boost oscillations

[thejean]

I get some significant boost oscilations and am wondering how to use the turbo dynamics tables to tune it out. I already cut way back on the WGDC but want to try the turbo dynamics since I read in the Hydra software manual that you can tune out oscillations using these tables. Thanks!

[ahains]

The ST manual has some info on tuning this:
http://www.accessecu.com/support/st-dbw-userguide.pdf

[thejean]

Thanks!

[thejean]

By the way... adjusting the burst and proportional tables totally got rid of my boost oscillations. Worked like a charm... just like the ST manual said.

[swifty]

Canyou post details of howmuch oscillation you were seeing and the before and after values in the burst and continuous tables please.

I have been through the same thing with my car and was able to tune out the oscillations but that was with the standard boost settings. I found that working out how much change in boost was made for each % change in WGDC enabled me to calculate better values for turbo dynamics.

I have started to modify my boost maps now and have been playing with turbo dynamics maps in order to try and improve boost response. I am currently trying rescaling the turbo dynamics maps to cover a wider range of boost errors.

Thanks

[thejean]

Actually, what you need to do is tighten up the WGDC compensation on thos maps. I think on the proportional I set the max to 2.0 and on the integral I set the max at 1.0. Big difference from stock.

If that doesn't work, take a look at the WGDC compensation with elevation. I think at 14.7 psi, its zero then everything else is set at 18%, which is ridiculous. That should be tapered so you only reach 18% at really low atmospheric pressures. Because I'm protuned for an atm press of 12.3 psi where I live, I went with zero all the way down to 11 psi atmospheric and then slowly tapered up to 18% at the lowest atmospheric level.

basically, the above tables make the actual WGDC look nothing like what you have mapped in the Max WGDC table due the crazy correction factors. Reduce the correction and the ECU actually uses a WGDC closer to what you actually have programmed in the Max WGDC table.

[swifty]

I am at sea level and atmospheric pressure is always at least 14.7psi so I have no issues with compensation tables, plus my logs show I am achieving the wgdc values in the table. I have them close enough to be hitting the target boost ok.

I have been working on boost response, how quickly the target boost is reached.

[Hobbes]

The boost map is a targeting value map.
The wastegate duty cycle is a rough value map
The turbo dynamic map is a pin-pointing map

Here is a IM of the ECU and the maps. (I'm a simplfying instructor)

ECU: whats the boost level
BM: 17 psi
ECU: Whats my starting duty cycle
WDC: 72%
ECU: I got 12 psi TDM whats my correction I'm off by 5 psi
TDM: use 4 % correction
ECU: My boost level is 14 psi
ECU: What should my duty cycle be now WDC
WDC: 72
ECU: I got 15 psi TDM what my correction I'm off by 2 psi
TDM: use 3% correction
ECU: My boost level is 16 psi now
ECU: What should my duty cycle be now
WDC: 72
ECU: I got 16.5 psi TDM what my correction I'm off by .5 psi
TDM: .2
ECU: My boost level is 17 psi
ECU: What should my duty cycle be now
WDC: 72
ECU: I got 17 psi TDM what my correction I'm off by 0 psi
TDM: use 0% correction
ECU: My boost level is 17 psi
ECU: I'm good now!

Here is a IM of the ECU and the maps with oscillating boost .

ECU: whats the boost level
BM: 17 psi
ECU: Whats my starting duty cycle
WDC: 50%
ECU: I got 12 psi TDM whats my correction I'm off by 5 psi
TDM: use 10 % correction
ECU: My boost level is 14 psi
ECU: What should my duty cycle be now
WDC: 72
ECU: I got 15 psi TDM what my correction I'm off by 2 psi
TDM: use 8% correction
ECU: My boost level is 16 psi now
ECU: What should my duty cycle be now
WDC: 72
ECU: I got 16 psi TDM what my correction I'm off by 1 psi
TDM: use 7% correction
ECU: My boost level is 17 psi now
ECU: What should my duty cycle be now
WDC: 72
ECU: I got 17 psi TDM what my correction I'm off by .5 psi
TDM: use -5% correction
ECU: My boost level is 18 psi now
ECU: What should my duty cycle be now
WDC: 72
ECU: I got 18 psi TDM what my correction I'm off by 0 psi
TDM: use -15% correction
ECU: My boost level is 18 psi
ECU: Oh shit!
Boost limit map: Drop it everybody cut back 20%!!
Then the ecu starts the hold process over again.

Understand that it is your turbos design that dictates how it spools up. The maps controls the solenoid the manipulates pressure on the wastegate actuator. open wastegate lowers the boost or keep it closed to build the boost. These map work together to hit the target boost.

You want to keep the WDC values lower and let the TDM adjust upward, because there are other maps that the ecu uses that may require lower boost or lower duty cycles at that time.

Hobbes

[MRF582]

You want to keep the WDC values lower and let the TDM adjust upward, because there are other maps that the ecu uses that may require lower boost or lower duty cycles at that time.

What ECU are you talking about? My 04 USDM WRX ECU has a MAX WDC map. So my wastegate doesn't ever go above what I've programmed into that map. Instead, I have to keep WDC slightly above what is required and tune the turbo dynamics map so the boost doesn't spike or oscillate. There are two turbo dyamics map in my ECU (burst and continous) or (integral gain and proportional).

Sombody explain integral and proportional gain to me please.

[Hobbes]

Remember the ECU also runs lookup tables. If you hit the limits of a value it will trigger other events.

You are using the MAX WDC is like ceiling instead of control limit. The MAX WDC map sets the duty cycle limits. It will triggers another look up for the ECU when the max level is hit.

What you want to do is have the ECU build the boost smoothly as possible directly targeting the value without having to dial back.

Burst and Continous
while you are cruising the contunious map is accessed punch it and the ecu accesses the burst map.

Ever hear the guy complain about boost creep while cruising on the free way. This is the map.

Hobbes

[thejean]

You want to keep the WDC values lower and let the TDM adjust upward, because there are other maps that the ecu uses that may require lower boost or lower duty cycles at that time.

Which maps to you speak of? And if so, why does PDX max out the WDC values in the protunes? Not being argumentative, just curious.

[Hobbes]

Look at the 04 STi for example. It has initial WDC and max WDC.

The map values are
IWDC is 80% throttle 5600 rpm and the duty is 57
MWDC is 80% throttle 5600 rpm and the duty is 65

Now if you are in 4 gear 100% throttle and all of a sudden something happen to the boost solenoid,

what should the ECU.

a. keep boosting
b. increase boost
c. back off throw check engine light

when the MWDC is hit the ECU will respone according the its program which will lead to a CEL.

Set MWDC to 100 then the ecu will continue to 100% now what can happen a or b who knows. But keep the limits and you can maintain the design integerty of Subarus programming.

As for the 100 MWDC Only PDX knows why they max it out. could be external boost controller or they could alter other map to compensate. Only they know for sure.

The only thing that I'm saying if you make changes within designed guidelines you will decrease your chances of failure/problems in your tuning.

Hobbes

[MRF582]

The 100% Max is for ultimate quick spool-up. It allows the car to reach target boost as quickly as physically possible. The only downside to that is that the other maps must be tuned perfectly to prevent over-boost or boost creep. If you can achieve that, you will have a car with much better boost response than what we're used to.

Theoretically the WRX can also run 100% WDC at WOT if and only iff the burst and continous maps are tuned perfectly to the turbo so you don't get boost spikes or boost waiver.

[m3n0ch3]

guys.. the ECU doesn`t throw a CELL if it reaches Max WDC...

only if it reaches limit BOOST.

tuning the wastegate with a 2 port WG is troublesome, especially when the spring is weak.

if the target boost is well within your turbo`s limit, the curve should almost always be nice.... as the wastegate behavior will nicely control it.

Problem arises when you play near the turbo max flow. Opening the wastegate juste a tinybit too much will have the turbo loose all it`s flow and will drop like crazy.

If your turbodynamics are giving too much adjustement, you will loose pressure when close to turbo limits..

not giving it enough might have you overboost if well under turbo limits.

A nice turbo curve will go a straight to desired boost or a little bit over with a smooth curve to level it back with NO waving.

If you have trouble reaching or eliminating waving:

1-make sure your wastegate is not sucked by a twin dump
2-make sure your wastegate is not pushed by high pressure the spring can`t hold
3-make sure your are not too close to turbo max flow
(you`ll have high pressure but too much heat which will reatard timing anywayz).

go GM BCS or lower boost a bit and play on timing to gain on torque