Carburetor Troubles Solved
By Corey Cassavant, EAA 675051
I had some trouble with my airplane. I think you might learn a few things from my convoluted troubleshooting and mistakes as well as maybe enjoy a chuckle or two. Last year when I converted my Rotax 503 to dual carburetors, I was having a problem with the carburetors overflowing with fuel when the engine was running. It was hard to miss. I had started my Quicksilver MXL II with a tailwind, and the fuel dribbled right down the back of my neck. The problem stopped immediately when I closed the fuel valve. Unfortunately, so did the engine.
At first, I checked all the usual stuff: float bowl level, float bowl gasket, cleaning the float seat, etc. The problem still persisted. After a minute or so of ground running, the front carb started streaming fuel from the vent line. The problem seemed to worsen with increased throttle. I had a hunch. When I converted to dual carburetors last year, the guys at Leading Edge Air Foils (LEAF) suggested I go with a larger flow rate dual-output fuel pump. I was skeptical that it was necessary, but the last thing I wanted was fuel starvation during a max power climb-out. I had bought and installed the pump back then. Now my skepticism returned.
I pulled the pulse line off the pump to stop it from pumping, but still allow it to gravity-feed fuel. The engine started and ran fine with just gravity-fed fuel. The fuel tank is well above the carbs in my airplane. Even better, there was no fuel overflow from either carb. Adequate fuel flow at high power settings was my primary concern, so I ran the engine as hard as my brakes and feet could hold it. I decided to taxi out to the runway for some full-power engine testing (Mistake #2). As I held short of the runway, the tower cleared me for my requested high-speed taxi test even though a Cessna 152 was just turning final.
I ignored the little voice in my head that said it would be better to just wait for the 152 to land and taxi behind him (Mistake #3). I accepted the clearance, lined up on the runway, and applied full power. All the temps looked good, so I wasn’t running lean. I was halfway down the runway, flying two feet above the ground at full-power level flight speed and feeling pretty good about my theory (and a little smug). Then the engine abruptly and completely quit. I gently set the airplane back down and started to time the runway lights to cut between them. I wanted to exit with enough momentum to roll well clear of the runway, but not with so much that I would skid and catch a runway light “in the shin.” When the speed and spacing were right, I kicked a bunch of left rudder and squeaked off the asphalt into the soft spring grass.
I rolled sprightly into the grass and missed the runway lights, signs, and all other obstacles by a good margin. Then I looked behind me and saw the C-152 was still about 400 feet up on final. I called the tower to inform them that I had a problem and had pulled off into the grass to clear the runway. They were kind enough to ask if I needed assistance, which I humbly declined. Then I removed my belts and headset and scampered around the airplane to try to pull-start the engine. At this point, I noticed there was a taxiway 40 feet in front of me that I could have turned onto from the runway. No one would have ever been the wiser that I had any trouble. Oh well. I executed the contingency plan that I made before taking the runway; I just failed to realize the better option that became available to me (Mistake #4).
After a few pulls, I realized the engine wasn’t going to start easily to allow me a quick retreat back to the hangar. I donned my headset again just in time to hear the tower send the 152 around as a precaution. Then the tower asked me to pull my airplane farther away from the runway, and I began to do so. It’s no mean feat for a 150-pound guy to push a 480-pound aircraft with 75 pounds of fuel on board through a swampy grass field.
As I really began to sweat, I pondered why most airports are swampy in spring. It also occurred to me that the fixed base operator guys were probably watching and getting a good chuckle from my predicament. Once I was well clear, I breathlessly called the tower to let them know. I began pulling weakly on the start cord to try to taxi the remaining half mile back to the hangar. It took a while since the carb bowls were empty of fuel. Eventually I got her started and was cleared to taxi back to the hangar.
Cory and his Quicksilver MXL II
Okay, so now I knew that the engine wouldn’t run at full power without a boost pump. I still wasn’t ready to give up on my theory that the dual pump was the problem. I dug around the hangar and found my old single-output fuel pump. I decided that I would reinstall it on the airplane and do some more testing. The clearances were such that I needed to remove the dual pump in order to connect the single pump. It took me almost an hour to get everything installed and plumbed up, but now I was ready to prove my theory correct.
I fired up the engine and didn’t see any leaking fuel. Woo-hoo! Then the engine abruptly died. (I find it remarkable how suddenly the engine goes from running fine to stopping dead when you forget to turn the fuel valve on.) I turned on the fuel this time and restarted the engine. Sure enough, after the bowls filled back up, a steady stream of fuel started flowing from the front carb. Defeated, I shut the engine down.
I had to abandon my treasured theory about the high-volume dual-output fuel pump overpowering the carburetor float bowl needle. I finally accepted the LEAF technicians’ advice that this is a well-tested configuration running on many airplanes without problem. Okay, time for some basic troubleshooting. I had a problem, I knew how the fuel system and carbs worked, but I was unsure of the exact cause. I started by exchanging the float arm and float needle valves between the two carbs. Aha! When I started the engine, the problem now moved to the aft carburetor. I had a new theory now. I was sure the problem was with the float needle valve. Upon close inspection, I couldn’t see any difference between the two needle valves. They were both perfectly clean, the taper was the same on the tips, and both plungers were equally springy.
I exchanged the needle valves back to their original carbs and started the engine. Dang! The problem stayed with the aft carb. I was learning to abandon theories more easily now, but I was still befuddled. The problem seemed to follow the float bowl armature assembly. I took everything apart and just stared at the parts for a while. I placed the two float bowl arms side by side and studied them one axis at a time. Gradually I started to notice slight differences. The problem arm seemed to be slightly off square in multiple axes. Perhaps someone had adjusted the height setting while holding the arm in his hand and bent it slightly. I began gently straightening the arm here and there with various kinds of pliers and clamps. It was only off by fractions of a millimeter, but I could see the difference. I eventually got everything square enough that I could no longer tell the difference between the two arms.
I reassembled everything with all of the components back in their original carbs and fired the engine up again. I fully expected to see a stream of fuel coming out of the offending carb, but there was none. This time, I was delighted to be wrong. I ran the engine through various throttle settings, and the problem appeared to be gone. Victory! Well, not quite done yet. Since the dual-output fuel pump didn’t seem to be the problem, I decided to stay with the LEAF advice and put the dual pump back on in place of the single pump.
By now, I was getting pretty good at this, so this exchange went pretty quickly. I got the dual pump reinstalled and fired up the engine again. I almost couldn’t bear to look at the vent lines, but when I did they were bone dry. Woo-hoo! No leaking fuel. The problem didn’t recur at any rpm range up to cruise. However, I did notice that the engine would bog if I advanced the throttle quickly.
The engine bogged at any power setting over 5,200 rpm. This pointed me to a problem with the main jets. I once again disassembled everything. I looked very closely at the main jets and noticed what appeared to be a fuel bubble on the aft main jet. Upon disassembly, I found a small (1/16-inch) clear piece of debris blocking the main jet. It almost looked like a fish scale. My guess is that during all the switching around I knocked some built-up scale in the fuel lines. I pulled all the jets from both carbs and cleaned them thoroughly. Then I removed and replaced all the fuel lines on the aircraft. I guess there’s no substitute for scheduled maintenance. I flushed the lines and reassembled everything once again.
Bing 54 carburetor with faulty part circled
The engine ran fine, but I wasn’t playing around this time. I wanted to really put it through its paces; I taxied over to the ramp and tied the airplane axle to one of those embedded-in-concrete steel tie-downs. I ran that engine for at least five minutes at every power setting, including wide-open throttle. The engine purred like a kitten, all temperatures stayed right within range, and there wasn’t one drop of leaked fuel. By now it was approaching dusk. The only thing left to do was a full-scale flight test. I did one fast taxi run, and then another with no anomalies. I was a little nervous as I lined up for a regular takeoff. As I poured on the coals, the airplane seemed to just leap off the runway happy to be free and flying again. With the cool air that evening and just me on board, we were at pattern altitude before the end of the runway. Just to be sure, I never left the airport perimeter, but everything was running great. After three full patterns, I began to relax and enjoy the calm that comes with flying right at dusk. It’s a big load off my mind to have the airplane right as rain again. Want to go flying sometime?Corey Cassavant is newsletter editor for Microlite Flyers EAA Ultralight/Light Sport Chapter #1 in Milwaukee, Wisconsin. Learn more about the Microlite Flyers on their website.