ATR tuning A BNR Stage 3

[Ziggo]

*Updated 10/27/2011

Now using load tuning. Targeting 2.4 load flat in the ABS load table. Boost is capped at 22psi, and the load error correction pushes it to 22.8psi before it stops climbing attempting to maintain the load.

I found that premium gas was severely knock limited @ 22psi to only 9* peak. I am now mixing in 3/9 E85/Gas to help this out with great results. This allows me to push to 17* peak timing and lean out to 12.0 AFR. Will be sticking with this tune for now. May attempt to move to a 4/8 mix and try to flatten out torque a little more on the top end at a future date.

Latest Log:
http://www.mazdaspeedforum.org/forum/foru...1020081102.csv

The torque "bulge" is from a slight downhill where I started my log.

• Share
  • Share this post on
  • Digg
  • Del.icio.us
  • Technorati
  • Twitter

__________________

Ask Me About My:
BNR S3 Turbo \ JBR WP 3.5" Intake \ COBB FMIC \ DNP EX MANI \ COBB DP \ MSCBE \ CPE HPFP
CPE Injector Seals \ JBR Thermal Intake & Throttle Body Gaskets \ BOZO EGR Delete

BC Coilovers (-2.5º F, -1º R) \ Saitek Front Endlinks \ SPC Rear Camber Arms \ Hotchkis FSB & RSB
JBR 88 Duro RMM TMM PMM \ COBB AP (Self Tuned) \ EGT \ Oil Pres \ Oil Temp \ DashHawk

---

Last edited by Ziggo; 10-27-2011 at 06:31 PM.

[Lex]

Those BATs are quite high for an FMIC ... but regardless the turbo's flowing quite well - esp past 5500 RPM.

The knock up top is probably noise - there are ways to tune some of this out by adjusting the sensitivity offset.

You will need over 20 psi to reach 2.1 load ... and with the WGDC being in the 50s you can certainly push it that far.

Consider load based tuning if you'd like quicker spool-up.

[Ziggo]

I will experiment with the timing to see if it is knock or just noise. I tend to think its real since I have never seen false knock before when tuning.

My intention is to have a load tune, but I didn't want to slap a 2.1 target in there at 6k and have it boost high enough to hit fuel cut, so I am walking it up there using boost tuning and seeing what kind of loads it is capable of hitting before I run out of room on the stock map sensor.

As for the BATS, yeah they are high for an FMIC, but its 105* in the shade out here, I bet the temperature over the asphalt is over 110, probably closer to 115, and I was only driving for a couple minutes before taking the logs after sitting and revising and flashing the maps.

EDIT*
BTW Lex, hows that guardian angel coming? I could really use one.

[drknyte]

I wish I knew what you guys are talking about

[laxplayermjd]

hows the butt dyno at that boost level ?

[Ziggo]

Butt dyno agrees with the results. Due to the current targets there is less torque, but higher HP. I want to stress that this is just my first cut at things. My AFRs are stupid rich, and I am not stressing the turbo at all.

I don't know that I am a huge fan of the new restricted WGDC dynamics. I can set my tables to be within 10, but based on previous experience this will drop significantly as temperatures drop. I bet I will see a drop of ~20 when temperatures drop from 105* to 70* and I dislike having my WGDC set too high as this could cause spiking.

Anybody have any experience using the load fuel cuts as a safety net? I used the boost - fuel cut before when I was tuning around my fuel pump and I was having creeping issues in below 50deg temperatures and I know that one works, but has anyone actually hit a load based fuel cut? I am trying to avoid replacing the MAP sensor for this, but I need a safety net incase a wastegate line blows off.

[cld12pk2go]

Looks like you need meth and a 3 BAR MAP sensor in a bad way...

Also have you modified your KR offset table? It is possible that the KR is real given the BAT, but it could certainly be engine noise as well. Either way it wasn't enough KR to be overly concerning.

That turbo should support 340 g/s, so it is barely working at this point...

[Ziggo]

I am trying to keep it simple, and will probably replace the MAP sensor in the future, I am just trying to see what I can get to without replacing it. Alcohol Injection is not allowed by alot of track orgs, so no meth for me

It just too damned hot outside. fuck this heat.

I know this turbo is calling me a pussy right now, this was just my first cut, I had no data to base this map on so its crazy conservative.

[Ziggo]

Its still too damned hot. Its also funny, I can hear the WG pretty clearly when doing the pulls and I am still not really pushing the turbo, but I am going to go ahead and switch to the ITV22s, that knock isn't getting any better and I need to figure it out before I increase the boost more.

Boost temperatures are getting crazy, I have to figure something out about them, climbing into the 150s is not acceptable even as hot as it is outside. Maybe the Cobb FMIC is too small for this turbo? I dunno.

[Lex]

Originally Posted by Ziggo EDIT* BTW Lex, hows that guardian angel coming? I could really use one.

Wish there was more than 1 of me to get this out sooner .... but it's coming along and should be here when it's getting cold.




BTW, the larger turbo should be keeping the boost temps lower ... so you should see something lower than the K04

[Ckmazdaspeed3]

Fuck, that thing hits full boost a lot quicker than my 2871... I should have gotten one of those!

What about e85? W 3.5 gallons and similar temps, I only had go stop increasing timing because I didn't want to go too far street tuning... I never see kr w 3.5 gallons.

[Ziggo]

I cant think of any reason it should spool faster than your ATP2871. The only functional difference is the ATP is a ball bearing and the BNR is a journal, so you should have the advantage there.

You are using an interrupt ECBS and an aftermarket manifold correct?

E85 isn't really an option, the closest station to me is 18mi away. If the temperatures don't drop before my trackday in September, I will be adding race gas to the tank just to be safe.

Lex, I know it will run lower boost temperatures at the compressor exit at the same flow as the K04, but it is also flowing more air, so it could be overloading the capacity of the FMIC to cool the air, particularly at these stupid high ambients.

Nobody out there knows if the load based fuel cuts are active and functional? It looks like I will have to push quite a bit of air to get to a 2.1 load at 6500, based on my last logs it will have to be something like 24/25psi.

[cld12pk2go]

Originally Posted by Ziggo I am trying to keep it simple, and will probably replace the MAP sensor in the future, I am just trying to see what I can get to without replacing it. Alcohol Injection is not allowed by alot of track orgs, so no meth for me It just too damned hot outside. fuck this heat. I know this turbo is calling me a pussy right now, this was just my first cut, I had no data to base this map on so its crazy conservative.

Any reason you cannot run straight water? It is still a shit-ton better than nothing.

Originally Posted by Ziggo I cant think of any reason it should spool faster than your ATP2871. The only functional difference is the ATP and the BNR is a journal, so you should have the advantage there. You are using an interrupt ECBS and an aftermarket manifold correct? E85 isn't really an option, the closest station to me is 18mi away. If the temperatures don't drop before my trackday in September, I will be adding race gas to the tank just to be safe. Lex, I know it will run lower boost temperatures at the compressor exit at the same flow as the K04, but it is also flowing more air, so it could be overloading the capacity of the FMIC to cool the air, particularly at these stupid high ambients. Nobody out there knows if the load based fuel cuts are active and functional? It looks like I will have to push quite a bit of air to get to a 2.1 load at 6500, based on my last logs it will have to be something like 24/25psi.

The K04 turbine housing could potentially be resulting in quicker spool relative to the GT2871. It is about the only explanation given the fact that the rotational hardware is essentially identical unless the ATP is BB instead of journal. If so then the BNR probably flows a little bit less on the top end as well.

The load based fuel cut does work, but in my experience you have to be a small amount over either the load/boost fuel cut for ~1 second before they actually kick in.


Looking at your log, it appears you still have the stock KR decay rate A/B of 640 Milli-seconds. I would recommend knocking that down to about 200. Also you could go with an offset table similar to mine to help filter out any engine noise up top:

0.16	0.16	0.16	0.28	0.36	0.40	0.60	0.80	0.80	0.80	0.80	0.80	0.80	0.80
0.16	0.16	0.16	0.28	0.36	0.40	0.60	0.80	0.80	0.80	0.80	0.80	0.80	0.80
0.16	0.16	0.16	0.28	0.36	0.40	0.60	0.80	0.80	0.80	0.80	0.80	0.80	0.80
0.16	0.16	0.16	0.28	0.36	0.40	0.60	0.80	0.80	0.80	0.80	0.80	0.80	0.80
0.16	0.16	0.16	0.28	0.36	0.40	0.60	0.80	0.80	0.80	0.80	0.80	0.80	0.80
0.16	0.16	0.16	0.28	0.36	0.40	0.60	0.80	0.80	0.80	0.80	0.80	0.80	0.80
0.16	0.14	0.16	0.28	0.36	0.40	0.60	0.80	0.80	0.80	0.80	0.80	0.80	0.80
0.16	0.12	0.16	0.28	0.36	0.40	0.60	0.80	0.80	0.80	0.80	0.80	0.80	0.80
0.16	0.12	0.16	0.28	0.36	0.40	0.60	0.80	0.80	0.80	0.80	0.80	0.80	0.80
0.16	0.12	0.16	0.28	0.36	0.40	0.60	0.80	0.80	0.80	0.80	0.80	0.80	0.80
0.16	0.12	0.16	0.28	0.36	0.40	0.60	0.80	0.80	0.80	0.80	0.80	0.80	0.80
0.16	0.12	0.16	0.28	0.36	0.40	0.60	0.80	0.80	0.80	0.80	0.80	0.80	0.80
0.16	0.12	0.16	0.28	0.36	0.40	0.60	0.80	0.80	0.80	0.80	0.80	0.80	0.80
0.16	0.12	0.16	0.28	0.36	0.40	0.60	0.80	0.80	0.80	0.80	0.80	0.80	0.80
0.16	0.12	0.16	0.28	0.36	0.40	0.60	0.80	0.80	0.80	0.80	0.80	0.80	0.80
0.16	0.12	0.16	0.28	0.36	0.40	0.60	0.80	0.80	0.80	0.80	0.80	0.80	0.80
0.16	0.12	0.16	0.28	0.36	0.40	0.60	0.80	0.80	0.80	0.80	0.80	0.80	0.80
0.16	0.12	0.16	0.33	0.45	0.50	0.75	0.90	1.00	1.05	1.10	1.10	1.10	1.10
0.16	0.12	0.16	0.33	0.45	0.50	0.75	0.90	1.00	1.05	1.10	1.10	1.10	1.10
0.16	0.12	0.16	0.33	0.45	0.50	0.75	0.90	1.00	1.05	1.10	1.10	1.10	1.10
0.16	0.12	0.16	0.33	0.45	0.50	0.75	0.90	1.00	1.05	1.10	1.10	1.10	1.10
0.16	0.12	0.16	0.33	0.45	0.50	0.75	0.90	1.00	1.05	1.10	1.10	1.10	1.10
0.16	0.12	0.16	0.33	0.45	0.50	0.75	0.90	1.00	1.05	1.10	1.10	1.10	1.10
0.16	0.12	0.16	0.33	0.45	0.50	0.75	0.90	1.00	1.05	1.10	1.10	1.10	1.10
0.16	0.12	0.16	0.33	0.45	0.50	0.75	0.90	1.00	1.05	1.10	1.10	1.10	1.10
0.16	0.12	0.16	0.33	0.45	0.50	0.75	0.90	1.00	1.05	1.10	1.10	1.10	1.10
0.16	0.12	0.16	0.33	0.45	0.50	0.75	0.90	1.00	1.05	1.10	1.10	1.10	1.10

If you are still then getting KR up top then you simply need one or more of the following:

• more octane
• less timing
• less load
• water or meth injection

[Ckmazdaspeed3]

Originally Posted by Ziggo I cant think of any reason it should spool faster than your ATP2871. The only functional difference is the ATP and the BNR is a journal, so you should have the advantage there. You are using an interrupt ECBS and an aftermarket manifold correct? E85 isn't really an option, the closest station to me is 18mi away. If the temperatures don't drop before my trackday in September, I will be adding race gas to the tank just to be safe. Lex, I know it will run lower boost temperatures at the compressor exit at the same flow as the K04, but it is also flowing more air, so it could be overloading the capacity of the FMIC to cool the air, particularly at these stupid high ambients. Nobody out there knows if the load based fuel cuts are active and functional? It looks like I will have to push quite a bit of air to get to a 2.1 load at 6500, based on my last logs it will have to be something like 24/25psi.

Firstof all, uou are right regarding spool. I Was looking at the graphs and it seems quicker on there. But when I reviewed the log, assuming this is 4th gear, you are right.

Also, fwiw, I am at 1.9x load at 6500 before E85 and super hot weather. This isat 20.5psi.

[Ziggo]

Thanks for the help with the knock tables, I will incorporate them in the next revision. I changed to the ITV22s and went out and logged this morning before it got crazy hot, but the KR still showed up so I am thinking its false. When the KR kicked in, it seems to have caused the WGDC to drop (and the boost)

I upped to boost target to 22psi at 6k and up since I am not much worried about boost spikes in this range. This is pretty much convincing me you are going to need to upgrade the MAP sensor to get the most out of this turbo. Its borderline, but I think it could push to 23-24psi to redline before running out of breath. EGTS are just starting to touch 1600* in this log. Like others have said, the DISI head flows like shit up top, gotta really up the boost to pack the air in. I fiddled with the load targets in this revision, it seems they were restricitng my top end boost before, and it seems they are keeping me from hitting my boost targets on the low end now. The load targets are 2.1, and I am hitting those before getting to the boost target on the low end. Still not sure what exactly the boost toggle does if this is the case...

[Ziggo]

For reference, a compressor map with a red dot at my operating point @ 21.3psi. 24psi will push it a bit beyond the current map. 22psi is probably all its got reliably as that would land right on the final shaft speed line, so I don't think upgrading the map sensor is really going to gain me much.

[cld12pk2go]

Originally Posted by Ziggo For reference, a compressor map with a red dot at my operating point @ 21.3psi. 24psi will push it a bit beyond the current map. 22psi is probably all its got reliably as that would land right on the final shaft speed line, so I don't think upgrading the map sensor is really going to gain me much.

You can still overspin it a little beyond the last speed line, but I agree you probably don't want much more than ~22-23 PSI at 6000 RPMs and maybe 21-22 PSI at 6500 RPMs.

You could push 24-26 at lower RPMs though and still be on the island. It just depends on what you are looking for.

[Ziggo]

I am staying away from higher psi at lower RPMs because I still have Stock injector seals.

My project next summer (assuming the child I have on the way that I just learned about doesn't interfere) is the intake manifold, injector seals, and stud/nut crows feet along with secondary fuel injection. Will probably replace the MAP sensor with a 3bar at that point and then start retuning to make more torque.

Still saw the knock up top in my new logs though the new decay rate makes it much more easy to isolate. I will just have to cut some timing and see if that fixes it. (back to 9* peak timing, down 5* from my K04 tune)


---
- Zigatapatalkreplacea

[Lex]

If the turbo has breath up top, try adding some intake advance via VVT in the upper RPMs and see if that increases flow across the motor.

[Ziggo]

V3 Log from earlier today. Only changes were to the knock tables, and some very small tweaks to the WGDC table.





Going to make some timing tweaks tonight to see if I can get rid of that top end knock and add in Dano's VVT tables. Also going to start logging VVT in preparation for making some changes there. This is all the boost I will be pushing at redline, starting to kiss EGTs of 1750* now. I want to spice up the midrange some, but I am not sure at what load the injector seals start to lift. My feeling is 2.1 is safe, 2.3 is risky, and 2.5 is eventually going to cause problems....

[Ziggo]

The degree of timing I pulled worked and I finally got a clean pull in with no knock. I am down to 9.5* peak timing now though, down from 15* on the K04. My loads are higher so I would expect it to come down some, maybe to 13*, but I seem to have reached the limit for 93oct gas.

Will start testing VVT changes with the following strategy.

Increase by +5 across the rev/load range all cells below 35 and take a WOT log.

In areas where I am limited by load, see if the boost psi drops, in areas limited by boost, see if the load goes up. Make adjustments to the VVT table accordingly at the measured load points and all load points above.

Once WOT is stable, repeat the process at lower load levels by using custom maps with nerfed throttle DBW tables and lower TRL tables New logs will be done at wot as well but the throttle plate won't open as much so I should be able to take logs that carry a constant load across the rev range even at PSIs below WG spring pressure. All these will be load limited, so benefit will be evaluated by boost psi.

Anything in error with this plan? My spring pressure is 16, so in order to evaluate this down to a 1.0 load I need the throttle to help limit boost.


---
- Zigatapatalkreplacea

[Ziggo]

Rechecked the rules today, and it seems NASA TT has changed its rulebook to allow methanol injection as long as the mixture does not exceed 50/50 and you don't use it as a fuel (whatever the fuck that means)

Things that make you go hmmmmm


---
- Zigatapatalkreplacea

[cld12pk2go]

Originally Posted by Ziggo Rechecked the rules today, and it seems NASA TT has changed its rulebook to allow methanol injection as long as the mixture does not exceed 50/50 and you don't use it as a fuel (whatever the fuck that means) Things that make you go hmmmmm --- - Zigatapatalkreplacea

So basically they want to keep it non-flammable.

[Fobio]

The local series (CASC - Ontario Time Attack) also associate methanol injection with nitro methane...lol...but once we explained to them it's water/alcohol injection, they were more than OK with it.

[Ziggo]

Yeah, now I just have to figure out out how to put a decent MAF based kit together. I never looked into it before because I got looked at like I was crazy when I asked if it was legal.

I think the hardest part will be finding a spot to mount a 2 gal tank. I don't think I would want to go less than that due to sloshing concerns and to give myself headroom for a 10 gal/hr nozzle.


---
- Zigatapatalkreplacea

[cld12pk2go]

Originally Posted by Ziggo Yeah, now I just have to figure out out how to put a decent MAF based kit together. I never looked into it before because I got looked at like I was crazy when I asked if it was legal. I think the hardest part will be finding a spot to mount a 2 gal tank. I don't think I would want to go less than that due to sloshing concerns and to give myself headroom for a 10 gal/hr nozzle. --- - Zigatapatalkreplacea

I run a 3QT tank with an M7 and only fill up about once per week.

Install thread with pics here including how to wire for MAF:

How to install an Alkycontrol kit with PAC controller.

[Ziggo]

To have the warm and fuzzies I need a big enough tank to run at WOT for 15 minutes. The progressive controller should keep the consumption down, but at the end of a long session I am already distracted by managing overheating tires and brakes, I don't want to be worrying about the amount of meth I have left too.


---
- Zigatapatalkreplacea

[Ziggo]

I found a local shop that rents their dynojet for 100$/hr so I will plan on getting on there sometime next week to develop the VVT tables. Will be posting results here so I am mentioning others that I know of with 2871s so they can take advantage. Should be pretty consistent across cars provided you have a TBE.

@Dano2010
@Ckmazdaspeed3

[Ziggo]

Originally Posted by cld12pk2go I run a 3QT tank with an M7 and only fill up about once per week. Install thread with pics here including how to wire for MAF: How to install an Alkycontrol kit with PAC controller.

How much should I expect to pay for an alkycontrol system? I think I will have to call and setup a custom kit as they don't have one listed for mazdas, but just curious on the ballpark figure.

[BlueStreak]

Originally Posted by Fobio The local series (CASC - Ontario Time Attack) also associate methanol injection with nitro methane...lol...but once we explained to them it's water/alcohol injection, they were more than OK with it.

Wow. You'd think an organization that governs car racing would have an idea of what WMI is...

[Ckmazdaspeed3]

Well, I have a catted DP and stock cbe. Will that make a big dif?

That should be pretty cool though. I only got half way through mine before the clutch issue... All I know is that 35° got me the best spool, but I never finished finding where to start to taper it or how quickly I could taper it off.

Send me your pay pal and I'll send u twenty bucks because I know that u will save me tons of time! Also, I know that you understand this stuff a lot better than I do, so u will get better info I'm sure.

[Ziggo]

Don't worry about the $$. Maybe I find some magical results, maybe it makes no difference, more likely I find a few percent here and there and improve spool in a range where only I care about (4.5k and up)

I would just do it on the street, but the idea of 20+ highway pulls in one area makes me nervous about police.

The stock CBE might be a bit limiting on the top end, I am using a catted DP as well, and I think Dano is too.

[cld12pk2go]

Originally Posted by Ziggo How much should I expect to pay for an alkycontrol system? I think I will have to call and setup a custom kit as they don't have one listed for mazdas, but just curious on the ballpark figure.

I would expect somewhere in the ~$500 range, but that is just a guess. Just call up Julio for a quote.

I have had my kit for 7 years, and I am still loving it. It is simply the best out there (in my opinion).

[Ziggo]

Update: I never made it to the dyno to tune, but it's holding up fine on the track. Not seeing any knock, but I am mixing 3 gal of 109 to 9 gal of 93.


Zigatapatalkreplacea

[cld12pk2go]

Originally Posted by Ziggo Update: I never made it to the dyno to tune, but it's holding up fine on the track. Not seeing any knock, but I am mixing 3 gal of 109 to 9 gal of 93. Zigatapatalkreplacea

1/4 mile time and mph?

[Ziggo]

Originally Posted by cld12pk2go 1/4 mile time and mph?

Wrong kind of track, lol. Don't have a better way to judge other than I was about even with a camaro ss on the straights. Still getting walked by Z06 corvettes, and one very stock looking V8 miata. (he missed his apex and I set him up for a pass, but he surprised the hell out of me when he it the go pedal)

My clutch is starting to get overheated though. Could be power, but I know part of it is I can't heel-toe this car for shit when wearing my driving shoes, and at 40k miles and lots of trackdays it could just be giving up the ghost. I am going to put on a fatter gas pedal and clutch before my next trackday.


Zigatapatalkreplacea

[Ziggo]

Trackday report & lap video:

MSR Cresson 3.1 9/10&11/2011


Zigatapatalkreplacea

[Ziggo]

FWIW, Virtual dyno says my shit is slow. I tend to agree, but think I fail at it somehow too. The curve looks like crap for some reason. I don't see anything in the log that would make it do that funny kick but who knows...



I may need to do a compression check. 22psi/300 hp doesn't really compute.

[BlueStreak]

Ziggo, my VD logs were doing the same thing. Turns out compression was way down in cylinder 3 due to a cracked ring land.

Hope yours isnt as bad.

[Ziggo]

Originally Posted by BlueStreak Ziggo, my VD logs were doing the same thing. Turns out compression was way down in cylinder 3 due to a cracked ring land. Hope yours isnt as bad.

That is not encouraging.