Quote:
Originally Posted by spooled240
..except we are and that's why the OP is experiencing boost creep? Just Google "boost creep backpressure" and you'll find numerous articles about why reducing exhaust backpressure (like your downpipe solution) leads to increased flow through the turbine wheel (mind you a larger one like a 2871r) resulting in a boost creep situation. As the RPM's rise there's additional exhaust flowing through the turbine in extremely high flaming velocity, like a jet engine's exhaust (I say this tongue in cheek) and the stock t25 wastegate just sucks for bypassing this when there's less backpressure.
https://cobbtuning.atlassian.net/wik...GR+Boost+Creep
https://www.bimmerforums.com/forum/s...h-backpressure
You're thinking in hypotheticals. Throw a full catless turboback on a stock FD3S and watch this not happen:
"So when the downpipe volume grows larger, and the turbine isn't using any of the extra space (same wheel speed with even less pressure) all that extra volume can be used by the wastegate." |
hehe, okay I see. In real life, boost creeps because reduced downpipe pressure causes reduced elastic collisions with the back of the turbine blade. Allowing the resulting collisions with the front of the turbine blade to apply the same exact force as before but with the result of moving the turbine blade slightly farther rotation in the same amount of time. If we could vacuum the exhaust downpipe it would super speed the turbine this way. Frontal collisions are effective- but rearward collisions are ALSO very effective. Both forces must be balanced in our equations.
This is why I specifically say that I am "holding mass flow constant". In control theory I am controlling a reasonable variable. Can we control wheel speed: yes. I can control it. I can measure wheel speed with a sensor and I can adjust the boost to achieve any number, even if I have to wire open the wastegate with no spring at all I can reduce is to any very low number which can barely impact engine function.
So there is no excuse not to be able to do this. If you unplug it watch what happens. It won't barely make 3psi let alone 20 or 30 overboost. The overboost condition is because you are ALLOWING it to happen by not correctly control the wastegate fully. When I say that I am holding mass flow constant in my equations, it is a completely plausible testing condition we can easily simulate at very low boost pressures. Under this condition, holding mass flow constant, we would see larger downpipe can just as easily help a wastegate flow as it can help a turbine flow. Both can benefit. Both will benefit. Therefore, by holding mass flow constant I was able to show that there is less pressure inside the wastegate tube (any design for any gate this will be true in this conditional statement). And as the gate becomes superior in design, this will become even more and more true, because less of a pressure drop will cause more of a flow potential energy and movement of molecules. Especially by taking advantage of acoustic tuning in which higher pressure can still experience increased flow over lower pressure situations, which is amazing and why I don't want to get caught up in the real life empirical situations.
The real life model has reduced elastic collisions with the back of the turbine blade (
same boost pressure, more wheel speed, more mass flow) whereas the constant mass flow model has reduced intake manifold pressure (
less boost pressure, same wheel speed, same mass flow)
Thus we avoid all differences in gate design and efficiency, they become constant. It's constant because the wheel speed is the same and the mass flow is the same, so every single atom in the engine system is absolutely hasn't moved a single atomic diameter. We can overlay the two situations absolutely perfectly if we admit to the fact that two identical engines could never actually do this. This is called an assumption in higher mathematics, we assume some things about our universe to help simplify the model and solve applicable equations pertaining to said model. Without making the assumptions of physical 'perfection' we would never be able to solve anything without collecting real world data and specifically restricting ourselves to direct testing with guess what: faulty sensor data. This is called empirical testing, it allows us to adjust for optimal control system and running output behaviors by manually and physically adjusting the inputs and watching output behavior. This is how our human brains use wideband data to tune the old school stand-alones before they had learning maps and auto-tuning features. You made changes to the VE Map and watch the wideband. I tuned so many engines and got so good at doing this I still use this method to this day to tune all stand alone computers< in conjunction with the auto learn feature but not relying on it. But I digress- empirical testing is alot of work and we can avoid it by making assumptions.
And this can be used very effectively and in fact is done to every modern engine configuration I am sure. We assume that two engines could ingest identical air molecule atoms all in identical positions and neglect their negligible differences so that there was no change in any of the flow or atomic positions of any air/gas molecules. Even the oil on the bearing is in the same exact position because that turbine hasn't rotated a single sliver of any infinitely small section, there is no difference in the friction of any gas or liquid molecules in our control volume to worry about, it simplifies the math. It's the only way you'll find me examining this situation because I am wayyyy far to lazy to look at the real thing. That is how I am saying wastegate efficiency is constant and getting away with it. It's the
only way to say that- otherwise I wouldn't bother. It's why this argument is worth making and why I posted the equation above showing that we can only solve for the force on the turbine blade when mass flow is constant because matter cannot be created or destroyed, whatever goes in has to equal what is coming out, that is the point all those variables- notice there are NO NUMBERS In the equation at all even density isn't assumed to be some number its a classic "row". That is irony, its why the formula is funny for this situation. You need all the actual numbers to solve the real thing and if you actually had real data it would be all over the place within some margin of error. Its really a meme now because in real life nobody is measuring and controlling their wheel speeds so nobody is setting up mass flow as a constant in their experiment to determine whether a larger downpipe really can improve wastegate function, they fail to measure wheel speed or acknowledge that any scientist could measure their wheel speed and control it to some previous number. News flash: If the gate was performing to X capacity with the old downpipe, then we install a new downpipe, is our variable X going to get better or worse? Will the new larger downpipe make wastegate function decrease, yes or no? That this is all you need to grasp to settle this argument in your brain, and understand the picture. I don't care what happens with the turbine- I am setting it to a constant by adjusting the wastegate to maintain the same exact turbine RPM as before, it became a convenient constant so we can look at the wastegate tube and see whether the larger downpipe is helping the gate or not. The answer is yes, a larger downpipe helps with a recirculated gate. It is a recommended additional gate volume when holding turbine wheel speed constant. Otherwise exhaust gas can be ejected to atmosphere or even some vacuum system if you like for even better performance (better boost control).
summary
Reduced pressure and temperature is molecules striking the back of the blades less (elastic collisions), so we adjusted the intake pressure to compensate and bring wheel speed back down in order to hold mass flow constant. To bring down pressure on both sides so that the forces net balance in both comparisons is the same exact number, it turns from a variable into a constant.
note*
There are some smaller forces we are neglecting to consider, some originally negligible and still negligible yet I think at that point I had better remind everyone of their existence and explain why they are negligible.
In real life if we measure wheel speed and set it to a constant, and ran both experiments, the true massflow would rarely if ever be constant. It would be very very very close, but not identical. In reality there is no such thing as a constant because even protons have a half-life. For example when we set the two experimental engines up and read their data, wheel speeds may match but they will never match perfectly for every sample RPM interval. Always some variation as the RPM number climbs as the control system(gate) attempts to hold wheel speed to some exact number. I did say these are negligible. Next, acoustics influences mass flow rate, so as the engine rpm is increasing and at specific intervals related to the timing of valve events and wheel acceleration and temp/pressure there will be acoustic affects to consider. For all of my examples I am neglecting these forces but in real applications they are highly sought to control for their potential to improve efficiency of anything, battery life, torque output, stability
Elastic collisions with atoms is one set of forces, and energy contained within sound/resonance tuning has its own set of pressure and force additional or subtractive forces as well spread out over frequency ranges of engine operating circumstance, which depend on header merge and collector design, as well of course as the wastegate design. And these will also vary from experiment to experiment. That is why we want them to become constants. We can neglect any differences in their design down to the single atom of identical twins which are identical except for one single atom.