A quick correction.
The engine doesn't care much about pressure or volume, what it needs is mass. To produce a given amount of power, the engine will have to burn X grams of fuel. In order to burn that fuel, you will need about 14.7xX grams of air (for stoichometric combustion anyway). So that can be accomplished by either low pressure high volume; or high pressure low volume.
Of course all three are related (mass, pressure, volume); but in this scenario volume is actually the dependent variable and is a function of the other two.
my $0.02.
Exactly right.
There's talk in this thread about the CFM being reduced with the head backwards. This is not really relevant.
The engine requires a fixed mass of air to produce a fixed amount of power to drive at a given speed (in this case, 90mph).
When the snorkel head is facing backards, the pressure in the snorkel tube is reduced. Therefore, the density is decreased. Yet, the engine needs the same mass of air. Therefore, the only way to achieve that mass is to increase the VOLUME of air. It follows that the simply measured CFM of air will actually INCREASE. This is counterintuitive to many in the engine-tuning community, but this is in fact how it works. This increase in volume is achieved by opening the throttle wider. If you could measure the throttle angle, you would see this is true. You have to step on the throttle more when you have the snorkel backwards.
Normally this will have no effect on engine efficiency. A fuel injected engine always maintains 14.7:1 Air/Fuel Ratio. The only exception to this is when it goes into open-loop mode. This usually happens sometime after half-throttle. There will be a measured change in the fuel efficiency of the engine because since you have to open the throttle a little more to generate the same power, you will cross the open-loop threshold more often on a given drive.
The only time the backwards snorkel will actually affect the engine power, is when comparing scenarios when the throttle cannot be opened anymore. ie: full throttle. In this case, the backward's facing throttle will achieve a maximum horsepower which is reduced by a similar amount to that of the air pressure. In the case of 90mph, the air pressure drop is 1.7% of atmospheric, and I would expect the horsepower drop to be similar.
Thank you Tucan for doing the experiment and getting the data. Did you have the sampling tube connected before or after the filter?