With the advent of the L flash Mazda have been making what are clearly moves to try and improve fuel economy.

They have made the WOT fuel ratios VERY lean up to a certain RPM. Like over 14:1 lean!

This will certainly improve fuel mileage to a degree.

What they also did, however, was to retard ignition in the lean range so as to avoid detonation, which cost some power.

Finally, what they have done is to make the point where it makes the transition from lower ranges at super lean to higher rpm more rich conditions occur at a different place in the rpm,
DEPENDING ON WHAT GEAR YOU ARE IN!

This is truly awful, as it makes tuning for performance much more messy and time consuming. One has to tune as if the car is entirely different in each gear. Typically the transition happens at a much higher rpm in lower gears, like 1st and 2nd, and at much lower rpm in third and up.
What would be a perfect tune for 1st gear would be way to rich and conservative in 4th, and what is perfect in 3rd would be too lean and too much advance in 1st.

So, we have had our work cut out for us, and we came up with a solution, which is now incorporated in the units we are now shipping.

To illustrate the fuel/air ratio under WOT in different gears, I have generated a series of graphs.

These are all runs done by starting in a given gear at around 2,500rpm, then flooring the throttle and holding to high rpm.

The AFR value is the legend on the right, while the rpm,, throttle position, and MAF output voltage are matched to the legend on the left.
In the case of the AFR, MAF voltage the values have been multiplied by 1,000 to put them on the scale properly for easy viewing.
The throttle position has been multipied by 100.
So, a AFR of 13:1 shows as 13,000 on the right hand legend, a throttle position of 100% shows as 10,000, and a MAF sensor output voltage of 4V shows up as 4,000 on the left hand legend.

Here is the run in 1st gear. Note how the AFR drops at 6,200rpm from 13.68:1 down to 11.8:1 at 7,000rpm.

 

Here is a run in 2nd gear. Again from around 2,500rpm to around 8,300rpm. At 5,800rpm the Air/fuel ratio goes from very lean to pretty rich.
To properly tune for this one would add a bit of fuel, then add some ignition advance. This will make more power.

 

Here is where we see it getting more messy. This time the transition from too lean, to too rich is at around 4,600 rpm.
This is a 3rd gear run.

 

Finally, we see another full run done in 4th gear.
Now the transition from lean to rich happens around 2,900rpm.
So, while we can tune easily for power for 4th gear andup, it is a problem as the tune will be too lean in the 4,000 to 6,000 rpm range for 1st and 2nd gears.

 

We had to do a lot of testing and work, and calculations to make the map safe across all gears, and to make more power as well, while not hurting the fuel economy.

The good news is that we are now done with this, and are shipping with a new and better tuned map for L flash and onwards.

What it takes is an adaptive logic, that compensates for the effect of the leaner mixture base in lower gears.

Tons 'o fun!

 

I imagine you have quite a number of STAGE1's that all will be shipping in mass. Look forward to hearing reviews as people get their latest units.

 

That looks very complex - crossover points all the way from 4k to 6.2k.... how does it know what gear it is in?
All I can find in the wiring is the neutral switch?

S

 

Maurice

Have you logged the ignition timing in each gear across the board? I want to know total advance and timing split.

 

Yes, that was a veryimportant point in tuning for this. Where we risked runninglean and pinging we took away small amounts of advance to compensate. Where Mazda went wildly lean, and we tune the AFRs back down to make more power we added advance back in.

I will try and generate some charts with the advance curves on them and post this week.

It all takes time..

 

Maurice,

Could it be more of a function of time versus the ECU actually knowing what gear is being used?

Seeing that the engine revs quickly in a lower gear the transition would happen at a higher RPM. Conversely the engine revs climb significantly slower when using a higher gear and when the same amount of time passed the transition would happen at a lower RPM.

Just food for thought.

 

have you been able to determine what effect, if any, ECU learning has on these graphs?

If you reset your "L" flash with the 20 taps on the brakes and drove fully WOT until the ECU learned vs. driving like my grandma after a reset?

Very curious as to whether there is some way to get more MPG or more power during the learning cycles and how significant of change one might see to the graphs posted.

 

Great work Maurice!

Thanks!

 

What if you have some unknow flash like d, e, f. I am not sure what I am running but it is def not L. Plus my Mazda dealer is balking and refusing to reflash to L. How will the new stage 1 work with earlier versions. Do we have to get the L flash to keep things safe?

Thnaks!

 

Maybe this will explain the issues I have seen with my car.

I emailed Graham earlier about a cable. Can I get the new map tossed in there as well? :D

 

I think CZ and crew are busy so I'll try to answer from the info that hegave me. I don't want to slow them down shipping my package

I believe the only way to get the maps correct, especially in the case of "L" they had to make custom maps. CZ had originally tried to make maps that would work for multiple PCM versions but this did not work well. I would suspect from the fact that they will be doing multiple maps, then you will have to specify which PCM version you have. I would suggest ordering the upgrade kit so you can update as nessecary.
=

 

How do I find out what version I am running. Also if that is the case do we need to contact the Canadian Pimps and let them know our ecu version.

Pimpin aint easy or you would have had the mod last week!

 

No speed, that would be quite silly for example what if you stay in 3rd gear for longer then the ecu expects you too. (That’s what I’m interpreting time to mean)

It’s a function of the MAF sensor and its voltage readings that determine what gear you're

 

I think you may be interpreting his idea incorrectly. What (I hear) him say is that [in these specific examples] at WOT, in each successive gear the car accelerates slower and slower. Perhaps a timer is set at liftoff and counts out some time, then the a/f goes lean ->rich.

In each gear it takes longer so the RPM where this happens is simply a f(x) of time and occurs therefore at a lower RPM in each successive gear. No maps - no fuss. Great if it were true. And this is for WOT :D, not anything else.

Simple way to tell is easy - just get time axis on the plots - if the switch occurs at the same time but at these dif RPM points - presto!

 

I'm going to post here just as a sort of "place-holder", but I think there are some dynamics in place that need to be tested before I'll sign off on the idea that the ECU remaps for gear.

I have done a LOT of runs with the "L" flash in place and I haven't noticed what Maurice has found to be present in any of my data logs.
What I have found is a definate "switch point" at 4200 RPM where the ECU goes several points rich at less than 100% throttle.
This mirrors several other TPS affected changes in the map that move around at different RPMs.

Plots like the ones that Maurice posted above can be the result of EGT and sensor response time.
Remember, there is a huge swing in A/F and ignition timing going on as you pass through the affected RPM ranges and the amount of time you spend in those ranges will greatly affect cumulative EGT which, in turn, screws up your WBO2S readings.

 

This is getting complicated. Is the latest canzoomer map designed to work for all flash versions? Or do you have to mate a specific map to a specific flash range?

If it's the latter then it seems like all maps require an easily identifiable version number that can be cross-referenced to determine which flash it's compatible with.
________
LovelyWendie99

 

The self-update cable and equipment is starting to look like a good investment if Mazda keeps changing their PCM versions like this.

With all the fancy new tricks you're having to pull, will this make the map any more sensitive to aftermarket exhaust and or intake systems?

 

My head is gonna explode.

 

I got the "M" flash last week and I can tell you It feels very different during gearshifts but seems to have a little more low and mid range responsiveness, The mpg went up to 18 mpg city from 14 as well. Hopefully the flashes will end soon so we can have some good tune, but I can't complain about the latest flash. Keep up the the hard work guys, I'm really looking forward to that stage 2.