was looking at two diffent things and got fuel mass and inj pulsewidth backwards

  3. #43
    Advanced Tuner WS6HUMMER's Avatar
    Join Date
    Feb 2006
    Location
    Pineville La. US
    Posts
    771
    Quote Originally Posted by Jay@HAP View Post
    HellcatANNSTFTlog.jpg

    Yes you copy it and multiply it into the the table in the tune, but you are only going to change the InjPW in both tables.

    In the pic, at .019 fuel mass the STFT is -2.3(roughly), so to achieve 0% you would reduce the InjPW by -2.3 to achieve the target commanded FA at that given fuel mass (This is just for example, I'd be happy with that)

    Does that make sense? It's no different than tuning a MAF or VE.
    How consistent is tuning PW vs fuel mass off cyl #1? I worked up a math parameter that averages fuel mass for bank 1&2 plotted against wideband error just assuming this would be the best way to tune the PW & then make the inverse changes. I haven't even tried it yet to know if it works, I've always just took the long road and turned off the NN.
    99 T/A WS6, original LS1 turbo

  4. #44
    Here is a simple question for the NN area.....
    If NN is enabled, we have to wait for the adaptives to adjust to the new tune?
    With NN disabled, there's no wait for adaptives, the tune goes into effect immediately?
    I'm just trying to figure out how people are able to make quick changes to there tune at a dragstrip without having to have there adaptives relearn the new tune. Or are they only adjusting WOT settings which the adaptives do not effect?

  5. #45
    Advanced Tuner
    Join Date
    Jul 2015
    Posts
    925
    Quote Originally Posted by WS6HUMMER View Post
    How consistent is tuning PW vs fuel mass off cyl #1? I worked up a math parameter that averages fuel mass for bank 1&2 plotted against wideband error just assuming this would be the best way to tune the PW & then make the inverse changes. I haven't even tried it yet to know if it works, I've always just took the long road and turned off the NN.
    From what I've seen it works fine.

  6. #46
    Advanced Tuner
    Join Date
    Jul 2015
    Posts
    925
    Quote Originally Posted by brr View Post
    Here is a simple question for the NN area.....
    If NN is enabled, we have to wait for the adaptives to adjust to the new tune?
    With NN disabled, there's no wait for adaptives, the tune goes into effect immediately?
    I'm just trying to figure out how people are able to make quick changes to there tune at a dragstrip without having to have there adaptives relearn the new tune. Or are they only adjusting WOT settings which the adaptives do not effect?
    In 3.4 you can reset adaptives.

  7. #47
    Advanced Tuner WS6HUMMER's Avatar
    Join Date
    Feb 2006
    Location
    Pineville La. US
    Posts
    771
    Quote Originally Posted by Jay@HAP View Post
    From what I've seen it works fine.
    Cool thanks, gonna give this a shot on one I'll have on the dyno tonight.

  8. #48
    But if you reset the adaptives, you would need drive time to get everything in check. So loading a tune where you need the adaptives set would be useless at the strip.

  9. #49
    Advanced Tuner WS6HUMMER's Avatar
    Join Date
    Feb 2006
    Location
    Pineville La. US
    Posts
    771
    I tried this last night and it does work but no where near as good as VE. You can get the fuel close but you almost have to trade off fueling for low rpm to get higher rpm lined out. As soon as you hit WOT it sees the same fuel mass & cell for the whole pull so you have one cell to adjust IPW to correct fueling. Someone please correct me if I'm wrong but this is the first time I did it and its the results I came up with.
    Last edited by WS6HUMMER; 03-30-2017 at 03:27 PM.
    99 T/A WS6, original LS1 turbo

  10. #50
    Quote Originally Posted by WS6HUMMER View Post
    I tried this last night and it does work but no where near as good as VE. You can get the fuel close but you almost have to trade off fueling for low rpm to get higher rpm lined out. As soon as you hit WOT it sees the same fuel mass & cell for the whole pull so you have one cell to adjust IPW to correct fueling. Someone please correct me if I'm wrong but this is the first time I did it and its the results I came up with.
    Did you do an adjustment with the NN active? like making a change while the adaptives still need to relearn?

  11. #51
    Advanced Tuner WS6HUMMER's Avatar
    Join Date
    Feb 2006
    Location
    Pineville La. US
    Posts
    771
    Quote Originally Posted by brr View Post
    Did you do an adjustment with the NN active? like making a change while the adaptives still need to relearn?
    I reset adaptives before starting & reset after every pull I made. The biggest thing I noticed was that as soon as you mat the gas you stayed in only 1 fuel mass cell for the entire rpm range. So you end up with only one cell to tune from 2500 rpm to 6000 rpm. I was able to add more fuel to that cell and it responded accordingly.

  12. #52
    Advanced Tuner
    Join Date
    Feb 2016
    Posts
    431
    Once you have the fuel mass about correct, you adjust the PE to get the FA nice across the Rev range.
    This is why people call it "hacking". Because the underlying NN fuel table is NQR and you are using PE to make it correct.
    Fuel is fuel tho, just feed the engine what it wants and get over it.

    The tuners that bang on about VE tuning being the correct way, often don't really get that the VVT engines are variable VE.
    That is why the NN is there, that is why it is difficult to get a VE tune correct as you are chasing a moving target with the factory cam tables.
    There is a warm up table, PT tables and WOT tables. It moves all around them and the cam(s) swing all over the place changing the VE.
    If you totally revamp the cam tables etc, and create a predictable VE engine, then the results with NN off are much better.
    So if you are leaving the cam tables stock or close to, NN on is probably the best option.

  13. #53
    Advanced Tuner WS6HUMMER's Avatar
    Join Date
    Feb 2006
    Location
    Pineville La. US
    Posts
    771
    Quote Originally Posted by Hemituna View Post
    Once you have the fuel mass about correct, you adjust the PE to get the FA nice across the Rev range.
    This is why people call it "hacking". Because the underlying NN fuel table is NQR and you are using PE to make it correct.
    Fuel is fuel tho, just feed the engine what it wants and get over it.

    The tuners that bang on about VE tuning being the correct way, often don't really get that the VVT engines are variable VE.
    That is why the NN is there, that is why it is difficult to get a VE tune correct as you are chasing a moving target with the factory cam tables.
    There is a warm up table, PT tables and WOT tables. It moves all around them and the cam(s) swing all over the place changing the VE.
    If you totally revamp the cam tables etc, and create a predictable VE engine, then the results with NN off are much better.
    So if you are leaving the cam tables stock or close to, NN on is probably the best option.
    Thanks, this makes perfect sense & explains what I was seeing. That's also a very well written explination. I've tuned VE with VVT on and off and what you're saying is very true.

  14. #54
    Tuner in Training
    Join Date
    Feb 2017
    Location
    Seattle, WA
    Posts
    24
    Regarding the injectors. You might have better luck if you rescale the tables. It makes no sense that just three datapoints define basically the entire operating range of the engine. Take a look at the thread I started here.

    And here's what my tables looked like before and after.

    injpw tables.png

  15. #55
    Advanced Tuner WS6HUMMER's Avatar
    Join Date
    Feb 2006
    Location
    Pineville La. US
    Posts
    771
    Quote Originally Posted by MasterG View Post
    Regarding the injectors. You might have better luck if you rescale the tables. It makes no sense that just three datapoints define basically the entire operating range of the engine. Take a look at the thread I started here.

    And here's what my tables looked like before and after.

    injpw tables.png
    This make very good sense, just looked at my last log on the dyno and you're absolutely right! Do you just reset the adapts after every time you make changes here & not wait for learn? I'm a little confused on this part, before I would reset and make another pull and it seemed to work good and respond accordingly to the 3 cells I did hit.
    99 T/A WS6, original LS1 turbo

  16. #56
    Tuner in Training
    Join Date
    Feb 2017
    Location
    Seattle, WA
    Posts
    24
    First, I just rescaled the table as shown. I also created a couple tables with Fuel Mass on the y-axis, one measuring error between commanded AFR and wideband AFR and the other showing LTFT.

    afr ltft.png

    I used these error perentages to adjust the injpw by a corresponding amount. Then I uploaded that tune and reset adaptives. That pretty much killed the error I was seeing. 8% before the tweak became 1%. Now I'm kind of a perfectionist so I continued tweaking right away, but I think it's better to wait for the LTFT adaptives to gather more data before making changes. The error percentages up top under WOT obviously respond to any changes right away.

    So that's where I'm at right now. This is the first car I've tuned so I'm figuring it out as I go. Take my suggestions with a grain of salt.

  17. #57
    Advanced Tuner WS6HUMMER's Avatar
    Join Date
    Feb 2006
    Location
    Pineville La. US
    Posts
    771
    Quote Originally Posted by MasterG View Post
    First, I just rescaled the table as shown. I also created a couple tables with Fuel Mass on the y-axis, one measuring error between commanded AFR and wideband AFR and the other showing LTFT.

    afr ltft.png

    I used these error perentages to adjust the injpw by a corresponding amount. Then I uploaded that tune and reset adaptives. That pretty much killed the error I was seeing. 8% before the tweak became 1%. Now I'm kind of a perfectionist so I continued tweaking right away, but I think it's better to wait for the LTFT adaptives to gather more data before making changes. The error percentages up top under WOT obviously respond to any changes right away.

    So that's where I'm at right now. This is the first car I've tuned so I'm figuring it out as I go. Take my suggestions with a grain of salt.
    Thanks for the input, I've only ever tuned this way once before, usually it's NN off and VE. I was just tuning the wot on the dyno so it was pretty much one cell lol. I plan on trying this out and adapting it for another car I have lined up. Just waiting on HPT to get the TCS disable button working so I can strap it off. It doesn't seem to disable on really new cars, just crashes the scanner.

  18. #58
    Advanced Tuner
    Join Date
    Jul 2015
    Posts
    925
    Adding more break points definitely helps. It's amazing how much ANN does in the background considering it only hits 2-3 cells at part throttle and WOT.

  19. #59
    very informative thread. thanks to all...

    as far as i understand, there's no need to disable the ANN to get the ve as smooth as possible, am i wrong?

    i never tuned a dodge before and about to tune a hellcat in the next couple days (sent the pcm to get unlocked). it seems a bit different than GM's. when it comes to fuel tuning, on GM's i basically go to the pe table, do the math and command my fuel and tune the ve and then i dial in the maf and good to go. Hemi's, however, have varied values which confuse a tiny bit... my question is how do i command the fueling and tweak the ANN

    i hope my question isn't stupid haha

  20. #60
    There's so much going on in the pcm we can't see that would scare you. The fact we can do what we are allowed to is amazing. The NN is more than a nuisance among a bunch of tables. Among Ford, Chevy, and Dodge the Dodge pcm is almost sentient. The NN is as close to AI in an engine controller as it gets. Dodge has built a brain that looks at everything going on and makes it's own decisions, regardless of tables or data. Think of it as a mini Watson in the background.
    Level 3 Master Chrysler/Jeep/Dodge/Ram technician
    20 years experience