I think it should be made very clear(for the sake of transparency) that the dyno sheet you posted is NOT representative of the files listed in your drop box account. In fact as I recall correctly, the top trace on that sheet is with a piggy back, the next one is with Advanced Performance working with the MazdaEdit software, while the two bottom traces are from work done by OV.

It IS interesting to note that Advanced performance had never seen the MazdaEdit software before the dyno day, and that OV had been working on this file for several weeks.


Now, I am all for open source tuning, and in fact all of us who do this as a profession started out as OS tuners and developers. But you need to be aware of what your getting into here, and not be misled into believing that a simple OTS file an a few days on playing around with a new software will net you a fully custom tune for your car.

I also notice that your post states that "Each file has dyno tuned maps and settings". Would you care to back that up with the dyno sheets for each and every platform? I only ask because you made the claim( and I suspect it was something you were told was true).


BTW, for those of you that do decide to take the foray into playing with your own car; We have been approached recently by several folks that got started down this path and decided it was not for them, and were looking for a well calibrated, dyno tested ROM file. So we have now started our C-n-C service for those folks. Don't worry, we can fix it, no matter how bad the file is!

 

"unfortunately the posting of results and progress have been hampered by some politics."

WHOSE results have been hampered by "some politics"? I've been following Crespo's tuning adventure, and while it seems to be taking forever and a day to get a final result (good or bad), I've yet to see a deliberate slowdown of posting results brought on by a "political" movement instigated by the mods or owners of this site.


"Since I'm not privy to said politics"

So you're taking "someone's" word on the political thing.....I'd be careful taking that stance. If you have proof of something, post it....otherwise, it would have to be considered as hearsay and not useful without solid evidence. Please, let's not muddy the water with innuendo and subtle-but-sinister warnings when it comes to tuning our cars.

 

And the shitstorm already starts. Sigh.... Can we just have Brian's as a locked post please @Frnchplayr15. PLEASE.

 

I'm just going to clean it up eventually. I mean to me it's nothing more than a product announcement. Like a new brand X exhaust available at brandxexhausts.com, so I don't see why ppl are up in arms about it. (Well I know why some would be)

 

Please and thank you @wingzero_.

@Dynotronics1 unless you have legal basis to complain, I'm going to ask you to stay out of this. I can better attempt to articulate the risks of open-source tuning but beyond that, it's on the user. And I specifically noted that the dyno results happen after tweaking of your own. We're providing exactly what it says on the can, files that you can use to tune your car, and nothing else.

 

@Dynotronics1 unless you have legal basis to complain, I'm going to ask you to stay out of this. I can better attempt to articulate the risks of open-source tuning but beyond that, it's on the user. And I specifically noted that the dyno results happen after tweaking of your own. We're providing exactly what it says on the can, files that you can use to tune your car, and nothing else.

No, that's not what you said;

What to Expect:

Each file has dyno tuned maps and settings, so expect some gains even with completely stock everything.

I simply ask you if you would like to produce the dyno sheets.

You also alluded to the dyno sheet in the post being results from these( or other files) done by OV. This is somewhat fraudulent is it not?

And since my post was aimed at "helping the community" which I believe is the battle cry of C-n-C, why should you be offended?

I simply wish to make clear some of the statements you posted. I believe I have done this in a very professional and courteous manner, have I not?

So again( and not to muddy the water here) can you produce dyno sheets for each platform( Malco 2.0/2.3/2.5 and denso 2.0 and skyactiv) that back up that statement? If so I am quite sure the "community" would like to see them, as would I.

And Wing, even as you and I have had our differences, I agree with you 100%. As stated before, we sell far more open source based tuning than we do closed, so the more educated the folks are, the better the understanding the difference between free and paid for work

 

What's up? What's the problem now?

 

No Joe I don't see your comments as respectful at all. You know the kind of work that Mat has put out over the months, yet you decide to interject here and advertise your paid services as an alternative to this free open-source release. This is not the place for that.

The information that Mat has given me states that the lines on the dyno plot reflect states of tune using the MazdaEdit suite. If you prove this is incorrect I will correct it. The maps themselves as released are also dyno tuned as per Mat's information. If you want to dispute that, you have burden of proof. PM me instead of posting to here as further discussion will simply take away from the purpose of this thread.

 

just so we are all on the same page here

http://mazda3revolution.com/forums/...9-mazda3-tuning-maps-tutorials-ovtuned-2.html

 

What to Expect:

Each file has dyno tuned maps and settings, so expect some gains even with completely stock everything. With a car with full bolt ons (intake, exhaust, header) and 93 octane gas, you can expect greater gains across the board. The benefits are far more than just "peak power": the throttle maps have been cleaned up and made more linear, the compensation maps, timing maps, load maps, engine load compensations, and many more have all been re-worked to transform the overall driveability and performance of your car. Here is one example of what can be done with a little tweaking to these maps: after several iterations, this user gained 28hp and 20lb-ft from just the tune alone, and you can see that it's proportional across the power curve. So with that, good luck, stay safe, and have fun!

Thanks for your help in bringing this to our forum! We finally have posted dyno results of what can be done to what I am assuming is a skyactiv 2.0, and the future looks bright :001_smile:

For the dyno, could you please clarify what is shown by the curves? My understanding is that the bottom curves are basically OV's files loaded straight-up, whereas the top two (looks green & purple) are dyno runs using OV's maps after they were "tweaked" to work with the specific car that was tested? Was it a Mazda 3 2012 or 13 skyactiv 2.0? And does "tweaked" mean that the user has taken the original safe maps from OV and performed changes with MazdaEdit software to optimize the timing for their car specifically (ie. adjusted to be a few steps before reaching detonation)? Or does tuned imply other big changes that would require the help of an expert tuner? Thanks very much!

 

Based on those numbers, it has to be a 2.5, not a 2.0L gen 2 sky. Stock whp for gen 2 is in the 155 range for 2.5L and our 2.0L skyactiv is more like 135.

Thanks for your help in bringing this to our forum! We finally have posted dyno results of what can be done to what I am assuming is a skyactiv 2.0, and the future looks bright :001_smile:

For the dyno, could you please clarify what is shown by the curves? My understanding is that the bottom curves are basically OV's files loaded straight-up, whereas the top two (looks green & purple) are dyno runs using OV's maps after they were "tuned" to work with the specific car that was tested? Was it a Mazda 3 2012 or 13 skyactiv 2.0? And does "tuned" mean that the user has taken the original safe maps from OV and performed changes with MazdaEdit software to optimize the timing for their car specifically (ie. adjusted to be a few steps before reaching detonation)? Or does tuned imply other big changes that would require the help of an expert tuner? Thanks very much!

 

I need to press Mat to get us a more recent dyno picture actually as @StaFForD for example has gotten significant gains from Mat's tuning but I don't know which Sky-2.0 has actually been put on the dyno.

 

Thanks, would be nice to get this info. For now, do you know what is meant by some "tweaking" of the maps required to get the maximum gains in power (top curves)? Is this merely adjusting the timing for the specific car that is being used, or much larger changes to OV's maps?

 

Given that these are OTS maps, they won't be perfectly optimized for an individual engine. So you could be close or far away from maxing out and I wouldn't be able to tell you off the bat.

 

Most likely you'll want to optimize an off the shelf tune (timing and afr's) to the quality of the fuel you have available in your area and the condition of your engine. If you have 93 octane available you'll be able to run leaner afr's and more timing than if you only have 91 in your area. As I understand it Canada is all about the fuel quality also. Some 93 up there contains no ethanol so although you get better mpg's the tune can't be as aggressive if you want to stay safe. The way to tune is to make small adjustments to the software and then log to see how the car reacts. Print the logs out and study them. Then you make more small changes upward or downward and log again. You do this several times till the tune is optimized without compromising reliability. Generally you let knock be your guide. Small changes to the tune lets you see a problem before it become an issue.

By small changes I mean changing timing in half degree increments, not whole degrees and afr change are made only in tenths. Move things in tiny increments and you'll have control over the tuning process.

 

Nice, I didn't realize ethanol helps prevent detonation... In Canada, I believe all our 94 blends have ethanol (10%). These high octane blends are basically premium with ethanol added to reach 94. Some of our 91 brands (eg. Shell) do not use ethanol, so I will watch out for them if I plan to stick with 91.

By small changes I mean changing timing in half degree increments, not whole degrees and afr change are made only in tenths. Move things in tiny increments and you'll have control over the tuning process.

Is this done in a specific order? First try to optimize eg. AFR throughout the rev range, then once that is done, work on timing, or do both at once?

If anyone knows any good website resources for open source tuning with NA engines, please post the links.

 

If you guys are interested, I can post of some of my latest logs so you can see how my car is reacting to the tune.

also if you have any questions on the process or any general questions please don't hesitate to tag me.

Edit: As I've stated before, after I get the Phantom in the mail and tuned for it, I'll be getting a dyno of stock, NA, and Phantom for a all in one comparison

 

This is good stuff guys. I decided to quit tuning on my car until there is a better understanding.

 

With moderator approval, we have been given a green light to continue on this thread posting information There will be a huge informational post coming soon

 

I think this the appropriate thread for @Derrick@Ovtuned to provide information, seeing as @color0 has been working with Mat on the open source aspects.

To all that participate in this thread:

THIS THREAD IS FOR THE EXCHANGE OF IDEAS AND INFORMATION ONLY.

 

I would like to open first with any questions anyone may have in regard to these ECUs. We have access to all of them including speed3 ROMs. Mat and I have discussed that we think the community should be re-given access to the OTS mapping that he was previously giving out. But, even better, we want to explain what it is that we are changing, and how it affects your car. As we have been instructed by the Moderators, We will not indulge in any vulgar attacks, or senseless arguing over what we share, or who may not agree with what w share, or the mapping that we do. All I can say on that part is we strongly encourage anyone who may disagree to do a bit of research, ask around, and please don't blindly take anyone's word for a matter we discuss.

To start, I will discuss some of the 2.5L ECU architecture, Mapping, and what is changed for power We have spent more time with this ECU than any. let me gather some data and information and I will be editing this shortly. This will be Naturally aspirated, and forced induction concepts. Thank you everyone.

Naturally Aspirated: For N/A tuning, the standard mapping provided with MazdaEdit has *nearly* everything you need to get the most out of the engine.
here is the main focus, on ANY of the ECUs. The Mass Airflow 2D Mapping.

The numbers in GREEN represent the Airflow Grams. However don't get caught up in the gram, weight, or any techincal representation. It is easy to understand this as.. The ECUs interpretation of X amount of air. Where each number is X amount. Thinking of it as a RAW form is easier to understand.
The outside AXIS in GREY represent the MAF voltage, in correlation to how much X air is being read by the MAF sensor. This map is where you do fuel trims. LTFT STFT. and Is a form to calculate engine load When this map is correct your Actual AFRs logged will be correct.
It is also important to understand that when trimming the MAF, it needs to be kept generally smooth, the transistion must be smooth. Take a look at this below. If this is what your map looks like it is incorrect.


Next is the VVT Maps. Most of the 2nd Generation ECUs have 2 VVT maps, skyactiv has many more though. These maps represent closed loop and open loop conditions. First let me explain that the numbers you see below on these maps are NOT the numbers you want to use, unfortunately we can't share that. Several people understand why. We change these maps, because it changes the actual cam ANGLE / degree in correlation to the crank and correlation to the exhaust cam. It is best to find the proper degree on a Dyno. Adjusting the degree changes how and when the cylinder fills with air, and more air correlates to more power. We will give a hint for anyone trying to get into tuning their own ECU, 31* is great. Follow the stop mapping, but you'll need to ramp the VVT up some in the top end :wink2:

Next is the Throttle mapping. I personally do not own a Mazda, but have been told there is a lag in the throttle. It's actually not lag, its requested throttle degree. you can play with these maps for an endless amount of time to get that truly...perfect throttle feel, if you want. For a simply, 1:1 linear feel, all you need to do is interpolate the mapping. The 2.5L ECU has several throttle maps for several conditions, and gears. Linear looks like this below. Our OTS mapping had Linear throttle angles.

Next is our CLOSED LOOP AIR FUEL TARGETING Map. This map DIRECTLY affects the desired, and request Air fuel target in Closed loop. You can use this map to target and dial in your Closed Loop Air Fuel targeting, so that you're not stuck in that.. "14.7" loop for Closed loop conditions. There are also TWO additional Air Fuel Targeting maps. However these are not directly correlated maps to direct AFR targets. From what we have personally tested, These maps have to do with Air Fuel target vs Cat vs Secondary 02 sensor. But again these maps are not directly affecting. You can actually see how these maps correlate if you Unplug your Rear 02 sensor, and watch how the AFR targeting changes accordingly. But with a rear 02 sensor, the closed loop air fuel target is the Main map. Generally, with these cars you will make best power targeting around 12.9-13.2 in high load conditions
CLOSED LOOP TARGET MAP

Additional Low det (low detonation condition) and high Det (high detonation condition) Air fuel mapping. We have tested personally at least 50 times, and these air fuel targets NEVER are directly targeted no matter the Knock or no Knock condition, for closed loop. Anyone who says otherwise I would sincerely request data on that :wink2:
I would encourage anyone self tuning to play around with these maps, and test for yourself.

Next is the OPEN LOOP AIR FUEL TARGET mapping.
There are two maps, for two different conditions. high detonation and low detonation. Can you guess which is which :grin2:? in OEM form, these are not singular in their target like shown below. These are simply Targets we have used. Now here is the KEY things to understand, the AIR FUEL target you place in these maps are NOT the air fuel you will actually get logged. you MUST trim the Mass airflow Mapping conditionally to get the air fuel you target. These are simply numbers the ECU uses with a complex algorithm to calculate the necessary amount of fuel. BUT, In OPEN LOOP conditions the ECU DOES NOT trim, change, or affect to request these targets. Anyone who says otherwise, I would sincerely ask for data to prove this :wink2:

An easy way to understand is that the number you place in the map. for instance 12.5 is taken + the number in the MAF map gram of flow, along with other things like the injector latency table, and other tables, ran through an algorithm which equals XXXX amount of fuel spray. The ecu does not care if you target 12.5 and want 12.5. You get whatever the ending algorithm number is, which is why you correct this to achieve your targets via the MAF map. You can easily test this to be true by simply unplugging your primary 02 sensor. The ecu CANNOT target what it cannot see, and it simply does not "see" in Open loop, the way it does in closed loop conditions.
Next is Ignition Timing. Ignition timing depends on several factors and we don't feel it is safe to give out any exact numbers, or exact timing degree that you should target. It, again, depends on several factors like AFR, load, cam degree, etc etc etc. However we will explain what is done to clean the stock mapping which is...ugly. General rule of thumb is that in high load conditions timing will increase with engine RPM. However the stock map does not do this, the stock map timing is considerably all over the place.:crying2:
You can correct this simply by smoothing some of the timing maps. however you will want to log, watch for knock, and adjust timing accordingly. From what we have tested, Best power comes just shy of your timing knock threshold, which may or may not be MBT. Just because you can run 35* without knock, doesn't mean it is making any more power than 29*, in some cases. So we don't recommend just ramping up the timing until you find knock, it doesn't always work that way, and you won't always find best power. However, as a hint, on 93 octane we have found best power just shy of 33* Which our OTS maps have adjusted accordingly.

There are several timing maps, and timing compensation maps, and you would to adjust and clean, properly, everything to optimize smoother power delivery. In some cases you can even achieve better MPG from optimizing the timing maps.
Here is some data showing how timing will generally Ramp up with engine RPM. This is not from a 2.5L

This is the general basics of what is changed in our OTS mapping, and what needs changing for Naturally aspirated tuning for these engines. There are some other tweaks that can be done, but this is a general outline. Although this is the 2.5L ECU mapping shown, these basic things are changed in any /all ECUs when it comes to tuning these vehicles.
Next will be Forced Induction, which is is quite complex and much more in-depth.
If you would like to take a look at any of the maps we have tested and revised on a Dyno and back to back street runs. Simply drop me a PM and I'll gladly share. I'd just rather not share here which may start a turf war with a certain character.

 

Forced Induction ECU / ECM Requirements. 2.5L:
Several things are required, which includes axis manipulation, engine load manipulation and scaling, rescaling sensors, among things. First thing necessary to do is achieve proper MAF scaling for your intake, without boost or with minimal amounts of boost. This can get quite indepth quite quickly and not recommended for any novice or inexperienced person trying to tune. But this thread is for information, and we love to share what we can.
You're going to want to make your BASE forced induction mapping, well this is what we like to do. :grin2:
You're going to need to get really into the guts of the ECU for this. You'll need to get into the hex, and define all load axis mapping, and axis correlations to which mapping. <--- This is extremely important that you make sure you have the PROPER load axis re-scaled for the proper maps accordingly.
If you don't have MazdaEdit pro, which I'm sure 99% of people don't. Simply drop me a PM or contact Mat, we can do 1 of 2 things: provide you with an un-encrypted ROM for you to hex out yourself, or we can Define your axises for you.
We can provide you with the Hex Address to your Axises, and provide you with the XML definitions to add in so you have them


By the way, the picture above is not all of the axises, several we have in the "hidden" mapping section not shown.
Our base F.I. file starts with load re scaling to 1.6. However there are times when it needs to be further re scaled to 2+. BUT there are hard fuel cut limiters in the 2.5L ECU for loads above 2.0+ which need corrected.

With the Loads re-scaled you wil notice that your mapping has has also been re scaled, like this example below, however that is not all that needs to be changed. As the ECU will still follow the load paths like stock, and it will throw off your mapping. Unfortunately explaining how to fix this is..well not for the novice, and not something easy to do. We can help you though

You are half-way there. to completing your BASE F.I. map
From this point you'll want to setup your base fueling, and base ignition timing, keeping the numbers nice and safe until you dial it in.
Since it has been advised to me, that this MAP sensor scaling information below has been taken, and potentially used without permission, I can share it. It might as well be shared instead of you paying for it. right? :surprise:
This is Map sensor scaling, for the 2nd Gen ECU / ECM for a Bosch 3 Bar TMap with last three in the product number "845".
These map sensors are $36 dollars online. Please don't be "suckered" into purchasing on with scaling for astounding prices. Pigtail adapters are relatively inexpensive as well. This Map scaling has been tested, to be as accurate as possible. Tested vs Baro and independent Boost gauge (analog and digital)

You will also need to input your Injector scaling. from the Injector Size table, and injector latency tables. Note, the "injector size" table is incorrect for true "cc" and if you input actual injector CC you will be incorrect.
For instance here is for one 2.5L ECU (note that not all have the same scaling, even though the injectors are the same) with SIEMEN DEKA 60LB EV injectors. 60lb or 630CC, our actual scaling was this. However there is an injector sizing algorithm that will get you close, I'll see if I can find it in my notes.

Next you will probably take a datalog of your car, with your new settings and realize in boost you will hit a limiter, a fuel cut. It will be instant and harsh.
This is where your Engine Load Compensation and Engine load compensation trim maps come to help. We aren't going to show you exactly what we've used to dial the Load in to stop this cut, but we will show you the necessary mapping to edit to resolve this problem. You CAN bypass this and bypass the 2.0 load limit fuel cut, but it is not really necessary as the ignition, fuel, and other mapping does not have enough resolution for 2.0+ loads. As compared to a Speed3 ECU. I'll show you the difference below.
These are the maps that need a special touch >

Notice about Map resolution N/A compared to a Speed 3
This is a 2.5L AFR target map:

this is a Speed3 AFR target map

Notice the differences, as the speed3 mapping is larger, with more resolution towards AFR and load. This is why you'd be best to stay away from too much load on the 2.5L ECU, as the larger the load incrase, the less the AFR resolution between breakpoints in load vs RPM mapping.
However, there is a way around this But it requires extremely diligent hex work to achieve, and patching the ROM with custom code. Which we are working on. likes this >
Similarly to how Subaru and EVO ROMs are custom patched and coded, we are working on the same. Here is what we are working on, a Pre-Alpha O.V. big mapping custom ROM patch for Boost

 

Does anyone know if this effort is ongoing or died a long time ago?

 

This is awesome. I would like to see the tutorials as well

 

Does anyone one have some goods guides for DIY tuning?

 

Does anyone still have the screenshots and information from this thread?