WOT fuel cut at ~5000 RPM

[mituc]

Without aux fuelling you will not have fuel for more than 25 (maybe 26) PSI of boost, and that's with revised fuelling tables.
On a big turbo setup you can chose the tapper up boost strategy instead of tapper down like for the tiny stock turbo. So you can set your boost limits to 22PSI at 3500rpm, 22.5 at 4000, 23 at 4500, 23.5 at 5000, 24 at 5500, 24.5 at 6000 to 7000. Add 14.5 PSI to these values so that you put the right MAP values in the tune.
Mind these are boost limits, not targets. In general I advise against boost tuning. Use the load tuning strategy instead because especially on a FWD car you will still be able to do boost/load by gear.

[SandSoldier]

Follow up:

I installed a 3.5 bar MAP and updated the MAP calibration values.

The car felt a little better, it didn't feel like I was rear-ending someone anymore but it was still cutting out a bit when in high load, high RPM scenarios.

After more research, analyzing logs, changing parameters like fueling and ignition timing, I actually threw a check engine light: P0303 - Cylinder 3 misfire

I swapped cyl 3 coil to cyl 4, and cyl 3 spark plug to cyl 2 to try troubleshooting it but was uncomfortable knowing i was misfiring at WOT and high load...

the theory at this point is i was hitting some kind of spark blowout. Car ran fine at 16PSI but was not happy at 22 PSI. After messing with my coils packs, I noticed two of the springs were already stretched by the previous owner and the other two were compressed a bit and didnt really want to hold their shape. They looked... idk... burnt? and weren't very springy so I ordered a set of Denso boots and springs,. I stretched the springs a bit before installing them and so far the car has been running great back at 22PSI.

Would like to do more testing at higher BATs to confirm the issue is gone for good but I am having another issue at the moment. I'll create a separate topic about that, though.

Additional info:

MAP STUFF:
BOSCH 3.5 bar TMAP
https://edgeautosport.com/bosch-3-5-bar ... -with-iat/
PN: 0 281 002 456

calibration:
MAP Scale Additive: -12.5
MAP Scale Base Multiplier: 75
MAP Scale Second Multiplier: 1

relevant links:
viewtopic.php?t=3898
https://www.mazda3forums.com/threads/tu ... ap.376920/

--------------------------------------------
(I know this one is from a different forum but has lots of good info on how the calculation works, heres a TLDR)

Input: MAP Voltage (0-5)
Output: Absolute Pressure (kPa)
(((MAP Voltage) * (Scalar A)) * Scalar B) - Offset
STOCK SETTINGS for 2.5 Bar STOCK sensor
Scalar A = 230
Scalar B = 0.23529412
Offset = 1.647
.
....

Hi Jeremy,

Sorry for the delayed response, it's been a crazy few days getting caught back up after the holidays.

Yes, the offset value value is subtracted. This is the equation: manifold pressure in kpa = (((MAP Voltage) * (Scalar A)) * Scalar B) - Offset

The values you came up with (75 and -12.5) appear to be correct. (.5v * 75 * 1) - (-12.5) = 50 KPa and (4.5v * 75 * 1) - (-12.5) = 350 KPa.

The issue at hand is a bug with our software that you just helped us find Some sensors require positive offset values (like the stock sensor) and some (like the one you're using) require negative offset values. I confirmed with an engineer that the software incorrectly sets the minimum floor to 0 (non-negative numbers only). We will be generating and releasing a new build of the software to address this ASAP; I will let you know when we have an update available.

Thanks for the heads-up on the issue and I think I should be in touch regarding a new software build by close-of-business today.

Regards,
Lance


Lance Lucas
R&D Calibrator
ph: 866.922.3059 Ext 1011
fax: 512.997.6400
[email protected]
-------------------------------------------------------------

ignition coil multimeter testing

test coil packs with multimeter:
https://mazdas247.com/forum/t/how-to-ch ... 123758745/
TLDR:
Set Multimeter to ohms
probe left and right pins: 0 ohms is not normal, should be infinite ohms (open circuit)
probe middle and right pins: same as above
probe left and middle pins: infinite or 0 ohms (open or short circuit) is not normal, should be several KiloOhms

note: after talking to my dad, who has opinions i trust, he said there should be two windings in the coil and this method will only test one winding. I don't have the knowledge or experience to say if this is true or not but something to keep in mind i guess

ignition coil spring stretch
...idk about this one...
it seems the consesus is:
*replacing the spring with a copper rod or putting some copper wire INSIDE the spring does nothing
*stretch the ends of the spring? stretch the middle? seems disputed. I found that if the ends are too short, it doesnt matter how much you stretch the middle, it will not make contact with the spark plug. this is how i tested mine:
*remove the boot and spring from coil, remove spark plug from engine, stick spring in the top of the boot, stick plug in the bottom of the boot. If the ends of the springs are long enough, putting the plug in the boot will push the spring up.
links:
https://www.youtube.com/watch?v=nqmgGi4j1iQ
https://www.youtube.com/watch?v=umyXK4h7PdU&t=450s
https://maxima.org/forums/4th-generatio ... retch.html
https://mazdas247.com/forum/t/coil-spri ... 123696474/
https://www.mazda6club.com/threads/stre ... ks.277185/
https://mazdaspeeds.org/index.php?threa ... ssue.2055/

new ignition coil boots/springs
https://www.rockauto.com/en/catalog/maz ... boot,10150
(I bought the Denso ones, NGK ones are prob good too)
also available here:
https://edgeautosport.com/ngk-spark-plu ... d_source=1

[SandSoldier]

Without aux fuelling you will not have fuel for more than 25 (maybe 26) PSI of boost, and that's with revised fuelling tables.
On a big turbo setup you can chose the tapper up boost strategy instead of tapper down like for the tiny stock turbo. So you can set your boost limits to 22PSI at 3500rpm, 22.5 at 4000, 23 at 4500, 23.5 at 5000, 24 at 5500, 24.5 at 6000 to 7000. Add 14.5 PSI to these values so that you put the right MAP values in the tune.
Mind these are boost limits, not targets. In general I advise against boost tuning. Use the load tuning strategy instead because especially on a FWD car you will still be able to do boost/load by gear.

Agreed, load tuning sounds like the way to go, im just trying to take this one step at a time. Currently trying to find a good WGDC/boost value without running out of injector, slowly increasing boost (and load as a result) and fixing issues that pop up along the way...

i like the idea of tapering up load/boost but that leads to another question ive been trying to find an answer to, hopefully you can help me out:

-What do people usually run for target AFR at WOT. Everything I'm seeing says the stock is waaaaay rich and ~11.5:1 is ideal. I am currently running 91 oct pump gas with plans to go ~E30, ~E28 idk something like that. I heard people run ~11.7:1. Does this sound right? I still dont know WHEN a good time to start transitioning to 11.5:1. right now at WOT, thats 3k RPM, blended down to 2k. Does that also sound right?

-What is a safe IDC? Ive seen anything over 100% is spraying fuel out the exhaust valves, do you think 95% a safe target?
Links:
https://www.reddit.com/r/mazdaspeed3/co ... the_ideal/
https://mazdaspeeds.org/index.php?threa ... 648/page-2

[mituc]

The target WOT AFR depends on the fuel used, its ethanol and other bio-fuels content, as well as on target boost, maximum air flow, and so on.
95-98% IDC may be still safe. Usually once you go above 90% you need to inspect the sparkplugs from time to time to make sure all the cylinders burn right and get the right amount of fuel. You still need to leave some IDC room in case the fuel demand increases momentarily (knock, inputs from stability control module, fuel enrichment because of high temps, and so on) so tuning your engine for more than 95 may have some implications in certain situations which can only occur like once or twice in the life of the engine (because after such events usually the engine is done).

[SandSoldier]

Great info!

Could you explain a bit more about target air fuel ratio? Is there a "rule of thumb" we can use as a starting point? And what should we look for if we were to deviate from that value (just read plugs? Or is there more we can look at from the data logs?)?

I assume increased load, boost, airflow, rpm all require a richer mixture but I expect that would taper off at some point. For example, I would expect 5 psi vs 10 psi of boost to richen the mixture more than 20psi vs 25psi but Im not really sure.

Looking back at the versatuner e85 guide
https://www.versatuner.com/article/vers ... ning-guide

They state that the max rich power afr is 12.5:1 on pump gas. Does this mean 12.5:1 should be the TARGET value at WOT/high load/high rpm?

After re-reading that article, I think I understand a bit more about how a increase in ethanol content changes our requested AFR tables

[mituc]

Is there a "rule of thumb" we can use as a starting point? And what should we look for if we were to deviate from that value (just read plugs? Or is there more we can look at from the data logs?)?

The "rule of thumb" or the point of reference is the factory tune. As we know that is pig rich because these tunes were designed back in 2004-2005 for the fuels of that era, and the tunes were not necessarily made for the best fuels. So in the mid/upper RPM range you can go up 1 AFR point no problem, even more.
Reading the plugs will not help you too much in this case because if you put a fresh set of plugs in the engine, do a WOT pull and pull the plugs right after that they will be clean (unless something is very wrong with the engine). You can use this approach more to determine if you're using the right amount of timing.

I assume increased load, boost, airflow, rpm all require a richer mixture but I expect that would taper off at some point. For example, I would expect 5 psi vs 10 psi of boost to richen the mixture more than 20psi vs 25psi but Im not really sure.

Well, of course. And how much the difference will be also depends on the fuels used. There's one thing on 91OCT and 95OCT, same for, say, 93OCT with no ethanol and 93OCT E10. You need to do some exploring.

Looking back at the versatuner e85 guide
https://www.versatuner.com/article/vers ... ning-guide

They state that the max rich power afr is 12.5:1 on pump gas. Does this mean 12.5:1 should be the TARGET value at WOT/high load/high rpm?

12.5:1 will work just fine for NA engines. You can go even in the high 12's or 13 on big V8 engines running NA or with very low boost even on 91OCT. 12.5:1 will not work for turbo engines that well even though you can force the note a bit. I suggest to stay in the mid/low 11's depending on the setup.
When adding alcohols (methanol or ethanol) you can keep the AFRs around 12.0:1-12.5:1 AFR (pump gas AFR, let's just keep the same point of reference the the multiplication of the lambda value) depending on the alcohol content.

After re-reading that article, I think I understand a bit more about how a increase in ethanol content changes our requested AFR tables

A little ethanol will help with the octane of the mixture (even though not necessarily with the actual octane rating of the fuel).
A bit more ethanol will help even more with knock resistance AND will be accountable for extra oxygen in the cylinder (alcohols contain a lot of oxygen, the normal hydrocarbons do not).

ignition coil spring stretch
...idk about this one...
it seems the consesus is:
*replacing the spring with a copper rod or putting some copper wire INSIDE the spring does nothing
*stretch the ends of the spring? stretch the middle? seems disputed. I found that if the ends are too short, it doesnt matter how much you stretch the middle, it will not make contact with the spark plug. this is how i tested mine:
*remove the boot and spring from coil, remove spark plug from engine, stick spring in the top of the boot, stick plug in the bottom of the boot. If the ends of the springs are long enough, putting the plug in the boot will push the spring up.

I used copper rods in the past. they corroded at the contact with the spark plugs and I had to file them after a month, and I just gave up and removed them after two more weeks.
In my case the spark blow-out issue was caused by a fuelling problem (too big WMI nozzle).
The OEM spring has the ability to conduct enough current for the spark to have the right intensity.

[SandSoldier]

I appreciate all the feedback and feel like I am learning so much. Your responses have really been invaluable.

This has brought up other questions but in the interest of trying to keep those questions more easily searchable, I have set up another post titled "commanded ARF not matching AFR tables in tune" (viewtopic.php?f=13&t=4589&p=17121#p17121)

I will probably need to reread this a few more times and do more external research before getting back to tuning (and figure out whats going on in the post linked above)...