You know that almost every car in existence leaves power on the table? This is why tuning exists. Mazda has factored elements into the tune that has to do with completely unrealistic scenarios, but because they are there they hinder the performance. For example cam timing maps are extended to 7500 RPM on the 2.5L. Can you rev that high? no. Cold start and temperature factors have reduction and addition factors for as low as -30 * C and high as 120 * C. for most cars out there they never see such temps. Another example is EGT function. These cars are very delicate about attempting to control exhaust temperatures. Why? for the catalyst of course. They factor this in, and it robs the engine of power because Mazda wants the catalyst to live as long as possible and continue to meet emissions standards. (not like tuning it does not though). There are so many things at play there is not a simple answer as to why mazda left power on the table between the two, or between any engines. You have to understand manufacturers rarely if ever try to get the maximum out of the engine with a stock calibration. Mazdas main focus was "long term reliability, excellent MPG, excellent emissions". Power was not a big concern for them and you can tell by how they designed the engine. You can simply remove some of the limiters mazda put in place and get the "power back" without sacrificing anything else. For example the stock ECU has certain controls that when activated will request AFR of 9.8. 9.8!! This is to control cylinder temperatures. Wanna know the negative side of that? Such enrichment causes large amounts of knock retard. We actually "cleaned" up this control and kept it safe by slightly leaning it out just enough to not cause knock retard, without losing the ability to cool temperatures, which we have tested monitoring exhaust temperatures. I'm sure mazda doesn't care as much about it knocking vs temperatures. The ECU is VERY GOOD at controlling knock once it has happened, so rather than try to stop knock, they let the ECU do it's job and enrich AFR as much as possible to make dramatic reductions in temperatures. I have even done testing with modifying how, when and how often the injectors fire vs crank degree to test differences in recorded temperatures. Extremely complex stuff.
here is the thing a lot of people need to understand. The reason why these engines get such excellent MPG and can do what they do is not because of the tune. It is because of the engine design and hardware. The cam profile, cylinder head, pistons, injectors, fuel pressure, etc etc all are what make this possible. Yes the tune is important in controlling it, but put this same style tune on an engine with different hardware components and you will not see anything like what the skyactiv achieves. Most of the controls in the ECU are about maintainin and controlling temperatures. Like EGT, catalyst temp, oil temp, coolant temp, cylinder temp, etc. Making marginal changes to these makes a big difference in how the engine performs. Now I did not just go at this blindly, I spent more than 3 months of datalogging and recording everything in these engines, every single day for 3 months straight before I made even the smallest of changes. Reviewing extremely large amounts of data. After 3 months of logging and recording, I spent another 6 months making small changes and again data-logging and recording output. I did not just create some "shelf tune" or canned map like tossing leaner AFR or throwing more ignition at it. Like I documented on Miata.net I have actually seen, in some cases where these engines make more power with LESS than stock ignition. Why would a tuner purposely retard ignition on a N/A engine to make more power if it is not knocking and not at MBT? that is virtually unheard of. The answer is almost always "advance, advance, advance." I took a different approach. I logged everything, made very small changes, recorded results. for all RPM breakpoints, load breakpoints, etc.
For example, here is a great one. At exactly 2000-2250 RPM the exhaust cam Valve timing must be dramatically reduced at a specific load range, for a very specific purpose, I'll mention part of it. Doing so stops on throttle knock retard in this RPM area. The stock tune does this, but not enough. And if you data-log a stock vehicle you will see it. SO why Reduce Valve timing in this range, because it also reduces torque? Easy, because of the knock control system, like I stated before. You get knock once, you lose power for a while, as the ECU will peak and hold that ignition retard for quite a while. So you can actually gain more power by reducing certain things in the OEM map, valve timing, and ignition timing being some of those things.
I have studied everything I possibly can about these engines for more than 3 years, every single day I don't even take sundays off. Every little control in the ECU, I have recorded data from skyactiv engines around the globe, and compared. For example do you know that in certain countries the target A/F ratio at full throttle is 12, but in other countries it is 12.4? there is a reason. I have studied, recorded and compared the calibrations of the 1.5L 2.0L and 2.5L and looked for key differences to understand why and under what scenarios these differences are necessary for. I am an enthusiast first, businessman second. I bought a skyactiv 2.0L engine practically brand new (100 miles on it) to tear it down just to take measurements, just to see the profile on the cams, just to see what the valves looked like. I have studied the injector spray patterns and firing order (multiple firing sequence per cylinder per fire event) to see WHY this happens and what it is necessary for (to control cyinder temperatures).
I have seen thousands of data-logs across the globe and have nit-picked every single little detail, in how these thinsg work. I expand my knowledge not just to "sell a tune" but because I love what mazda designed. These engines have so much potential and like I said I am an enthusiast first. I love cars, love modifying them, love tweaking them. But I do so very slowly and carefully.
I have gone out of my way to find controls in the ECU that are not necessary to change or even bother with under any scenario ever! ( for example, the crank tooth timing sequences). WHY? Because I enjoy it. Under no circumstances would you ever change the crank tooth timing sequences if you are still using a stock trigger wheel. So why would I bother to find this stuff? Or other things like MAF flow conversions for load and temperature. these are fine tune from the factory and would rarely if ever need to be changed. But I found them anyway.
Or Cylinder heat efficiency for fuel calculations. These are very sensitive maps that calculate the efficiency of each cylinder, never need to be touched, but I found them anyway. I could go on and on.
Like I said before, I have already turbocharged one of these things. 91 octane, 12:1 compression and boost. That is not easy for any tuner, let alone doing it on a stock ECU calibrated for natural aspiration only. I also spend a lot of time in IDA disassembling the logic in these ECUs. Contrary to the belief of a single forum goer here, it is very helpful to understand how these ECUs work. Once I have mapped out enough I plan to alter some controls and add a few of my own custom features to these ECUs. (yes this is possible!). Name a feature. something you think would be cool that the engine can do. It's possible by manipulating code.
The engine design does most of the job in preventing carbon build up, like I posted (probably got deleted). Mazda designed the cylinder head to keep the intake valves as hot as possible (400* +). The stock tune only gives small amounts of help, it's mostly the hardware.
I want to add this as some "proof" of my ability.
in the 2nd GEN MZR ECU, 2.5L a car I personally owned. I documented, tested, confirmed and showed proof of manipulating the controls to run full time closed loop with boost.
http://forum.miata.net/vb/showthread.php?t=588644&highlight=closed+loop+boost+tuning
But even further, I extended the load ranges, adjusted EQ targets so that the ECU would target the EXACT AFR I listed in the map. Lets say I wanted exactly 11.67 at full boost, I manipulated the closed loop controls so that depending on load and rpm it would request that EXACT AFR in closed loop. here is a video showing complete stock like driving of boost on the N/A ECU in full time closed loop : https://www.youtube.com/watch?v=etTO_kZPKTQ
I went further and changed the maximum and minimum allowable trim percentages for recalculating fuel to ensure enrichment was always exactly what I requested.
Now that sounds "simple" but that is no easy feat.
After finalizing this I was able to run over 35psi on the stock ECU in full time closed loop when this ECU left the factory tuned to run a 167HP naturally aspirated engine.
Last edited:
