MY ATR Journey...

[cld12pk2go]

So I got my AP and the ATR software.

Now time to tune.

I started with the Stage2+SF 93 v103 map. My mods obviously don't match the map notes all that closely (I have an PG exhaust manifold, CP-E CAI, Alky control kit, and PG Turbo Inlet beyond what is covered in the notes and a stock CBE).

To start, I needed to calibrate the MAF table A/B. To achieve this I logged my LTFT and MAF flow as as per the ATR help file after putting a gentle ~100 miles on the flash (methanol injection disabled for initial testing):



The 5-16 g/s and 16-28 g/s areas where close enough to zero to not alter. The others were adjusted.

I smoothed the transitions between the regions that were shifted as follows:



And what I got after:





Final MAF A/B Values:

0	0	0	0	0	0	0	0	0	0.02	0.03	0.06	0.09	0.12	0.17	0.23	0.3	0.37	0.45	0.55	0.65	0.77	0.89	1.03	1.17	1.33	1.52	1.71	1.92	2.15	2.41	2.7	3	3.31	3.68	4.09	4.55	5.22	5.7	6.14	6.62	7.16	7.75	8.37	9.07	9.82	10.59	11.42	12.29	13.17	14.08	15.05	16.1	17.21	18.38	19.61	20.9	22.26	23.68	25.17	27.5	30.29	32.07	33.9	35.81	37.79	39.85	41.98	44.19	46.5	48.89	51.35	53.84	56.37	58.96	61.61	64.33	67.13	70	72.97	76.07	79.24	82.52	85.86	89.29	92.8	96.36	99.97	103.67	107.45	111.33	115.3	119.38	123.57	127.86	132.34	136.96	141.7	146.56	151.54	156.65	161.87	167.21	172.68	178.26	183.94	189.73	195.65	201.71	207.88	214.18	220.62	227.2	233.93	240.8	247.81	254.96	262.26	269.69	277.29	285.04	292.93	300.98	309.19	317.55	326.08	334.77	343.62	352.64

[cld12pk2go]

Since the entire section I increased by 7% in my MAF tables results in higher calculated load (effectively lowering my load and yes it is noticeable), I next moved to analyzing/tweaking the Throttle - Req. Load (TRL) tables.

Here is the stock map for reference.



It is pure weak sauce...Notice that the profiles are identical for gears 3-6.


Here is the Stage2+SF 93v1.03:



What is interesting is that Cobb is greatly pulling power in gears 3-6 in this map. This is something I have noticed. 2nd gear would rip and 3-4 would be much softer. Does anyone have a clue what Cobb was doing here?



I am planning to work the loads up slowly in the upper gears towards the 1.9-2.0 range. My plan is to stop where I hold ~18 PSI through most of the power band. I also am currently limiting torque in the 2500-3000 RPM range to save the rods and not attempting to fry the turbo by holding high loads above 6500RPMs. This is obviously a work in progress. I plan to load the map with these changes tomorrow and play with it over the weekend.




One thing to note is that I have a stock HPFP and it is rock solid at 1600 PSI+ even before the +7 MAF adjustment.

[cld12pk2go]

I didn't get much testing done over the weekend due to my car leaking antifreeze.

I was pretty concerned at first thinking it was a head gasket; however, I quickly found that during the recent install of my PG Turbo Inlet I appeared to have dislodged my reservoir hose from a clamp by my radiator fan. It had bumped up against the fan blades until a pin hole was worn through.

I fixed this today by cutting out the 1'' thin spot and inserting a 3/8'' to 3/8'' barbed fitting with hose clamps.

When I first smelled the sweet ethylene glycol sent, I was just in shocked disbelief since I haven't had too much boost yet and haven't had more than 1° of KR in my testing... That would have mega sucked.

Anyways...


Loaded this TRL combo this morning and just letting the computer learn:



The intent is to have something that resembles traction in 1st and 2nd gear coupled with not too much torque at 3000 RPMs.


I also modified the all of the fuel tables to something similar to the following in this version:



The intent is to aim for 11.5:1 any time I am in boost except for around 2000 RPMs when the old values were targeting 13.3:1 I dropped to 12.5:1. This will slow my spoolup very slightly; however, I will happily take that with the added protection against detonation.

[cld12pk2go]

1st-4th gear run this morning...



AFR's were below 10:1 at the top of 2nd, but my fuel press dropped off to ~1100. I am not sure if this is significant since I spun through 1/2 and my TC/DCS was active...




On this one I had a definite FP drop to about 1050 PSI briefly. AFR increased to 12.2:1 during low pressure excursion. I have approximately 20 logs with no FP drops below 1500 PSI prior to aiming for 11.5:1 AFR...




Everything was good on this run except my BAT (not on this plot) topped out at ~150°F. Stock IC FTL. This is why my AFR dropped in the top of 4th.




So I decided it was time to engage my methanol injection again. The car responded by losing traction from a 3rd gear roll for the first time. Nice. Very nice. Torque FTW!





Another meth enabled run. My BAT was ~75°F at the end. Meth = awesome (as is the DaskHawk sample rate!).

Looks like the meth was flooding in around peak boost in 3rd gear (AFR dropped below 10:1). Still pulled like a mofo though...

So since I appear to be encroaching on the limits of the stock fuel pump, I am going to lean out my AFR maps to target closer to 12:1 and keep the meth active. I am also going to lower my load targets in 1st/2nd since traction is just not happening with my current limits. I will also increase my 4-6th load targets slightly.

[cld12pk2go]

New TRL tables:



Load reduced 0.05 in 1st/2nd to limit wheel spin. Load increased 0.03 in 4-6th.

[cld12pk2go]

A pull from this 7/21 (the label on the charts is out of date, these were with flash v1.04c not v1.04b):



Everything looks pretty swell...

One thing I believe I have found out is that the "Engine Load (%)" parameter that we can datalog with the AP appears to be the "Abs Load" parameter in many ATR tables * 100.

I have added it to my graphs (divided by 100) and it lines up well with the TRL table values.

For example I was between around 1.8x in most of 3rd and 4th gears.

My Boosted Air Temp dropped to about 70°F and increased to 85°F at the end of the run. Stock IC FTL!

A little info on my meth system... I have it wired up as progressive off of the MAF voltage. It turns on at 2.8v and ramps up pressure as the voltage (mass flow) increases. The 2.8v turn on means it will be active anytime I am in boost except in low boost situations in 5th/6th gears at ~2000 RPMs.

More info on the meth system: http://www.mazdaspeedforum.org/forum/foru...ontroller.html


A pull from 7/23 (Dashhawk):



A little knock (0.7°) but nice an clean up to 6700 RPMs. AFRs were very good. Fuel pressure never dropped below 1600 PSI. I would like to get my boost pressure to taper a bit more above 6000 RPMs. So I will modify my TRL tables in my next flash.

Another pull from 7/23 about 3 min later:



This is odd. It looks like at my CL/OL transition the AFR dropped from mid 12's to mid 9's, which resulted in my fuel pressure dropping to ~1300 PSI.

The only fuel table that has AFR's in the 9's is the "Fuel Commanded EQ Max Enrichment Allowed", which still has the default 9.072 in all RPM values. I will increase these to 10.5:1 in my next flash.

I guess I will start shopping for a fuel pump tonight in case I cannot tune my way out of this...



Updated TRL Tables:



Changes:
Max of 1.40 load in first gear (down from 1.45)
Max of 1.55 load in second gear (down from 1.65)
Max of 1.75 load at 6000 RPMs (down from 1.86)
Max of 1.45 load at 6500 RPMs (down from 1.50)
Max of 1.25 load at 7000 RPMs (down from 1.30)


7/24/09 Update: Well, I had another dip today on my fuel pressure to 941PSI on a second gear pull...so I just ordered a KMD fuel pump kit. I will not beat on my car anymore until I have the pump installed hopefully next Friday.

[cld12pk2go]

So I got the KMD v2 installed yesterday. It appears to be working quite well thus far:



Pressure was 1700-1800 for most of the run and was a little over 2000 between shifts.

You can clearly see how my lower load in the TRL tables for second gear compares to the higher load in 3rd. I am still spinning the tires some around the boost spike in 2nd gear. I will see if I can tweak the wastegate table some to prevent the spike. The 14-15 PSI range appears to be perfect for second gear.

Updated TRL, which i will load tomorrow:




The stock v103 WG DC table and my first edit of the WG DC table. The intent here is to reduce the boost spike around 3k rpms...





I used Dreads APP transition table and Throttle Requested Load tables A/C to try to smooth the jerky throttle situation:





Here is what the v103 Throttle Requested Load tables A/C looked like (jerky!)



I applied SLS MS3's Closed Loop to Open Loop tweaks and lowered the Closed Loop - Max Load D table to below my 1st gear load target of 1.40. This should keep me in open loop anytime I smash the gas.




Updated Boost Target Table:




Upgraded Fuel Target Tables(I changed all of them, just a generic shown here)

The intent here is to reduce the AFR in the 1.1-1.3 load range so that during spool-up I have ~12:1 AFR instead of 13:1+.


So, how did it work? Here are some pulls from today:


A 4th gear pull. The AFR's dropped to 12 very quickly and stayed there. Not shown on this graph but I was in open loop at 2 PSI of boost around 2400 RPMs.


2nd-4th gear:


Looks like the 2nd gear boost spike is still there and I cannot achieve my load target in 3rd gear with 92°F ambient temp (WG DC up to 97%).


Another 2nd-4th gear:


AFRs are rock solid. My BAT dropped to 78°F before rising up to 98°F by the end of the run (stock IC = heatsoaked). I would say 100% meth with an M7 nozzle = win (BAT 14°F below ambient).


3rd gear run:


No boost spike. Nice. I was actually over my load target and boost target at ~2700 RPMs though. I will need to adjust the WG DC some more later.


8/10/09 Update:

Passing in 4th gear sure is fun now:



8/11/09 update:


Pulled hard for this entire run. Looks like the combination of reduced TRL, boost target, and WG DC is keeping the boost spikes under control.



8/13/09 update:

Updated my WG DC table some more here:



3rd gear pull. I still have a nicely controlled boost curve without the spike and the ramp up after 3000 RPMs.



This one is a 4th gear pull from low RPMs. You can see that I am hitting 1.4 load at 2000 RPMs (vs my TRL table target of 1.20) and 1.7 load at 2250 RPMs (vs my TRL table target of 1.40) and then it achieves and holds the 1.65-1.75 load target until after 3000 RPMs when it starts to ramp up the boost.

It cannot control the 2000-2500 RPMs loads to the targets due to the physical constraints of the system (spooling too soon coupled with WG spring pressure being ~10 PSI or so). However, it stilll limits the boost at low RPMs to 12-13 where I would have been spiking to 15-20 prior to my AP.




Just a nice 2nd to 4th pull here. You can see the boost tapering down to 14 PSI at 6500RPMs in the top of 3rd. I was achieving my 1.90 load targets in the cooler air. MAF was up to 4.62v. WG DC ended up in the low 90% range in the top of 4th.

I think I am getting really close on my map to being complete.



3rd gear pull. WG DC maxed out around 5500 RPMs.




2-3rd pull with WG DC also maxed out around 5500 RPMs.




2nd-5th pull.






Here is my current tune (still based on the Cobb Stage 2+SF 93 v103) v104e:
Here is my current tune (still based on the Cobb Stage 2+SF 93 v103) v104f:

[cld12pk2go]

So after reviewing my datalogs from v1.04f, I have made the following tweaks to create v1.04g:



Here I am trying to reduce the WGDC being pegged at 100% at 5500-6000 RPMS by slightly decreasing my TRL targets. Also, I am upping my TRL targets where the stock snail has potential to support some additional torque (3500-4500 RPMs), and I am upping the 3000 RPMs load 0.03 as well in 3rd-6th.

In addition, I got some new tires this week. 215/45ZR-18 General Exclaim UHP XL from tirerack.com. They are a great deal; however, they are not quite as grippy as the stock tires. Thus, I am slighly lowering my load targets in 1st/2nd to limit wheel spin.

These tires do give a much smoother ride (not nearly as jarring over small bumps). A very nice trade IMHO.





Just trying to clean up the load and boost dynamics tables here to allow better PI loop response.




Added launch control and flat footed shifting. These should be fun!




New boost targets that are aiming for higher load where the stock turbo can support it.




My first edit of the "Boost Table - Throttle Close" table. Obviously, I am using this as another layer against too much boost at 3000 RPMs or below.

[cld12pk2go]

New tweaks:

New TRL table. I was never able to hit the real low calculated loads around 2500-3000 (even with relatively low boost), so I bumped these up a little.



I am trying to see if this will help my loads more accurately follow my TRL tables.





I decided to drop the AFR aim from 12:1 to 11.8:1 for a little more piece of mind. (Still getting used to Excel 2007 charts).



Still tweaking the throttle close table to keep boost in check down low:



Dread's throttle tweaks did smooth out a lot of the jerkiness; however, there were some "off" spots, which I am going to see if can be fixed by returning to stock APP Translation tables...v104h = v105 stock now.




Results: (engine load is now /10 to be able to see with higher resolution, and I forgot to change the early chart titles for v104h)



Morning pull. No problem hitting the load targets with 60°F temps. WGDC well in check.



Morning pull. No problem hitting the load targets with 60°F temps. WGDC well in check.



Afternoon pull. Looks like the throttle closed around 2500-3000 RPMs a little to prevent overboosting. Nice. WGDC is pretty much maxed out on the high end and I never quite achieved the 1.86 load target above 3500 RPMs.



Afternoon pull. 4th gear roll from about 3k RPMs. Looks very nice here except for not achieving the 1.92 load targets.



Afternoon pull on 8/27 in 4th from ~3400RPMs. Passing was fun today at 21 PSI. :-)



From 8/31: These are my first two attempts at FFS. Pretty neat. The AP log doesn't even show a boost dip on the second chart. Feels good. It will take some getting used to not to lift my right foot on shifts though. I have 15 years of muscle memory to fight.




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

9/10/09 Update:

Changes going into v104i:

Boost target table updates:


Basically just want to be allowed to go over 20 PSI where the turbo can support it and I am not achieving load targets.

Boost limit tables updates to support the higher boost targets:

I am knocking down the boost limit - throttle closed in the lower RPM range to further limit the potential for too much boost down there.


Fueling changes: Basically, I have observed that my meth activates around 0.85 load, which means that my old settings would aim for 14.68:1 AFR in closed loop when the meth was active.

Since meth has a stoich of 6:1, this will result in a very lean mixture with my M7 nozzle.

My car hasn't run real crisply in the past when I am holding the load just above the meth activation point. I attribute this to the surmised leanness.

Evidently, since I have not had any issues this is not a major problem at these loads (low load + high % of fuel from meth at 116 octane + evaporative cooling effect = no detonation even while very lean). I should probably run the calcs on the % meth fueling I am running at this these loads and determine appropriate AFRs to aim for. I can do this by measuring the g/sec of meth injected and the amount of gas and g/sec MAF airflow and run the equations.

However, I can also just richen it up until the car is happy, which is what I am doing in v104i.

So I have added fuel down to 0.75 load, which should give much more margin for error here.


9/11/09 update: The car feels much crisper now around the meth activation point. I did some more logging today, and I will probably richen it up a little more pretty soon.

TRL updates:


----------------------------------------------------------------------------------------------------------------------------------
9/12/09 Update:

Here is a log from 9/11:


FFS still rocks. Pulled up to 21 PSI at around 3500RPM after being limited to 18 PSI at 3000 RPM. What is weird is that my load is flat at 1.873 from 2800-5200 RPM in 3rd gear. It looks like the ECU is controlling it to sustain this load, but I do not understand why it isn't targeting the values in my TRL table.

Also of interest is that it appears that I can sustain more than my TRL target of 1.80 at 5500RPMs and 1.75 at 6000RPMs. I will likely bump these in my next flash.

I did have 1° of knock retard briefly, but this isn't enough to be concerned about.




Changes in v104j:


Adding some additional fuel down low. I determined that during normal highway cruising, my load is ~0.4-0.5. So all this additional fueling should have only a minimal impact on fuel economy but should offer considerable additional protection for the engine. Seems like a no-brainier to me.


Tweaking the load up in the 5500/6000/6500 cells.

I also copied my fuel cut values down into some lower barometric pressure lines just in case I take the car into higher elevations at some point.



9/24/09
Changes for v104k: I lowered the AFR in the 1500 RPM cells and increase AFR back to 14.68 in the 0.69 load row. This was due to being leaner than I would desire when boosting below 2000 RPMs and occasionally hitting the 0.69 load under cruising conditions.

I have pretty much ran a tank on v104k now and have gotten no significant decrease in mileage (still about 320 a tank). I also changed my meth turn on point from 2.8v to 2.9v on the MAF signal to make it a little easier to not hit it when driving mildly.


9/27/09 data logs: (running 80/20 meth/water instead of my previous 100% meth)


3rd gear run. Pulled nicely.



2-4th gear run. FFS still rocks.

Everything is looking pretty good. Just too bad the turbo is maxed out.


9/29/09 Update: Cooler weather arrives. Traction in 3rd gear goes away...


No issues.


No issues.

I will likely increase my boost target at 5000RPMs on my next map from 19 PSI to 20 PSI. I currently drop from 21 PSI at 4500 to 19 PSI at 5000. This change will smooth it out and produce a little more power.


---------------------------------------------------------------------------------------------------------------------------
9/30/09 Update

Made some additional tweaks to my tune for v104l:

Up to 20 PSI from 19 at 5000 RPMs.


Dropped my 2250RPM WGDC from 50 to 38 to help keep the boost smooth from low RPM rolls in high gears.




More tweaking of boost/load dynamics. Specifically, I am trying to see if more aggressive values can keep the loads in check from lower RPM pulls in higher gears.

10/2/09 results:



I still do not appear to be targeting load since I am .08 above my load target at 3000 RPMs where my load dynamics table calls for -28% WGDC. My actual WGDC was about 48% and my WGDC table calls for 50%.

It appears that I am controlling boost here since my boost target is 17 PSI and my actual was 17.2 PSI. The 0.2 PSI error should only call for about -0.6% WGDC from my table value.

However, this was a nice clean pull...

[cld12pk2go]

10/4/09 update:

So I am still trying to understand when/why the ECU targets boost and load.

I am sure there is some type of underlying logic (hopefully other than a random choice generator) as to which is chooses to control .



Here is one run from yesterday with comments (4th gear):






And another (3rd gear)





The only thing I can conclude after 30 min of analysis is that my WGDC table for 3500-6000 RPM under WOT are set too low based upon my current TRL and boost targets in those ranges. I suspect this coupled with some max adjustment to WGDC in logic is preventing me from achieving my load targets.

I will correct for this in v104m to see if it helps.


It had been since v104f that I have altered the WGDC table in these ranges. I brought them up between 2-5 in the 3500-6000 RPM range.


10/5/09 Update:





Two clean runs that pulled hard. However, I still am not getting the 1.95 load target. Not sure why. It appears to be controlling load very well at the 1.91 level and then tapering as per my TRL tables (and turbo being maxed out).

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
10/9/09 Update: Dyno Time: http://www.mazdaspeedforum.org/forum/foru...58-0wtq-39636/

2008 MS3

Mods: CP-E CAI, PG turbo inlet, PG V1 manifold, CP-E DP with Cat, Alky Control Methanol injection, KMD V2, Denso ITV22, Cobb AP ATR

Notes: stock CBE, stock IC, stock BPV, running 100% methanol with M7 nozzle, and 93 octane. I am actually specifically limiting power at 3000 RPMs or else the torque might have been higher.

Runs at 65°F ambient. SAE correction factor 1.02.

The tires spun on the pull 001 (wet from drive) so that one was deleted.



I guess that solves the mystery of my lack of traction from 3rd gear rolls.


I am actually surprised that the HP stays so flat all the way to 6700 RPMs. Nice.

Pull 002 datalog:



Pull 003 datalog:




Video of pull 003:

" target="_blank">

" target="_blank">

I was only getting ~1.83 loads today. I had been getting 1.92 as recently as two days ago.

What is interesting is that my max MAF voltage on these runs was 4.51 and 4.52. I have hit 4.62 when it is ~10-15°F cooler outside and the WGDC isn't maxed out so badly.

Just scaling my MAF g/s at 4.52 to 4.62 would be an increase from 255 to 273 g/s. Or an increase of 7%.

This means if I was to dyno in cooler conditions I would expect roughly 291.5* 273/255 = 312whp (Max torque would likely stay the same since the WGDC isn't pegged at peak torque).

I think I might be done tweaking my car for awhile now. Time to enjoy.

Tune attached in dyno thread.

[cld12pk2go]

When to the drag strip on 10.25.09 to see how much I suck at driving.

Here are the results:



Run 1: 14.0 @ 106.39 mph
Run 2: 14.1 @ 106.51 mph

Neither was a clean run. I used LC (2800 RPMs) on the first run and spun badly in first gear.

On the second run I just tried to launch around 2200 RPMs and bogged badly and then hit the rev limiter in 3rd.

I was not shifting into 5th.

In hind sight, I would have been better shifting around 6k instead of trying for 6500 based upon my dyno.

I was hoping to get more runs in; however, 2 cars puked their guts up and I only got 2 runs in 3 hours of waiting so I decided to head home.

While I wish I had gotten a good pass in, I think the decent mph shows that the power is there.



FMIC FTL! At lauch, my BAT was ~150°F. The meth brought it down to 75-90°F.





FMIC FTL! At lauch, my BAT was ~150°F. The meth brought it down to 80-90°F.



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

11/15/09

More tuning tweaks:

I changed my meth activation point to about 2.9v so I have tweaked my fuel table to go rich a little later.




My AFRs in OL are not quite as stable as I would like. Specifically, in the 3.8-4.4v range I am a little lean (~11.8-12.0) and above 4.5v I am often a little rich (~11.3). Thus I am tweaking the MAF curve in these ranges slightly +3% and -1% to try to drive these to 11.5:1.

Latest tune attached:


---------------------------------------------------------------------------------------------------------------------------------
1/3/10 update

My adding to the MAF curves did not give me any noticeable effect on the AFRs, which is very odd. I made map 104r with another 3% added and still no change. This is very strange.



Map v104s will revert to the previous MAF curves to gain the load back in the 3.8-4.5v range.

I will be installing a FMIC in the near future and will have to see what needs to be tweaked once gaining some data.

I am tempted to add some timing in the 4500-5500 RPM based upon this thread: Discussion regarding power limitations of the MS3

[bf360]

nice work in progress gl

[Speedn3]

good stuff...thanks for sharing.

[cld12pk2go]

01/23/10 Update

So I got my PG piping kit and CXRacing IC0035 FMIC installed. I had some issues to work though as documented here: http://www.mazdaspeedforum.org/forum/foru...install-46831/


So now it is time to tune.

For starters I had a pull in 4th gear from a 3000 RPM roll and wrapped the boost gauge around to 24PSI! Obviously, the IC0035 core is a shit-ton less restrictive than the stock TMIC! Notice that the logged boost pegged out at 22.63 PSI.



After about 1/2 a second I got off the throttle since my tune is obviously not ready for prime time with the new FMIC. No audible knock was heard on this pull, so I am currently associating the large knock spike with the normal "let off the throttle" knock that often appears. This is when a higher sample rate from the AP would be damn helpful...

This datalog concerned me greatly because both safety nets failed to keep boost under control. The throttle close and fuel cut. After examining my tune, I believe the failure of the fuel cut was due to it being set at 22.5 PSI. There might simply not be enough voltage difference between 22.63 and 22.5 PSI to reliably trip. So I am going to cut this to 22 PSI on v104t. The trottle closed was set to 22 PSI, and didn't appear to do anything. This will be dropped as well.









I am also tweaking (for the first time) the "Load Tables: Abs Load Limits - Fuel Cut"








Hit the 22 PSI fuel cut on this pull with tune v104u:


What is interesting is the MAF voltage pegs at 4.99v on fuel cut. Odd.



Here is another pull where I just miss fuel cut at 22 PSI. Pulls damn strong...




Looks like I am overboosting so I will need to pull some more out of my WGDC table.










Other tweaks for v104v, which I will install this morning:

Based upon this thread, I am going to see if I can limit load further in 1st/2nd by lower my TRL values and making my "Abs Load Target" table match my "Throttle - Requested Load" A/C tables.







I am also increasing my TRL and boost target tables based upon the extra potential to carry boost farther out into the RPM band with the lower restriction core:










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

3/10/10 update


A fun pull from 3-5th. Looks like I am holding boost a little higher and longer with the FMIC. I will probably re dyno pretty soon to see what the FMIC got me. I expect to see a decent increase in area under the curve.

I am also contemplating getting the Racing Beat exhaust, since it seems that the stock CBE must be a 10+ HP restriction at the ~300WHP level.

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

5/30/10 Update:

I recently updated my AP software/firmware to the latest versions and examined the Stage2+SF 93 v106 FMIC timing map. I noticed that Cobb essentially added 1° of timing to this map over the previous one I had been using.

Here is my previous map used for both the High Throttle/OL (Knocking) and High Throttle/OL (No Knock)



I made the following tweaks:



I pulled some timing around spool up (2500-3000 columns). The logic is that I occasionally have a little knock around spool up. I think this could be related to my going to OL around 1.1 load. The CL timing maps have much less aggressive timing in the spool up range compared to the OL maps. I pulled the 0.5° in the 3500RPM range to smooth the timing decline with increasing load. This results in timing that is basically 1° more advanced than the stock map in these ranges.

The 4000RPM column had a goofy anomaly where the timing was increasing at the last couple of rows, so that was smoothed out.

I left the 1° of additional timing in above 3500RPMs except the last couple rows in the 4000RPM column and where I smoothed the timing decline with increasing load in the 5500RPM column.


This resulted in my current 106b OL timing maps:




I also noticed that I occasionally hit fuel cut around 2k RPMs, due to my ABs Load Limits - Fuel Cut having 1.33 in the 2000 RPM column. This was altered to 1.60.


New map posted.

------------------------------------------------------------------------------------------------------------------------
New updates start next on post 43. This page is getting too freaking long...

[SpdFreak]

Post some raw logs too, not just the mumble jumble graphs.

[Lex]

Thanks for posting your experience. I know you know this but I suggest looking into a pump going forward.

[kgb]

Originally Posted by cld12pk2go New TRL tables: Load reduced 0.05 in 1st/2nd to limit wheel spin. Load increased 0.03 in 4-6th.

Great...I was hoping to find someone else upping load in 3rd gear and upwards....I am not as far as you but do have my 3rd and 4th gear hitting. 1.83 around 3500 - 5000. If everything goes well, I will up it somemore! Oh, I also reduced load below 3000.

[94jedi]

wow, excellent work and thanks you for documenting all of this. I have a long way to go and a lot to learn/read lol.

[wakeboarddude]

Nice! This is good reference and training aid! Good job bro!

[spd frk]

This thread FTW. I am fooling around the same with my ATR/AP but I have less mods and htis is helping me understand how some of the variables are changed.

J

[cld12pk2go]

Originally Posted by SpdFreak Post some raw logs too, not just the mumble jumble graphs.

I have no problem posting the raw data by exception if there is a run you are interested in; however, I will plan to continue posting the graphs pointing out what I consider the important points.

[cld12pk2go]

Originally Posted by Lex Thanks for posting your experience. I know you know this but I suggest looking into a pump going forward.

Yep, I will continue to monitor closely. The runs I saw the dips on where worst case (from the HPFP's perspective) with high fuel load being requested.

As of this morning I have leaned out the fuel tables to target 12:1 instead of 11.5:1 (~4% less fuel) and am adding a significant amount of meth (M7 nozzle at ~200+ PSI).

However, if I see any more dips I will absolutely have to give very serious consideration to a replacement or backing off on the load targets.

[cld12pk2go]

Originally Posted by 94jedi wow, excellent work and thanks you for documenting all of this. I have a long way to go and a lot to learn/read lol.

Originally Posted by wakeboarddude Nice! This is good reference and training aid! Good job bro!

I am glad you all are enjoying. I am documenting my journey as a guide for others.

One thing I am surprised about is how crude the Cobb cookie cutter maps are.

They are certainly better than stock, but leave much to be desired.

ATR is really essential to obtain something that approaches our car's potential.

My car is really ripping now. I am very content with the power level. I just need to knock off a few rough edges (throttle can be jerky/unstable, which I will address after the AFR/load is straightened out) and enjoy for a long time.

[cbusby]

Thanks so much for this, I've been following along and making a few changes myself. Finally starting to get the results I've been hoping for, these load tables are magic.

[Speedn3]

can you post a graph showing how you smoothed the data point dislocations between break points?
it would help if you could detail it like the first graph in this thread.
thanks

[cld12pk2go]

Originally Posted by Speedn3 can you post a graph showing how you smoothed the data point dislocations between break points? it would help if you could detail it like the first graph in this thread. thanks

I added an update to the first post in this thread to show how to perform the smoothing. I hope this helps.

[Trader]

When I did my MAF calibrations, I adjusted the voltage to account for the trending STFT as well as accounting for the LTFT.

For example,

Looking at your 28 to 77g/s area, your LTFT is at +7, but when you look at your STFT it is ranging from -8 @ 28 g/s to +6 at 77 g/s.

So if you were to add your STFT and your LTFT you get the following adjustment.

@ 28 g/s = LTFT + STFT = +7 -8 = -1 (no adjustment)
@ 77 g/s = LTFT + STFT = +7 + 6 = +13 (provide this adjustment)

And then you should interpolate all data points in between.

I am thinking this would more accurately reflect the required adjustments for the MAF, because when the ecu learns LTFT it is going to take the average value from the 28 to 77 g/s range, but only the STFT can adjust the values between these points. So you could require a + adjustment at one end of the range and a - adjustment at the other and still register 0 LTFT.

Am I making any sense here??

[cld12pk2go]

Originally Posted by Trader When I did my MAF calibrations, I adjusted the voltage to account for the trending STFT as well as accounting for the LTFT. For example, Looking at your 28 to 77g/s area, your LTFT is at +7, but when you look at your STFT it is ranging from -8 @ 28 g/s to +6 at 77 g/s. So if you were to add your STFT and your LTFT you get the following adjustment. @ 28 g/s = LTFT + STFT = +7 -8 = -1 (no adjustment) @ 77 g/s = LTFT + STFT = +7 + 6 = +13 (provide this adjustment) And then you should interpolate all data points in between. I am thinking this would more accurately reflect the required adjustments for the MAF, because when the ecu learns LTFT it is going to take the average value from the 28 to 77 g/s range, but only the STFT can adjust the values between these points. So you could require a + adjustment at one end of the range and a - adjustment at the other and still register 0 LTFT. Am I making any sense here??

I understand what you are saying. However, my STFT do not appear to repeat closely run to run, which would make that type of correction unlikely to improve my STFT's.

I just looked at my last 3 data logs and they range from 0 on an entire 1st-4th gear pull to +10 on the next run for about 2 seconds.

I will probably get some more data logs in the near future and see if I have an improvement opportunity...

[cld12pk2go]

Updated. Ordered a KMD fuel pump to eliminate my pressure drops...

[cld12pk2go]

KMD v2 internals arrived on Wed. Palerider = win.

I had family visiting this weekend so no time to install.

Will plan on doing it sometime this week or weekend.

[cld12pk2go]

KMD v2 internals installed and updates added.

[cld12pk2go]

Lots of tables updated in map v104e.


Check post 7

[austin102085]

check my mods in my sig. do you think this map would work good for me? thanks.

[cld12pk2go]

Originally Posted by austin102085 check my mods in my sig. do you think this map would work good for me? thanks.

It would probably be as good or better than starting with an OTS map. I found them quite crude honestly.

Just make sure to get your MAF scaled properly.

I personally would knock the TRL load targets down about 0.1 in 3th-6th just to play it safer though until you see how it works on your setup.

Of course YMMV. Don't come crying back if your engine blows as I take no responsibility.

Do come back and post any comments if you try it.

Tks,

[dread]

how did you like my trl and app changes for your map?

[cld12pk2go]

Originally Posted by dread how did you like my trl and app changes for your map?

So far I have to say it appears to have removed ~80-90% of the jerkiness that I had with the OTS map.

A very nice mod I would say. Tks again.

8/11/09 Updates added to post 7. Looks like the stock snail is pretty much maxed out.

[austin102085]

Originally Posted by cld12pk2go It would probably be as good or better than starting with an OTS map. I found them quite crude honestly. Just make sure to get your MAF scaled properly. I personally would knock the TRL load targets down about 0.1 in 3th-6th just to play it safer though until you see how it works on your setup. Of course YMMV. Don't come crying back if your engine blows as I take no responsibility. Do come back and post any comments if you try it. Tks,

cant get the map to open in ap manager or atr???

[cld12pk2go]

I have ATR v1.7.3.1.

Not sure the nuances required to open another person's map; however, I have seen it posted on this forum. Just check around.

I am assuming that you uncompressed the .rar file prior to trying to open it? If not download winrar from www.rarsoft.com and convert to a .ptm file.


8/13/09 Updates added on Post 7.

[austin102085]

Originally Posted by cld12pk2go I have ATR v1.7.3.1. Not sure the nuances required to open another person's map; however, I have seen it posted on this forum. Just check around. I am assuming that you uncompressed the .rar file prior to trying to open it? If not download winrar from WinRAR archiver, a powerful tool to process RAR and ZIP files and convert to a .ptm file. 8/13/09 Updates added on Post 7.

yea that could be the problem. thanks

[cld12pk2go]

Added updates to post 8 detailing v1.04g changes.