Check the petcock for sure. Sometimes they fail and allow fuel to back flow down to the carbs from the vacuum line.

I strongly suggest measuring fuel level using the clear tube test. This is super easy to do. Get a piece of 6mm OD soft plastic clear tubing and use a grinder or similar and create a taper shape on the end of the tube. Then remove the floatbowl drain screw and cram the tube into the hole, twisting and screwing it into the threads. This will provide a leak free attachment so you can check fuel level. Needless to say, if you have rust flakes in the tank this can foul the float needles so check this too.

Good luck

Might have hard to see oxidation or flakes on the seat. I like to wrap steel wool around a tiny drill bit and polish the seats for a good seal

Be careful you dont nick one. Its easier than it sounds?

Thanks a lot for your tips.

Contemplating the whole thing a bit, I'm leaning towards something sticking, either the floats or the needle:

• It's not constantly overflowing, but rather in bursts
• Bursts stops some time after the engine has been shut down
• Whenever I had the tank off, the petcock didn't leak anywhere

But since I don't yet know how the petcock works internally, I will have a look at it as well.

I'll update this thread as soon as I know more.

My OEM peacock only leaked after shutting down the bike. I would check it before running the bike and it was fine. Finally thought to check it after and discovered it was leaking and not shutting off properly.

I recently had some good luck getting a float valve to quit sticking on my Skunk. The whack had been working, but it was temporary, and carb 4 would go back to sticking. I took the drain plugs out from the carb, and put it on Prime and let it flush for a bit into a catch pan. That has worked up to now. I haven't been using the bike that much since though.

It doesn't take much of a piece of crap in the fuel to cause a problem at the float valve.

I have a 750 carb set doing same thing after rebuild... Actually, I got this extra carb set with purchase of two bikes bit when I popped the bowls, I found new or perfectly cleaned jets and bowls cleaned too)... Anyway on the bike carb 1 will overflow periodically in spurts like 2-3 seconds long. Like a little gremlins is turning on a valve then turns it off. I adjusted the float level and then cleaned the needle and seat with alcohol and q tips but problem persists. One oddball thing I did notice is that these bikes have all metal float needles (as compared to the big on tipped needles) and metal needle seats without an o ring to seal the needle. I thought all metal needles seal against o rings and cut in tipped needles seal against metal seats??

The float valve needle itself could be worn. If you can feel a groove with your fingernail, that might be enough to keep it from sealing. I replaced mine with K & L parts.

Instead of guessing about the condition of the float needle and seat a simple fuel height measurement using the clear tube test will provide the facts.

Measuring float height is just to get you close to the proper fuel level in the carbs, and some fine tuning after that is typically needed, particularly if the float needle is getting old and the spring is weak. When new the float needle spring will support the weight of the float but if the spring is weak it won't. That's the complaint I have with K&L needles, they won't hold up the float properly (at least on the BS CV parts anyway). The OEM needles will.

Thanks both! I could barely see and fingernail feel a groove in the needle and wondered if needle and seat wear as a pair. Again, I thought there was always a rubber component (needle tip or oring) in this fuel stop (all small engines are like that and I think the last three carb sets I rebuilt had rubber needle tips.). And good point on actual fuel level measure.. I did it gross by float height because I didn't have a fitting to screw into to drain plug. I hadn't considered tapering a hose and jamming it in.. And on the needle springs, if the K&L are softer, then the float levels will cycle more often but is that a problem? Maybe they bounce shut with vibration???

Yea my latest 550 has kL float needles they were already in there and I had to completely change the way the float is bent so the tang didn't contact the seat before the needle bottomed out. Then I think after clear tubing them, then checking the height again with the bowls off, they are at around 26mm. Lot different than the spec but they work now.

I recently redid some VM26's and set the floats at 24mm but the fuel level was too high by about 3mm. Bending the float tang so the float height was at about 26mm provided the proper fuel level in all but one carb, which required more bending. Measuring the fuel level proved to be the critical process.

You replaced the needle valve seat O rings . . . . .

Alright boys, solved it. And laughed my ass off when I found out what the issue was, but Pictures will follow later. In any case, the valve was obstructed. Not by rust, not by varnish.

The needle and seat are in tiptop shape, did the fingernail test. Petcock is also working fine, couldn't suck even the tiniest drip of fuel through the vacuum line.

Fuel level seems fine, did check it on the bench, not the machine, shoving the tubes in there was to fussy for me to do on the bike, in the underground parking with bad lighting and all.

Update will follow.

So, about this overdue update.

The reason I was laughing:


In case you don't recognize it, that's a single strand of hair, slightly nibbled at by gasoline, but of course still more than capable to set up the needle valves. That hair originated from my beard - I assumed that must've gotten in when I was rebuilding the carbs, hence my laughing.

Little did I know that this didn't get in while rebuilding.

I had to take off the carbs twice since (I haven't updated this thread for mixture of lack of time, frustration, and various other problems to chase):

First, I found the bowl gaskets to scrape against the floats.

Second time around, the needles valves were obstructed again, so I knew the problem had to be further up the fuel system:


However, the petcock wasn't leaking, and worked properly with the vacuum tube.

So, before I set up to debug my petcock, I disassembled and cleaned the petcock on my other bike, and it was there when I realized that the nozzle of the petcock is a) a separate piece and b) may be a bit loose. (This post here: Clicky).

Lot's of grime. I realized that the paint job on this tank either wasn't done very well, or some fuel got under the paint. What originally seemed like a weld to my untrained eye, turned out to be either excess paint, or paint that was attacked by fuel. The latter seems unlikely, since the petcock didn't leak anywhere.




Cleaned the area superficially with some WD40 and set out to remove the petcock:


Something's missing here.


There you are!


This picture doesn't do any justice to the amount of fiddling, shaking&turning the tank, and cussing I had to do in order to get the nozzle out.


(hit picture limit, continues in next post)