i am running the factory 5.7 t-body
i am running the factory 5.7 t-body
The intake manifold and injectors - are they stock 6.4 Ram or from a 392 car?
6.4l ram 2500 truck. intake,fuel rail, and injectors
Last edited by jeepingsteve; 06-26-2023 at 12:17 PM.
Recommendations:
1. Check for intake leaks, maybe on the manifold or throttle body. Sometimes the intake gaskets smash the wrong way and will not seal properly. Just due diligence.
2. Set [ECM] 44539 (ECT Rear) to 300 or your favorite unattainable temperature (effectively disables rear O2 sensors - cats deleted). You can probably put P0420, P0421, and P0430 back to the way they were originally.
3. Turn the ANN back on (did you try running the ANN first?). The neural network is probably closer. With VVT enabled, the PCM can't adequately estimate VE using those little tables without the ANN, especially during transients.
4. Match [ECM] 12003 to the Ram 6.4 intake manifold volume: 15.74 L.
5. Try putting [ECM] 34050, 34052, and 34070 to match Ram 6.4. You can adjust the fuel injector PW/fuel mass tables to help zero out your fuel trims. If fuel trims are WAY out of whack, then the Neural Network Trainer is the way to go.
This thread has a 6.4 Ram tune: https://forum.hptuners.com/showthrea...t=Ram+6.4+tune
6. You may need to adjust [ECM] 34000 to help with the very rich spike as the engine fires up.
7. Start logging: Commanded AFR; Exhaust Cam Center Position; and Burn Fraction Inverse.
8. [ECM] 34210: This is the table that may need the most adjustment to get the AFR down before closed loop kicks in and the fuel trims start correcting. This table (and Burn Fraction Inverse) has been one of the more tricky ones to tune. First, the horizontal axis doesn't appear to be "airmass," but rather Engine Run Time. In addition, I've found that my "airmass" breakpoints (the axis values that are 2, 3, 5, 10, 15...) needed to be multiplied by about 2.5 for the PCM to run through the table in REAL time the way it should. Same for the Burn Fraction Inverse table. It's like the clock runs faster in my PCM or something... If you tune the NNT successfully, you'll find [ECM] 34210 may need some adjustment back to something resembling the stock table.
9. Take another startup log and post it and the tune.
what is ex cam center position??
ex cam angle is one of those channels that dont slow down scanner so its what i have been using
what is the difference??
Ok cool. I will try that. With Ann on the truck starts and idles good. I've seen the Ann trainer. Thought about trying it. I've looked for more info on using it.
Thank you for your help. When Ann is off what else is turned off. I've also read that I need to return the injector tables back to stock with it on. Not sure how true that is. i did play with the burn fraction inverse, that did help, it set my power enrichment down to .49 lambda. i wasn't to sure on what that table is or does let alone how to tune for it so i put it back to stock. when my ve started getting close(plus or minus 5) is when the cold start issue started.
Last edited by jeepingsteve; 06-27-2023 at 06:55 AM.
If you disable ANN you should also disable VVT, put your injector tables at stock (Ram 6.4 in your case), and zero out your fuel trims using the two little VE tables. The problem with that approach (especially with a stock cam) is that you will lose hp/torque at the top end when at wide open throttle.
When you have ANN and VVT enabled and you're not tuning with the Neural Network Trainer, you will adjust the fuel injector fuel mass/pulse width tables to zero out your fuel trims.
The web has some good discussions on burn fraction in a cold engine, but it's basically a fuel mass multiplier that compensates for some fuel condensing on the walls and not burning in the combustion stroke. Mopar uses the inverse to express the burn fraction as a multiplier that is applied to the final fuel mass calculation. The actual BFI applied at any instant does not strictly follow the table, so I think there is a complex hidden algorithm using the table as maybe a maximum value to apply at a given ECT and engine run time (not airmass).
If you look at your fuel trims prior to startup, you'll see your LTFTs in that log you posted is around -10%. The PCM uses the saved LTFT at your last idle before your last shutdown and applies it to the next startup. This means that your're running around 10% rich at hot idle, but the cold start suffers because it looks like you needed that 10% (maybe more) for startup. Tuning cold start is tricky, and you'll find as you get your tuning dialed in at normal ECT that you'll have to revisit the cold tables. A little bit of chasing thy tail but it will all work out.
have you used the ann trainer. if so would you recommend going that route. is there more performance gains doing that vs tuning with it off
Sorry I don't mean to be a pain. I am new to the tuning world and of course I had to pick a hemi to start with. Not alot out there when comparing to gm or ford.
how long did it take you to figure out BFI is just a maximum??????
great stuff thx again
Yes I have used the NNT and it works very well. I'd say it can't be beat for getting the engine to run the way it was meant to run from the factory - drivability, response to fast transients in throttle, being able to command the camshaft to whatever you want to and the fueling stays dialed in... That's also considered performance. You can optimize WOT HP and torque using any method, but the NNT is definitely the best for all engine running conditions with VVT.
I'm not sure you need to use the NNT because your setup seems to be pretty close to stock, and I haven't tried tuning a setup like yours. If you want, you can make the above changes, and run another log, with lots of cruising, mid throttle, maybe WOT, try not to stab the throttle but make slow throttle changes, from startup through warm-up and normal ECT, and let's see what it looks like. Just make sure you log those additional channels I noted above.
Just log your BFI and you'll see it go up and down as the PCM responds to transients during warm-up. The applied value of BFI won't necessarily match the corresponding value set in the tune's BFI table, but I think the applied value won't go above the table's values if that makes sense.