Interesting
Interesting
EAST TEXAS DYNO
SERVING ALL OF EAST TEXAS
Send both if you canOriginally Posted by 14HemiRam
Ok I added a log in with them. Thanks for the help!
Do you have the stock tune? Could you send that one too? I have a stock 2015 Ram tune, just not sure how much is different...
OK first a couple of recommendations. I don't yet have experience with tuning spark and knock sensors for an aftermarket camshaft, but I'd recommend returning them to stock values while you're tuning the ANN. If you know what you're doing with your spark and knock sensor sensitivities, then disregard. Also, it looks like you did a cat delete? If not, then be careful while you're tuning and put the COT back when you're done.
OK now for the ANN tips:
The correct cam position Training Values will allow the Network to determine which of the five VE tables it should be using at any instant, or which two VE tables it should be interpolating between to get the correct airflow estimation.
- Take a look at your Training Values under the Neural Network tab. The first four settings tell the Network what your cam sensor will be indicating when the cam is at its extreme values.
- Since the hemi has only one "exhaust" cam, those four settings define only two cam positions: {Exhaust Max <=> Intake Min} and {Exhaust Min <=> Intake Max}. You modified the exhaust settings, but you will need to also modify the intake cam settings. We don't have a way to modify some of the cam position information in the PCM, so the corresponding training values must maintain the built-in lobe separation of the stock cam, which in your case looks like it is 230 degrees. So using the values in your tune:
- Your Intake Cam Max value should be 118 ["Int Cam Max" = 230 - "Exh Cam Min" = 230 - 112 = 118]
- Your Intake Cam Min value should be 97 ["Int Cam Min" = 230 - "Exh Cam Max" = 230 - 133 = 97]
It is very important to get these training values right. You will be tuning five VE tables corresponding to five specific cam positions. This can be a bit tricky, because the Network was designed to handle independent intake and exhaust camshafts, and it doesn't "know" that your lobe separation is always constant. For the Hemi, we have to "lock" the intake and exhaust values together in the tune, so that no matter what the exhaust cam position is, the intake number used by the Network is correct.
We also cannot change the cam angle numbers that the PCM will see, so for tuning, it doesn't matter what angle the aftermarket cam is making at max advance - only the indicated value is important. If you want to tune the Network to properly run for your full camshaft range, then try this:
- Adjust your Desired Angle - Normal table so you will not only see max advance at idle, you will also be able to see the minimum exhaust / maximum intake value while cruising.
- Take another log and watch your "exhaust cam center position" and "intake cam center position" channels. Use the values you see in the logs to define your Training Values in the tune. You will also be able to see the lobe separation in the log data (add the intake and exhaust angles at any instant); if it is not 230 degrees, make sure you use whatever the data shows you.
The actual VE tables use RPM and Pressure Ratio. The max and min values for each are defined by the four Training Values near the bottom of the list (just above "Mult Bank to Bank"). When you collect and analyze your log data, you will want to create one or more VE graphs that use the same axis values that are in your Neural Network Trainer VE tables. I've attached a graph using your values - load it into your VCM Scanner.
VE - NNT B1 TOTAL FT PT.Table.xml
Other tips you may already know:
1. Return any fueling adjustments to stock before tuning the ANN. Looks like your fuel injectors are already at stock values.
2. ECT should be at normal temperature for good data.
3. BFI should be run out to 1.0 (if you do a hot start, BFI may linger above 1.0 for a few minutes).
4. Try to move the gas pedal slowly, move from cell to cell in the VE table slowly to get the best fuel trim data. Moving too fast may cause some acceleration enrichment or deceleration enleanment to kick in which will affect fueling.
5. Get a wideband O2. Everyone needs one, can't tune WOT or Open Loop without one. Don't go WOT any more until you can tune with the WB. Put it as close to one or both of your upstream O2 narrow bands as possible. I have the AEM 30-300 with the LED gauge. I've shocked the O2 sensor with cold starts hundreds of times and it still seems to be working well.
6. I had best results by setting my cam to one position in the tune - setting the Normal, Max TIP, WOT, and Max Airflow tables to one value - and then collecting lots of VE - fuel trim / WB AFR error data. I started with the minimum exhaust angle, collected data, saved the log, and then moved the cam to the next position, collected data, etc. until I had one set of log data for each of the five cam position VE tables defined in the Network Trainer. The rest is pretty similar to tuning speed density - use the fuel trim / error data to modify each of the five VE tables. It took me about four or five days to collect enough data, analyze, adjust VE, and upload to the server for one total ANN adjustment.
7. If you collect data like I did (#6 above), you will want to apply a filtering function to your VE graph to remove any idle data (the PCM will put the cam back to full advance near idle, regardless of what the tables command).
Let me know if this helps.
First of all thanks for all the help greatly appreciate the help! My cats are deleted, I'm building off the original tune I had got from hemifever which is where some of the knock sensors sensitivity's where adjusted from. I will turn them back to stock then while messing with ANN. I'm actually out of town at the moment will be back later today I'm going to mess with this tune when I get back. Just wanted say thanks and a lot of what you have mentioned makes since I'm sure ill have a few questions but with what you have already explained I'm confident Ill be able to work through it now! I do have a question #3 you mentioned "BFI" I'm not sure I understand what that is?? Also I do have a wideband installed I've been using it to tune. Looking forward to getting home to try out what those methods! Will get back with you oh and Im going to send that stock tune right now as well! Thanks again!
BFI is your "Burn Fraction Inverse." It's basically a fuel trim adder (EDIT: I think it is technically a fuel mass multiplier) for cold engine conditions; some good articles on the web describe it (burn fraction) in more detail. If you log it, you will see it relatively high at cold start, and it will decrease and hold steady at 1.0 close to when ECT reaches normal operating temperature. The burn fraction is dependent on several things like ECT and other transient conditions. There appears to be a complex algorithm in the operating system that will vary BFI when, say, you accelerate while ECT is below normal - so you'll see BFI taper down some, then go up a little, then back down again etc. You can see in the BFI table that the factor decreases to 1.0 (i.e., no additional fuel added) as engine run time and ECT both increase. I think Chrysler uses the inverse of burn fraction in a table to make it easier to adjust by using fuel trim or afr error.First of all thanks for all the help greatly appreciate the help! My cats are deleted, I'm building off the original tune I had got from hemifever which is where some of the knock sensors sensitivity's where adjusted from. I will turn them back to stock then while messing with ANN. I'm actually out of town at the moment will be back later today I'm going to mess with this tune when I get back. Just wanted say thanks and a lot of what you have mentioned makes since I'm sure ill have a few questions but with what you have already explained I'm confident Ill be able to work through it now! I do have a question #3 you mentioned "BFI" I'm not sure I understand what that is?? Also I do have a wideband installed I've been using it to tune
If you see either rich or lean at low temperatures, I'd recommend adjusting the BFI table before adjusting the VE ECT Factor table. But don't worry about the low temperature tuning until you get your normal ECT tuning dialed in.
Last edited by rockystock; 05-30-2021 at 12:30 AM. Reason: Hopefully more accurate...
A simple question is...should I use the 6.4 cam angles in the 5.7 tune after changing the cam ?
Most people say no, but I noticed an alternation from the fuel trims everywhere.
No, you have to log some data to figure out what values your PCM sees at max advance and max phase, and compare to stock 6.4. Then if you want to use the stock 6.4 cam positions at Normal and WOT, you'll need to copy the tables over from stock 6.4 tune and subtract all table values (not the breakpoints - leave them at stock 6.4) by about 7 to 10 degrees. I've seen some small differences in the indicated cam angles between Ram 5.7 and various car 5.7s, so you'll need to see what your setup is doing to get it right.
Yes, your fuel trims will be off, until you adjust your VE tables in the NNT. Different cam = different VE.
Reply With Quote