peloponisios

Looking forward to it. Hope everything turns out ok.

d-LoiX

170 WHP on these conditions. I thought it would be worse. Finishing the cooling mods (vented hood + battery to the trunk) and after that NA tuning starts. Found a beautiful 30km flat straight and will do a lot of logging and start fooling around with it. Graph looks rough on open loop. But will also change the stock header to a 3-1 with nicer curves and will start from there.

A little shameful TBH since the guy before me had a 800+ WHP 2019 Dodge Challenger SRT

TeamRX8

iTrader: (25)

it seems to me those are corrected numbers to standard values, which would make the elevation and such not so relevant

MincVinyl

So bit of an update, tuning/logging will soon begin again.

Where we left off was a strange dip in MAF g/sec. Which I believe was due to tight side seals binding and causing a loss of compression when at temperature. Even with a compression check conducted within a minute of shutting the engine off I was still reading healthy compression numbers. After tracing down the vacuum leaks I wanted to check various faces on the Lower Intake Manifold, and at the same time decided it was just as easy to drop the eng/trans and crack it open again to see. Wound up finding that the rotor bearings were rather cooked and there were signs of blowby between sideseals. Opened the side seals gap further, and dug through testing my piles of springs to pick only the healthiest springs. I can reassure that it would have been much cheaper to buy a reman engine, but I would not have learned as much as I have by now.

List of replaced components:
List of Various other projects
Current ongoing issues:

So far after breaking the engine in again, my temperature issues appear to be resolved. With the AEM intake and space free behind the radiator, I might make some shielding to prevent heat from soaking into bottom side of the intake tube. Tomorrow I will get some idle logs again at the stock idle rpm to compare with my previous logs.

MincVinyl

Good news,

I went all through swapping around various plugs, injectors, ignition coils, cleaning maf, and messing around with knock sensors. I ended up doing the 20 brake stomp reset and the misfire issue has been resolved. Car has noticeably more power.

Car runs beautifully now so I took another idle log. Currently running stock injector scaling, and a rescaled maf for the AEM intake cross sectional area. I suppose I won't have to mess with idle much considering the trims. Cruise logs are up next.


wcs

iTrader: (2)

d-LoiX

That looks good @MincVinyl I did some mods to the front bumper and intake which yielded a high trim on idle but seems ok on rest of RPM range. I am finishing the hood now and will try some adjusting after I get it back. How's temps on high load (track)

MincVinyl

As always getting sidetracked

Power steering blew on the rx8, always had been finicky on this car, but finally gave up the fight. It is comical how many weekends I talk about junking the parts car and then saving hundreds/thousands from it.

Also picked up a clean 80k mile 04 black rx8 soon to be stored for the winter. Still contemplating either selling or rebuilding the motor for a turbo to go in the red rx8. In the end the plan would be to clean and protect this black car while prepping for a rew swap.


---------------------------

I did some data logging on the Red rx8. When plotted it appears I'll have to tweak both the maf and Pri injectors. I may try to do some more logs next weekend before doing any changes

RPM...............STFT+LTFT
1500rpm......... -0.54
2000rpm......... -1.38
2500rpm......... -2.12
3000rpm......... -4.09
3500rpm......... -6.34
4000rpm......... -6.4

MincVinyl

Well the snow is gone and I have run the red one for a week. Found a slight oil leak, the sohn kit needs a new gasket. Turns out oil dripping and burning on the exhaust manifold smells a bit. Otherwise everything checks out. Power-steering is still working. I ran some idle logs and will end up taking low rpm logs again.

I have to look through other tuning threads to see if this +50-50 rpm bump happens to other Rx8s....or I suppose I can now log the completely stock black rx8. Checking older logs this appears to be on a 110 second cycle.

MincVinyl

Here's the log for my black fully stock rx8. Which appears to still show the bump in the maf every 110 seconds.

This weekend is supposed to have some nice weather, so I will be working towards making gaskets for my red8 sohn omp to stop oil from leaking onto the exhaust manifold.

MincVinyl

Well made a new gasket and so far no leaks or smoke coming from the engine bay. Pulled apart a bunch of rust along the pinch welds. Bent my obd plug to be able drive while the dongle is plugged in. Spent about 2 hours driving around searching some spots for flat unpopulated roads, so far unsuccessful. My new 22 club RF miata arrived and I got to see it at the dealer, which is exciting.

I had tuned the maf based on the previous low end fuel trims(post 33) raised idle to 1000

Reading through 20 or so logs the fuel trims are within 0 to 1% with an idle at 0.09% in the 1000 to 3500rpm range. Which I believe getting any better would require me to find a better road. I did log a little at the 3500-6000rpm range and was able to see the fuel trims cut out and go to 0.00. Which I believe this is the ecu swapping between closed and open loop. I was curious at first what was happening, but looking through the maps on mazda-edit it looks like the closed/open loop transition is based on engine load and in the 4500rpm range. Any veterans can chime in on this.

Next steps are to compare afr differences in the 4000-6000rpm range to continue tuning the maf and the secondary injectors.
Then on to 6000-9000rpm

kevink0000

Watching your posts with interest. Have you determined what that 110 sec thing is?

MincVinyl

So far not yet, although I'm fairly certain it is just a factory check for any related engine codes. Considering my other bone stock car does the same check I am not too worried. With my logs while driving I have not seen signs of this check, it likely only happens at idle.

MincVinyl

Picked up my clubRF daily, its been great so far.

Last weekend I attempted some higher rpm logs to look into the secondary and tertiary injector sizing. I had a friend with me operating the computer. I tried logging in a few places, and in different gears. Trying 2nd gear high rpm logging I was noticing coolant temps start to climb slightly. After finding a few other locations for 3-4th gear logs it was not an issue. From my buddy reading out the measured vs demanded afr it sounds like a linear difference. (which points to injector sizing instead of MAF) However I have to redo the data since mazda edit apparently will not record graphs for the demanded afr and only records the measured afr data.

More logging will have to wait, as I am fighting the rust on the Red rx8. Already pulled a mouse house out of the rockerpanels. The entire rear wheel well panels are rusted out. Already welded in a patch for the massive trunk hole. Might use this to test some rust products on the red rx8 to be able to preserve the clean black rx8.

Mazda edit is kind of a pain in the sense that there is a **** ton of variables to "record" but more than half of them do not do anything. On top of that there are ones that appear to not actually record data. I was curious about trying versatune for the newer UI, however I am also saving up for a rew8374 build which would just go to an adaptronic anyways.

Spoiler Club RF pic:

wcs

iTrader: (2)

Correct.
I have not used ME, but with Cobb ATR and VT there are several tables that control the transition from CL to OL.
Closed Loop Max Load
Closed Loop Max RPM
Closed Loop Max Throttle

VT suffers for this exact same issue, it might even be worse because it doesn't seem to even work in CL.

Other than the aforementioned Desired/Demanded AFR in VT not working, I don't recall having an issue with any of the other parameters not providing data or not recording data.
For the most part I'm happy with VT and would recommend it, that said, again I've not used ME so ... not really a fair comparison.

Keep up the good work, I too have been lurking your posts.

MincVinyl

Thanks, sorry this is taking so long. Honestly I think if anyone just took a weekend with a buddy you could bang this calibration out. Always getting side tracked in projects.

I also think I lucked out after just adjusting the maf to my aem intake. So this really just left me with the injector sizing to adjust. To be honest I keep forgetting which way to adjust, whether I need a "larger" injector scaling or "smaller".

Next time I hook up ME to the car I will grab a screenshot of the possible variables.

Brettus

iTrader: (2)

I've done hundreds of NA tunes and never touch injector scaling for stock injectors. It's being pedantic for the sake of being pedantic ................. IMO
But at least it will give you some sense of what the adjustment actually does ...so worth it from that perspective.

MincVinyl

Yeah I wanted to go through them anyways mainly just to go through all the steps.

MincVinyl

Been busy with many various projects.
-Assetto Corsa Racing sim rig, I run a server with all of the initial D courses and many personally modified cars.
-My brother is making an actuated sim rig, so we are essentially welding up that looks similar to a formula SAE frame.
-My BRZ friend had gotten coils, wheels+tyres, swaybars, tuner, exhaust which took up a whole weekend.
-I had to fix some wiring issues on the red Rx8, the ign key fuse kept blowing when the trans was put into reverse

Otherwise:
I figured out why mazdaedit does not log the "commanded AFR" after the computer leaves closed loop it just reeds 14.77afr. So I had to change my method. What I ended up doing to finish calibration was to log in 500rpm intervals (Absolute engine load, AFR S1E2, and Engine speed. Then looking back at the log find where on the fuel maps your target afr is to compare. Comparing your (recorded-target)/target*100 you get a %error. Which then you can see the difference between %errors across 500rpm intervals. Below is a screenshot of a random data log where I have selected a point around 7500rpm > then looked at the load I was at > then selected the cell on my fuel map for the target afr.



Comparing the differences it appears that the secondary injectors are off by about 2% from 5000-7000rpm. From 7000 and up it drops rich and then leans out. I feel as though the 3rd injectors should not be this far off, and so far the maf has been fairly on point.

Now on to the issues:
While out testing different methods my friend noticed that I was not at the correct rpms. When my ECU reads 8krpm the tach is reading almost 9krpm. (unrelated but I also noticed the 04black rx8 and this 04red one has different colored LEDs)
I wanted to do some higher rpm testing to see what the afrs look like during pulls. Did several pulls and cooling times between on flat roads. Most of the pulls appear to go 5% rich and then go 5% lean by 8500rpm.
While shifting at a high rpm I had lost all feeling in the clutch pedal and couldn't go into any gears. Was able to get it into gear and drive back without using the clutch, which wasn't difficult at all tbh. Fairly certain the throw out bearing went or the slave went. (Will take apart next weekend to order parts)

MincVinyl

Well fairly quick update, new Exedy clutch went in two weeks ago. Yet again got side tracked, a while back I had done a full custom oil cooler/thermostat setup. However I never mounted my oil cooler fans.

So I ended up taking off my coolers and started designing my own mounts with fan shrouds incorporated. Also plan on wiring them to an oil thermal switch so that they will operate themselves if the oil temp before the coolers reaches 200F. I will also be fixing my oil pan/bolts that are leaky.

MincVinyl

Got about 1000 miles on the car since the clutch change. Oil system is running well with no leaks. Really happy with the pineapple racing oil pan gasket. I changed out alot of the A/C system outside of the firewall, so now it is only moderately warm in the car. I am still having an issue where it appears that my driver side vents are not getting as much air, not sure why as all the diagrams I found showed no flaps distinguishing the driver vs passenger side. Perhaps my blower motor is just weak and the driver side is just the farthest point? Otherwise I also started on ducting for the radiator which now provides mounting for the belly pan. Soon to begin making more underbelly pans for the rest of the car.
--------------------------------------------------
I looked into the gauges issue more. Before I was seeing that my tach was excessively off, hitting the rev limiter (9krpmECU) at 11,000 rpmTACH. I pulled down my other gauge cluster and swapped them in place. Using mazda edit logger dashboard I could compare the TACH's rpm and ECU rpm easily. Inevitably both gauges were reading around 8500rpmTACH when the ECU was at 8000 rpm. Assuming mazda has this in place for safety to make people think they should shift earlier............I am curious if its possible to adjust this, but this is a lot of work/research for a small annoyance. I did come across someone selling a kit to make the oil pressure gauge and coolant temp gauge actually work, so may investigate that further.
---------------------------------------------------
Tuning wise

Where we left off I had pretty much dialed in the MAF and primary Fuel injectors to give us fuel trims within 1 to 3% under 4krpm. My laptop also **** the bed and I picked up a new one solely for tuning use.

I dug through mazdaedit some more and learned about the tracer and log link functions. These are found in the view tab of the fuel maps. Tracer lets you watch what cells on the table you are hitting as you datalog. Log link lets you take a compare a log to the fuel table you have open. To use log link you have to have a log open where you recorded at least engine speed, load, and AFRs. Since mazdaedit has a slew of repeated and unnecessary variables I used the variables listed below.

Engine Speed (RPM)
AirFuel Ratio S1 E2 (OBD)(AFR)
Absolute Engine Load (OBD(%)
Short Fuel Tream B1S1 (%) >>>>>>>>>>>> for the 4000-6000 RPM range having the fuel trims helps to see when you are in open or closed loop as the ECU can sort of mess with you.
Long Fuel Tream B1 (%)

For logs I recorded in 3rd and 4th gear holding 3500rpm -6500 rpm in 500rpm intervals. This data was about 30 mins worth so I had to cut down a lot of the data to avoid drastic load changing events. The closed loop vs open loop crossover also seemed to mess with data so I cut out any data where the fuel trims were not static.
Below is an image showing the loglink function done on the 3rd-4th gear fuel table. The log link view at the bottom can be opened by double clicking on any of the cells with data. Then you can click on a cell again and use the arrow keys to move around. The last value in the log link view shows our % difference between the table's value and our recorded. For this log some of the data for 5000 and 5500 rpm was not usable. The lower load cells were lean by 10-14% and the higher loads were a mix of 3% or 0% lean.


In my case I ranged from a load of 25 to 50%.
>>> I had noticed that the 25% load cells would tend to be more lean as I was probably letting off throttle slightly. These cells were around a 106% AFR across all the rev ranges.
>>> The 50% load cells were the opposite where I likely was on the throttle a little extra and could have been adding fuel. These cells were around 100% AFR across all the rev ranges.
>>> Our middle ground cells for 31%, 38%, and 44% sat at around 103% AFR across all the rev ranges.

This leads me to believe the injectors are off by 3%, however the notes I have from Kane say the secondaries should turn on at 5k rpm and we see this 3% lean out happen suddenly at 4000 rpm and continues onward.

I also did some WOT logs just to see. It appears that in the 4000-6000 range under WOT I was about 10-12% lean. In the 6000+ rpm ranges I was about 5-7% lean.
I will redo these logs and try to get more pulls for averages. Some older WOT logs I had showed a 3% lean out across the board from 4000krpm+

I should also mention that there appears to be no exponential trends which implies that my maf calibration is pretty on point.
----------------------
QUESTIONS I shall keep investigating.

1. Can anyone confirm whether the 3 fuel maps are actually linked to what gear you are in? Mazda edit lists them as Gear1-2 Gear3-4 Gear5-6. However it seems people on versatune were talking about the same 3 maps and had conflicting thoughts.
2. Can anyone confirm that the secondary fuel injectors turn on at 5000rpm? At what RPM do the tertiary injectors turn on?
3. When it comes to adjusting the injector scaling. If I am 3% lean am I correct to say that I should decrease the injector scaling. If the ECU thinks the injectors are smaller it will add fueling to compensate.

wcs

iTrader: (2)

1 - Yes there is some discussion about the gear based fuel tables. I can't confirm anything just to say I've had success making the Fuel tables same. Right or wrong this how I do it

2 - I don't believe the secondaries are rpm based but load based. I typically log the injector duty cycle. You can tell when the secondaries/aux come online as a dip in the duty cycle.

3 - Correct

Brettus

iTrader: (2)

To add to what WCS said ...............................
There is one map that does most of the work for most gears, but the other maps are also used so making them all the same prevents mucking around trying to figure out which map is doing what. Kinda sucks because fuel by gear would allow finer tuning. In general the higher the gear the richer it runs. What I do is logs in both second and third and split the difference between the two.

If you feel like it you could do your own experiments to verify (like I did 15 or so years ago).

MincVinyl

Another question I bet most other new tuners will have is about the lean out bump in the center of the map at the 44% 5000rpm cell. Is this to compensate for the secondary injectors turning on? The torque dips happens at around 6250 so I'm not so sure if it affects that,

wcs

iTrader: (2)

More about the intake