Hybrid Boost Control - solving the problem of minimizing spool and stabilizing boost.

[BrandonDrums]

Intro

The Subaru factory boost control system is a unique take on controlling the boost levels on a turbocharged engine. When tuning WRX's and STI's using the factory 2 port and even with an upgraded 3 port boost control solenoid, the tuner continuously has to walk a fine line between achieving the quickest spool and providing stable and predictible boost in all conditions.

Essentially, with both a 2 and 3 port boost control solenoid, the ECU applies a correction value via Turbo Dynamics to "Waste Gate Duty Cycle" or WGDC to correct Boost error. WGDC is the percentage of boost signal that the boost control solenoid diverts away from the wastegate actuator on the turbocharger. Turbo Dynamics is a ECU table that checks the raw pressure value difference between target boost and actual boost and applies additional WGDC based on the amount it's off.

What results is a somewhat laggy system of boost control that is difficult to tune for maximizing boost and guaranteeing boost stability in all gears in all conditions.

This issue is primarily with the 02-05 16-bit WRX ECU's and the 06-07 32 bit ECU's. STI ECU's have separate boost control tables for each gear which allows much better control for maximizing spoolup without sacrificing boost stability.

[SIZE=4]A little more explanation
In order to control boost, pressurized air from the turbo itself is applied against a little diaphragm controlled arm that opens a flap that allows exhaust gas to bypass the turbine wheel during an overboost situation and then bleeds pressure away from reaching that diaphragm to keep it closed when trying to get more boost out of the turbo.

In a Stock Subaru WRX, there's a solenoid controlled valve that essentially controls what % of the pressure coming from the turbo reaches that arm. This is called Wastegate Duty Cycle or WGDC. 0% duty cycle means the solenoid is applying all of the turbo pressure to the wastegate actuator, 100% duty cycle means it's bleeding all pressure away from the actuator in an attempt to maximize boost.

The ECU system that controls that solenoid works continuously in a feedback control loop or Closed Loop process which means you can't set how much duty cycle is applied at a given set of criteria. Instead, you only set the boost you'd like to achieve, the maximum amount of WGDC you wish the system to apply at a particular Throttle input and RPM speed and the increment in which the system adds in WGDC per correction cycle as it tries to achieve target boost.

The reason subaru uses this system is because weather and altitude affect a turbo's performance and the car is designed to operate reliably in all conditions. For that reason, the boost control system is pretty good but it sacrifices quite a bit of turbo performance to achieve more day to day reliability.

You can speed up the control system by upgrading to a 3 port boost control system but that is still difficult to tune and only reduces the balance between spool and stability.

The method below allows you to maximize turbocharger spool performance while coming pretty close to providing the same level of all condition boost stability from the factory unit on top of making it

How the Hybrid Boost Control method improves this sytem

Using a Manual boost controller in conjunction with this system, you can essentially 'trick' the ECU to run in "open Loop" where the Manual boost controller provides a mechanical limit to the peak boost you want to apply to your engine and the Electronic system controls partial throttle boost levels. Instead of guess and check, you set the iTurbo Dynamics correction values to 100% for all underboosting error levels (on 16 bit ECU's 02-05 wrx's) and then use the Max WGDC table to set the throttle and rpm dependent WGDC's (if the correction is 100% but the max is 75%, then the applied WGDC will be 75% regardless of the 100% correction).

On 32 bit ecus (06+ wrx's, STI's and 03+ legacies) you instead set the initial WGDC and turbo dynamics becomes irrelevant. Either way, you then have full control over the WGDC being applied in the system at all throttle and RPM levels which remain static.

Tick #2 is to set the partial throttle target boost levels a little bit higher than the WGDC will allow the turbo to achieve i.e. at 75% throttle, you apply 60% WGDC but leave the boost target at 18 psi. A TD04 on a Ej20 using a 3 port BCS would never achieve 18psi with only 60% duty cycle, instead is only reaches 16 psi in the winter and 15 psi in the summer (example figures, not actual data) Many tuners use this method on a 3 port without a manual controller, the issue is setting the max target at WOT higher and only relying on max WGDC to limit boost allows for massive changes in boost pressure depending on ambient temperature and engine load. It's dangerous.

Conclusion

SO Not only does this Hybrid Boost Control method improve performance, but it also makes this process MUCH easier to tune for all weather conditions (temperature affects boost a whole lot), it improves the stability and safety of the system by eliminating boost spikes, boost creep and boost oscillation (overshoot, overcorrect into underboost, overcorrect into overboost etc etc) and leads to a more driveable and more reliable car.

We'll be driving these 818's hard and many will be driving them on the racetrack. Add better power to weight ratios and less grip and you've reduced the amount of time the ECU has to apply corrections during acceleration and de-stabalized the amount of load the engine sees as the tires struggle for grip. You'll need a faster boost control system to keep up.

The hybrid is effective for both internally gated and externally gated wastegate systems and I think every future 818 owner planning on performing any power mods or tuning should consider.

References

NASIOC Sticky on hybrid boost control
http://forums.nasioc.com/forums/showthread.php?t=602674

One of the best dyno result demonstrations of Hybrid Boost Control
http://forums.nasioc.com/forums/show...2165560&page=2

Explanation of the factory boost control system to compare
http://www.cobbforums.com/forums/sho...l-System-Works

I guess this counts as my "Subaru Tidbit of the Day" although it's a bit long.

I'd love to hear comments and questions. This is one of my favorite topics so fire away!

[Steve91T]

I'd love to comment, but that's way over my head!

I'm used to a simple manual boost control. Turn it in for more boost, out for less. Waste gate stays closed until it reaches the desired boost. Pretty accurate and reliable.

Steve

[BrandonDrums]

I'd love to comment, but that's way over my head!

I'm used to a simple manual boost control. Turn it in for more boost, out for less. Waste gate stays closed until it reaches the desired boost. Pretty accurate and reliable.

Steve

Haha, well, this is essentially setting the boost to run only on the Manual Boost Controller everywhere you want it to and to have separate limits everywhere you don't. MBC's work, but turbochargers create heat so with an MBC you can see partial throttle full boost which is fine if you tune for it. However, Turbo's make lots of heat, so asking for full boost when the throttle plate is closed puts larger turbos at risk of compressor surge and smaller turbos will just heat up like crazy and shoot out extremely hot air aiding in heat soak.

I firmly believe you'll make more power if you limit boost a bit at partial throttle, cooler air and less heat soak. Not only that but you'll have better throttle modulation which is good for a RWD car. MBC only on a smaller engine like a subaru means power is way more dependent on boost levels than a big V8. On a V8, the engine swallows more air (higher VE) so a turbo would have to be much much bigger to have this issue, I bet you don't run into many partial throttle full boost situations even driving down the highway in your car. Happens all the time with these smaller turbocharged engines.

For your Camaro I doubt you've ever had to consider the pitfalls of using a MBC only setup. If you build an 818 or get a Subaru, you'll revisit this post ;-)

[Steve91T]

Not my Camaro, but I used to have a couple of MR2 turbo's for about 10 years. You are right, they can cause a high boost at partial throttle, but if you have a proper IC, it isn't a bad thing. One thing to consider is that high boost at partial throttle means the turbo is spooled and ready for you to slam the throttle open. With a lower boost, you will have to wait for the turbo to spool, while it'll do it quickly, not as quick as one that's already there.

[skullandbones]

Don't want to interupt your conversation; it's pretty interesting. But I have a couple of general questions since I have an interest in turbo technology. It's a very attractive way of boosting hp without "selling the farm". I would be adding to a small block Ford.

Ques 1. Do you think the overall engine compartment temp would be reduced using the hybrid system? Heat control is a big issue (shielding , etc).

Ques 2. If the Subaru system is "guess and check" what would you call the hybid system, "variable feedback" or what?

Thanks, WEK.

[16g-95gsx]

On a V8, the engine swallows more air (higher VE)

VE=not equal to CFM

What you're describing is how all ECU based boost control systems are done, including all modern standalones. Effectively you have a PID control feedback loop that is constantly comparing current boost to a target boost map, and then adjusts the duty cycle up or down based on another seperate "boost error" table. There is potentially a lot more that can go into it than that on standalones, but that's the gist of it.

As for electronic solenoids I would highly recommend not going with a GM unit if you can find a MAC valve. I have a good 30 of them sitting in my house, used to sell them all the time.

[16g-95gsx]

single port flapper style WG's aren't the nicest to work with.

[BrandonDrums]

VE=not equal to CFM

What you're describing is how all ECU based boost control systems are done, including all modern standalones. Effectively you have a PID control feedback loop that is constantly comparing current boost to a target boost map, and then adjusts the duty cycle up or down based on another seperate "boost error" table. There is potentially a lot more that can go into it than that on standalones, but that's the gist of it.

As for electronic solenoids I would highly recommend not going with a GM unit if you can find a MAC valve. I have a good 30 of them sitting in my house, used to sell them all the time.

Well VE determines how many CFM an engine can consume. VE is the exact variable that is used to graph turbo compressor maps and changing throttle position I believe effectively changes the VE.

I don't have tremendous knowledge of other boost control systems but I believe most modern boost control systems are load based or at least load inclusive rather than throttle and RPM based so they are much more easily tuned to be stable. Perhaps "guess and check" or "error" based isn't the limiting factor on Subaru ECU's as much as it's the lack of load-based boost control tables. Without Load based mapping (STI's Do have Gear-based which helps) tuning boost for quick spool in gears 1-3 will result in unstable boost or overboost in the higher gears as the engine observes higher load at the same RPM in 5th vs. 1st.

Load based control systems as found on the Evo would allow you to adjust for that parity.

Regardless, guess and check systems although sophisticated and fully integral to the ECU without any mechanically actuated boost adjustments, are less ideal when it comes to maximizing turbocharger performance and they are far more difficult to tune and understand. This Hybrid method provides a whole lot of dummy protection so to speak while still providing the benefits of controlling boost at all throttle positions and RPM ranges without the need to find the invisible line between quick spool down low, sustained boost up top and stable boost under changing weather and changing load conditions.

It's a big topic for Subaru tuners, not so much for Mitsubishi tuners I guess.

[BrandonDrums]

When I say most modern boost control systems, I mean factory systems. Stand alone boost control systems I don't know much about at all. My bud's RX7 has a stand-alone system and I think it's blind of throttle rpm and engine load. It has boost target, trim and gain all of which I believe are analog settings not quantified digital settings. Other than that, it's greek to me.

So, this discussion is just talking about one way to improve upon the Subaru Factory boost control system easily and cost-effectively.

[305mouse]

go EWG, makes response so much better and scares deer late at night so I don't hit them.

[BrandonDrums]

This is a poor man's EWG and some argue it's just as effective. Some even use the Hybrid setup with EWG.

I'm not sure what the exhaust options will be for an 818, I imagine doing an external wastegate will take more custom exhaust work since there's a chance the 818 will have a completely different turboback exhaust than an impreza.

EWG is best but this is a good option to get EWG like results using the stock exhaust configuration.

[BrandonDrums]

Here you go http://en.wikipedia.org/wiki/Boost_c...ontrol_details

Not all electronic boost systems are Closed Loop, the wiki article has a pretty good explanation of the pros and cons for both methods. It does leave out how even with an interrupt controller, in a Closed loop system you have limited ability to fully interrupt all boost signal to the wastegate.

It does talk about the limitations of internally gated turbos, even with no boost signal the exhaust gas pressure can begin to open up the flap. At least with the hybrid control you can remove the boost signal from the equation. An EWG with a center hinged flap wouldn't open with exhaust gas pressure which is something to consider.

[305mouse]

With an EWG you would just have to worry about the dump tube, and since it's using the pieces, companies like Grimmspeed make a EWG with a dump tube that avoids the axle and exits free and clear. Now if you wanted something plumbed into the DP, then yes, that's custom, but who wants that. Also with a EWG you could have the internal flap welded shut. With Grimmspeed they make a bracket to keep it shut.

I'm running a VF35 witha EWG and 3 port boost solenoid currently. What'll be fun is with the 818 I plan on running a twin scroll and I really don't want to mess with 2 EWG's on that.

[BrandonDrums]

I'm bumping this thread, I think this will be relevant to all the guys now considering which donor to buy and are concerned about spoolup on the 2.0L models.

[Flamshackle]

loved the info thanks for sharing mate

[Viper1982]

MBC+OEM2-port.jpg