Dano's Boost Tuning 101 + Cool Stuff

[Cataphract_40]

Sure here you go. It's still a work in progress of course. Note that I have tweaked this map for my deleted EGR system. It's cold here right now so that's probably why I'm seeing just shy of 18psi

[SRTie4k]

I finally got my map fully sorted out as a baseline, now just need to add BD back into the equation. Right now my car spikes to 17psi at 3000RPM and holds 16.5-17psi to 5500RPM on WGDC alone. I can't go much higher with the 10-20* weather and my stock pump, but I'm thinking the Boost Comp IAT will push that up to about 17-17.5psi in the summer automatically anyways.

But so far, this is easily the most stable and easiest to modify map I've created yet; the car pulls like a freight train in 4th now, and doesn't hesitate constantly like it did on my load-based maps.

Thanks for the help Dano!

[Dano]

Glad to help where I can guys. This is certainly a work in progress for me as well. One thing about the IAT comp table Christian modified is that its only set to reduce dc below 32* and above 140* so keep that in mind.

I actually negated my table as well [1.00 in the 32.00 cell] until I get the WGDC table finialized to remove that variable.

[DSIT995]

This is a great guide to get a solid understanding on how to tune with boost.......

I just started making my own winter map for 16.5psi & Ill also make a summer map (or slight changes to the winter map) targeting 18psi

Dano for gearhead?

[acidbbg]

Something I noticed about Boost Dynamics during experimenting (stock turbo & OEM boost control).

With BD setup as per Christian:
Boost Dynamics
0.09 0.04 0.02 0.01 0.00 0.00 -0.01 -0.10 -0.15 -0.30 -0.70

I noticed that boost would quickly drop off. In other words, I would hit my target psi (~17psi at 3500 rpms) but than boost would almost immediately begin to taper down to 15 psi at 5000 rpms. So I began to increase WGDC up top... so that if it punched it down low...I would be able to hold boost up top (instead of it tapering).

BUT than when I punched it up top (~4800 rpms) I would occasionally see boost spike ~20 psi (when I am only targeting 18psi).
* This was exaggerated when I used FFS.

So I lowered WGDC's up top and went back to OTS Boost Dynamics:
0.35 0.15 0.08 0.05 0.01 0.00 -0.01 -0.10 -0.15 -0.30 -0.70

Now when I punch it down low, it smoothly increases boost to ~18 psi at 5500 rpms.
Also, more importantly, if I punch it up top (~4800 rpms) it spikes to ~18 psi (instead of ~20 psi) and tapers per my boost targets.

Originally Posted by Dano2010 I have learned that it is better to have to little WGDC values set and have BD add DC than to have too much WGDC values set and have BD remove DC.

This helped me to decrease boost spikes and hold a more consistent boost curve.

IMHO, this thread should be stickied by a mod.

Now how about some more info on reducing boost in the first 3 gears.

Anyone care to share some of what they found... or maybe we should open a new thread? Just wondering if I am better off using Boost Comp X Gear or App Translation X Gear. Has anyone has figured out the logic behind these tables.

[Dano]

yes the BD table values make quite a bit of difference in how the turbo reacts. That is of course, expected but I haven't figured out the close relation between it and WGDC tables yet. I am still working on and will post up when I have conclusive results.


As far a limiting boost per gear, below are my current APP values for 1-2-3rd gears. 3rd gear is now limited to 14PSI but 1-2 still have too much power for the 30-40* temps here now. Depending on a ton of factors [AMB temps, tire size and compound, motor mounts, etc] you may need to adjust your table values. Keep in mind you want to keep the 1st half of the table close to OTS so APP between shifts etc remains consistent. I multiplied the entire table by some values and APP/clutch combos to shift were way off and didn't work well at all.

1-2nd

0.00	0.25	0.65	1.68	3.25	4.50	6.00	7.50	9.00	10.50	12.00	13.75	16.00	17.50	19.50	21.50	24.50	27.50	31.00	33.90	37.21	38.60	40.00	41.41	42.80	43.24	43.68	43.90	44.12	44.56	45.00

3rd

0.00	0.25	0.65	1.68	3.25	4.50	6.00	7.50	9.00	10.50	12.00	13.75	16.00	17.50	19.50	21.50	26.50	29.50	33.00	36.00	37.40	38.80	40.20	41.60	43.00	44.40	45.80	46.50	47.20	48.60	50.00

And for reference my 4th gear APP which is OTS I believe



oh and I haven't messed with the Boost comp tables but the B tables would be the obvious choices of ones to change. If someone wants to jump in start out with .25 across the board and see what happens.

Exaggerated values are the best way to weed out the ECU logic.

[cld12pk2go]

I love the idea of limiting power in 1st/2nd gear via APP trans tables.

That should work brilliantly.

Just remember that Cobb did some work where beyond (~50% IIRC) very little additional power was obtained when playing with the throttle blade, so you might need to restrict further.

[Dano]

yes actually my 3rd gear is a bit too limited.....this weekend it got up into the 50s and I couldn't break 3rd gear lose at WOT so it is very temp dependent. My APP setup for 3rd was done at ~40 and I could just barely get it to spin...that's the sweet spot.

The other good thing about using APP is that it will work for MBC as well. Albeit the curve might have to be a bit different but it should still work.

on-the-fly/partial flash tuning would shine here to flash some pre-defined APP tweaks for temp changes.

[cld12pk2go]

I think I am getting closer on my tune with the grimspeed.






After this series of logging I am at the following values for my next map:







Boost control seems pretty decent so far, I need to install my ACT clutch so I can get the boost back up to 22 PSI...

[Dano]

Its interesting to see the differing WGDC strategies used to control boost...I would guess that once you have your K04 spooled that you are trying to prevent a spike around 4k?

Here is my latest DC table and you will notice I now employ a much lower DC early in the RPMs which coupled with BD makes the turbo spool faster that using DC values that alone would make target.

It doesn't make much sense to me but I mimicked these values off of one of Christian's tunes and suggested by him. He actually uses 0 in the 2.5k range.

Logs to follow.

[cld12pk2go]

I currently experience boost creep around 4000 RPMs when it is cold out. So I can pretty much make these low targeted boost pressure at that RPM with ~0-10 WGDC.

[SRTie4k]

Originally Posted by cld12pk2go I currently experience boost creep around 4000 RPMs when it is cold out. So I can pretty much make these low targeted boost pressure at that RPM with ~0-10 WGDC.

I'm getting the same thing now strangely. I zero'ed out my BD table and now that I'm making runs in 2* weather I suddenly creep to 18.5psi from 4000-5000, and I can't seem to get control of it.

I'll try copying your values between those RPM's and see if it helps.

[cld12pk2go]

Originally Posted by SRTie4k I'm getting the same thing now strangely. I zero'ed out my BD table and now that I'm making runs in 2* weather I suddenly creep to 18.5psi from 4000-5000, and I can't seem to get control of it. I'll try copying your values between those RPM's and see if it helps.

If you are creeping above your boost target and your WGDC goes to zero (like mine did when I was targeting 15 PSI recently), then tuning won't help (unless you are going to close the throttle blade).

The creep is due to physical flow limitation issues where insufficient exhaust flow can go around the turbine wheel instead of through it...

The solutions would be either to go to an external wastegate, port out the stock wastegate port, or just run boost higher than the creep point...

[Dano]

I would try lowering WGDC in the RPMs that you are creeping before going back to a zero value BD table. The zero BD method is just a starting point to get WGDC close then you HAVE to have it enabled to complete the process.

Edit: actually lower WGDC about 300 RPMs before your creep starts [or the nearest 500RPM cell] Remember think just a bit in the past with WGDC tuning.

[Dano]

So here is my latest revision of map and logs.

I actually lowered DC in the 2.5k and 3K ranges which along with the BD table, actually made boost come on quicker than with the higher DC values from my older revisions and zeroed BD values.

I also found that around spoolup, its better to target 3-4PSI of difference for each cell than to try and gradually ramp the boost up. You can compare my BT from older revisions to my current one between 2.5-4K. I suspect the reason for this is that spoolup is so fast and our table resolution is so small [500RPMs] that overshooting target is very easy which kicks in the BD table to compensate. I also built in a slight spike at 4K to help smooth out the transition which worked out great.

So I have to go back on what I said in post 1 about targeted boost not having to match the spool characteristics of the setup. It does matter what you request and the closer you match that request to the natural spool the smoother your curve will be. Its still not near as hard to match as Load is


In the attached log I was right on target 50% of the time so more work is needed. You will see where sometimes boost hangs out at 18 and 19 PSI at spoolup. This is due to my BD table having 0 value in the -1.00 underrun cell AND having WGDC very close to what it needs to be. This is where I came up with the conclusion it is actually bad to have DC right on target. I either need to put in a .01 value there or lower or raise DC so the boost error will be greater than 1. make sense?

I am also getting a slight overrun around 4700 up to 21.xx. Again I think the 0 value in my BD overrun cell is allowing this. I am afraid to put a .01 value in there as it might be too reactive.

I think my next revision will include the .01 in my -1 and +1 cells within BD and I'll ramp up DC in the upper RPMS to see if BD prevents overruns. I suspect if I left my WGDC table as is, warmer temps would leave me coming up short.

So here is my Map, screen shots and logs. These runs were not full runs as I was only working on spoolup

[amoosenamedhank]

I'm learning very fast that it's a little too easy to spool up this K04 at the lower RPMs. I'm going to do my best to keep this under wraps in my tune but at the end of the day, it will just take pedal self-control to keep this short block together.

I'm at work so I don't have my tune in front of me... but IIRC I don't have 18.5psi targeted until 3,200rpms and this damn little thing spiked immediately to 18psi by 2,700 rpms where my BD table kicked its dick in and caused a crazy looking boost curve.

[djuosnteisn]

I just skimmed most of this and had a couple thoughts.


I believe the main reason people are seeing "smoothed" throttle response with the Boost Targeting method is because the Abs Load Target table was the main culprit of the overly touchy throttle. And it's likely being neutered with boost targeting (along with all the other load based tables). That's probably why you have a much more intuitive throttle feel now.

And for the gear based boost control. How come your having to use the APP mapping tables? Don't the gear based Boost Comp tables work? If they do, they may be the better option for limiting boost in lower gears, you can be aggressive with the limitation in the low rpms (where traction is a premium), and relax it some in the upper rpms.

Seems like there's a table for gears 1-2, 3, 4, & 5-6. 1-2 & 5-6 sharing a common table.


Also, one last thing i just figured out a couple weeks ago is that the various CL Max Load tables actually govern when certain features transition to open loop. I'd have to go back and look, but i noticed that one table (with lower values) governed when the STFT zeroed out... and another table (with the higher values) determined with the LTFT finally zeroed out. I'm sure there's similar definitions for the remaining tables.


One of these days, i intend on playing with them to see if you could possibly leave the STFT functional to redline... and potentially serve as an additional fail safe...

but that's best saved for another topic down the road.



Good shit here, seems like boost based tuning is pretty easy and effective.

[Dano]

Thanks for sharing your thoughts Dustin!

I know back on my load tune I adjusted the Boost Comp with no results but suppose that would be expected.

so I just quickly adjusted APP to see how it worked. Then spent the rest of my time on the actual boost tuning. I also have a Halman MBC and was thinking along the lines of boost limiting even with it being used.

Yes I totally agree with you and if you are running EBC the Boost Comp tables would be the best way if they work LOL

When I finialize my boost curve my next task is to work on the boost comp tables and my results will be posted here....where do all the hours go LOL

[Frequentflyer]

I figured out why I wasn't reaching boost targets and why it was tapering off too soon. I copied my max targets one more line up to the "68.75" line in the Boost Targets table since my maxes started at 75 and above. Since my throttle pretty much peaks at 74'ish, I wasn't quite reaching my max. DOH! It was that easy and what a difference that extra 1 psi makes.

I went back to stock on my Boost Comp gear tables. I haven't decided what the better method of traction control is. I am assuming the A and B tables are like any other where A is normal and B is when traction control engages just like the Load tables, but we still have the old DBW A, B and C tables and of course the APP tables. It's all a friggen mystery how this ECU works, but is the ECU using the DBW tables anymore with Boost tuning? I wouldn't think it's trying to control torque with the throttle anymore.

I copied your APP tables for 1st and 2nd and they seem to work pretty well.

[SRTie4k]

The only thing I can recommend from my couple of days playing with boost targets is to make sure you make all your map revisions and test runs within similar ambient temps. The hugely fluctuating temps here in the northeast are sending my WGDC values all over the map from day to day.

Once the temp hits about 30* here, I'm going to spend as much time as I can trying to get WGDC as close as possible to the psi I'm shooting for, then implement the BD values. After that, I'm going to leave it alone.

[m4tic]

have we figured out what to do with the boost RPM comp A/B tables?

[cld12pk2go]

Originally Posted by SRTie4k The only thing I can recommend from my couple of days playing with boost targets is to make sure you make all your map revisions and test runs within similar ambient temps. The hugely fluctuating temps here in the northeast are sending my WGDC values all over the map from day to day. Once the temp hits about 30* here, I'm going to spend as much time as I can trying to get WGDC as close as possible to the psi I'm shooting for, then implement the BD values. After that, I'm going to leave it alone.

You should be able to adjust for this with the "WG Duty - IAT Comp." table.

Mine is a work in progress, but I am using the following values currently:

-40.00	-4.00	32.00	68.00	104.00	140.00	176.00	212.00
0.80	0.85	0.91	0.97	1.00	1.01	1.03	1.05

[Frequentflyer]

Originally Posted by cld12pk2go You should be able to adjust for this with the "WG Duty - IAT Comp." table. Mine is a work in progress, but I am using the following values currently: -40.00 -4.00 32.00 68.00 104.00 140.00 176.00 212.00 0.80 0.85 0.91 0.97 1.00 1.01 1.03 1.05

I do a lot of highway driving, so my intake temps stay relatively cold (about 6* above ambient off the outside temp gauge in car). Even driving around town in some stop and go only gets me up into the 50's and 60's for intake temps and as soon as I stop moving, they quickly come back down again. So, here are my numbers: Quite similar to stock up top, but I adjusted for colder weather accordingly. Since I overshoot my WGDC's 2-3% from what I need as it is, I figured I'd make 68* my 100% mark and at 32*, I'll have it run 98%.

My current WGDC's are attached.

-40.00	-4.00	32.00	68.00	104.00	140.00	176.00	212.00
0.81	0.92	0.98	1.00	1.02	1.05	1.12	1.20

[Dano]

I have actually set my IAT comp to 100 for the usable range 32-104 until I get the other tables settled.

I do log within 10* of the same AMB temps though at between 45-55* so it wouldn't matter if I hadn't changed the values.

Just like to remove as many variables as possible until I see the ECU do what I want. LOL

So my method so far as been:

zero out BD table values [ This would not be needed if running OEM turbo and BC]
negate the WGDC IAT comp table by using 100% in the temp ranges I'll be tuning in
Set BT down to the 68 TPS row [ or alter DBW tables to run 100% vs. 80%]
get WGDC values close to target [within 3-4 PSI I would say]
re-enable BD to at most 1/2 of OTS values but I used Christian's very limited values
Fine tune WGDC and BD to achieve targets

Next in line would be

Alter Boost Comp gear tables to limit boost per gear
re-enable WGDC IAT comp table values as desired.

I think that sums up my thinking for now...subject to change

Oh and FF DBW will always be used regardless of load or boost tuning method as it is what the ECU uses to determine throttle blade position based on load and RPM.

[Frequentflyer]

Originally Posted by Dano2010 I have actually set my IAT comp to 100 for the usable range 32-104 until I get the other tables settled. I do log within 10* of the same AMB temps though at between 45-55* so it wouldn't matter if I hadn't changed the values. Just like to remove as many variables as possible until I see the ECU do what I want. LOL So my method so far as been: zero out BD table values [ This would not be needed if running OEM turbo and BC] negate the WGDC IAT comp table by using 100% in the temp ranges I'll be tuning in Set BT down to the 68 TPS row [ or alter DBW tables to run 100% vs. 80%] get WGDC values close to target [within 3-4 PSI I would say] re-enable BD to at most 1/2 of OTS values but I used Christian's very limited values Fine tune WGDC and BD to achieve targets Next in line would be Alter Boost Comp gear tables to limit boost per gear re-enable WGDC IAT comp table values as desired. I think that sums up my thinking for now...subject to change Oh and FF DBW will always be used regardless of load or boost tuning method as it is what the ECU uses to determine throttle blade position based on load and RPM.

And then figure out what the Boost Tables: RPM Comp A and B do.

[m4tic]

Originally Posted by Frequentflyer And then figure out what the Boost Tables: RPM Comp A and B do.

http://www.mazdaspeedforum.org/forum/foru...tml#post694273

[Dano]

one step at a time...LOL

[XLT_66]

sub'ed for reference.

[m4tic]

this is awesome... i have my WGDC and BD set where it works for me.
now im just inching up my high throttle/OL timing table.

the car feels much better with each timing revision. it gives me a lot more confidence in my car actually working because for the longest time i thought something may be wrong with my it.

now i want FMIC and 3071/6

[Dano]

Well I think I've reached the end of my journey with setting up boost control on my pressure based map. Next I will work on Boost Comp xGear tables.

Conclusions:

1) If you are running OEM Turbo and Boost Control there is no need to zero out your BD table nor your WGDC table. Just throw the boost toggle and then make adjustments.
2) If you are running big turbo and or aftermarket EBC then starting out with BD zeroed and very low WGDC values is helpful.
3) Just keep BD zeroed out until you start to see WGDC ramp boost up to target, then set your BD table to your desired values. Keeping it zeroed out too long will just have you chasing your tail with WGDC table changes. You will end up with too much DC in the table and will need to go back and reduce it.

The changes I made since my last revision:

BD: I added -0.01 to the 1.00 over run value to help keep boost overrun in check

WGDC: As you can see I lowered just about the entire table. This is where I went wrong before by keeping the BD table zeroed out for too long. Lesson learned.

BT: I bumped up my target value at the 3.5K range since it was hitting ~19PSI anyway. This keeps the BD table from jumping in and adding oscillation during spoolup.

Now that I have the table values set, I went back and cleaned them up a bit by interpolating values in the BT and WGDC tables.

You can see in the logs that I still have a slight over run around 5.5K and I will put a final tweak to my WGDC and lower DC at that RPM by 1 which should put it where I want it.

I am not sure how my limited BD table will react as temps get warmer but I hope it can compensate and I won’t have to either modify it or change WGDC table values but time will tell. Even if changes are needed they should be minor and only take a rev or two to complete.

Well here are my existing values and logs.

Two map revisions are included in the log as I only changed WGDC value in 1 RPM range to prevent the boost overruns.

[djuosnteisn]

Looks great man.

[Dano]

Thanks

Yeah it will do for now. We'll see how the limited BD table works as temps rise this spring/summer


Sent from my iPhone using Tapatalk

[DSIT995]

very nice, not a big prob to tweak a couple tables when the temps rise

I still like this better than Load tuning... small price to pay & I cant help it I like to tweak shit all the time lol

EDIT: hmm is anybody else having trouble seeing Dano's screenshots?

[speed23]

I cant see them anymore either.

Dano, do you see any advantages to running the boost targeting maps if you already have dialed in load based map? and lastly looking forward to your boost per gear tuning.

[Dano]

my hosting site is down but the pics should be back up by the end of the day.

yeah if I have to tweak BD or WGDC it will be quick and easy...hell I'll be tweaking to add more boost in the spring/summer anyway. Trying to stay in the mid 20s now to keep load values down. I don't want to become a member of the zoom zoom boom club for being greedy

[Dano]

Originally Posted by speed23 I cant see them anymore either. Dano, do you see any advantages to running the boost targeting maps if you already have dialed in load based map? and lastly looking forward to your boost per gear tuning.

Well the first thing that comes to mind is smoother throttle response with Boost tuning. Bringing on power seems to be more controllable...I described it a few posts back IIRC but if you have your load based tune down and like it then no, but I like to tweak things constantly

I may get some time this week to try out the Boost Comp tables. They will either work or not so it shouldn't take long to find out if they do...LOL Getting them completely dialed in, on the other hand, is a different story

Temp will play a key roll and as it changes, so will the need to tweak the tables...This is another reason I may favor the APP method...I just don't know yet.

I suppose you find some middle ground on the Boost Comp tables and control the rest with your right foot.

If the IAT Comp table had higher resolution THAT would rule for controling power by looking at IAT.

hum....

[Fobio]

I'll be getting on this!

[Dano]

hahah

yes for me it literally took 1/2 the time to get boost tuned with pressure based tuning vs load based tuning. Boost curve is way smoother as well. I had a lot of trouble with boost occiliations on my load based map...finally got it worked out but then wanted to give the pressure based tuning a try.

man I wish I had used pressure based tuning long ago!

jussayin

[Dano]

Boost Comp Test

Well I gave my Boost Comp 1-2nd gear table a tweak today and have disappointing results.

I set the A table to .50 from 3-5.5k and it does limit power in 1st gear very well but doesn't do any limiting for second.

I tried it about 5 times with the peddal on the floor and it crawled in 1st gear, shifted at 5K then blew the tires off in second gear. WTF..why make something that 1/2 works!

my tires are about 1/4" shorter than OEM so I wonder if my gear ratio is off enough that the ECU doesn't know what gear I am in after 1st....nah it does b/c my TRL gear tables worked so IDK.

I didn't alter 3rd gear comp table but will give that a try later this week.

I will be interested to here other's results with this.

[Frequentflyer]

Originally Posted by Dano2010 Boost Comp Test Well I gave my Boost Comp 1-2nd gear table a tweak today and have disappointing results. I set the A table to .50 from 3-5.5k and it does limit power in 1st gear very l!well but doesn't do any limiting for second. I tried it about 5 times with the peddal on the floor and it crawled in 1st gear, shifted at 5K then blew the tires off in second gear. WTF..why make something that 1/2 works! my tires are about 1/4" shorter than OEM so I wonder if my gear ratio is off enough that the ECU doesn't know what gear I am in after 1st....nah it does b/c my TRL gear tables worked so IDK. I didn't alter 3rd gear comp table but will give that a try later this week. I will be interested to here other's results with this.

Did you try both a and b tables?