I was starting to get really bummed about the bike Saturday morning, it was starting, then surging and dying, and no matter to screw-tweaking seemed to make a difference. I decided that for some reason maybe that overflowing carb was involved and that instead of trying to “tune the whole guitar” maybe I should “replace the broken string” first, so to speak.
This was a good move in that it got me in the right mindset for the job. Visions of riding the bike were now far away and out of focus and all I was looking at was the carb, and why fuel kept coming out of the overflow. As far as I know, the only way fuel flows out of this port is when too much fuel is in the bowl. So I turned my attention to how fuel gets in the bowl, which works surprisingly like a toilet.
The bowl is at the bottom of the carburetor and inside this bowl is a float (really a pair of connected floats). When fuel fills the bowl, these floats move upward and at a set point, engage a valve which cuts off the fuel. As fuel is used by the engine, the floats drop as the fuel level drops and at a certain point, fuel once again flows into the bowl. If this mechanism fails to do its job, there is an overflow tube that lets the fuel run out the bottom of the carburetor. This is what was happening to my carburetor.
The first thing I did was pull the bowl off. I was really afraid that the gasket was going to jump out and I was going to have to wait another week for it to shrink back to normal size so I could put the carb back together again. To my surprise, it just stayed in place, so I let it be and I was careful not to disturb it the rest of the day.
First I wanted to test the shutoff valve, since a failure or leak here would explain everything. I manually lifted the floats to their maximum “closed” position and turned on the fuel: no leak. I slowly dropped the floats with the fuel on and as expected, the fuel began to flow as soon as the floats allowed the valve’s needle to fall away from its seat.
The next test I did was of the floats themselves. One common cause of this is a hole in the float, so that it is no longer buoyant and doesn’t lift when the bowl fills with fuel. I did a crude test of this by removing the floats and placing them in the removed bowl. I added fuel to the bowl and the floats lifted as expected. Examining the floats after this test, there was no evidence that they had taken on any fuel, or had any holes or openings that should not be there.
Another cause of this problem can be a mis-adjustment of the cut-off point for the fuel valve. Like a toilet, if the valve waits too long to close the fluid will continue to flow even if the float is at its maximum height. On the carburetor this is adjusted by bending a small metal “tang” on the float assembly which contacts the needle portion of the fuel valve.
There is a recommended procedure for adjusting this tang in the manual, but it requires you to tilt the carburetor and take some measurements. Obviously this is impossible with the carb attached to the engine. Removing the carb from the engine is surprisingly allot of work on this bike due to the position of the exhaust and how the throttle linkage works so I thought instead that I would take a “cut-and-try” approach to adjusting the tang.
Either I thoroughly underestimated the precision necessary for this adjustment, or there is some other cause for the overflow because I tried this adjustment no less than 27 times through the course of the day, with not satisfaction. I was able to adjust the tang to the point where no fuel flowed, and adjust it the other direction to cause the overflow, but I was unable to find the “sweet spot” where the float action correctly metered the fuel in the bowl.
I’m surprised that the “window” of adjustment is as small as it appears to be and I have a feeling there is more to this than I’m understanding.
At this point it’s clear that another approach is necessary if I’m going to correct this problem. It is unclear if correcting this will make the bike run right, but I’ve decided to fix the things that I know how to fix, and since this engine running problem is more complex and elusive, I’ll focus on the carb for the moment.