A very influential person in the tuning community just wrote me a scathing email questioning my motivations for doing this “Open Source” or open file tuning thread.

Here is something to consider-

In my business life, I did not openly discuss the intimate details of our technical advantage apart from the surface information found in our marketing literature. The company had seventy five mouths to feed and the integrity of our customer’s projects to protect.

In my personal life, I have tried to share what I have learned about subjects I find fascinating. A good example is www.lolachampcar.com. I went to school on ChampCars, high downforce chassis tuning and competitive ChampCar engines. I spent a good bit of time documenting how to prep a ChampCar for safe aero balance so the average guy could learn how to drive one safely. I detailed how to avoid the two biggest oops factors in learning how to drive the cars, surviving cold tires and big aero braking. Cosworth holds their motors for four years then sells them to the public. I rebuild them myself and you will find a lot of pictures of things you will not see anywhere else, details of the inside of a 1999 specification Cosworth XD (XBs as well).

People came into historic racing with ChampCars and ultimately ended up faster than me. I even gave some of them my Motec data so they could do overlays and better understand both braking and super fast corners (172 through the kink at Road America) I hope some of the information on my site helped them beat me and stay safe in the process.

The point to all of the above is that I am doing this because I enjoy it. The files have to be open so that any programmer can use them. If not, I get tarred with being an EFIDude fan. I am, but that should not prevent people from using other products to develop files that will benefit everyone. It really is that simple.

On the flip side, I really do not have a desire to divulge information that would damage people like Hymee’s ability to produce, ship and make money from what I am certain will be a very good product. I think there is room to do both.

 

I am certain it can be done but then what would the motivation be for the EFIDudes to continue on? Not only that, but the whole dongle thing is just $150 which lets you log with a little plug in the OBDii port. It sure beats the damage I have done to my PC with radical data collection using my CarDaq! Lastly, the same little dongle does reflashing (you do not have to believe me, just ask NaarLeven).

ALSO
Forget logging for mapping. Try using that puppy on a track day. Mid corner speeds, straight speeds, throttle pickup and so much more all for the viewing!

 

Just because there's an "open" tune file out there, doesn't mean that RB, Cobb and Hymee will go out of business... There's a significant advantage in going with one of their products, including customer service, reputation, and (probably) less risk.. What fraction of the RX8 owners even looks at these forums?

Ignore the haters, you're on the right track..

 

There are teachers who instruct us without switches or rods,
without slaps or anger, without notice of rags or riches.
If you approach them, they are not asleep;
If you ask a question, they do not hide;
They do not mutter at you if you make a mistake;
When you are ignorant, they do not know how to laugh at you.

"Nil Illegitimus Carborundum"


I've always believed owning many tools was the path to success...it teaches you a lot.... and also that no one has a lock on good inventions.

 

I would like to circle back around to the reason for the thread and start developing some tuned files.

A decision needs to be made regarding long term trim as it has thrown a bit of a monkey wrench into the works. If we are going to shoot for a 13:1 air fuel, should this be done on each car with the long term trim that each car develops or should the MAF be corrected prior to putting a “standard” tune on every car? Each approach has a little “unique to the car” element.

If we work on tunes allowing for different long term trims, there could be a set of files for each tune solution. It could be something like one version for every 2% of long term trim from -6 to +6 percent, or six files total. This approach would allow someone to check their existing long term trim or clear then log the creation of their long term trim and then pick the correct file to provide them with close to the desired 13:1. The person could then log a wide open rip to verify the results and be down the road (hopefully with a smile on their face).

If we choose to correct MAF to get near zero trims and then use one file (the zero LTFT file from above I guess), then each person must undertake the calibration of the MAF meter for their own car. This is not terribly difficult but it is a task that requires logging data at several different steady state points so you can make a good stab at recalibration. Some just apply a global shift to the MAF calibration to account for the averaged LTFT. I prefer to spend just a bit more time and make sure that the LTFT built at 15, 25, 35, and 45 gms/sec are the same or that I correct the right amount for each.

Which approach do you guys (and gals) think makes the most sense? I tend to lean towards the multiple files per tune approach so people do not have to go down the MAF calibration route. It really would not be that much work to generate the different files once we prove the tune and they could be “tweaked” once people get some experience with them (or the person could just choose the next file over to get them dead on the desired target).

 

Or one or more of us could market an accurate replacement MAF, each with it's own test certificate. Then the numbers could then be plugged into the tuning tool of choice.

Cheers,
Hymee.

 

And then there are those with aftermaket CAIs and/or with non-stock MAF tubes. Doesn't this make any MAF no matter how standard or accurate subject to calibration errors in practice?

Would it be that hard to develop a standard MAF calibration procedure that if followed by everyone (regardless of intake or MAF) would allow the application of a standard tune as the MAF output would have been 'normalized' first?

Then the calibration (and adjustment procedure) would be the 'pretune' that anyone could follow to get to the correct point before applying the open source tune of their choice (like CAT or no CAT tune, or whatever)?

One adjustment procedure perfected would make everything after that much simplified. Just a thought.

 

Wow. I wonder where you got that idea? How original.

https://www.rx8club.com/showpost.php...&postcount=550

 

If you were really concerned that the MAF calibration was incorrect, or not accurate enough, you could always flow test the MAF and get it's calibration, or you could buy one that had a known calibration, and perhaps a known error tolerance.

Cheers,
Hymee.

 

Jeff, you know I do not read all those posts out there so, if I am unoriginal, the other guy had a good idea. No matter what the source of the idea, people need to see a logical path to the creation of these files so they can understand how they were created and how best to make use of them.

The engineer in me wants to do a MAF cal before installing a tune file. It just feels like the right way to do the job. The marketer in me thinks the easier it is for someone to start using files without having to do things like MAF cal, the more people will use the files and thus the group effort will bring more value to more people.

Why do we not just do both. We will build a set of files for each tune (I can do this work) and we can generate a MAF cal procedure for those who have the time and interest in doing the work (if someone else wants to bite this off). Whatcha think?

As for a standard cal for an aftermarket MAF, my guess (and this is ONLY a guess) is that a given MAF will act differently in different installations. I would think that the current state of the engine and the surrounding components (filter, adaptor tubes, meth injection, etc.) would work to thwart attempts at a universal calibration. I would think that Mazda did a good bit of work on their MAF cal and yet small changes to the intake appear to have a dramatic impact on building long term fuel trim. My opinion aside, there is no reason not to give it a try.

 

Hymee, since I've not seen anyone do either of these two things you mention, is that a likely senario for future development? I don't know, but it's not done as of yet that I have seen. If it was, there would be no need for at least some of the process MM is doing creating a multitude of tunes and tunes on top of tunes, e.g., yes?

What I was responding to is below, and as I don't personally have the experience to know this to be a accurate, I was simply taking lolachampcar and others word that it needs to be accounted for and voicing an opinion as to the better solution.

In my case and as an example - my car will soon have a RB intake and duct, RB flash and midpipe, MazSport Ignition Solution. Now RB discusses the difference betweem their tube and the OEM tube saying how sensitive the MAF is to changes like this (new tube) and how they tried to do a good job in matching so as not to mess up the MAF output, but it's not exact. Others might have other aftermarket CAI with different effects on LTFT and therefore the resultant problem with any random end tune on any particular car if you impose a "predone" tune on the car (and esp. if you take the tune to the 'edge'). Plus I have these other mods to adjust for (perhaps).

Bottom line and given that, I vote for developing a standard procedure for anyone to measure (say with efidude dongle) their LTFT and with these standard offsets, modify the predone tune they want to flash. Or they could start from scratch and make a complete custom tune, but the topic was about predone tunes.

Sorry to be so verbose, I was trying to be helpful, but I emphasize I don't have practical experience like the rest of you to base ideas from, it's purely a mental expercise until I get my efidude (coming) and later AccessPORT and some bench time on my car. If I'm not helping, or cluttering up the discussion, I can also stay out of this and let you guys have at it.

 

Spin,
You say "will soon have". Can you describe your current intake and then the exact changes you will make? The reason I ask is that I would like to see before and after LTFT data to see if these changes are driving LTFT shifts or if the car has a set LTFT and the intake changes are not causing a shift.

It would be really helpful if you could-
log data the way the car is now
clear trims and then log trim building the way the car is now
clear trims and log trim building with your new set up

It is a bit of work but it would go a long way to shedding some light on this subject.

 

Well the efidude dongle shipped yesterday. Engine is currently totally stock and only mod is CATback. So I'll get base data as soon as I get the dongle and get the car back on the road (now laid up for winter storage). Between the efidude and my Traqmate I should have enough data of current performance to see improvement (hopefully) as I add mods.

Currently somewhere between on backorder and in route and shipped are:

- Mazsport Ignition Solution (better midrange power from better burn?)
- REVI Intake and Intake Duct (new MAF tube and better flow)
- RB Race Flash (better Hi-End power and rev range)
- Mazsport mid-pipe (stop worrying about CAT temps)

also
- CorkSport CATback (already installed, better flow)

I can't think of any more ways to increase NA power than that (anything I've missed?). And I will be doing what you suggest anyway, elsewize how else will I be able to tune the thing for using all that I've added/changed in the combustion equation?

 

Spin,

I wasn't meant to be appearing to argue with you, and I'm sorry if that inadvertantly happened.

I'll gladly go and flow test a stock MAF. I'm pretty sure the results will look something like this:



I'd have to take a break from sCANalyser Pro Tuner development, and S/C development. I think it would be more beneficial to test more than one OEM maf tube. I'd probably test it as a whole air box. Then while I'm there, flow test the same thing with the air-box chopped off. Then I could test some aftermarket MAF tubes, such as the RB or Cam's. It could go on an on... Are not the RB and Cam's MAF tubes the same dia as stock?? Please forgive my ignorance. Then I could go and get an LS1, LS2 or LS7 MAF and flow them, and see if the output signal is compatible.

So there is quite some work there. I would love to do it, and be more than happy to. We did a stock throttle body the other day. It is just time, effort and money. So I need y'all to by exhausts (MK3 coming soon) and scan tools and pro-tuners and S/C's to help fund the R&D. Sorry for the completely selfless and shameless plug!

I'm half surprised that this hasn't been done, as from my point of view, it is the most technically correct answer - not that there is anything wrong with the "bracketing" method mentioned by Lola and MM.

It just makes sense to me to get all the computer to read the correct values to start with, then there is no need to fudge anything else. I have said this from day one (2003), when CZ argued with me about needing an extra wideband (I said there was nothing wrong with the stock one), arguing the MAF was wrong and doing "adjustments" to its signal with the CZ box. Even MM has been guilty very recently of questioning the stock O2 sensor. Now all of a sudden, people are realizing the worth of factory computer. Logging with the Dude Pro Logger is a very good idea that I fully support. It is the next logical step from what we provided in a crude but effective manner in 2005 with sCANalyser Live, and even in the first prototypes in early 2004.

So my vote for the "standard procedure", although maybe not the cheapest way to go, would be to calibrate aftermarket MAF's accurately. If I (or someone else) productizes it, then the costs would be mitigated by economies of scale. Lets see if someone jumps on this idea . Not that I am too worried. I'd (probably) rather buy / resell an OEM sensor with a "test certificate" or known good calibration curve, rather then develop something new. In fact, I think the RX-8 one comes close to achieving that goal anyway The real problems have been with "modded" ones, have they not?

So Spin - how about a Pro Tuner ??

Cheers,
Hymee.

PS - Sorry about the long post. It is about time I made one again. Hopefully it is all good info and thoughts to digest.

 

I think I'd be in deep trouble if I took you away from full-time sCANalyser Pro Tuner development or SC development ...so let's think about that for a bit

otherwise...

Interesting read what RB says about this. Sorry it's not possible to directly link to RB's site to their "Frequently Asked Questions Regarding The REVi Intake System " but if you go there and select the REVi and then choose FAQs you'll find lots of info including this (hope they don't mind the quote)

also pm sent (shortly)

 

For the most part, I am a bad engineer. I tend to focus more on the goal than the method which is not good engineering. If I can be allowed to continue, I would like to take this opportunity with Spin to see the difference (with just one sample, I know) between stock and aftermarket X intake solution. If LTFT changes, Spin can pick his poison on fixing it. If it does not, then we have learned that it is a no problem problem.

I volunteer to
(1) help Spin address any LTFT problem he finds be it by edits that meet A/F target without MAF cal or MAF cal then a standard file that will work across multiple correct MAF cars.
(2) add any Hymee generated proper calibration to anything I do as an option to any files we generate here until such time as Hymee releases his software (at which point you can suck in the Open Source Tunes and edit them in a nice graphical environment).

I am also open to taking on other tasks if it would be helpful. Spin, bring on the data and let’s start developing files. If you want to Dude flash, they tell me you will have the capability to change dongle personality in the ProLogger software (same dongle just different programming) by the time you get before and after data collected! I am going to hold them to that. If you want to flash using a different tool, let’s find out what file format that tool accepts and you will get the files in that format.

 

NaarLeven’s first pass file mods are available to anyone. If memory serves me, they are good for 8 bhp (see https://www.rx8club.com/series-i-engine-tuning-forum-63/reflashed-today-using-efidudes-138881/page3/ for dyno plots). We are heading back to the dyno soon for a multi pass session to head to 13:1. I suspect we will do so using the LTFT the car shows up with so we stand a better chance of staying on target. I will try to get Naar to log data over time and post it so we can track our ability to stay on target.

 

Its actually very easy and doesnt take much time at all, i did the before and after logging with the efidude logger. As soon as i get the midpipe on (hopefully Sunday) Bill and I will begin to make steps towards the tuning capabilities of the tool and how to address the LTFT level im at right now (+5%-6%) in order to obtain the best possible tune for my application. I will share all the info re: what changes are made and what maps are formulated and we will post them up for others to view and use should they need/want to.

 

The stock O2 sensor is fine.
The problem is the range is limited to .75 lambda. For an FI application, you need to be able to accurately tune down to .70 lambda.

Scaling the MAF is easy. Knowing the trim tables for said adjustment is the trick.

 

Wow, .7 lambda or approximately 10.3:1 A/F gasoline. That is a fat puppy!

 

I've got some MoTeC logging of a 26B 7 second run @ 180+ MPH pass where I can check out the Lambda. Pity I'm not at liberty to disclose the MAP or Lambda readings, cause they are not mine to disclose.

Cheers,
Hymee.

 

Rotary motors in boost like it wet.
High compression rotary motors under a LOT of boost on pump fuel like it very wet.
I'll go as low as .65 lambda at the torque peak.
Because of the high BSFC, you often have to go a whole point richer at the tailpipe to get the equivalent true A/F that you would look for mathematically.
Its not a piston motor.

 

So today I went through every thread that I could find by searching for "ultimate" and "emu".. One in particular would be helpful for people looking for an NA tune..

https://www.rx8club.com/showthread.p...&highlight=EMU

 

A bit like your famous avatars!

Cheers,
Hymee.

 

Let's go to work. I want files with anything near those results posted!