Need help controlling boost

[Mchart]

As can be seen from the datalog the engine is doing pretty good. Have everything modded on the fuel/intake side. Exhaust side is stock. Map is the OTS Stage 1 with higher load values.

However, i've never had much luck controlling this boost. I got it a little bit under better control by changing the Stage 1 OTS WGDC table, but it was only a little bit better and the car was slower to build boost then as well.

I appreciate any feedback.

[jack_hammer]

when i have trouble controlling my wife, i bitch slap her. that usually helps. have you tried bitch slapping your boost?



dangah-zone!

[phate]

Killing me here. What are all the parts that you have on the car, because it affects boost control, and what year is your car?

[Mchart]

2010.

SRI/TIP, KMD's, Cork TMIC.

Will be running another log soon, lowered the WGDC values at 3500 RPM's and 4500 RPM's. Raised throttle close to 19PSI so I don't get the throttle drop.

[phate]

What's your elevation?

I presume you are trying boost tuning?

If you are boost tuning, I would first eliminate the LEC table as the culprit - zero it out. Populate all BEC over and underboost values so you allow the table to work.

[Mchart]

Elevation is basically sea level.

I have the 'boost tuning' box checked. My ECU seems to be one of those nut hugging ones that ignores that for the most part though.

I don't know what LEC and BEC means.

[phate]

Originally Posted by Mchart Elevation is basically sea level. I have the 'boost tuning' box checked. My ECU seems to be one of those nut hugging ones that ignores that for the most part though. I don't know what LEC and BEC means.

So you probably don't need to worry about the baro comp tables.

BEC = WG Boost Error Comp

LEC = WG Load Error Comp

They are both active with the boost tuning box checked. LEC often interferes with the pure boost tunes.

[Mchart]

Ah. Yeah that table had a bunch of shit in it. Zeroed out now. That would make sense it forcing the overboost to reach the higher loads I set.

Are there any 'reccomended' values for the BEC table beyond OTS? I don't want to be potentially overboosting now that i've removed the LEC tables.

[phate]

Cut down WGDC by ~20% across the board to make sure the car will compensate properly with just BEC. Some cars do not, so making things safe first is recommended.

I would start out small with the BEC values, maybe something like:

0.14 0.10 0.06 0.03 0.01 0.00 -0.02 -0.07 -0.15 -0.25 -0.50

If the car runs exact WGDC target values, you can try upping the BEC values, but it could be the car going straight to the LEC table. If the latter is the case, start playing with LEC, but you'll need to set your TRL xGear tables appropriately.

[Mchart]

Thanks. Hopefully it was just that LEC table.

Once I get the boost decently stable I can finally start ramping up timing.

[Mchart]

Alright, attached you'll find my datalog from today. 3rd gear run and couldn't get up as high - But it shows me all I needed to know.

That fucking LEC. I had been chasing the WGDC ghosts for half a year because of that fucker. So i'll be able to dial in this boost to exactly where I want it (18 PSI) now.

Dropping the OTS WGDC table by 20% also was a great piece of advice. Allready pretty close to what I want.

Not really sure why the virtual dyno is showing so much more power on this new datalog. It was a little bit colder outside today, and maybe because this other one is 3rd gear?

[zpele]

Looks much better man. Your 4500 values for WGDC still seem a bit high but its no where near as bad.

Also, dont ramp up timing unless you know what you are doing. It's fairly easy to do but it's also easy to do damage if you do it wrong.

[Mchart]

I was allowing 80 as I did a 20% reduction of the OTS map which usually requests 100% at that point. I was also targeting for 17 psi at that point.

Frankly, I don't know if I want to touch that value given the fact of how close it is to 17PSI allready.

Whoops, was looking at 5500 not 4500. Yeah, i'll probably lower the middle-band regions of WGDC a bit more.

As for timing, I only plan on pushing half a degree each datalog. I'm not a pro at this, but I can't see what the harm would be in only pushing it half a degree each time would be. I don't have any KR as it is, and i'm running 94 Octane gas. I also plan at staying at the OTS values of 11.4 AFR.

[phate]

Glad to hear that's helping.

Another note, since you are using strictly BEC. The OTS maps only allow 10% additive compensation from the BEC table (it's one of the 'High Limit' table). You may need to open that up a bit, as boost tunes need more WGDC range to cover wide temperature ranges.

If you're tuning when it's cold out, you can have WGDC targets pretty close to actuals. If it's hot out, low ball the WGDC and get BEC/LEC to work up to the targets, so you don't overboost in cold weather.

[zpele]

Originally Posted by phate Glad to hear that's helping. Another note, since you are using strictly BEC. The OTS maps only allow 10% additive compensation from the BEC table (it's one of the 'High Limit' table). You may need to open that up a bit, as boost tunes need more WGDC range to cover wide temperature ranges. If you're tuning when it's cold out, you can have WGDC targets pretty close to actuals. If it's hot out, low ball the WGDC and get BEC/LEC to work up to the targets, so you don't overboost in cold weather.

I ran into this issue myself when I was first learning how to tune.

btw @phate, why can't you interpolate Horizontally in the OL/High Load timing table? It makes me mad that I cant do it as it would smooth out my timing.

[phate]

Originally Posted by zpele I ran into this issue myself when I was first learning how to tune. btw Phate, why can't you interpolate Horizontally in the OL/High Load timing table? It makes me mad that I cant do it as it would smooth out my timing.

I've never had that problem!?! I've had a couple instances where certain tables wouldn't copy/paste, but it was an issue with the ATR install. Maybe something similar for you...that's really weird!

[zpele]

Originally Posted by phate I've never had that problem!?! I've had a couple instances where certain tables wouldn't copy/paste, but it was an issue with the ATR install. Maybe something similar for you...that's really weird!

Yea its really weird... I can do it vertically but it just wont let me do it horizontally. I thought it was a safeguard COBB had stuck in there but it is incredibly annoying. I also might have a different version of ATR than you do. It's the latest MS6 version and I know back when I was learning, my tables were a bit different from Buckers.

[phate]

Yeah, I have the MS6 ATR...Lemme go check this, lol.

Edit: Horizontal Interp works on mine.

[zpele]

Originally Posted by phate Yeah, I have the MS6 ATR...Lemme go check this, lol. Edit: Horizontal Interp works on mine.

Thanks for checkin. What happens is I will highlight the cells I want to interpolate and right click. The menu comes up but only a single cell is highlighted instead of the multiple ones I just did.

Does it over and over. It works fine for vertical cells and horizontal works fine in other tables.

Sorry for the thread jack btw.

[Mchart]

Originally Posted by zpele Thanks for checkin. What happens is I will highlight the cells I want to interpolate and right click. The menu comes up but only a single cell is highlighted instead of the multiple ones I just did. Does it over and over. It works fine for vertical cells and horizontal works fine in other tables. Sorry for the thread jack btw.

I've had this happen to me a number of times. I usually have to fuck around with right clicking and reselecting a bunch of times and it will do it finally.

Originally Posted by phate Glad to hear that's helping. Another note, since you are using strictly BEC. The OTS maps only allow 10% additive compensation from the BEC table (it's one of the 'High Limit' table). You may need to open that up a bit, as boost tunes need more WGDC range to cover wide temperature ranges. If you're tuning when it's cold out, you can have WGDC targets pretty close to actuals. If it's hot out, low ball the WGDC and get BEC/LEC to work up to the targets, so you don't overboost in cold weather.

Yeah this is England. Hottest it ever usually gets is in the 70's. Been doing these datalogs in high 40's.

[phate]

press 'v' for vertical interp, and 'h' for horizontal???

[zpele]

Originally Posted by phate press 'v' for vertical interp, and 'h' for horizontal???

Tried that the same thing happens. It changes from several cells being selected to 1.

[rfinkle2]

Left click to select and forget the right click menu.

Use the "V" and "H" hot keys.

Good stuff guys.

[Mchart]

I think @rfinkle2 should tune my timings. Dangerzone and all.

[rfinkle2]

Once you get your boost dialed in, I'd be happy to help.

Phate>rfinkle2.

[2k4_8]

Good stuff. Do you have any issue with overboosting/boost spike when you ffs? 1-2 and 2-3 there is little to none, 3-4 there is a small spike, but 4-5 I go 2.5 psi over target? I am also still getting boost levels under control....and cold weather fucks me it seems.

[phate]

Originally Posted by 2k4_8 Good stuff. Do you have any issue with overboosting/boost spike when you ffs? 1-2 and 2-3 there is little to none, 3-4 there is a small spike, but 4-5 I go 2.5 psi over target? I am also still getting boost levels under control....and cold weather fucks me it seems.

The BEC table is a little slow to respond with small values, so you can try larger values on the overboost side, but it may cause some little pressure oscillations.

The other option is to enable the overload side of LEC, dial in your TRL xGear tables to give them just a bit of headroom, and allow it to knock down spikes when they occur. If you do this, realize it may (depending on where you set the load targets) lower boost when the temps get cold out, as it takes less pressure to flow the same air mass/generate the same load with the denser air.

[Mchart]

Alright, well i've gotten my boost under decent control now. Needs to be tweaked a little bit more but i've run into a new issue.

From about 2800 RPM's to 3600 RPM's the car is backing my throttle off. This happens in pretty much every WOT run now. So it wasn't me letting off on the throttle for a small period. I've looked through all the tables and I can't really figure out what would be causing this. My load caps are set to 2.3, and my throttle close for overboost is set to 19PSI.

Any help is appreciated.

@phate @rfinkle2

[phate]

DBW Throttle A/B/C (or whatever they're called) tables need to be set to 90 or higher to pin the throttle open. I think the OTS values use 80 - you can change those to 90 and then smooth the transition with some interpolation.

[Mchart]

If my throttle was pegged to the floor, along with the calculated load being basically 2.0 why would the throttle dip down to 68% if were set for 80%? As can be seen in the datalog it was at the max position of 75.6% but the moment the load came near being 2.0 it dropped to high 60's.

[silvapain]

Originally Posted by Mchart If my throttle was pegged to the floor, along with the calculated load being basically 2.0 why would the throttle dip down to 68% if were set for 80%? As can be seen in the datalog it was at the max position of 75.6% but the moment the load came near being 2.0 it dropped to high 60's.

You were approaching your load targets I'd guess. The ECU closes the TP based on load.


Tapadatass

[Mchart]

Originally Posted by silvapain You were approaching your load targets I'd guess. The ECU closes the TP based on load. Tapadatass

I've upped my load targets to the 'reccomended' values of 2.3 though. I only ever hit 2.1 at most with my current parts and what not. Maybe I need to throw them up higher as the ECU was scaling or something?

Regardless, I've upped my throttle values and smoothed out those tables a bit.

[phate]

Do you have load dynamics zero'd??

[Mchart]

Yes

Just noticed that my max load for the 2500 RPM cell's was 2.12. The points where the throttle drop is from 2500-3500 RPM's have load nearing this value. So I upped this value to 2.4 as well.

[Mchart]

Well, that fixed it. Need to do a couple days more worth of tweaks and my boost should be dialed in good. Picked up some decent power gains now that the throttle plate isn't shutting some.

[Lex]

I find it useful with this ECU to never completely ignore boost or load - keep realistic targets for both even if predominantly using load tuning or boost tuning.

[Mchart]

I have my load caps set to 2.3 currently. The problem is that it was 2.12 for the 2500 column which meant load cap. Besides the fact that my DBW was only 80% so I was getting random moments where the throttle was dropping a notch.

[econman]

Keep in mind that if you are zeroing out the LEC table that you need to adjust WG Duty RPM Comp tables so that when you are really overboosting the ECU won't skip over the BEC and go to the LEC table.

If you zero out the LEC table and don't make the adjustments to the RPM table, you might end up having zero compensation when you overboost a lot. That would be bad.

Post 1:http://www.mazdaspeedforum.org/forum/foru...ml#post1009814

Post 2:http://www.mazdaspeedforum.org/forum/foru...7/index10.html

[phate]

More often than not, those RPM comp tables don't do a damn thing.

[Mchart]

Most recent datalog attached.