Nothing like that on the maps I have ..............screen shot it.
You are overthinking the process IMO.

 

Circled in blue. This is on my stock unmodified map.

I was more or less curious if someone knew the reasoning behind this section of the fuel map as it stuck out to me.

 

Interesting .... that lean spot isn't on JDM tunes . But I see it on the US and EU maps

 

Right? Perhaps it was for emissions reasons/numbers.

 

Alright well I have been playing around a bit more. I tried a similar map to what Brettus had previously posted, except I followed Kane's advice to lean out to 12.1afr. This actually made a huge difference in how smooth first and second gear was. The car was much less jerky.

Otherwise I have gotten within 3% of my afrs across all the ranges. For the Higher RPM ranges it is very beneficial to do multiple tuning runs and adjustments on the same day and on the same flat road.

I had some trouble for a while where I had a linear error in the second injector band. Then the error was about halved in the tertiary injector band. In the end I had reduced the secondary+tertiary injector scaling from 475.93 by 8.2% to 436.96. This leaves me running 0-3% rich instead of 5-8% lean at high rpm

Something that may be screwing with me in the top end is that I had opened up the APV ports slightly and port matched the valves. I played with the VE map to test a few things and bumping up the higher rpm section after 6250rpm(APV opens) appears to have had a positive effect afr error wise. I remember a few threads/people saying to avoid touching the VE map though, I will look around again for why.

-------------
I am actually going to work through my Clean black body Rx8. Already have mazdaedit setup for it. Perhaps I should have gotten versatune to be able to compare, but I already had the dongle setup for mazdaedit. Inevitably though I will be looking to have a rew block built and begin a swap thread. Its renesis would likely be kept as a backup engine for the Red rx8, or perhaps rebuilt for a turbo renesis.

 

Been messing with this first car alot more recently. Since the last post in this thread I have put together a used greddy turbo kit. By no means am I planning on this being a perfected build, I partially wanted a poverty type setup to mess around without fear.

So we ran 30k miles on my above NA tune and it was great. Now I am on about 10k miles on the turbo build, mostly babying it still. I would like to find someone with a dyno who does rotaries up here in New England.

Changes to the car include:

• dw300c pump (I had left over from my REW build.)
• d585 coil kit (about 35k miles)
• 2x Yellow Primary Rail injectors
• 4x Blue Secondary Rail injectors (chosen from info on this thread: https://www.rx8club.com/series-i-eng...-turbo-263157/ )
• I had to replace the greddy intercooler with a shitymoto equivalent.
  Amazon
  Amazon
• Ported the IWG hole due to boost creep
• Second Engine is all prepped and ready to be put together. Keeping it apart to ensure it is coated and not rusting.

Current Issues:

• I still have boost creep starting around 4500 rpm. Which is likely caused by the full 3in decat exhaust providing no back pressure compared to the stock 2.5in with cat. Even with the WG completely jammed open it still builds boost.
• I have a breaking up issue that sounds like ignition cut around 6k rpm with 6psi boost. When I stuck the wastegate completely open I can climb higher rpm, but stop myself when I see boost creep hitting ~3-4psi. I may try some ideas in this thread: https://www.rx8club.com/series-i-tro...iagram-247395/

Plans going forward:

• I cant address the boost creep for now until my REW build stops hogging the lift. I'll probably weld in an EWG.
• For the Breakup issue I have already tried the colder Rx7 BUR9EQ plugs. It seemed like they had a slightly harder time starting, but that could be due to a dying battery. Otherwise no change. I logged the fueling+afr before and didnt notice any glaring issues. I will capture some logs on the way home from work today.
  Next I may try gapping the plugs as that above thread suggested. Perhaps try the 4x Rx8 Trailing plug idea.
  Then its a matter of questioning the coils, I was hoping to grab a sakebomb IGN1a kit next sale, but I may just buy it sooner.

Here was a pic I grabbed today of the profec. I went to a 6.5psi setpoint. Creep begins curving up around 4500 and the breakup happens at 6000. Which I then let off.

• 

 

You need to really hog out the Greddy WG to within a mm or so of the flapper to get creep under control. You don't say if turbo is stock or upgraded ... but creep is common on the Greddy with a free flow exhaust like you have.
Test all your coils per Teams HEI coil test thread!
Make sure dwell times are not excessive - see thread on dwell settings for d585s.
Make sure you use quality 585 coils not cheap Chinese knockoffs!
Run race plugs with smaller gaps (not the BURs) - see the turbo spark plug thread.

 

• I wish I knew where my pics went, but below is a crude drawing of how I gutted it. Basically left a little ledge for the flap to sit on (in green). Next time the lift is open I will see if theres more I can do.
• I believe it is the stock turbo, the whole rest of the kit appears to be of the original kit.
• I double checked the dwell last night and it matches the below table from the end of this thread: Link
• The 585 coils I had gotten from Bennet Built I think about 7-8 years ago. They didn't have an easy life tbh, I definitely beat the **** out of them through college. If they have given up, Ill just go to the IGN1A coils.
• I have 2x NGK R7420-9 and 2x NGK R7420-10 plugs from my REW that I could swap to. I see that they have a 0.7mm(0.027") gap. I sure didn't enjoy the idea of gapping the rx8 plugs.

 

ffs ......................TEST THEM !!!!!!

Dwell is a little higher than what I have settled on but not too bad.
Hopefully you are blocking or disabling the APVs ?

 

I plan on testing this weekend. Team's test link

Yes apv and vdi are disabled.

 

Looks like my logs didnt post for some reason. I will have to do that later again.

Car died on me when I was out and about, dropped to 500rpm at idle as if running on one rotor. On the side of the road I swapped the leading and trailing coils wiring. The car fired up like it was brand new and was as smooth as could be. Tester will be here tomorrow morning.

 

Ok ran the HEI test. One of the primary coils was not firing at all. The other primary coil was not running right either.

I swapped back to the stock rx8 coils for now while I wait for the columbus day sale on sakebomb in a 2 weeks.

 

When you get a good set they work well and last a good length of time - but finding duds in brand new sets is very common and in a boosted setup they become a a huge issue. Not so critical in an N/A.

 

Swapped out the sakebomb IGN1a coil system. What a world of difference that is, so smooth. Only complaint I have is that the harness they have for the rx8 doesn't seem to be that well thought out. I have nearly 3ft of extra wiring wrapped up. Perhaps they use the same loom for other rotary kits?

Only a few remaining issues on this car.

1. boost creep (gotta gut out the turbo WG more)
2. APV motor kicking an engine code P2010 (Maybe melted or wiring issue....it is disabled anyways, but I may have to check if it is stuck open)
3. On cold starts it is hesitating to start and I believe it is flooding with fuel. I am going to be checking fuel pressure. The larger fuel pump may be over pressurizing the injectors. (need to investigate other's return style setups)

Tuning wise from here I still read within 7%afr across my ranges that I can drive safely. Before with Kane's methods on the NA rx8 he had you avoid touching the VE map and only had you adjust the injectors and MAF. I assume the intent was that mazda would have flushed out the VE map based on the engine and focusing on the injector and maf scaling would be where most people have variations. This method got my NA car within 1% for the majority of my logs.

Going forward I was going to just set and forget the injectors and then play around with the VE map for all of my tuning. For the MAF I still have it in the previous AEM cold air intake tube I did the NA calibration on, if I remember correctly I barely had to adjust the maf. I assume that I will have to recalibrate at different boost levels since I would never reach those flows before.

• MAF calibration check
• VE table tests and adjustments.

 

I got a simple fuel pressure gauge using the aeoromotive 15120 adapter. With the DW300c pump I sit at 58psi which checks out. We also got an injector test stand, so I should be able to apply pressure on them and see if they leak. Otherwise DW said that you should expect pressure to bleed off quickly at first and then slowly bleed down after. After shutdown it goes from 58psi to 40psi within 10 mins. Then down to 20psi over 30+mins. If I do not have an injector leak, perhaps I changed something that dumps fuel on startup now.

Been playing around with the VE map. What I found strange in logs is when trims are negative(pulling fuel to lean) but at the same time the car is leaner than target. I'm not sure why it would be doing this. Its almost like there is a few % threshold where it wont keep trying to trim to the target. Like if it is within 2% of target it wont bother to try to be within 0%.

Also I took my adaptronic from the REW build and set it up a turbo REN basemap. Plugged it in and it ran great, really only had a sluggish startup. Idle was amazingly stable compared to the OEM ecu. The stock ECU for some reason has these dips every minute or so where rpms drop by 50-100 for 10s or so and then pop back up..... Evap purge cycle or something? Valve checks?

I'd assume the p2010 code wouldn't be kicked from disabling the APVs? Maybe them not moving caused carbon buildup that made them get stuck, then the shutdown APV motor checks freak out.

 

Still working through issues with the car:

1. Loud piercing whining noise, sounds electrical, doesn't change pitch with rpm. Maybe alternator? I had smoke tested the car and cant find any vac leaks, MAF reads fine.
2. Cold start flooding with fuel, just got an injector tester. Otherwise maybe I need to see if cranking pulsewidth changes help. (maybe this needs to be adjusted for the larger injectors)
3. Boost creep

So think I figured out my method for VE tuning. In short: log and apply your STFT, LTFT and AFR differences to the VE map.

• Datalog to collect load, rpm, AFR, long fuel trims, short fuel trims.
• Edit the log to cut out any log of 20afr off throttle and the shortly after on throttle dip..
• Open up your VE table and loglink your load vs rpm and select the LTFT.
• Select the whole map (ctrl+A)
• Tools>apply from map
• Select VE, average datasource, and data datatype. Using the expression [map]-[x]/100
• Repeat the loglink for the STFT and apply from map again to apply your short trims from the log.
• Now for AFR you need to open up your Target Fuel map
• Loglink this map and select your AFR readings.
• Back on your VE table select all
• Now apply from map choosing: Open loop fuel map, Average data source, %difference datatype, then use the expression [map]*[x])

Repeating this 2-3 times should bring close to 1%. I will be repeating this process to make sure there are no issues. With my luck one of the injectors will have issues and Ill need to repeat this again.

 

Well repeated some tests on VE summed up with an example

Base tune = trim of -2.29%
Add VE = trim of -4.86% (adding ve added fuel thinking more air was entering the engine..... thus the trims reflect having to remove that fuel)
Sub VE = trim of 3.8% (subtracting ve removed fuel thinking less air was entering the engine...... thus the trims reflect having to add fuel)

 

Well just my luck, one of the new IGN1A coils from sakebomb swelled up and I caught it in time. Sakebomb immediately mailed me out a new one which was nice.

I flow tested my injectors and the set of blue injectors(aftermarket) were matching my yellow set of injectors..... so gotta fight with that vendor now.

 

I also replaced my alternator and found that my mystery rpm bump every 110 seconds that was happening went away. Nice since that has been going on for a while now.

 

Wanting to have an autocross car I started playing with this red car again. Basically wrote up a map from scratch based on what I learned from studying basemaps for the REW swap. Now with the IGN1a coils I changed the dwell to 3ms (256*3ms in mazdaedit). Combined with this I wrote up a new target fuel and ignition map to start. I tried running with my boost controller set off so only on stock ~4psi spring and noticed I got no boost creep all the way to 8krpm where i stopped(previously 6krpm). Maybe I wasn't getting a good burn in chamber and burned in the manifold? I suppose I haven't tested going higher rpm since I fixed my coils either?

My current efforts are to find a way of recording data and applying those recordings to the VE map. It seems like closed loop is making this difficult. In the higher load open loop sections I can just adjust based on AFR. I will be updating my original post on the process I figure out. Currently it goes as follows:

• Datalog to collect load, rpm, AFR, long fuel trims, short fuel trims.
• Edit the log to select usable data
  
• Open up your VE table and loglink your load vs rpm and select the LTFT.
• Select the closed loop range of the map under (5krpm)
• Tools>apply from map
• Select VE, average datasource, and data datatype. Using the expression [map]+[x]/100
• Repeat the loglink for the STFT and apply from map again to apply your short trims from the log.
• This seems to work best if you get a good log with not alot of transient throttle responses. Try to gingerly drive around and hit cells. Use an idle log in the idle area, etc.

• Now for AFR differences you need to open up your Target Fuel map
• Loglink this AFR map and select your AFR readings.
• Back on your VE table and select the areas you want to edit. For example I would do anything out of closed loop or in boost if turbo'd.
• Now tools> apply from map choosing: Open loop fuel map, Average data source, %difference datatype, then use the expression [map]*[x])
• This should essentially use your target afr and measured to figure out the %difference [x] and then multiply it against your ve [map].

Ideally I find a way in mazdaedit, my other thought was to just throw in the adaptronic temporarily and using that to make a ve map quicker. After alot of playing around I ended up doing a clutch bleed because it has felt off and while I was in there I swapped the MAF and Throttlebody as I have had strange no start issues on rainy days.