Multiple throttle bodies = slight pumping loss reduction?

[serialk11r]

Hey guys, when I was thinking about how a positive displacement pump would work as an "air motor", I realized that an important part of the equation is intake plenum volume and acoustics...If the intake plenum volume is very small compared to the displacement of the engine, then most of the losses for an underdriven Roots blower are eliminated as the piston more or less directly does work on the rotors.

Then I realized that ITB setups have an extremely small distance from the throttle to the cylinder...and thus a very low volume between the throttle body and the cylinder! Moreover, since you're flowing a smaller net quantity of air through each throttle the throttle plate should be closed more. Then each piston will draw a slightly deeper vacuum during the intake stroke, with less adiabatic free expansion going on at the throttle body. After the intake valve closes the air pressure after the throttle will start to equalize with the air before the throttle, but since the volume of low pressure air is smaller, the free expansion loss should be lower.

Of course, getting ITBs solely for this purpose would probably be stupid since I can't imagine the difference being very big, but I wonder if anyone's bothered to document this...

[drmiller100]

you are missing the point of it all.

you are doing work over time which is power loss. Somewhere there is a restriction. Somewhere there is a vacuum. The SIZE of the vacuum area is meaningless.

The number of TB's is meaningless.

[t vago]

Originally Posted by serialk11r

Then I realized that ITB setups have an extremely small distance from the throttle to the cylinder...and thus a very low volume between the throttle body and the cylinder!

This is correct. The amount of work needed to draw a vacuum is proportional to the product of the intake manifold volume multiplied by the vacuum.

Originally Posted by serialk11r

Moreover, since you're flowing a smaller net quantity of air through each throttle the throttle plate should be closed more. Then each piston will draw a slightly deeper vacuum during the intake stroke, with less adiabatic free expansion going on at the throttle body.

This may well be, but increased vacuum is not good for freeing up mechanical energy. Refer to the below diagram.



This is a representation of the work generated by an ideal gasoline engine with a throttle. The white enclosed area represents the total mechanical work produced by the combustion of gasoline. The grey area represents the pumping work that the engine must perform just so it can run. Essentially, you're wanting to decrease the grey area in this curve.

Lowering vacuum will cause the P(man) value to rise, causing the grey area to become smaller. Conversely, raising vacuum will lower P(man) and cause the grey area to expand.

Originally Posted by serialk11r

After the intake valve closes the air pressure after the throttle will start to equalize with the air before the throttle, but since the volume of low pressure air is smaller, the free expansion loss should be lower.

May well be. It depends on if the increased vacuum developed offsets the initial air pressure as the intake valve opens.

Originally Posted by serialk11r

Of course, getting ITBs solely for this purpose would probably be stupid since I can't imagine the difference being very big, but I wonder if anyone's bothered to document this...

Might be worth a shot. I, for one, am certainly not going to try it for this purpose. I'd need 8 throttle bodies!

[TheEnemy]

The improvement I see would be more in the ability to better match the air flow between each cylender.

[serialk11r]

Originally Posted by t vago

May well be. It depends on if the increased vacuum developed offsets the initial air pressure as the intake valve opens.

Yes, this is why I speculate that any sort of possible improvement would be very small if existent at all.

drmiller, the idea is that the cylinders do not draw air in a constant stream, but a large plenum plus single throttle tends to average out the pulses.

If we're going by the pumping diagram, the difference is that normally, when the intake valves open the piston is immediately exposed to manifold vacuum, whereas in this situation the cylinder is at atmospheric pressure. As the piston draws air in, the pressure will fall more slowly because it starts out closer to atmospheric, but then you end up with deeper vacuum. So what it does is sort of chop a "triangle" off the bottom left of the pumping loop, but extends its height downward. It gets a bit more complicated as if the temperature changes then the compression curve looks a bit different as well, so it's rather hard to say, but I have a hunch it might have a noticable effect.

I think it would be cool to see data from someone who has swapped ITBs into their car.

[t vago]

Originally Posted by serialk11r

Yes, this is why I speculate that any sort of possible improvement would be very small if existent at all.

I gave this idea some more thought on my commute home, and have come to a different conclusion.

Recall that the whole point of creating a vacuum is so that air can be effectively metered into the cylinder to vary the amount of generated mechanical energy (which is then used for various things like pumping work and driving engine accessories and overcoming friction and ultimately pushing the car forward). Therefore, for a given power output, the final in-cylinder vacuum must be the same when the intake valve closes, regardless of throttle system being used.

Normal single-throttle systems for gasoline engines use a relatively huge volume, compared to the volume of a single cylinder. This is to provide simple control, and dampens out variations in vacuum created by each cylinder on its intake stroke. However, it takes away some of the mechanical energy from the engine.

For ITB systems, to generate the same amount of mechanical energy per cylinder as the single throttle system, again the final in-cylinder vacuum must remain the same. So, I don't think that the vacuum is necessarily higher with an ITB system, than with a single throttle system. There will still be pumping work associated with an ITB system.

However, at the beginning of the intake stroke for an ITB system, the intake valve opens against a pressure much nearer to atmospheric than with the single throttle. This will make the pumping loop look more like a triangle, as you have hinted at.

Potentially, an ITB system should see significant gains in fuel economy, compared to a normal single throttle system.

Originally Posted by serialk11r

I think it would be cool to see data from someone who has swapped ITBs into their car.

Yes, it would...

[jakobnev]

Let me draw it out so everyone can see:



I've been pondering on a similar idea, but with a Helmholtz resonator for each cylinder. (Allowing the manifold pressure to even exceed atmospheric at the beginning of the intake stroke)

[BHarvey]

I would think less throttle bodies, or carbs, would be better for the simple fact of lessening the "start/stop" nature of the air in the intake stream due to the valves opening and closing.

IE- If the air is constantly moving and has velocity retention, is has less loss...

My guess anyway.

[t vago]

BMW seemed to think well of this idea, too.

The BMW S65 series V8 engine, with 8 ITBs! - BMW S65 - Wikipedia, the free encyclopedia It's a production engine, and has been going into cars since 2008.

[EdKiefer]

If you try an add very short runners with each cylinder having TB, this tends to opimize high rpm usage . If for example to modified a vehicle that has single TB with one plenum and long runners you will lose a lot of low end torque and throttle will be extremely sensitive . only way around will be extremely small TB .

If you look at that BMW engine ain link above its optimized for high rpm use, look at were peak HP is , this is a short stroke engine . So the engine needs to be engineered as a whole not as parts .