Engine technology thread.

[serialk11r]

I understand you have experience with turbocharging cars, but I want to say again, I don't think the advantage over supercharging is as big as you say it is, for the reasons I mentioned already that I won't repeat. I'm not denying that it's there.

Basically I'm just saying the same thing as madfast, if your goals are modest, you won't see much of a difference, and the extra complexity of a turbo could be a bad thing.

[Dimman]

Is anyone able to get into detail about Helmholz (sp?) resonance tuning?

[serialk11r]

Helmholtz. Where would you find helmholtz resonance in the engine?

EDIT: I see on wikipedia that an airbox is supposed to exploit helmholtz resonance...I don't really see how. Could someone explain where resonance comes into play? I can do some physics explaning after that.

[Dimman]

Originally Posted by serialk11r

Helmholtz. Where would you find helmholtz resonance in the engine?

EDIT: I see on wikipedia that an airbox is supposed to exploit helmholtz resonance...I don't really see how. Could someone explain where resonance comes into play? I can do some physics explaning after that.

Maybe if you could start with the physics and we could extrapolate an application?

I think it has to do with sizing the plenum of the intake manifold.

I'm having an idea on a TVIS-style (two runners per intake port, secondaries are ECU controlled) application of an intake manifold for a boosted applicatin. Except instead of having both runners identical (which allows for inertia-based velocity gains) and sharing the same plenum, I was thinking of two plenums with ECU controlled DBW throttles and having different length runners. Idea is this would improve low-rpm response through velocity, possibly more tumble and less lag from filling a smaller plenum. Then at high rpm with both open it would flow more and allow a (weaker) wider spread of pulse options to deal with reversion etc...

Problem is where the two runners merge before the port. What happens if the throttle to the low rpm runner/plenum is open and the other closed. The intake air will be flowing (hopefully rather fast) over/past the path to the second runner/plenum that is closed. From a flow perspective it will probably cause some turbulence and act sort of like a TGV, which can help low rpm combustion/mixing.

All I've heard about Helmholtz is that it's like blowing air over a bottle to make a sound, and the above situation of the two runners but only one active sounds similar.

See sort of what I'm getting at? If it could be exploited for more mid range it would be nice.

With both open it would be pretty 'normal' but maybe the pulse tuning will be more balanced from the different runner lengths spreading out that part, but regular flow will be conventional.

[serialk11r]

So I mean it's actually not that complicated, Helmholtz resonance is just air in a closed area acting like a spring, as air is compressible. The behavior is not exactly like a spring obviously since it's more complicated.

Okay so the examples I can think of would be:
1. Blowing over a bottle makes sound, or any other wind instrument.
2. Opening your car windows at speed creates that buffing noise.
Bernoulli's principle => blowing over the opening creates slight decrease in static pressure, so the air inside the space pushes out, but the air rushing out has "inertia" so the pressure manages to drop below equilibrium and the air moves back into the space, and you get this resonance.
3. If you blow straight into a soda bottle in a quick burst, you may be able to detect sound for a short period afterwards. This is like playing a note on the flute really shortly, the sound doesn't cut cleanly off, but rather continues with diminishing amplitude.
The decay is faster than on say a stringed instrument because steel strings store much more potential energy by comparison.

Where it could possibly come into use in an air intake would be at the intake valves. The front of the car can be assumed to have approximately constant pressure. The side of the intake tract that is changing is at the valves: when the valve closes the pressure wave will hit the intake valve and "bounce back" since it would create an area of high pressure. Now we have a closed cavity with an opening to the atmosphere. While the intake valve stays closed the wave will travel to the opening of the intake and then for the same reason the wave will reflect back. If you open the intake valve just as the area of highest pressure reaches the port, the wave keeps travelling into the cylinder (where the velocity of the wave reduces and hopefully the high pressure zone stays inside the cylinder). I guess this could be called Helmholtz resonance except this "resonance" lasts only 1 period of the wave, so the use of that term is slightly unnatural.

If the engine speed and the period of the wave match then this works well, but you'd need a way to manipulate either the length of the intake or the speed of the wave to make it work. Exhaust tuning should actually work by exactly the same principles, I think Honda stuck "throttle plates" into the exhaust of a bike before to get the velocity of the exhaust gas wave so that it would reflect back and have the area with lowest pressure be at the valve when it opens. One idea I had for engines that cannot use variable intake duration (hello rotary) is to stick an "air motor" (variable geometry turbine) into the intake, as a way to reduce throttling losses. You could combine this with a second intake tract as a bypass, and it would be a more fancy variable frequency intake.

As far as your idea goes, I think that's about right. I think it would work with just one throttle plate, although if you have 2 different length runners you could get a wider range of frequencies. You would be right to say something bad happens where the 2 meet, because if you have one throttle plate closed, the other open, then the tract with the closed throttle plate now exhibits the same effect, and the pressure wave will start to travel up that tract. If the throttle is completely closed this isn't a problem, now the effective runner length is the sum of the open runner and the distance between the place they meet and the closed throttle. But if this throttle is partially open, the wave will "split" I think, and you get interference, which becomes harder to deal with. First of all having a pressure drop across the throttle plate would make your 2 separate waves end up with very different amplitude (not to mention overall decreased maximum amplitude), and it would be a much more difficult task to compute the way it behaves. Not that it can't be done, but...

Oh and I just realized, this explains why some people say certain intakes make more noise. If the intake is either tuned well (high pressure zone of wave hits intake valve on time) or very poorly (low pressure zone hits intake on time, high pressure gets reflected back), this will create noise. So my guess is the reason stock intakes don't perform well is not because of restriction, but because the manufacturer wants to tune it so the pressure is as close to atmospheric as possible and that the air flows more continuously instead of in pulses, thus reducing noise.

[Dimman]

Hmmm, all that sounds like 'normal' acoustical intake tuning, which is good as I think I have an okay grasp of it. It's about figuring out lengths of intake tract to the valve, given where a wave will change and then working with the speed of sound in the temp/pressure of your intake and the timing of your cams, plus reflection numbers, etc...

I'll probably have to diagram out something to give a better idea of where I think there may be issues.

[Dimman]

Switching back to the turbo-compounding for a moment, one of the F1 proposals is the 1.6L turbo, but also turbo-compounded.

In this case instead of the hot side of a turbo attached to reduction gears to the crank for the compounding part it looks like they may use an axial (jet engine-style) turbine spinning a generator that feeds the KERS battery and the power boost comes from the KERS motor.

Axial turbine is apparently better suited for performance motors as they rely on exhaust gas velocity (as does inertial scavenging) instead of pressure differential. However they are significantly more money than half-turbos. But we are talking about F1 here...

Other thing that I couldn't quite figure out from the article was whether or not that specific style of turbo-compounding was being proposed for F1, or simply any turbo-compounding in general.

[serialk11r]

That's not true, jet engines aren't velocity (impulse) turbines, they're more pressure turbines. The ducts in a jet engine serve to efficiently slow down the air and pressurize it, and the compressors increase the pressure. The air under pressure is then combusted and allowed to expand through the turbine, which only needs a small portion of the pressure to provide enough power for the compressor. The gas is then finally allowed to expand fully in the nozzle, and exits with high velocity to provide an impulse. sidenote: The reason afterburning is not efficient is because it attempts to raise the pressure by heating when the gas is not at its peak pressure, and at the typical speeds aircraft see, using mechanical power to compress gas is more effective than using just heat to do it, but the advantage is that afterburning is like turbocharging, it only increases the pressure behind the exhaust and not direct mechanical stress.

Axial or radial, you can optimize either more for velocity or pressure. I think at higher flow volumes a radial turbine is not so efficient anymore because the gas needs to make all these turns but I'm not sure. I'd actually think an axial might be worse for a piston engine compound turbine since you can't really separate exhaust gas pulses into 2 groups or more like in a twin scroll turbo. In F1 I'd imagine 2 or more turbine stages since they have money to do that, with a compressor to improve specific power output.

Turbine design is inherently very complex, but there are lots of people who work on this stuff, and F1 can cough up the cash to make it happen the way they want it to.

Anyways back to intake tuning, I saw a picture of someone who put some kind of resonator cavity into their intake and was talking about tangential waves, and said on whatever car they were using it increased power by a few hp, but significantly improved at relatively low rpm. I don't really understand what he was doing though.

[Slide]

Wow that shit too long to read and hurts my head :s

[serialk11r]

lolol what shit?

[Slide]

By shit I mean all this wonderful information lolz

[Dimman]

Originally Posted by serialk11r

That's not true, jet engines aren't velocity (impulse) turbines, they're more pressure turbines. The ducts in a jet engine serve to efficiently slow down the air and pressurize it, and the compressors increase the pressure. The air under pressure is then combusted and allowed to expand through the turbine, which only needs a small portion of the pressure to provide enough power for the compressor. The gas is then finally allowed to expand fully in the nozzle, and exits with high velocity to provide an impulse. sidenote: The reason afterburning is not efficient is because it attempts to raise the pressure by heating when the gas is not at its peak pressure, and at the typical speeds aircraft see, using mechanical power to compress gas is more effective than using just heat to do it, but the advantage is that afterburning is like turbocharging, it only increases the pressure behind the exhaust and not direct mechanical stress.

Axial or radial, you can optimize either more for velocity or pressure. I think at higher flow volumes a radial turbine is not so efficient anymore because the gas needs to make all these turns but I'm not sure. I'd actually think an axial might be worse for a piston engine compound turbine since you can't really separate exhaust gas pulses into 2 groups or more like in a twin scroll turbo. In F1 I'd imagine 2 or more turbine stages since they have money to do that, with a compressor to improve specific power output.

Turbine design is inherently very complex, but there are lots of people who work on this stuff, and F1 can cough up the cash to make it happen the way they want it to.

Anyways back to intake tuning, I saw a picture of someone who put some kind of resonator cavity into their intake and was talking about tangential waves, and said on whatever car they were using it increased power by a few hp, but significantly improved at relatively low rpm. I don't really understand what he was doing though.

That velocity/axial part was paraphrasing what it said in the article.

My understanding is high velocity going in a straight line becomes pressure once it runs into something like a turbine blade.

Maybe what I didn't make clear was that there is also a turbo in front of this second generator turbine. So pulsing is not an issue.

[serialk11r]

Originally Posted by Slide

By shit I mean all this wonderful information lolz

lol did you start reading from my post 2 pages back or so?

Originally Posted by Dimman

That velocity/axial part was paraphrasing what it said in the article.

My understanding is high velocity going in a straight line becomes pressure once it runs into something like a turbine blade.

Maybe what I didn't make clear was that there is also a turbo in front of this second generator turbine. So pulsing is not an issue.

Ah I see. So the wikipedia article on turbine is pretty helpful imo, in illustrating reaction vs. impulse turbine. If you have gas going at high velocity in a straight line and it hits the turbine, it gets deflected and it transfers momentum to the turbine. An axial turbine makes it easier to add stages to more effectively recover additional energy since you just stick a new row of blades behind it, whereas with a radial turbine you'd need a pipe that bends around and channels the flow into the second turbine, which would cause the gas to lose some energy most likely, which is perhaps why they'd be thinking of doing that.

If this shows up in F1 I will be SUPER excited. However in the ultimate quest for fuel efficiency, I am not sure how useful this is vs. late intake valve closure + larger displacement, which is much cheaper to do and theoretically more efficient, although the larger displacement has additional losses associated with it. One thing that will change for sure though is that spectators won't be blasted by the scream of engines that just release all the pressure straight into the air, which might be a disappointment to some people lol.

I wonder how much power they can recover though, racing engines have pretty high compression ratio...on old aircraft they were making ridiculous amounts like 500hp on a 1400hp engine since they had pathetic compression ratios like 6:1. Or wait I think the wikipedia example was a diesel...anyways pretty low compression ratios. This can theoretically bring the efficiency of the engine well past 40%, but that needs to counteract the drop in displacement...you can only boost so much.

[Dimman]

Originally Posted by serialk11r

lol did you start reading from my post 2 pages back or so?



Ah I see. So the wikipedia article on turbine is pretty helpful imo, in illustrating reaction vs. impulse turbine. If you have gas going at high velocity in a straight line and it hits the turbine, it gets deflected and it transfers momentum to the turbine. An axial turbine makes it easier to add stages to more effectively recover additional energy since you just stick a new row of blades behind it, whereas with a radial turbine you'd need a pipe that bends around and channels the flow into the second turbine, which would cause the gas to lose some energy most likely, which is perhaps why they'd be thinking of doing that.

If this shows up in F1 I will be SUPER excited. However in the ultimate quest for fuel efficiency, I am not sure how useful this is vs. late intake valve closure + larger displacement, which is much cheaper to do and theoretically more efficient, although the larger displacement has additional losses associated with it. One thing that will change for sure though is that spectators won't be blasted by the scream of engines that just release all the pressure straight into the air, which might be a disappointment to some people lol.

I wonder how much power they can recover though, racing engines have pretty high compression ratio...on old aircraft they were making ridiculous amounts like 500hp on a 1400hp engine since they had pathetic compression ratios like 6:1. Or wait I think the wikipedia example was a diesel...anyways pretty low compression ratios. This can theoretically bring the efficiency of the engine well past 40%, but that needs to counteract the drop in displacement...you can only boost so much.

The article mentions something like 7.3% power gain over same mass fuel flow in simulations which worked out to an increase of thermal efficiency to 39% from 36.95%.

Quote from article regarding F1 application:

The exact specification of the new engines is yet to be revealed but it will be a 1.6 litre turbocompound V6, and they will likely appear in 2014.