Gen 2 Cobb ATR discussion

[Ziggo]

Originally Posted by Nataphen Great news about the grimmspeed, ck. I was hoping that it would help, but I just wasn't expecting it to, LOL. I'm all over 19psi to redline at less than 90% WGDC!

You do realize that the lower WGDC is just an effect of running an upgraded EBCS like I discussed in my earlier post right? The lower WGDC has nothing to do with how hard the turbo is being worked, it's just that the upgraded EBCS allows it keep the WG closed against those higher manifold pressures with a lower duty cycle.

Just mentioning it because you were so worried about being hard on the turbo earlier. Lenny ran 19-20psi to redline and his EGTs got high enough to crack his turbine housing. (1850+) I don't expect the K04 to last long at all like that, and you better have an FMIC to bleed off the massive amounts of heat it's going to create.

[Nataphen]

When I get to a computer, I will. I normally hold within .5psi of my targets, but I haven't had to touch the BD table from the s1+ OTS map.

Well, 16psi it is. Thanks, Ziggo. I will be running meth and fully bolted ASAP, so maybe that will be enough for me in spite of low boost in high RPM. I wonder why the Cobalt k04 is more efficient at higher boost. Different housing perhaps.

[Ckmazdaspeed3]

Originally Posted by Ziggo You do realize that the lower WGDC is just an effect of running an upgraded EBCS like I discussed in my earlier post right? The lower WGDC has nothing to do with how hard the turbo is being worked, it's just that the upgraded EBCS allows it keep the WG closed against those higher manifold pressures with a lower duty cycle. Just mentioning it because you were so worried about being hard on the turbo earlier. Lenny ran 19-20psi to redline and his EGTs got high enough to crack his turbine housing. (1850+) I don't expect the K04 to last long at all like that, and you better have an FMIC to bleed off the massive amounts of heat it's going to create.

That's great information. Regarding the EGTs, what is your opinion of a safe limit for the ko4?

[Ziggo]

Originally Posted by Ckmazdaspeed3 That's great information. Regarding the EGTs, what is your opinion of a safe limit for the ko4?

High EGTs are the result of lean fueling and retarded timing. Both of which Lenny was running. If that was the only problem then you could get away with it by richening up the fuel mixture and using more advanced timing. The real problem is with the turbine shaft speeds. As for the "reliable" limit of those the answer is probably stock taper. Everything above that is just shortening the lifespan of the turbo. The real question is, is the extra power worth the extra shortened lifespan? If at redline, 9# gives you 230g/s, 16# gives you 290, and 19# gives you 305g/s where you choose to balance the power vs turbo lifetime is for you to answer for yourself. (I just took a shot in the dark at the 19# number)

[Ziggo]

Originally Posted by Nataphen I wonder why the Cobalt k04 is more efficient at higher boost. Different housing perhaps.

Its mostly because the cobalt is a 2.0l engine. They might be able to run 19psi to redline, but I bet they still only flow 280-300g/s

[rfinkle2]

I'd say that Cobb is pretty conservative when developing their maps, and know what the compressor map looks like for our car etc.

Keep in mind, Cobb tapers the boost down towards redline.
from the ATR helpfile:
We have found it helpful to establish proper boost targets in the Boost Targets tables in order establish these various safety limit values. Reasonable maximum boost targets for 91 octane are 18-19psi, and 19-20psi for 93 octane fuel.

The amount of taper can be seen in this excerpt from the map notes here:

*Note difference in targets and final taper from the stock intercooler vs. FMIC...*

Stage2+SF 93 BTv100 - Intended for otherwise stock 2010 MAZDASPEED3 vehicles with a COBB DP with one high flow catalytic converter, a cat-back exhaust, and a COBB SF
Intake system, and an upgraded CDFP ONLY. 93 or 94 octane fuel. Boost Targets:
~18.5psi peak boost pressure tapering down to ~17.5psi by the 6700 RPM redline, +/-
1.5psi. Pressure-Based Boost Tuning enabled.

Stage2+SF+TIH+FMIC93 BTv100 - Intended for otherwise stock 2010 MAZDASPEED3 vehicles
with a COBB DP with one high flow catalytic converter, a cat-back exhaust, a FMIC,a
COBB SF Intake system with TIH, and an upgraded CDFP ONLY. 93 or 94 octane fuel.
Denso ITV22 spark plugs and an upgraded bypass valve highly recommended. Boost
Targets: ~20psi peak boost pressure tapering down to ~19psi by the 6700 RPM redline,+/- 1.5psi. Pressure-Based Boost Tuning enabled.

[indianaryan]

Originally Posted by Ziggo High EGTs are the result of lean fueling and retarded timing. Both of which Lenny was running. If that was the only problem then you could get away with it by richening up the fuel mixture and using more advanced timing. The real problem is with the turbine shaft speeds. As for the "reliable" limit of those the answer is probably stock taper. Everything above that is just shortening the lifespan of the turbo. The real question is, is the extra power worth the extra shortened lifespan? If at redline, 9# gives you 230g/s, 16# gives you 290, and 19# gives you 305g/s where you choose to balance the power vs turbo lifetime is for you to answer for yourself. (I just took a shot in the dark at the 19# number)

Exactly. It's all about finding a happy medium. If the extra 15 g/s is worth it to you, then go for it. For me personally, it's not about pushing the turbo to the limit. It's about getting more power out of the turbo, knowing that it's decreasing its lifespan, but still staying fairly reliable. I can't afford to have stuff break on this car.

[rfinkle2]

Just happened to notice this from Cobb's recommended upgrage path for the MS3:

below is an excerpt:


5) Upgraded turbo, at this point in time, the stock turbo is working at is maximum and you can see this through the WGDC values that you log through the AccessPORT. The WGDC values will likely be at or above 90% from the mid RPM range to redline...letting you know that the stock turbo is spinning ts head off. To help increase turbo life, you can target lower boost levels where the WGDC is around 80% by redline.

If anyone is interested in #1-4, it can be found here:

sticky suggestion thread (gen2)

[Ziggo]

Originally Posted by rfinkle2 I'd say that Cobb is pretty conservative when developing their maps, and know what the compressor map looks like for our car etc.

We have the compressor map on this site. The final shaft speed line works out to about .23m^3/s or ~278g/s at 77deg. It's not insta-fail to exceed that or we would have alot of busted turbos on this site, but that is probably the presumable "reliable" limit. I reach this number at about 5k rpm I think, then taper boost down from there and the maf stays relativly constant. I didn't do this to save my turbo, as I said earlier, I hate it. I did it because thats about the limit of what the stock BCS system can commmand. That it roughly coincides with the reliable limit of the turbo probably isn't an accident.

Cobb is conservative with thier timing and fuel as the OTS mafcal is variable and both of those things are affected by that and can cause insta-boom, but they are not going to loose sleep over turbos wearing out after xxxx miles. I bet there is a disclaimer in there somewhere about risks of tuning and increasing performance of your motor.

[rfinkle2]

Originally Posted by Ziggo We have the compressor map on this site. The final shaft speed line works out to about .23m^3/s or ~278g/s at 77deg. It's not insta-fail to exceed that or we would have alot of busted turbos on this site, but that is probably the presumable "reliable" limit. I reach this number at about 5k rpm I think, then taper boost down from there and the maf stays relativly constant. I didn't do this to save my turbo, as I said earlier, I hate it. I did it because thats about the limit of what the stock BCS system can commmand. That it roughly coincides with the reliable limit of the turbo probably isn't an accident. Cobb is conservative with thier timing and fuel as the OTS mafcal is variable and both of those things are affected by that and can cause insta-boom, but they are not going to loose sleep over turbos wearing out after xxxx miles. I bet there is a disclaimer in there somewhere about risks of tuning and increasing performance of your motor.

Agreed on all points...

With that being said, Cobb is / has been a tuner of many platforms for many years, and has an interest in the relative long term reliability of the Mazda, Nissan, Subaru, BMW, Mitsu and Honda platforms they develop / have developed the AP for.

I think that you and I are on the same page, considering your targets are 20psi, tapering some (12- 15 psi) to redline.

[indianaryan]

Originally Posted by Ziggo Cobb is conservative with thier timing and fuel as the OTS mafcal is variable and both of those things are affected by that and can cause insta-boom, but they are not going to loose sleep over turbos wearing out after xxxx miles. I bet there is a disclaimer in there somewhere about risks of tuning and increasing performance of your motor.

This is the key here. Cobb mentioned on a thread that @rfinkle2 started about 100% WGDC that they usually have to replace the turbo on their cars about once a season...

[rfinkle2]

Originally Posted by indianaryan This is the key here. Cobb mentioned on a thread that @rfinkle2 started about 100% WGDC that they usually have to replace the turbo on their cars about once a season...

The thread referenced above can be found here: http://www.cobbforums.com/forums/sho...ning-solutions

That is on their track cars, and if you track your car, You will see way more time in high boost, extended boost use than the average daily driver.

[indianaryan]

Originally Posted by rfinkle2 That is on their track cars, and if you track your car, You will see way more time in high boost, extended boost use than the average daily driver.

True. I don't expect to have to replace the turbo on my car every season lol. But the point is the turbo's life expectancy is definitely reduced even on the OTS maps.

[Speed3eak]

Okay, 1st thing for you guys worrying about EGTs

WMI - water lowers your egt's by quite a bit, don't run 100% meth like lenny did if you want your turbo to last longer. IDK about turbine speed or anything, but getting a k04 rebuilt is like... what, $50, $75? If your turbo shits, either get it rebuilt with higher quality parts or go big turbo. I keep saying it but everybody either goes ATP or gets a 3071 or 76 with a t3 housing - USED HOLSET HX-35. I called bullseye, and they'll make you a t3/v-band housing for it for ~$200, and you can pick up a used hx-35 for ~$200 +/- $100. Getting one rebuilt usually is under $100, and they come with an iwg designed to work with huge ass diesel engines.

That's less than $1k for a turbo that is designed to last for over 100k miles in diesel engines, and will make the same kind of power people see with 3071 and 3076 turbos, especially with those .55 a/r stock exhaust housings. Plus, with those, you're going to get an ex-mani and DP anyway, so I can't wrap my head around the fact that there aren't more people going with holsets.

lol, got a bit derailed there. tl;dr - WMI will cure your egt ills.

[kylec_2003]

So I just started reading up and working with ATR. So far all I have done is leaned out the stage 1 map to 11.5 AFR. I am still waiting on my HPFP to show up from cp-e but can you guys give me some ideas of where to go next with my tuning. Here is my log from tonight.

[rfinkle2]

I'm thoroughly convinced that there are different tables in the gen1 ATR vs. the gen2 ATR, and some of the load tables DO affect boost tuning.

I'm hoping (when I get home from work) to create a thread about the available tables in gen1 vs gen2, and how those differences affect gen1 and gen2 tuning.

Originally Posted by kylec_2003 So I just started reading up and working with ATR. So far all I have done is leaned out the stage 1 map to 11.5 AFR. I am still waiting on my HPFP to show up from cp-e but can you guys give me some ideas of where to go next with my tuning. Here is my log from tonight.

The absolute best advice I can give you is to take a look @ the threads in the first post.

As far as I can read, it is recommended (after your sure that your maf curve / calibration is on point) that you tune your fuel tables 1st, boost tables 2nd, and timing last.

[kylec_2003]

I have been reading all of the threads, and thanks for giving me an order of attack.

[xxdjfate76xx]

ive been doing some changes to get a smooth throttle response, ill post up the tables tonite if they work out.. i basically used Dread's throttle request load chart, and APP gear chart. Tweaked them a bit..... all the info here is great, we shoud definitely start putting up some examples of our charts, keeping in mind that they are specific to our mods

[rfinkle2]

Originally Posted by rfinkle2 I'm thoroughly convinced that there are different tables in the gen1 ATR vs. the gen2 ATR, and some of the load tables DO affect boost tuning. I'm hoping (when I get home from work) to create a thread about the available tables in gen1 vs gen2, and how those differences affect gen1 and gen2 tuning.

I spent 2 hours trying to get some good ATR screenshots, and can't get them to show up properly (too small), see my gallery, so I'm going to go beat the shit out of my car and hope to get this frustration out.

Thanks for listening to this absolutely useless rant.

[Fobio]

I know someone was asking for my BD table etc...

Considering that there are some major difference between Gen 1 and Gen 2 boost control in terms of tables used etc, I do not feel it's right to post my Gen1 tables here and add to any confusion. I may update my WGA/EBCS thread as weather warms.

[xxdjfate76xx]

yea ive been looking through the gen1 atr stuff, hopefully its not too far off.. i guess we can find out the hard way

this is what i came up with as far as the APP Translation for gears 1-6

0.00	1.79	2.84	4.24	5.70	7.21	8.85	10.44	12.01	15.21	18.65	21.95	25.82	29.38	34.16	45.29	55.96	70.70	89.38	100.00

[Nataphen]

I did direct APP trans. in neutral and 1-5. I love the way that the car feels with that setup. In 6th, I left it like the OTS map had it because direct APP in that gear caused the throttle to go crazy when using cruise control. I made a map for someone else using that, and he liked 3-5, but thought that it was too touchy in 1-2, so just mess with it and figure out what you like is really the best solution for throttle response.

I also switched my Speed Comp. APP table to 1.00 in all cells. When it had 1.30 in some of the low speed cells, taking off was too touchy IMO.

EDIT: The guy that I was talking about e-mailed me tonight, and he's decided that he likes the 1-2 throttle feel now. After he got used to it, he said it made the car more fun to drive around town. Werd.

[xxdjfate76xx]

by any chance does anyone know for Throttle Requested Load, do we use A and B or A and C... cuz the gen1 guys use A and C except for the 2009s use A and B

i tried my tune and the throttle felt great, but i have to go back to the ots map and only tweak the throttle stuff cuz i had messed with the boost tables and forgot to change it back, got some detonation

[Ckmazdaspeed3]

So it appears my struggle with the ebcs may be due to incorrect req- load normal bat tables. I am testing an edited table for gear number three tomorrow.

If anyone has experience with this and thinks I could be putting my motor at risk, please chime in. My BD, WGDC and boost tables are listed on page 3, as is a 3rd and 4th gear pull.

[rfinkle2]

Originally Posted by Ckmazdaspeed3 So it appears my struggle with the ebcs may be due to incorrect req- load normal bat tables. I am testing an edited table for gear number three tomorrow. If anyone has experience with this and thinks I could be putting my motor at risk, please chime in. My BD, WGDC and boost tables are listed on page 3, as is a 3rd and 4th gear pull.

This makes perfect sense considering the gen1 guys have their maps dialed in, but did a great deal of load based tuning before hitting the toggle switch to boost based tuning in ATR.

[Ckmazdaspeed3]

Oh shit, I was a little buzzed last night and forgot to post the results.

So, igot impatient (nothing new there) and decided to do a pull w the new req load table, and it works... Kinda. My boost is SO much smoother and maintains 18+psi to red line. The downside is that my wgdc is maxing out as early as 4.7k! I don't get it because even with the oem bcs, I was hitting targeted boost at these RPMs w/o 100% wgdc... Fuck!!!, I wanted to dyno today and put on the bt this weekend!

Well, I've decreased my astronomical wgdc values to see if that helps w/o sacrificing too much boost... I'm trying the 2X+9 formula, which I had been ignoring.

I would post a log, but I'm on the iPad and dropping logs off and I'm bout to leave for work. I will post more results soon.

[rfinkle2]

Originally Posted by xxdjfate76xx by any chance does anyone know for Throttle Requested Load, do we use A and B or A and C... cuz the gen1 guys use A and C except for the 2009s use A and B i tried my tune and the throttle felt great, but i have to go back to the ots map and only tweak the throttle stuff cuz i had messed with the boost tables and forgot to change it back, got some detonation

This is courtesy of @wolly6973 from Dano's boost tuning thread.

"Genpu - A & B are equal, C is lower... "

[xxdjfate76xx]

Thanks finkle, I'll try it out tonite

[rfinkle2]

I think that the load dynamics table is definitely having an effect on the wgdc, and gen2 is seeing this effect more than gen1 users. (given gen1 users had been load tuning before pressure based tuning)

----from the ATR helpfile (pg17/71) :
The Boost Dynamics and Load Dynamics are used to fine tune boost or torque control characteristics. These tables give the ECU authority to remove WGDC when an over boost or over load condition occurs, and add WGDC when an under boost or under load condition occurs. These tables are used to help correct boost and torque targeting values when an over boost (or over load) or under boost (or under load) condition occurs. These tables are calibrated to help control the smaller stock turbo, if you have changed your turbocharger we suggest you modify these table settings in order to fine tune the boost control
characteristics.

----my guess is this is why Christian recommends that calculated load is within .05 in the logs for proper boost control.

[Ex_WRX_driver]

Dont suppose anyone knows if I can use 2010 ATR on 2011 tunes?

[Nataphen]

That question is above my paygrade. Cobb should be able to tell you though.

[triphop631]

Originally Posted by Nataphen Start by leaning out your AFR targets to about 11.9 from 3k+, this will help with power and fuel pressure. It's the single most helpful thing that you can do to keep your stock pump in check.

what table do you do this in?

[Nataphen]

Originally Posted by triphop631 what table do you do this in?

Fuel OL/WOT Commanded EQ (No Knock A)
Fuel OL/WOT Commanded EQ (No Knock B)

[Speed3eak]

Wow, this discussion is really taking off - I wish I could chime in more, but I've been out of town and away from the car for a few days.

I'm curious about how the load tables work with the boost tables - that's funny that CK noticed his wgdc jump and his boost smooth out with the updated load table. I'm sorry as I don't know what the ECU has at its disposal to achieve a target load, or how EGR works, or any of that stuff really. I got stuck modifying boost and wgdc tables, then my wga decides it won't hold boost, so I've been waiting to the weekend to get under the hood and see if there's a mechanical fix, and now I feel like I'm in over my head haha.

DC needed to achieve a certain psi boost changes over the RPM range - I imagine this speaks to both the efficiency of the turbine and housing (compressor side basically) itself to achieve boost, as well as to the amount of exhaust gas flowing though the hot side. Is it possible that some of the parameters the ECU has authority to change also affect how much gas the exhaust side sees?

CK, what else did you notice that was different about those logs? Spark advance? Throttle Position? exhaust gas temps? If the exhaust gas is cooler, it'll flow more slowly, so could that affect what you're seeing?

I do remember you saying at some point that your stocker was beginning to die, was that the case during your most recent observations? Or were these taken with the 2871?

How did you populate the load table? I'm curious about the part-throttle stuff because that seems to me to be especially tricky. Thanks in advance, Hopefully I can get back to learning about this before too long and you guys leave me in the dust.

[Ckmazdaspeed3]

I haven't fucked with the part throttle yet...as long as it's safe, I'm happy for now.

My turbo may have just been a random crazy ms3 case, cuz it hasn't happened again... Smoking. Also, at the time, there weren't as many people having problems holding boost, so between the two, I figured it was bye bye turbo... May have been wrong.

So I got third gear worked out, but it ended up fucking fourth gear up. So I was advised to just put the req load normal bat tables up to 3. I tried it in fourth gear and the results are fucking beautiful. I hit 288 g/s, which is about 15 g/s higher than my old best (which was at like 30° colder temps)! Also, the boost was fucking smooth. I am going to apply this approach to all gears.

If you look at atr help file, you will see that this approach makes sense. This is because the ecu is targeting one of three different tables (that are mentioned anyway) to determine output (load, boost, idk). It chooses the table with the lowest torque output, which in my case was the req load table. So setting these values too high makes the ecu choose a different table and takes the req load table out of the picture. Whichever one it chooses gives much smoother and higher results than the req load table.

So hopefully with a little wgdc adjustment it'll be perfect... Knock on wood.

[Speed3eak]

So, wait - for your 3rd gear requested load table, it's all 3's across too?

I'm on ubuntu right now so I can't open up ATR - is requested load a per-gear table? I can't remember off the top of my head.

You think that could be a way to de-tune 1st and 2nd gear? I <3 the APP tables for that, but it'd be interesting to see.

I'm gonna shut the hell up now until I can get some more logging/tweaking done and actually have something to say.

[Ckmazdaspeed3]

Yeah, it's by gear, and yes, I will use it for third because I'm fairly certain I will make more power.

I'm also going to apply it o second and see what happens... If its too much power, I'll address that then. Same with first.

[Ckmazdaspeed3]

Ok, so I switched all load values after 3k rpm to 3 on the req load norm bat tables for each gear except first... I'll probably do it there too, but I wanted to see what happened in second.

So, I was afraid that I might lose control of second gear, but no, I did not. I would post a log, but it didn't record for some reason, oh well. I will get another one later.

3rd behaved exactly like fourth! This is fucking win, because before I maxed out these tables, the two gears were always different regarding boost, and for me, it was impossible to tune that way... I'm sure the other gears would have all been different too.

Now if g/s can be used to compare how much power your making on your own car, then I am making more power than I ever have. 288 g/s in noticeably warmer weather than my old highest numbers.

I still need to adjust wgdc a bit, and if anybody has advise on how to get rid of my 22+psi boost spike, I'm all ears... I'm not so worried about it on my set up, as the 2871 should go in next weekend, but I would like to hammer this shit out for others that would like to be able to make more boost on the ko4.

Here is a pre max load value 3rd and 4th, and a post 3rd and 4th log. Look at the decrease in oscillation and the similarity in boost in the post logs.

The pre 3rd gear pull looks pretty good, but the corresponding 4th gear pull has lower boost and is all over the place.

[jwilkins88]

Originally Posted by Speed3eak You think that could be a way to de-tune 1st and 2nd gear? I <3 the APP tables for that, but it'd be interesting to see. I'm gonna shut the hell up now until I can get some more logging/tweaking done and actually have something to say.

No sir... this isn't a way to detune gears. What keith and I were seeing before was boost oscillation. It wasn't so much that he wasn't hitting his target boost as much as his boost was oscillating up and down. This fixed it.

[jwilkins88]

Stupid phone won't let me edit my post.

Edit to the previous post: these tables are how you control boost by gear while doing load tuning. So this is how you would detune your car were you using load based tuning.

Basically, at this point, I'm not sure what the hell the boost based tuning does. It's supposed to make it so that the ecu disregards the load tables, but that doesn't seem to be the case since we still have to fuck around with the load tables.