New ATR is out Jan 25 2012 build: 1.9.1.0-3289

[Lex]

I assume COBB will follow up with a description of the new tables.

If you guys are tuning with me, update your ATR to the latest!

[superskaterxes]

jan 25th?

can someone post the help docs/log to see what changed?

[cld12pk2go]

Finally.

I have been scanning everyday for a thread like this!

Can't wait to get home now...

[atvfreek]

Just updated my ATR as well, and It looks like the help file is all new. It does explain what most of the tables do, including the load tables that I've looked at so far.

Also has a new table:
Engine Volumetric Efficiency (VE) (BETA)
Table Description - This table will re-scale the MAF and other inputs for the calculated load used in all ECU calculations.
Tuning Tips - None at this time.

[rfinkle2]

-Coarse and fine wgdc adjustments (seperated for load and boost based tunes)

-DJ will be happy: knock retard - max allowed per rpm

-gen2 load ignition load axis fixed (@G26)

@superskaterxes

[Lex]

I am very interested in the VE table.
VE as a table generally only exists in speed density (read map based) systems so I asked David for some input here.

[djuosnteisn]

Weeee.....


Thanks for the hard work David.

[Cobb Tuning]

Originally Posted by Lex I am very interested in the VE table. VE as a table generally only exists in speed density (read map based) systems so I asked David for some input here.

The VE table exists in this ECU to "normalize" the tables you look at. Because the engine doesn't have a linear efficiency map, your fuel and ignition tables will look funny. Standalone ECU tuning displays this very un-fun property.

The Mazda Engineers have spent countless hours making this VE map for the engine, so that they can design fuel/ign maps interchangeably between makes. Hence "normalizing".

If you vastly change your Engine setup (headwork, intake manifold/turbo manifold, etc) this exposed table would be of use to you.

-David@COBB

[cld12pk2go]

So does this release resolve the Load Cap V2™ issue?

[Cobb Tuning]

Originally Posted by cld12pk2go So does this release resolve the Load Cap V2™ issue?

Not particularly. That effectively requires a large modification of the lookup routines the ECU, or "Load Scaling". It's something that's popped back onto the list of things to do as of recently.

-David@COBB

[Lex]

Originally Posted by David@COBB The VE table exists in this ECU to "normalize" the tables you look at. Because the engine doesn't have a linear efficiency map, your fuel and ignition tables will look funny. Standalone ECU tuning displays this very un-fun property. The Mazda Engineers have spent countless hours making this VE map for the engine, so that they can design fuel/ign maps interchangeably between makes. Hence "normalizing". If you vastly change your Engine setup (headwork, intake manifold/turbo manifold, etc) this exposed table would be of use to you. -David@COBB

So increasing the values in this table will effectively increase fueling? Or calculated load?

Essentially what and how will changing this table affect the calibration?

We certainly don't have the tools to calculate actual VE so we might as well know what we should expect when changing this table.

[Cobb Tuning]

Originally Posted by Lex So increasing the values in this table will effectively increase fueling? Or calculated load? Essentially what and how will changing this table affect the calibration? We certainly don't have the tools to calculate actual VE so we might as well know what we should expect when changing this table.

Increase in efficiency = More air = more fuel delivered.

Decrease in efficiency = less air = less fuel delivered.

That in turn will define your load, after the fact. It does directly affect your airflow in the fueling calculation, so be very careful if you choose to decrease it.

This is why it is reserved for "big mods" to the manifolds, heads, or block.

-David@COBB

[cld12pk2go]

Originally Posted by David@COBB Not particularly. That effectively requires a large modification of the lookup routines the ECU, or "Load Scaling". It's something that's popped back onto the list of things to do as of recently. -David@COBB

Total buzz kill there...

But I am sure some neat features have been unlocked for us to enjoy until then.

[Lex]

Originally Posted by David@COBB Increase in efficiency = More air = more fuel delivered. Decrease in efficiency = less air = less fuel delivered. That in turn will define your load, after the fact. It does directly affect your airflow in the fueling calculation, so be very careful if you choose to decrease it. This is why it is reserved for "big mods" to the manifolds, heads, or block. -David@COBB

So essentially this is a non linear multiplier to the MAF curve. I was under the impression that having a MAF on a vehicle dealt with this issue quite effectively - in other words if the motor had a high VE, it would ingest more air g/s and that would be enough to calculate fueling.

This seems to deal more with how efficient the motor is at taking that air and combusting it (versus losing it due to cam overlap) for example.

It's nice that it is RPM dependent but in most cases it seems that tweaks to regions of the MAF curve would cover 99% of people's needs. This would be useful if we couldn't change the MAF table and had to use the same MAF housing in all cars - which probably happens in the OEM world.

Kind of like in speed density systems where a pressure of 10psi on 1 vehicle means something very different than on another but you can only read pressure so you have to drastically change the VE table.

Also wanted to add, when load tuning (boost toggle off) are the boost based wgdc tables completely ignored and vice versa?

Finally the WG Duty Baro Error Comp doesn't make sense to me. Where is the baro axis? All I see is a boost error and a WGDC% compensation.

[Cobb Tuning]

Originally Posted by Lex Finally the WG Duty Baro Error Comp doesn't make sense to me. Where is the baro axis? All I see is a boost error and a WGDC% compensation.

That table is what has been plaguing some of you on the Load tuning side. When the WG Duty Throttle Close Baro Threshold is reached, it branches from using WG Duty Load Dynamics to using this WG Duty Baro Error Comp. Notice as well that all the values are negative, as the ECU suspects that it needs to pull WGDC.

There are two ways to combat this. First of all you could do it the easy way and make it all 0's. At that point, it wouldn't pull WGDC at all when going over there.

The more CORRECT way of doing it is to use the WG Duty Throttle Close Baro Threshold values to raise the threshold of when it changes logic branches on you. You could raise threshold of Load/RPM area to see if helps.

-David@COBB

[djuosnteisn]

Originally Posted by David@COBB That table is what has been plaguing some of you on the Load tuning side. When the WG Duty Throttle Close Baro Threshold is reached, it branches from using WG Duty Load Dynamics to using this WG Duty Baro Error Comp. Notice as well that all the values are negative, as the ECU suspects that it needs to pull WGDC. There are two ways to combat this. First of all you could do it the easy way and make it all 0's. At that point, it wouldn't pull WGDC at all when going over there. The more CORRECT way of doing it is to use the WG Duty Throttle Close Baro Threshold values to raise the threshold of when it changes logic branches on you. You could raise threshold of Load/RPM area to see if helps. -David@COBB

Was the WG Duty Throttle Close Baro Threshold table all 1's from the factory? Or is this just the default value with the latest upgrade?

If it were all ones on previous tunes, then sure we've been exceeding this threshold all along.... which begs the question, why haven't wee seen this issue early? Is it possible that the Load Dynamics and Baro Error Comp were both running at the same time? And "fighting" each other? And only when loads become significantly higher does the baro table begin to over power the load dynamics?

[econman]

Originally Posted by djuosnteisn Was the WG Duty Throttle Close Baro Threshold table all 1's from the factory? Or is this just the default value with the latest upgrade? If it were all ones on previous tunes, then sure we've been exceeding this threshold all along.... which begs the question, why haven't wee seen this issue early? Is it possible that the Load Dynamics and Baro Error Comp were both running at the same time? And "fighting" each other? And only when loads become significantly higher does the baro table begin to over power the load dynamics?

It is unlikely that the threshold is being exceeding unless your Boost Tables -> Boost Limits - Throttle close values are low.

The values in the WG Duty Throttle Close Baro Threshold table are multipliers. So at 14.71 PSI baro pressure, the multiplier (1.0) times the Boost Limits - Throttle Close table would be just the values in the Boost Limits - Throttle Close table. Then if boost > (multiplier x Boost Limits - Throttle close values) the ECU will switches from using the WG Duty Load Dynamics (Fine) (Load Based) table to the WG Duty Baro Error Comp (Fine) (Load Based).

Basically the Baro Threshold table modifies the Boost Limits - Throttle Close table based on the baro pressure. The relationship between the modification and the actual pressure determines which load dynamics table is used.

At least that's how I interpret the ATR help file... I may very well be wrong

[djuosnteisn]

My throttle close tables have always been set to the same values as the fuel cut. So i'm not sure how a 1.0 multiplier would ever convince the ecu to switch load error tables.



I'm still at work, but will read the ATR file tonight.

[econman]

Originally Posted by djuosnteisn My throttle close tables have always been set to the same values as the fuel cut. So i'm not sure how a 1.0 multiplier would ever convince the ecu to switch load error tables. I'm still at work, but will read the ATR file tonight.

From what I understand it causes the ECU to switch from the load error tables to the baro error tables. I can't see it happening too often either

Added: Here's what the ATR help file says

WG Duty Throttle Close Baro Threshold (Load Based) Table Description - This table is a threshold multiplier of the Boost Limits - Throttle Close table. When the result of the multiplication is larger than the Manifold Absolute Pressure, the ECU will choose WG Duty Baro Erro Comp (Fine) (Load Based) over WG Duty Load Dynamics (Fine) (Load Based).

[djuosnteisn]

from.... is what i meant.





The point stands though.... no switching should be going on since my throttle-close boost tables are set higher than ever seen.

[Nliiitend1]

Originally Posted by Lex I assume COBB will follow up with a description of the new tables. If you guys are tuning with me, update your ATR to the latest!

Done!

[superskaterxes]

david,

what happened to the E85 scaling table?

[Ziggo]

Based on the initial description it seems to me that the VE table could be abused to compensate for E85. Just scale the whole thing up by 30% or so. Need to poke around at it more first though.


Zigatapatalka

[superskaterxes]

Originally Posted by Ziggo Based on the initial description it seems to me that the VE table could be abused to compensate for E85. Just scale the whole thing up by 30% or so. Need to poke around at it more first though. Zigatapatalka

i dont think thats the purpose of this table. i think even minute changes (seeing how its out to 4 decimal places) will greatly affect fueling/airflow.

i could be completely wrong but im not gona touch it til david can shed some more light lol

[cld12pk2go]

So I am doing the following...

• Going back to load tuning.
• Cutting the values in WG Duty Baro Error Comp (fine) (Load Based) in half for starters.
• Increasing my throttle close table to fuel cut +1 PSI (they were about fuel cut -1 PSI).
• Keep watching this thread.

[djuosnteisn]

Originally Posted by cld12pk2go So I am doing the following... Going back to load tuning. Cutting the values in WG Duty Baro Error Comp (fine) (Load Based) in half for starters. Increasing my throttle close table to fuel cut +1 PSI (they were about fuel cut -1 PSI). Keep watching this thread.

Are you going to leave the "boost based" toggle enabled? I kinda prefer that now... load tuning with the toggle.

[Lex]

Originally Posted by djuosnteisn Are you going to leave the "boost based" toggle enabled? I kinda prefer that now... load tuning with the toggle.

@David@COBB

Dave, what did the Boost Toggle do? Did it just change the tables you now revealed?

[Cobb Tuning]

Originally Posted by Lex @David@COBB Dave, what did the Boost Toggle do? Did it just change the tables you now revealed?

Boost Toggle On:

Use WG Duty Boost Error Comp. (Fine) (Boost Based)

Boost Toggle Off:

Use WG Duty Load Dynamics (Fine) (Load Based)
Use WG Duty Baro Error Comp (Fine) (Load Based)
^Above tables switch back and forth with below threshold table.

Use WG Duty Throttle Close Baro Threshold (Load Based)


Hope that helps.

-David@COBB

[superskaterxes]

david,

anymore comments on how we can adjust and play with the VE table? how much/where/when/why lol

[djuosnteisn]

Originally Posted by David@COBB Boost Toggle On: Use WG Duty Boost Error Comp. (Fine) (Boost Based) Boost Toggle Off: Use WG Duty Load Dynamics (Fine) (Load Based) Use WG Duty Baro Error Comp (Fine) (Load Based) ^Above tables switch back and forth with below threshold table. Use WG Duty Throttle Close Baro Threshold (Load Based) Hope that helps. -David@COBB

Is WG Duty Load Error Comp. (Course) active regardless of the toggle? This is how i interpreted your earlier explanations. And I ask, because I've "load tuned" several cars with the toggle enabled... including my own 3. All TRL tables work great, and WGDC modulation is very fast and converges on load targets nicely.

[G26]

Dammit, just when I thought I was done.

Right now all my load targets are set to 3.00, and load error comp/load dynamics zeroed. Should I keep it like this?

I might end up load tuning this pos after all.

[rfinkle2]

Originally Posted by G26 Dammit, just when I thought I was done. Right now all my load targets are set to 3.00, and load error comp/load dynamics zeroed. Should I keep it like this?

You wanted to start working on your timing once the load axis were fixed so I tagged you.

The load axis are fixed. LOL.

You shouldn't have to change any wastegate related tables, etc.

[Cobb Tuning]

Originally Posted by djuosnteisn Is WG Duty Load Error Comp. (Course) active regardless of the toggle? This is how i interpreted your earlier explanations. And I ask, because I've "load tuned" several cars with the toggle enabled... including my own 3. All TRL tables work great, and WGDC modulation is very fast and converges on load targets nicely.

Yes, that table is always active regardless of boost tuning toggle or not.

[cld12pk2go]

Originally Posted by djuosnteisn Are you going to leave the "boost based" toggle enabled? I kinda prefer that now... load tuning with the toggle.

That is what I get for drinking decaf...I meant to include that this morning...

I unchecked the boost targeting toggle...

Originally Posted by djuosnteisn Is WG Duty Load Error Comp. (Course) active regardless of the toggle? This is how i interpreted your earlier explanations. And I ask, because I've "load tuned" several cars with the toggle enabled... including my own 3. All TRL tables work great, and WGDC modulation is very fast and converges on load targets nicely.

I haven't tried this before. Can you elaborate on the methodology required?

[djuosnteisn]

Originally Posted by cld12pk2go I haven't tried this before. Can you elaborate on the methodology required?

I think you have actually, it's what i suggested a while back... leaving the boost toggle enabled, but still load tuning like normal. As i understand it from David's descriptions, the boost toggle is supposed to remove 2 of the 3 load error sources from the logic, and pretty much just leave the usual TRL tables we're used to. But i'm not sure how this plays into the whole IDC over 85% correlation Lex and you have been exposing. There may indeed be something else going on deeper in the logic.

All i know at the moment is that leaving the toggle enabled doesn't adversely affect load tuning... but that's not to say it fixes Load Cap v2.

Hopefully a combination of tweaking these newly exposed tables and possibly leaving the toggle enabled will remedy everything.

We'll know soon enough i guess.

[Lex]

From what I have seen, checking the toggle stops the WGDC from dropping when IDC reaches 80% plus. However load cap V2 at much higher loads still remains.

[cld12pk2go]

There does not appear to be any salvation to load tuning at least in my first attempt...

My throttle started closing and WGDC was pulled over 20% down from the table value...


Load target was 2.35 from 3-7k. Load dropped from the 2.35 aim down to 2.1 by 6300 RPMs and I was never over 75% IDC.

I have dropped down to 5 gal E85 per fillup now, which is most likely in the E26-E30 range.



I was up to ~2.3 load at 6500 RPMs on my boost targeting such as this and was holding ~22 PSI when targeting 23.5 PSI:

[cld12pk2go]

Originally Posted by djuosnteisn I think you have actually, it's what i suggested a while back... leaving the boost toggle enabled, but still load tuning like normal. As i understand it from David's descriptions, the boost toggle is supposed to remove 2 of the 3 load error sources from the logic, and pretty much just leave the usual TRL tables we're used to. But i'm not sure how this plays into the whole IDC over 85% correlation Lex and you have been exposing. There may indeed be something else going on deeper in the logic. All i know at the moment is that leaving the toggle enabled doesn't adversely affect load tuning... but that's not to say it fixes Load Cap v2. Hopefully a combination of tweaking these newly exposed tables and possibly leaving the toggle enabled will remedy everything. We'll know soon enough i guess.

Specifically what do you do? Populate the boost target table with something higher than what will satisfy your load targets in the Abs Load targets, TRL, and TRL baro tables?

[djuosnteisn]

Originally Posted by cld12pk2go Specifically what do you do? Populate the boost target table with something higher than what will satisfy your load targets in the Abs Load targets, TRL, and TRL baro tables?

Not necessarily higher or lower... just "close". The only two tables i make sure are set higher are the throttle close and fuel cut. Then i just kinda set the targets to something close.... and turn my attention to load tables.


I have one guy currently who's WG is dropping on me... so i'm gonna take a close look at his tune tonight and see what i can come up with. I'm hoping there's something obvious (like boost based toggle empty, or throttle close table too low), but if not i'll let you know.

[Ziggo]

I would really like to experiment with this and try to re-re-enable by Gear based load tables but I am afraid that I should just leave well enough alone.