A small part of the supercharger workings that its unrealistic

which the GM does

 

Some of those things are new and i find logic in it, others dont make much sense.

"This is because whereas a roots-style blower is an air pump, the centrifugal type is a glorified fan"

Aren't both technically both air pumps and air fans? How does that means that the supercharger pushes air with a different pressure compared to the root? I mean, if this explains why the S/C wont be able to provide positive power at lower RPM, its a thing about physics, so what's exactly the difference in the way they both handle air?

"if you hand-spun a centrifugal supercharger it wouldn't compress anything because you're not spinning it fast enough. But at higher RPMs, the turbine incrementally begins to produce boost"

But wouldn't that be solved by changing the dimension of the gears that move the pulley of the supercharger? Why bother with that kind of things when you can have a turbine that is moved at lower torque but has better mid end and low end RPM?

 

They both function as pumps, but only one is a "fan". It's not really a fan, but it's my oversimplification to explain it. Here's the difference, or why they are not both air pumps and air fans. With a positive displacement supercharger, the air drawn in cannot escape - not matter the pressure, since the screws do not allow air back out. It grabs a predetermined amount of air, compresses it, and pumps it into the engine. However, with a centrifugal supercharger, it's alike a fan (hence my use of the oversimplification) because air can escape backwards between the blades; so you need to be moving large quantities to overcome the pressure of the intake manifold and start actually compressing the air.

Try placing a computer fan against a surface, blowing towards the surface. It can't exert very much pressure on the surface because the pressure just immediately flows backwards out though the blades. But if you could keep jacking up the RPM of this little computer fan, it'd eventually move enough air to overcome its own losses and start exerting pressure on the surface.

At low RPM, a centrifugal supercharger loses almost all of the pressure it generates because it just escapes or cannot be compressed because the internal manifold pressure it too high. It must generate a significant amount of boost before it can break even and start generating meaningful pressure to overcome the huge loses and really power the engine. And, like I said, large amounts of boost can only occur at high RPM.

At low RPM for a roots positive displacement supercharger, it loses basically nothing to backpressure, so the small amount (but consistent ratio) it produces at low RPM can be consumed by the engine, since the pressure simply cannot escape. The backpressure exists, yes, and this is what blows superchargers a dozen feet into the air on drag strips, but it's a sealed unit that has no choice but to be pushed into the engine.

It only scales the effect. You could absolutely do exactly what you said and indeed the supercharger would produce proper posivitive-displacement-esque boost at low RPM. However, you'd be overboosting like crazy at mid- and high-rpm situations. If a roots style supercharger produces 2psi of boost at 2000rpm, then a tuned centrifugal charger as you have described could produce 2psi at 2000rpm, but would make 8psi at 4000rpm and 32psi at 8000rpm. With a roots blower, it'd only be 4psi at 4000rpm and 8psi at 8000rpm. Thirty-two psi is enough to blow most engines apart, but at the low end, 2psi is very modest (in other words, low) amounts of boost for a proper forced induction setup.

Though, as I said before, blow-off valves CAN be used on centrifugal superchargers, since they're so much like turbos - they may need to release the pressure they're still generating when the engine slows down suddenly whereas a positive displacement supercharger always feeds the engine as its optimal rate.

I see what you're thinking; it seems the roots blower is better in every way, so why does the centrifugal type even exist? Well, they can be much easier to fit in the engine bay, they produce more peak boost, they are more efficient in terms of boost/heat output, and they have less parasitic losses. If your car already makes good torque down low, but you need more power up top, then centrifugal superchargers are the right choice. This is why many Crown Vic owners use them, since the hefty V8 could haul at low RPM but for better acceleration the top end can be beefed up with one of these. I suspect the dev team implemented them on the GM because of this trend.

Please, point out any inaccuracies, I'd like to rectify them.

Why is this funny? That's to be expected, this discovery does not disprove anything here and is to be expected with the millions of centrifugal setups/configurations out there in the world.

 

in game duh

 

"because air can escape backwards between the blades; so you need to be moving large quantities to overcome the pressure of the intake manifold and start actually compressing the air."

What? Of course air escapes between the blade, that's what happens with pressure. The movement of the twin screws and the turbines generates places of higher and lower pressure which forces the air a way, that's why it forces it. But yes, air does *escapes through the blades*. Because if not, you would generate an actual vaccum. Same thing happens with roots, its not like there is no way for air to go one way no matter what. Unless you meant generally that its easier for a turbine to have a different air flow, which is something i agree with.

"It can't exert very much pressure on the surface because the pressure just immediately flows backwards out though the blades"

eh... No. The reason why it cant do a lot of pressure its because of a lot of things, material of the blades, RPM of the blades, angle of the blades, design of the blades, but most importantly, the area around it. if you pick up your typical fan and place it in a tube that closes as close as possible to the side of the blades themselves, the pressure itself would increase by a lot. Angle itself also has a lot to deal with that.

"But if you could keep jacking up the RPM of this little computer fan, it'd eventually move enough air to overcome its own losses and start exerting pressure on the surface."

Yeah but its not the same deal with planes and mach speed. Its not like "you get to a certain RPM now its easier", increasing the RPM itself does increases pressure and the wind it moves, but not because it managed to do something easier, but rather just because you increased the RPM.

"At low RPM, a centrifugal supercharger loses almost all of the pressure it generates because it just escapes"

But wouldn't the same be happening with roots too?

"or cannot be compressed because the internal manifold pressure it too high"

That would not be something that blocks the supercharger from having a lot of torque, it would break the manifold tho. Then again, having a custom manifold when you make a car with a turbo bolt on its a big part of it.

" You could absolutely do exactly what you said and indeed the supercharger would produce proper posivitive-displacement-esque boost at low RPM. However, you'd be overboosting like crazy at mid- and high-rpm situations"

..... Which is what the guy who wants more power wants? i mean, sure, he could not have the money to have the car ready to take a shit ton of boost but wouldn't changing the gear ratios of the pulley and the S/C pulley AND making the S/C smaller increase low/mid end torque while also limiting high end torque? remember that air pressure that enters the car is also limited by the size of the place it goes. If you reduce the size of the turbine and pipes you are also restricting the amount of air.

"I see what you're thinking; it seems the roots blower is better in every way, so why does the centrifugal type even exist?"

Because we just talked about it moving a lot more air due to their design? i mean, i think its the design, if it wasn't the turbine design then the only think that i think that could be making the centrifugal have worse low end torque is just the pulley ratio and that's it.

"Well, they can be much easier to fit in the engine bay"

eh.... i would like to agree but on V-engines superchargers aren't that hard to fit either. Since all kinds of forced inductions require similar preparations (new intake manifold, new valves, bigger headgasket to reduce compression, etc.) A centrifugal could be just a think of attaching, placing the crank supercharger pulley, placing everything together and be done with it, but roots are also quite easy, just bolt it on top of the manifold (if you dont want to use a different carb, its a great idea to keep it UNDER the carb, not on top of it) and that's it.

"they are more efficient in terms of boost/heat output"

Yes. That's because you COULD use an intercooler with the centrifugal since its position its the same as a turbo, right? while the closer the end of the root supercharger was to the cylinder head itself the better, so people didn't bother to make a way to fit an intercooler in it.

"Why is this funny? That's to be expected, this discovery does not disprove anything here and is to be expected with the millions of centrifugal setups/configurations out there in the world."

That's because 6677 said that they work exactly like that, with low-low and mid end RPM torque.



So, considering that it is possible that a turbine design could be able to move more air in the intake than the twin screw design, which could be part of why turbos of the same psi and similar size could give more HP than superchargers apart from their parasitic power (otherwise, i am SURE someone would make a root supercharger driven by exhaust gases, and if no one did, i would love to make one), the ONLY thing that comes to mind as the real reason why they generate less parasitic power, generate less power on low and mid end RPM, and maybe also why they have different amount of boosts per RPM, its because of the pulley ratio.

Either that, there's something else that i am not getting, or a lot of people were wrong and the centrifugals give just as much boost as roots (which at this point i consider rather not very likely)
--- Post updated ---
Apparently we've made a small mistake. Root and twin screws are different superchargers.

Also, could it be that you can pick up from between having a different gear ratio on the centrifugal to have different kind of powerbands? Maybe if you adjust the turbine to run easily at high speeds it can take more air at low and mid end RPM but it will kinda choke on high end due to the turbine fighting the incoming air.

 

No, precisely not the case. With centrifugal superchargers, pressure can actively escape the high pressure region (manifold) and the low pressure region (atmosphere) - this does not happen with twin-screw/positive displacement superchargers unless you run them backwards, as they are forming sealed air pockets. To reiterate, manifold pressure CAN escape with centrifugal superchargers because of this thing, as I said before, the high pressure zone can escape to the low pressure zone through the blades, should they not be producing enough airflow to overcome this loss. This is NOT the case with twin screw/positive displacement.

Of course those factors influence it; this is just an example to show how an artificially high, high pressure zone (a solid surface) causes the air to simply fail to compress, escaping backwards instead. A twin screw supercharger would not have this problem - air from the high pressure zone cannot escape, so it would compress the air and build pressure until, well, something breaks and the pressure escapes or it becomes to difficult to crank it.

Not sure what you're saying here. I never said it'd be "easier", but there is a threshold a centrifugal supercharger has to overcome (manifold pressure) before mad boost happens. I think I agree with what you're saying though.

Definitive no, as I said above. A twin screw/positive displacement supercharger (or roots, you've pointed out there's a difference) has no boost loss in the circumstance that manifold pressure > atmospheric pressure (whereas centris do, until they generate pressure greater than manifold), and thus could produce the perfect amount of boost for a theoretical engine running at just a few RPM. In my first post I explained why.

I guess, you just have to remember you're fighting physics here, as centrifugal superchargers produce exponentially more boost at higher RPM.

Maybe my own opinion go in there, I just think linear boost is so much better than exponential boost.

Have you ever seen under the hood of a crown vic? Not much vertical room for a medium-sized blower to sit on an aftermarket plenum. But it's a long car, there's plenty of room in front of the engine. That said, you could fit a house under an F-150's hood.

I think he was generalizing, and he's not wrong about his generalization.

Hmmm... twin screws driven by exhaust... I'm in, though I imagine it'll be like a very responsive turbo. I wonder if it's possible in BeamNG, even if it involves "cheating" physics.

Oh really? I guess I need to research that better.

And I TOTALLY agree with the last idea. Different pulley ratios would be a hell of a lot of fun, might be worth suggesting in a thread.

 

Since reading this thread I've decided to play around with the Grand Marshal's engine superchargers (aka all that centrifugal commotion).
So, is there really any use for this on the car because it only kicks in the last 500 RPM and gives a short speed boost/wheelspin, but you lose all that when shifting.
What advantage does having this type of supercharger have besides size and a short burst of excitement? It doesn't seem to help my laptimes sitting idle there 3/4 the time.

 

Well, i got proven wrong (after i posted that i checked up some more) And found out that you are indeed right, turbine-style fans DO have air escaping and require a shit ton of RPM before being able to provide boost.

And in case you didn't knew, and to save up the trouble for you to look it up:

Root supercharger: Works like oil pumps, there are 2 ovals with thin middle sections (basically 2 oval gears made in such way that if you rotate one the other one also rotates, or anything in the like, like a triangle, works) and one of them is connected with the crank pulley. Due to the movement of the crank, to the root supercharger, and from one of the pieces to the other, there is a "suction" effect happening at the intake of the supercharger, forcing way in the same way oil gets forced. However, twin screws work very similarly, but they not only draw air through vacuums but also compress it before sending it, which means that twin screws achieve higher power. Since my mental image of how a twin screws internal works, i would like to see the effects of having different patterns in it.

And funny enough, the only thing that i can get from the fact that the centrifugal gets so much power its because its just an oversized supercharger, its not much to do with it being a turbine itself, since its yet to be confirmed that high speed turbines can draw more air than a twin screw (not counting on the parasite power drag)

But even though that's all said and done, my doubt is still there.

Cant you modify the pulley ratios to make the turbofan spin even faster at lower RPM? this would increase overall high power and if it gets too much to be realistically possible you could just make a shorter supercharger turbine to make it draw overall less mass of air.

And even if you couldn't, IMHO, the supercharger for the grand marshall has way too little power on low and mid end. Like Instant Winrar said, having power at the last 500 or 700 rpm is not a good thing. Even race car drivers with their own modified short gear ratios transmissions lose around 2k RPM when shifting up, and the difference in power makes the grand marshall too unstable.


Although this entire thread could work as an example to why it would not be a terrible idea to have 2 or the 3 types of superchargers in the game (per car), and them acting differently.

 

RE: Separate carb and blower swaps - the actual reason for that is that there can be only one horsepower-affecting parts subslot, the "tuning level". 2bbl carb is a tuning level, 4bbl carb is a tuning level, 4bbl + supercharger is a tuning level. So if you wanted to have the various different types of forced induction represented separately from the different types of fuel systems, you'd need a combinatorial explosion of different-but-equal tuning levels. While I know the original point was about different setups within a fuel system type, such as 2bbl and 4bbl carbs, that would probably be better off rolled into the tuning level system since the fuel system tends to be upgraded along with the rest of the engine anyway; when a low-spec V8 is being hopped up, for example, the 2bbl carb is usually one of the first things to go.

So, for example, the first stage in the "NA carbed" tuning path for a 2bbl V8 Moonhawk could include a low-end 4bbl with some performance adjustments, an intake manifold to go with, and a glasspack in place of the muffler. The second stage might build on that with further enhancements to the same carburetor and perhaps a mild cam. On and on it goes.

In my head I've been thinking something along the lines of 15 stages for each path, done roughly as in Tokyo Xtreme Racer 3, with stage 9 (the highest in that game barring engine swaps) giving about 670-690 HP for a car that had about 300 from the factory; stages 10-15 would be where things got really crazy, with 1500+ HP not out of the question for the biggest engines.

 

I get it. Basically you think that it could be way too complex? i dont know, not that i dont think that its easy, but ive seen so many games do engine tuning, hell, Street Legal Racing Redline, an hungary indie studio in the early 2000's doing one of the most accurate representations of engine tuning even to this day.

having a dynamic physics-based torque output might be too much of a hassle, considering that the devs have to input the torque by hand, but it would allow for things like more than 1 type of power-affecting tuning and it would just be nice.

 

I'm not sure where you got that I said it would be too complex. It is an interesting point to consider; other games have indeed done much more complex engine tuning systems, but and the way this game handles power and torque might indeed make it rather complicated to have more than one tuning slot on a car (I'm absolutely a non-coder, so I wouldn't know personally; I'm just taking shots in the dark here). All I meant, however, was that the game as it is now is set up to use a fully stage-based tuning system. There's nothing particularly wrong with that, either.

In any case, I suspect that trying to replicate all the different fuel systems and forced induction options within such a stage-based system wouldn't be so much too complex as too tedious and labor-intensive. Multiple stages each for NA carbed, NA MFI, NA EFI, single turbo carbed, single turbo MFI, etc. could require shocking quantities of time and effort depending on whether you want all engines to be tunable in this manner, whether you want detailed models for each stage, how anal you want to be about making the power curves all nice and theoretically perfect, what kind of tools you have access to to construct those power curves, etc.

 

Well, no, there isn't really anything wrong with a linear way to make tuning, but it makes too obvious what's better for you

Youve got an 2.5 SBR4 engine N/A vs a 2.0 with a turbo, there is no doubt that the turbo would win, it just annoys me that (being a mechanic) the game decides to take such a simple route.

I am not that anal about power curves being 1:1, but i would love to see a difference between ways to set up the car, say, monopoint injection, multiport, direct, carbs, forced induction, higher compression, higher RPM cylinder heads, exhaust that allows a better airflow, etc.

 

Okay, here's an idea of mine. I don't know enough about these centrifugal superchargers to either know if this is already done or if this is even possible, but take a listen:

So we take your idea of a crazy ridiculous pulley ratio that spins the turbine at a high enough speed at low RPM to generate some meaningful boost. Then, we use a bleed valve to let off all the pressure that the s/c produces above a certain boost threshold (pick a number, basically) for higher RPM. Since the excess pressure is being let off, it won't produce any more resistance than would be made producing the threshold amount of boost, so it's a win-win.

Like:

1000rpm = 2psi
2000rpm = 8psi
4000rpm = 10psi bleed open (down from 32psi)
8000rpm = 10psi bleed open (down from 128psi)

 

reason for not overdriving a supercharger that much is that the parasitic losses become simply insane.

 

Then have like randomshort guy said, but modify the blow off valve to just open to a certain degree, so that it keeps working, but the more air it moves, the more power the engine will take, just on a reduced note compared to no blow off valve.

 

it would be the loss in driving the compressor at all, not the compression efficiency but the actual work done to spin the compressor itself.

 

But it would be no different to having any other kind of supercharger. If you mean because of the ratio, ideally there would be a different ratio to allow the turbine to spin faster, with less torque, which would allow to provide better power at low and mid end RPM, while maybe not having enough torque to force air at high speeds.