BMW's approach to boost control
 

Hi guys,

I've been playing around with our new 335i and found a few interesting bits. One thing I noticed, after much datalogging, probing and head-scratching is its boost control system. It's pretty unique. I haven't seen anything like this employed before in any turbo application, twin turbo or otherwise.

Unlike other turbo systems (evo included) which have a positive mechanical wastegate pressure (that keeps the wastegate normally shut); the wastegates on the turbo are designed to remain fully extended in its natural state. This means if you disconnected all forms of boost control, the 335i runs 0psi of boost. On the other hand, if you do this to an EVO, it will run 11psi of boost. To get more boost in the EVO, the factory boost control solenoid bleeds off the wastegate pressure so the wastegate stays more shut. If the wastegate signal hose, for some reason, were to pop off or rupture, the wastegate would get no boost signal, and you'd get an engine-grenading overboost condition.

With the BMW, to get more boost, it regulates a *VACUUM* signal to pull *CLOSE* the naturally open wastegate. During normal off-boost driving, the wastegate is fed 10" of vacuum which keeps the wastegate partially open. During WOT, but before full boost is reached, the wastegate is fed 20" of vacuum, which completely closes the wastegates for maximum boost response. Once full boost is reached, the wastegate sees 3-5" of vacuum used to regulate the wastegate opening for sustaining the desired boost. Once fuel cut is reached, the clever folks at BMW decided that boost needs to go to 0psi so as to not result in a big, uncivilized "bang bang bang" against the rev limiter. To do this, it cuts all vacuum to the wastegates and they go fully open (no boost!).

The upside to this approach, as i see it at least, is as follows:

1) During start-up, the wastegates are open and the turbos are effectively bypassed. This means that much of the hot exhaust coming out of the engine doesn't go through the turbine wheel and cool off. Instead, it goes right out the wastegate and right into the catalyst. This gets them up to temp quicker and makes the boys at the EPA happy. This also means that BMW can push the cats back downstream a little bit instead of cramming them right up against the turbo where backpressure effects are most deleterious.

2) There is no way to induce an overboost since the natural state of the wastegate is fully open. In other turbo cars, something as simple as a wastegate line rupture will cause the car to overboost.

3) Its just cool engineering. Leave it to the Germans.

The downside to this approach, as i see it at least, is as follows:

1) It requires the use of a separate vacuum tank that needs to be smuggled under the hood somewhere. This takes up a little bit of space and costs extra money. Not much but we all know how OEMs like to shave off every extra bit of material and cost.

2) It confused me for a good part of the morning. :lol:

Very interesting read. So what type of potential do you think the new TC 3 series has ???

+1

That's what we're going to figure out over the next few months. The car is certainly different with all its fancy automation and freighteningly clever logic. From what I'm hearing and have been told, we should see a lot of this stuff on the Evo X. The next several months are going to be interested :)

shiv

i like safety. any way to reverse engineer an external gate on an evo to work like this?

Shiv,


Thanks for sharing. I have been soaking all the info regarding the 335I. I have seen a few videos posted on various websites as well. The stock turbos(IHI) are set to 9PSI max but I am not certain if this car posses a overboost delay.

Would you say that a conventional MBC running a "y"/ "t" fitting is still aplicable to these turbos?

The one thing I already notice is that both turbos has precats right after the turbo however the car is pushing full torque at only 1400rpm with absolutely no lag.

Either way I see already huge margins of improvement for this turbo set up. Have you heard the exhaust note? very deep and throaty!!

Carlos

I yanked the boost gauge out of the supercharged 06 civic today and plumbed it in to the bimmer. The car makes 6psi easily before 2000rpm. It then ramps up to 9psi by 5500rpm. Then down to 6psi just before fuel cut. Then to 0psi while whacking against the rev limiter.

And nope, you won't be able to use a conventional MBC with this boost control set-up. Thank god. Two MBCs on a $50k car. Yuck. :o

-shiv

Uhmmm, guess what I was just thinking about? if BMW enginers are so smart to cut off boost by red line then my question is how did they plan on rebuilding boost back up?

In a convencional turbo car whenever you reach red line you only loose a portion of the boost and no all the boost. On the newer BMW you start to boost from 0psi again right? It is no difficult to understand that a turbo car that respool from 10 psi should respond better than a car that has to respool from 0psi...

That is just me finding the logic but I am not any genious just a guy that owned 12 turbo cars:D I already see the BMW new boost aplication as a discrepancy, well at least a disadvantage.

Carlos

Intresting read. I guess we'll have to wait and see as a little more times goes by.

This is suck!!No boost at redline:crap:.How to boost up,Shiv??:DWHat your plan with this Bimmer??:confused:

Hey guys... there is no real performance related downside to the way that the BMW controls boost. If anything, it's a more sensible, albeit more costly, approach. Whatever compromises that have been made towards safety/refinement/etc,. can be eliminated just the same as it can be with any other stock turbo car. The method to do so, however, is going to be a bit different.

Uhhhh, couple questions... how does one go about "feeding" vacuum to an actuator ? And, why would they need a vacuum canister ? Why not a simple pump ?

Hey Shiv, looks like you gotta rewire your turbo brain for this car huh :), How do you plan on raising boost levels with this setup?? The vacuum has to be controlled by the ecu correct? so ecu manipulation should be able to adjust how much vacuum and when hopefully. or you going mechanical route maybe a stiffer spring in the actuator, or a new vacuum system with an adjustable regulator? (assuming its still a spring type actuator, and its a internal wastegate)

That is crazy.. Lol

I like their approach. It is easy for people (Shiv) to remove vacuum or slow down the rate that vacuum is pulled. But it is much more difficult to add vacuum, when you don't have a source with the level of vacuum or greater than the pressure that you are looking to emulate. If there was a -20 psi vac that was there all the time you could just regulate the waste gate off of that. Hence the tank.