Bizarre fuel starvation problem

[Egaeus]

I got stranded for a while on Saturday on my first legal (and therefore extended) ride.  After a few minutes of riding, it quit running.  I coasted into a mall parking lot, tried some roadside diagnostics, and couldn't figure it out.  It was behaving like it was out of gas, so I put the petcock on PRI.  I opened the gas cap and there was no pressure difference, and further examination shows the cap vent to be working.  I left the petcock on PRI.  After about 15 minutes, it started and seemed to want to run, so I headed home.  A few minutes later, it quit again.  So I got a bottle of water, had a refreshing drink, and waited.    Again, it started, I made it home, and parked it. 

Yesterday, I proceeded to take the entire fuel system apart.  I drained the tank (plenty of clean gas).  I checked the screen and tank petcock (in great shape and clean).  Checked for rust (light surface rust, but nothing loose).  I checked my new fuel lines and fuel filter (clear).  I checked the frame-mounted petcock (diaphragm is intact, petcock is clear).  I checked the fuel line routing (correct).  I'm at a loss.  Any ideas?

Oh, and I didn't have any problems yesterday driving an even further distance to test it.  The only changes I made were a minor float height adjustment on the left carb (fuel level was too high), and the proper needle valve adjustment for my carb jets. 

[Cal Amari]

Just a few suggestions; first, place the GS on the centerstand, switch the petcock to PRI if it isn't still in that position, open the float bowl drain screw a few turns, and check to be sure you have good gas flow from each carb. I attach a clear vinyl tube (think it is 1/4" I.D.) to the drain tube, then run it into a gas can, so I can observe any interruption in the fuel flow. If gas flows well for several minutes from each carb, you don't have a fuel-delivery problem.

If it shows you DO have a problem with fuel getting into the float-bowls, check to be sure that the fuel filter is installed correctly. If it is reversed, it can obstruct the fuel flow enough to cause this type of problem. If that isn't the culprit, check to make sure no lines (fuel, vent, or vacuum) are pinched when the tank is in position. Spotting pinched lines isn't always easy, so look carefully.

No matter what else you do, be logical and methodical, and you'll find the problem. Obstructed fuel flow should be fairly obvious; it means fuel can't get to the carbs because of a blockage, but disrupted fuel flow from a pinched line can be a lot more difficult to find.

[Egaeus]

Okay, I forgot a couple of things.  I did check that the fuel filter was oriented correctly, so that's out.  I also used the draining of the tank as an opportunity to check things.  First, it was all of the fuel system from tank to carb.  I noticed that the fuel would eventually stop flowing out of the tank.  I methodically removed parts of the fuel system until I was coming straight out of the tank petcock to see if each was causing the problem.  They weren't.  All I had on was a single line attached to the reserve and it was still a bare trickle. I didn't check no lines though, because I didn't think about it, I mean, I could see through the line!  So I put the fuel filter back on just to filter any possible garbage out since I didn't know the condition of the tank screen at that time.  I was convinced that it was the tank screen (or lack thereof), so I was suprised as crap when I pulled it off and it was in pristine condition.  When I put it back together, I tested the tank flow, and it gushed out of the tank petcock with no lines attached.  Maybe I fixed something inadvertently, but I don't know what it would have been.  I guess I'll have to test the fuel flow by riding and riding and riding some more.

[scratch]

I noticed that the fuel would eventually stop flowing out of the tank.

Was the tank shut off valve partially closed?

The only other thing tht I can think of, even if the rubber fuel lines are new, is if the interior tore, making a flap, when the upper end is put on the tank outlets.  You may not see the flap unless you poke around.

[Egaeus]

No, it was open.  Next time the fuel lines are off the tank (valve adjustment as soon as I get the tool), I'll poke around on the hoses.  I'm at a loss myself.  I consider myself a fair diagnostician, but this one has me scratching my head.

[Mandres]

You sure you didn't develop a vacuum in the tank?  When you pulled the petcock off it would have relieved that pressure and fixed the problem.  Try blowing in the vent holes in the gas cap with a can of air with straw attached to make sure they aren't blocked up. 

-M

[Egaeus]

Well, when I was trying to get the tank to drain, I shook it.  The gas cap went psssshhh (and I spilled gas all over the place because I had the other petcock attached to an open tube pointing up).  I'm pretty sure it's working the other way too. 

[Egaeus]

So I finally figured out my fuel starvation problem.  It's the fuel filter that I added.  I added an automotive filter, thinking that the higher surface area would restrict fuel flow less.  However, I wasn't thinking about the fact that cars have fuel pumps.  The filter material is much better, but much more restrictive to fuel flow.  In the gravity-fed system, it restricts fuel flow to a trickle, causing the carbs bowls to empty.