Fuel Tank Pressure By  James Goss        Have you ever wondered just how much pressure your muffler returns to the fuel tank while the engine is running?    I have thought about this subject many times in the past, but could not find much information about it.  What is happening  inside the fuel tank when the plane is inverted for a long period of time, like it is in an  inverted flat spin, that could cause the engine to loose rpm's?  For short duration inverted flight,  the engine will respond with plenty of thrust and acceleration. During the inverted flat spin my engine, an OS .46 FX, would start reducing its rpm; the longer it was inverted, the slower it would get. I tried several engines with the same results.  So what changes about the engine or fuel supply when the plane flips over to inverted?      The other day at work I was calibrating a bubble detector. This is a device used in industry to determine the amount of liquid in a storage tank. Some electronic students would be using it later on that day to take  instrumentation measurements. The device works by applying pressure through a small tube at the base of the storage tank and watching for bubbles to start rising in a steady stream. The system works because it is known that it takes 27.7 inches of water to equal 1 psi  of pressure.  The weight of a column of liquid is referred to as the head. That is to say that if the water in the tank was 27.7 inches and the input pressure was .9 psi, there would be no bubbles until the pressure reached 1 psi. At this time the pressure is greater than the weight of the liquid and will allow bubbles to flow up through the liquid.   An example of this would be to have a water tank  100 feet high and is full of water. The pressure exerted at the base of the tank is equal to 100 feet multiplied by 12, to convert into inches. This is 1200 inches.  Now divide this by 27.7 to get psi. This is 43.32 pounds per square inch. It is interesting to note that the width or diameter of the tank does not matter. It could be 10 ft wide or it could be 100 ft wide, the pressure would still be 43.32 psi.  The 43.32 is the weight for a 1 inch square column of water 100 feet high.  So remember that 27.7 inches of water column is equal to 1 psi.      The atmosphere that we fly our planes in also has weight.  At sea level it is 14.7 psi.  This weight comes from all the various gases the atmosphere is made up of.  If you had a 1 inch square column of these gases that reached from earth to the top of the atmosphere, it would weight 14.7 lbs.  When you measure pressure, this 14.7 psi is normally not thought about and the pressure measurement is said to be gauge pressure.  If you include the 14.7 psi in your measurement, it is said to be absolute pressure.  So if you measure 1 psi  gauge pressure at sea level, you really have 15.7 psi  absolute.  About 50 % of the atmospheric mass will be found between sea level and 3.5 miles up.  If the pressure distribution is linear it would calculate to about .000397 psi per foot of altitude. If you fly at 500 feet the pressure change on your fuel tank would be from 14.7 at sea level to 14.50 at 500 ft. A change of .2 psi.  This figure would vary depending on your location in reference to sea level.  The higher you are above sea level determines how much the 14.7 psi is reduced. If you were below sea level the 14.7 psi would be increased. For example, if you were flying your airplane in Death Valley California, on the ground it would be at - 282 feet below sea level. If you were flying at 282 feet in the air, you would be flying at sea level.   This pressure change could be one of the factors why our engines sometimes run good on the ground and not so good in the air. Now .2 psi does not sound like very much change, but as I have found through some experimenting, it is.      While setting up the bubble detector I thought about how the height of the fuel in our fuel tank would affect the tank pressure while being inverted. When the plane is upright everything is ok because the fuel is below the pressure input to the tank. While inverted the fuel is above the pressure input, this will tend to reduce the pressure in the tank. Just like the bubble detector, the input pressure will have to be greater than the head of the fuel before the tank can receive pressure from the muffler. In other words the pressure coming in must pick up the weight of the fuel before back flow from the muffler can occur.  If water is the liquid being used, based on 27.7inches equals 1 psi, it would be .036 psi per inch of water. I think alcohol has a density of about .8 that of water so it would be about .028 psi per inch of fuel. In our small 8 oz tanks the vertical head is about 2 inches. This equates to .056 psi that the muffler pressure would have to over come before gasses from the muffler could flow into the tank. I have noticed that when the tank is about empty my engine does run better inverted.      Having wondered about this for some time now, I decided to experiment and find out, once and for all, how the pressure varies in my fuel tanks. As a rule, I normally remove the baffle from the muffler to get a little more rpm's. I just figured the baffle was only used to make the engine generate less noise.  After the following experiment I got to thinking that maybe I need to replace all those baffles. Here is how the experiment went: OS .46 FX      Without Baffle      Standard Muffler  # 873      12.25  x  3.75  prop      15 % fuel      I connected a manometer in the muffler return line by using a tee connector. A manometer is used when measuring very low pressure.  The U- tube manometer I used was calibrated in inches.  I also used one calibrated in millimeters and got the same results. Remember one inch of water movement in the U tube manometer is equal to a pressure of .036 psi.  With the engine at an idle of 2,220 rpm's, the pressure was only .4 inches of water.  This converts to .0144 psi of muffler pressure in the tank. If you were inverted at this time, the head would have to be subtracted from this figure which would really lower tank pressure down to almost nothing.  May even have negative pressure on the tank at this time and fuel might flow out of the tank and into the muffler. To have negative pressure on the tank would mean that the weight of the fuel is creating a head that is greater than the pressure coming in.      At maximum throttle the rpm's went to 12,180. Remember I am using a 12.25 diameter prop at 3.75 pitch.  At this speed the tank pressure was .606 psi.  This is a good positive pressure but still much lower than I thought it would be.  If you return the engine to an idle, the pressure will drop back to almost zero with no delay. The tank will not hold any reserve pressure because it is connected to atmosphere pressure through the muffler. So if you are doing an aerobatic maneuver and reduce the throttle, you will loose most of the tank pressure and the engine may hesitate when you want it to pick back up.      The same test with the baffle installed showed an increase in tank pressure.  At an idle of 2220 rpm's, the pressure was .028 psi.  This is double that without the baffle.  At 12,180 rpm's the tank pressure was .82 psi. It doesn't take much here to make a difference. This is about a 20 % increase in pressure. This alone should improve my rpm problem. With the baffle installed the engine could reach almost the same rpm range as that without the baffle.  So I don't think I will remove anymore baffles from my mufflers.      While running this experiment I also got to thinking about what if the fuel tank had more pressure. I know some manufactures like YS uses 4 to 6 psi on their tanks, but they have a fuel regulator ahead of the carburetor.  I tried a check valve in line with the muffler feedback line and it seems to really work well.  Just in case someone does not know what a check valve is, it will let liquid or gasses pass in one direction but not in the other.  For those into electronics a check valve is like a diode which passes current in only one direction. Any time the pressure in the muffler is greater than the pressure in the tank, the check valve will open and let gasses flow into the tank from the muffler. Without the check valve in line with the pressure line, each time the piston goes up a pulse of pressure is forced into the tank. When the piston goes down, some of this pressure is removed from the tank.  Also during the down stroke of the piston, it pushes fuel from the crankcase into the ports to get fuel on top of the piston so the engine can fire on the next cycle.  The pressure on the tank varies with the incoming and outgoing pulses from the muffler, and what is left in the tank is the average of these pulses.  The check valve allows only incoming pulses and has no out going, so the pressure adds and is regulated quite well.      With a check valve installed the tank pressure went up to 1.5 psi at max rpm's. When you crank the engine and give it max throttle for a few seconds, the tank will build up maximum pressure. You can now idle the engine but the tank pressure will stay max. This was something I did not expect to happen. I thought it would bleed down faster, but it maintains a high working pressure of 1.5 psi.      When you first start your engine with the check valve installed, it will not run good at all.  You will need to adjust both the high and low end for leaner operation than before.  Apparently the engine must not be using as much fuel with the check valve as it does without it. It probably gets a better spray from the spray bar in the carburetor when using higher pressure.  This means you will have less unburned fuel pass through your engine. You will get a lower idle rpm and great acceleration all the way through mid range.        There is only one disadvantage I see at this time when using a check valve. If your engine stops running before you cut it off, the tank will still be pressurized so some fuel will still pass on into the carburetor.  This condition may allow fuel to over flow out the carburetor venturi. If the engine is still running when it is time to stop it, just remove the vent tube from the check valve first to bleed off the tank pressure. Remember that the fuel tank will hold this pressure for a long time.  As long as these check valves have been on the market you would think we would read more on this subject.  Tower has these valves under the fourmost brand.  # FORQ2050 for about   $ 5.  I plan on giving it an actual flight test next week. If anyone has tried the check valve  please let me know.     Thanks!    James Goss      Update… It has been about three months since I wrote the above article on fuel tank pressure and I now have the outcome of that project. The check valve has been installed on several engines with the same results in each case, improvement in the overall operation.  A more reliable idle and better acceleration is common. With the reasonable price and quick installation time, about one minute, why not use one. If a cowling surrounds your engine, you will need to insert a plastic tee on the tank side of the valve and then plug its take-off line. This line will be used when filling the fuel tank. You will unplug this line so air can escape the tank as fuel goes in. Also when you kill your engine removing the plug can exhaust the pressure in the tank.  Good luck!      The End

Back