Fuel tip in tables

[mfinlay04]

Anyone running RDX injectors care to share their table values?

I've been playing with it and have greatly improved tip in fueling, but still is richer than I'd like and not sure how low I should go on these tables.

**Still in the middle of recalibrating the fuel tables, so don't mind the dip at VTEC

Full mods: Hybrid CAI, PLM V1 ... Header, Full Race Exhaust, RDX injectors, TB Spacer.

[EFICU]

Are you attempting to tune the WOT fuel via closed loop? If I am seeing your short term fuel trims as the blue line in the top graph, your short term trims are getting into the -30 range. At that point, the tip in fuel may not have much impact since the ECU is trying to pull so much fuel. These should be tuned via open loop for WOT, but it's entirely up to you of course.

Do you have the proper RDX injector data in the fuel tab? If you can post a datalog this screenshot was from, and perhaps your calibration, we might be able to lead you in a direction.

[mfinlay04]

No, I have use fuel trims checked so that my OLI continues to work. But I have OL loads around 85" on the low cam and <40" on the high cam. I'll post up my tune and my last catalog when I get home tomorrow.

Ps. My fuel tip in has been better as of late, while decreasing those table values. I was just curious as I haven't seen a lot of people posting about them. Maybe just a downside of speed density tuning...

[EFICU]

I see, okay well it certainly appears the ECU is trimming fuel under WOT which is not really the ideal way to do it. If your primary o2 sensor were to fail, and the ECU isn't transitioning into open loop at WOT, your WOT fueling will be a long ways off.

As far as your tip in, I am more interested in seeing how much fuel tuning you've done in the tables instead of trims and tip in adjustments. If the fuel is far from tuned, I would think you will be chasing the tip in settings quite a bit trying compensate one for the other. If you have tuned the fuel tables, then tip in would be less important IMO.

Either way though, I'll wait to see your calibration and hopefully a datalog so we can try to help you. Either way though, under WOT in your screenshot, the short term fuel trims should be 0, and certainly not -30. That means it's running WOT in closed loop. I hate to assume, but I would assume your WOT lambda table values might be too high to cause the open loop transition.

[mfinlay04]

Here is the most current revision of my tune. As I said before working all the cam angle tables again when the weather permits since installing my new injectors. up to 30 degree cam angle.

[EFICU]

Gotcha, yeah I am not sure exactly what kind of help you want and how much. But to start, I would highly recommend getting proper RDX injector data in there so you're tuning it with the proper settings. Granted it can be done with the wrong information, but then you're really having to tune the tables to remove fuel to compensate for the wrong data. I can input that data into your file if you want, but it will cause you to retune the tables.

To my original concerns over tuning WOT in closed loop, you do have it set to do so. While this is fine when the primary o2 sensor is working, if it were to fail as I have found roughly fifteen that have in the last couple years, your WOT fuel will be very far off. When the o2 fails, they generally read lean, which means your ECU would then go in and try to add 10-20% fuel (depending on how bad the failure is) resulting in very rich WOT fuel. When you tune the tables in open loop, as Honda and Hondata intend, the tuning is done with a working o2 sensor, then in the future if that sensor were to fail the proper fuel would still be delivered to the engine via your tuned fuel tables. When in closed loop at WOT, the fuel trims working make it seem as though everything is good, but you're dependent on that primary o2 sensor working at 100% accuracy all the time. Just something to think about, not trying to tell you how to do it of course. The only concern is your o2 sensor were to fail and send a very rich reading to the ECU, it would then go in and try to pull a bunch of fuel leaving you with a very lean AFR. And the worst part is, most o2 sensor failures don't result in a check engine light, so you would never know unless you were datalogging

The only other thing I would recommend is to go into the VTC tab, and check the box "used VTC tables upon decel fuel cut". That will help it drive a little smoother and help you tune the tables. If you're not sure how this works let me know and I can elaborate.

Are you datalogging through a laptop, or are you datalogging through the FlashPro and opening files from the device itself? The sampling rate seems low, which is why I ask.

Anyways, just let me know how much help you want, if any.

[mfinlay04]

I appreciate you looking over the tune. For my own edification, what values or tables do I have set that would cause it to run CL in WOT? I assumed that the "MAP WOT determination pressure" and "TPS WOT" tables would be the ones to trigger OL/CL. Comparing my tune to some of the others you have posted the only major difference I see is that I have the two "Long term fuel trim" tables populated with a value other than 0. TIA

[EFICU]

Open loop and closed loop operation is controlled by the "WOT lambda adjustment low" and "WOT lambda adjustment high" tables. Any value higher than 12.46 may not allow the ECU to enter open loop operation. I think the number from the help file is 12.5, anything higher than that and open loop will not engage. It appears you have your WOT tables setup to command the AFR you target, but that just means the ECU will trim fuel at WOT to hit that target. With your tables at 13.16 in WOT columns, it will never transition to open loop operation.

Check out the help file here https://www.hondata.com/help/flashpro/i ... meters.htm , scroll down to the WOT lambda table and read the "important note" below it. "Important note: Do not use values above 12.50 (AF) or 0.85 (lambda), otherwise the ECU will always run closed loop." Also mentioned by Hondata here in this thread viewtopic.php?t=24369 .

I would bring you tuner's attention to the following line in the help file:

Important note: Do not use values above 12.50 (AF) or 0.85 (lambda), otherwise the ECU will always run closed loop.

So the vehicle has been tuned to run on closed loop at WOT, which is not normal and can cause many problems. I'd also like to point out that the lambda tables are correction tables, not target lambda tables.

Finally 12.5:1 is very rich for a NA vehicle. Peak power is around 13.2:1 and we've run vehicles at 13.8:1 for 25 hours continuously. Honda ran their turbo F1 engines at 14.7:1 @ 58 lbs boost...

Closing this subject, essentially setting these values to 12.46 will allow the ECU to enter open loop as desired, then you tune the fuel to hit your goal. If your goal is 13.1, the WOT tables need to be set to 12.46 in your WOT columns, then adjust the fuel tables to hit your target.

MAP to determine WOT means that the MAP pressure is determining WOT instead of the pedal movement percentage. That doesn't control open loop operation, that just tells the ECU when to determine WOT. In stock form, or if desired, WOT isn't engaged until the pedal percentage is met. With Hondata, we can choose manifold pressure to determine WOT instead of the pedal movement.

Long term trims are not needed during the tuning process or for tuned cars. If you want to run them you can, but there is no need when tuned properly.

[mfinlay04]

Thanks so much. I appreciate that and totally missed them when reading the help file before tuning. I'll correct these tables right away, as it was not my intention to run in CL at higher loads. As for the injector dead time, I'm going to look back and see if the company I sent the injectors out to have service before I put them in the car did any dead time testing. If not, I guess I'll just have to guess as I'm not capable of calibrating these tables the way that Hondata suggest since I'm far down the modification road, and do not have many/any of the stock parts.

[EFICU]

Sounds good. Be careful, when you make the WOT lambda tables work properly, you may have very rich or very lean conditions. So perhaps watch the data while making the first pull to make sure you're not way out of bounds.

Are the RDX injectors factory OEM Honda, or did you purchase them from Amazon or something? If they are OEM injectors, I have the dead time data I've used on dozens of cars that works well. Totally up to you though, just let me know.

Look into the VTC following cam angle mapping too, and pedal mapping. The pedal mapping is one of the biggest differences I noticed when I had my 8th gen. It's incredible how lazy the factory pedal map is.

[mfinlay04]

Thanks for the help. Car feels way better with it transitioning to OL at WOT now. I'm sure you'll know what I mean by the AFR snapping lower when you hit OL. My trims weren't bad the last run I did before fixing the table that prevented me from entering OL, so my first few pulls were a little rich (12-12.5) but nothing that I couldn't tune up really quick.

And yes, the RDX are factory, I pulled them from a wrecked RDX with pretty low miles and then had them sent off to be serviced (cleaned and flow tested). I looked back through your post and copied the dead times from one of the other tunes you helped with. Definitely, makes idle smoother and I notice the afr while cruising does seem to wander as much as it did with the stock values.

I'll check on the VTC and throttle tables as well. I also have a Speed6 that I drive and tune, the throttle tables on those from the factory are super soft to prevent the average consumer from feeling the torque steer. Just by making them linear made a huge different in drivability. So far Hondata has been a real joy to tune with. I used COBB and now Versatune on the Mazda and while they are great in their own way, the live tuning and active tuning capabilities of Hondata are amazing and save significant time. I'll keep tweaking and the. Post up some more logs later this week. Hopefully have everything finished up by weeks end.

[mfinlay04]

Almost done with my retune. Thanks EFICU for mentioning Evans Performance, I totally forgot that they were a big tuner for the 8th gens back in the day. I went ahead and bought access to his live tuning training videos, they have been immensely helpful. Ran into some trouble with the car last week, actually figure out what has been rattling on my passenger side (blown motor mount) and is down until my replacement shows up.

But I can say, so far my high cam does not like lower RPMs. I'm guessing it's my header that is giving me such a slope, PLM V1.

High Cam 45.png

https://www.youtube.com/watch?v=JyOG16bQgDw X-D

[EFICU]

Yeah just be sure you're getting enough data samples in those cells to make good decisions with. Since the data samples can be very low, the average data may not be very good.

Also be mindful of your vtec rpm with that dip in the low rpms. You will want to look into the vtec rpm point and maybe hitting vtec at 40* and taper up to 45* then back down. I don't know what your 40* table looks like so I'm just making assumptions. But definitely make sure at vtec rpm the high cam is demanding more fuel than the low cam, or you may feel a dip in power.

[mfinlay04]

The dip is pretty consistent across all my high cam angle tables. I'm thinking it's probably the flow scavenging characteristics of the PLM V1 header.

Since I obviously don't have a dyno to find the best vtec crossover, I planned on making a few runs on the same flat part of the local freeway. Set vtec at 3K or so and do a few runs at 40* and then at 45*. Then change my vtec to 6 K and repeat. Then use Virtual Dyno to plot and see where the low and high cam cross over on the TQ curve and set my vtec at that RPM. Obviously, this adds more external factors than using a dyno, but should give me a roughly picture where the ideal vtec crossover should be.

Ps. I totally get what your saying about fuel map tables and setting vtec crossover. Based on table values alone, I'm guessing my best will be in the 4.5-4.6k range (and the butt dyno agrees with this as well) but interested to see how the vdyno plots the tq curves.

vtec crossover.png

[EFICU]

Sounds like a plan, when done right you can get some comparable VD plots to work on that stuff. I had one guy with a 9th gen who had the perfect stretch of road, there was a noticeable and repeatable difference in cam angle mapping in his readings. So yeah, if you can find a nice smooth and flat stretch that might work. If nothing else, the fuel tables will tell you. Wherever it's demanding the most fuel it's making the most power. Which is why the cam angle mapping is so important.

Good luck with it. It'll be interesting to see what you get on the VD pulls.