Page 12 of 12

Re: Tune development

Posted: March 17th, 2023, 7:53 am
by .p4nzer
I had some fun and made plots showing the power and torque estimations of the various tunes, from data collected in various tests.

How did I calculate the estimated power and torque? I used the 175 hp tune as point of reference, comparing the injected fuel quantities against the maximum torque at 2000 rpms (420 Nm) and the maximum power at 4500 rpms (172.6 hp actually, as 175 is the power in PS). As expected, the torque correlates directly with the amount of fuel injected but at high rpms a bit more fuel is required to produce the same amount of torque (at 4500 rpms you get around -9% of torque with the same amount of fuel injected per cycle). Having 2 known points (max torque and max power) is enough for making an estimation of power.

This is not going to be a perfect power estimation, we will see that it actually predicts poorly the performance of the 1.08 tune at high rpms, but big-picture-wise should be accurate.

The plots contains a lot of datapoints, basically all the ones collected during the sessions, so it's not like dyno run. Still it shows what is the max power and torque per engine speed range you can maximally achieve.

The first 2 show the 175 hp tune vs mine:
Image

Image

More striking to see, these 2 plots show the stock 150 hp tune vs mine. It's literally another car:
Image

Image

Finally, the 1.08 tune vs mine. Here we can see that 1.08 is mighty, pushing a huge amount of fuel through all the range. Looking just at the fuel injection amounts, the power should reach 220 hp over 4000 rpms:
Image

Image

But.. virtual dyno says otherwise (the data come from the same logging sessions):
Image

Basically, the 1.08 tune reaches around 200 ponies (flywheel ponies) a bit above 3600 rpms, waaay before my tune. But it loses steams from 3900 rpms on despite still injecting a high amount of fuel.

I don't fully understand why, to be honest. Turbo oscillations start in that range but a crazy wastegate behaviour should not impact power so much, in my opinion. But I'm no expert. Perhaps not enough air gets into the cylinders? But the ECU should cut the fuel injection if the oxygen rate in the exhaust gets too low or if there's not enough air. It indeed adjusts the quantities when needed as I saw these cases in the data, but this is not one of those. Also boost levels are not bad even with the oscillations.

Timings should be advanced a bit given the high amount of fuel at high rpms? Perhaps it could help.

At the end we cannot fully trust both virtual dyno and my estimations but it was cool to play a bit with the numbers. I will test the car on a real dyno sooner or later and see what are the results. As of now, I measured 4th gear accelerations on the autobahn from 80 to 180 km/h. The best time was 19.2 seconds, which is compatible with segment-C diesel car performance in the 200 hp range.

Re: Tune development

Posted: March 18th, 2023, 5:10 pm
by mituc
If you inject more fuel you also have to start injecting it earlier, meaning a bit more timing advance.

Re: Tune development

Posted: March 18th, 2023, 9:28 pm
by .p4nzer
Absolutely, but the injection quantities we are playing with are already mapped by Mazda as we are not requesting crazy amounts. They are not advancing the timings from 70 to 100 mm3/cycle in the top end area.

This can be for several reason, as advancing too much increases Nox.

But when I calculated the start of injection point at high rpms I saw that, even with the quantities I was requesting, more than 35% of fuel was sprayed before top dead center, which sounded decently advanced. That’s why I didn’t bother to advance them, as any change would have been tiny.

Re: Tune development

Posted: March 21st, 2023, 8:11 am
by arno.cigolotti
Nice job !

I've flashed the stock 175hp over the MKv6.05 tune this morning (for testing purposes) and I can clearly say it lacks acceleration/power !

I don't know if you've managed to get it and if what you've done is similar but yeah it feels realy good!

Re: Tune development

Posted: January 19th, 2024, 2:06 pm
by Christian89
Hi,

Anyone still here? Would either mk or p4nzer be willing to share their tunes with me? I haven’t bought the software yet as I haven’t heard great things about the 1.08 map.

Thanks

Re: Tune development

Posted: January 23rd, 2024, 8:01 am
by mk_
Christian89 wrote: January 19th, 2024, 2:06 pm Hi,

Anyone still here? Would either mk or p4nzer be willing to share their tunes with me? I haven’t bought the software yet as I haven’t heard great things about the 1.08 map.

Thanks
I've stopped with tune development. Unfortunately I wasn't able to get rid of the oscillations, so I'm back at 175 tune.
Send me your email in PM, I will send you my last tune.