rotary.enthusiast

orly? Based on my experiences it seemed to directly affect the amount of fuel injected, but that could have been a coincidence caused by my simple setup. I didn't notice a change in ABSload, but I wasn't really looking for it. Thanks for the tip

morkusyambo

So what are some good #s to set VE at from 100% to 160%load?? I have my load set up in 10% increments.

-Yambo

rotary.enthusiast

With out a lot more information there is no way to answer that question. I have mine set to 1.00 everywhere; the whole thing seems like a giant band-aid to me, and I try to avoid using it. I've never tried to tune an 8 with FI though, so I'll let somebody more qualified give you advice on this one.

r0tor

iTrader: (1)

what is this VE table we are talking about??

morkusyambo

Its in the fuel subset about halfway down. If your tune is good allready I would recommend leaving it alone.

MazdaManiac

iTrader: (3)

You won't see any.
The change comes after the reported ABSLoad number.

Do not bother with the Ve table. It will just screw up your tune.

You have to have a dead-accurate correlation between the fuel tables and the resultant lambda before any changes there are useful.

morkusyambo

But for those of us with FI, you have to at least set the loads scale above 100%, right??

rotary.enthusiast

I've always assumed that the PCM performs a look-up of the closest value contained within the table to the current parameters (haven't tested this, and could be wrong here). So if you end up with a load of 2.00, and the highest load cell in your table is 1.00, it will use that one.

MazdaManiac

iTrader: (3)

You have to set the load limit proportionately to the actual load curve. You could set your maximum load to 1, but you would have to scale everything under it accordingly.
So, from 4800 or so and up, you will be at 1.75 or more until 5600 or so where it needs to be 2.
In my case, I actually exceed what would be 2, so it clamps to the top value.

I rescaled my MAF on my current calibration to hit 450 g/sec or so by 5v since it was topping out at 376 g/sec at 4.68v and I was regularly seeing 4.8 and 4.9 v.
I think I saw quite a few excursions into the 400 g/sec territory on the track this weekend.

morkusyambo

I want to rescale the MAF based on my AP datalogs. Besides changing the flow #'s, is there a reason I would ever mess with the voltage #'s across the top??

When I change the flow #'s, should I input the highest # for any given voltage?

Should I input the #'s that are allready closest to a STFT of -0, or just log STFT and adjust the #'s based on the negative or positive value of the different trims??

-Yambo

rotary.enthusiast

If your setup can max out the MAF, then you need to extend the calibration. For example, if you're seeing MAF voltages of 4.8-5.0, then your MAF calibration table should extend far enough to include air flow numbers for those voltages.


Huh?


What is the highest value you're seeing for a short term or long term trim? Do your trims vary wildly across the entire MAF range?

In all honesty, the MAF calibration isn't something you should be playing around with if you're aren't sure what you're doing. I would bet that the OE MAF calibration will be "good enough" for most people, even those with FI. Focus on extending your maximum load, and any tables that use load as one of the axises (there are a lot of important tables that do).

morkusyambo

I get what you're saying. But I do not have a stock intake, therefore I must calibrate the MAF because my AFRs are off at higher rpms even with the stock MAF calibration

rotary.enthusiast

What kind of FI do you have on your car? Does your car have stock injectors? If not, have you set up the injector scaling and latency appropriately? Are you using a quality aftermarket wide band for AFR measurements?

SebtownRx8

anyone gotten it to work on 64bit vista? keep getting the error cannot load communications module.

morkusyambo

Pettit SC. Stock injectors. I've been playing with the MAF scaling and I "think" i'm getting closer.

Flashwing

So, when MazdaManiac and I were not dealing with the home invasion, I had the opportunity to work with him regarding the accesstuner software and able to apply some things I learned to my own vehicle.

I'm not here to give away all of Jeff's secrets but instead I'd like to share a couple of interesting tips which might aid anyone tuning for a N/A application.

I realized pretty quickly that much of what I was doing with the accesstuner software was WAY too much. I was attempting to do too many things at once and making sweeping changes instead of smaller ones. The new data logging feature on the accessport is a vital tool that I suggest everyone use.

I do realize now that the real skills of a tuner overall come into play if you were to design a completly new tune from scratch for a vehicle. With as complex the RX8's PCM is, I cannot see the benifit of doing that since Mazda has clearly spent millions in the development of the stock tune. That's good news for those of us novice tuners as small changes can have great desired effects!

MazdaManiac and I started with his new base calibration. There were several steps we took right off. First was to take note of the STFT at idle once the vehicle was warmed up between 175 and 180 degrees. We logged the Mass Airflow readings and the MAF voltage with the trims.

Next was a run down our local "test track" area. If you can find a nice quiet, long stretch of road to do some street tuning on it will benifit you greatly and make the data much easier to read. We logged Load, RPM, Speed, Air/fuel, and Mass Airflow.

The final step was checking the air/fuel in cruise while in open loop. This meant bringing the car to approx 4200 RPM's in 4th gear (speed limit was 55mph) and noting the air/fuel. This helped us understand if the primary injector scaling was accurate or not.

In my case, we noted that the Primary injector scaling was off and my injectors are about 2% - 4% smaller than the OEM rating.

Now, I really have come to love the data logging in CSV. It's simple and very detailed. My dashhawk makes pretty graphs but the data is hard to apply to the tuning software.

We adjusted my primary injectors to the size we thought they were. This ensures that our calculated load values are accurate since load effects nearly every table in the PCM. Next, we adjusted the Mass AirFlow Sensor Calibration. This got tricky as we were noticing that lower intake amounts (under 40 grams/sec) the PCM was adding a lot of fuel. Our final amounts resulted in about a 6% increase on the grams/sec area. Still, the STFT smoothed out as we got into WOT and through the upper parts of closed loop so we split the MAF curve and made a smaller increase above 40 grams/sec.

The whole goal in the end was to set up the MAF and injector scaling such that your calculated load values and MAF readings are as accurate as possible. At no point did we ever touch the fuel tables.

The awesome part about CSV is being able to line up an air/fuel readout with a load and RPM value. You can then go straight to the fuel table and make direct changes. In the end, the idea is to have a constant air/fuel amount through the RPM band with exceptions made around the torque peak.

I think the most valuable lesson I learned is the need to make small chances and test them before moving on. Doing too much at once will result in skewed or dangerous results and having to start over. Having a laptop or a place nearby your residence also helps as this can be a time consuming process.

There's still a ton for me to learn but I'm becoming more familiar with the software. At first it was rather scary and overwhelming. I then found that even the fuel table ranges you would need to alter are only about 6 cells.

It's just a trial and error process once the basic configurations are set while making minor changes. You take more data, make your changes and go test the car, taking more data etc.

MazdaManiac

iTrader: (3)

lol

Arrrrex-8

I just recently got an AccessPort. I'm looking to start playing around with the tuner software in the near future. I've read all the threads I could find on it and the PDF help file. I taught myself how to tune on my 1G eagle talon. I program and burn my own chips for that. On that car for maf calibration in closed loop I monitored ltft and adjusted the maf compensation to where my ltft's were at 100, this car was OBD1 so 100 is the same as 0 on an OBDII car. And for open loop I monitored WB afr's and adjusted until the target afr matched the actual afr. For the fuel map I found the afr the car liked the best and adjusted for that, on that car it happened to be 11.3:1. For the timing map I set up my timing curve to match the inverse of my torque curve. Lowest timing at torque peak, then as torque fell off I raised timing. During this I monitored my knock sensor, I ran the max timing I could with 0 knock. I did all this with seat of the pants street tuning, always third gear pulls preferably up hill to make sure there was antiquate load on the engine. This method seemed to work really well. I ran 30psi on 92 octane and stock engine internals, never had any problems related to the tune. I had to be a little careful about were I got my gas. It made ton's of power, I ran consistent 7.2's in the 1/8th mile.

I've never done any tuning for rotary motors. Will this method translate to my RX8?

What's the max safe rev limit? I don't care that power may fall off. I'm an autocrosser so a higher rev limit can get me a few more mph in those situations when I'd normally be bouncing off the limiter. A lot of times in autocross its not worth the time to shift into 3rd gear.

What are the ball park afr's N/A rotary's like to be in?

Is the calculated load grams per rev?

Is there a rule of thumb for the timing delay between the leading and trailing spark plugs?

In one of the threads I read it was mentioned that the stock knock sensor isn't sensitive enough. This would potentially make my method of adjusting timing unsafe. Is there anyway to turn up the sensitivity of the knock sensor?

With the talon there's a lot other really cool features you can do with the ECU. Such as stutter box rev limiter, no lift to shift, fuel cut based off knock threshold, usage of stock solenoids for other tasks such as nitrous activation, boost control. The factory boost gauge didn't display actual boost, just an approximation the ECU calculated. So I had it display knock instead. Can something similar be done with the oil pressure gauge?

MazdaManiac

iTrader: (3)

https://www.rx8club.com/nw-rx-8-forum-33/interest-flying-up-mazdamaniac-ap-dyno-tuning-weekend-160927/

shaunv74

He's already on the list for the session.

MazdaManiac

iTrader: (3)

Perfect.

Flashwing

What? I didn't talk about how you sprinkled ferry dust on the accessport or the fact that you hooked your super computer brain directly into my PCM.

Did I mention Jeff is a cyborg from the future?

Arrrrex-8

Yeah, I'm already on that list. Just want to learn and have fun with it for my own enjoyment. One of the nice things about the AP is you can have multiple maps. Use yours when it counts and have fun with mine.

MazdaManiac

iTrader: (3)

Fer sure.
I was just pointing out that the best way to get answers to those sort of questions was coming up in three weeks.

3DZukini

iTrader: (1)

I'm going to subscribe to try and learn something, but after reading this thread and hardly understanding anything being discussed, I feel like an idiot Is there a starting point for me? Like from an "intro to tuning" or "tuning for dummies" perspective? I bought my AP used without MM's maps and I was hoping to learn enough over the winter to get a decent NA tune going on my own. I haven't started using the Accesstuner software yet because I want to get a basic understanding of what I can/should do with it before I go in and screw it up I wish I had bought my AP from MM, but it was a local purchase from a friend and he kinda needed the money at the time to buy a turbo kit. Anyway, sorry for the long-winded post, carry on and I'll try to learn what I can