Yup, it can put pressure on the banjo bolt threads and make them harder to turn, possibly cracking the fragile alloy housing.
Floats were fine so that wasn’t it… I blew out the needle valves and can now observe that they seal well by blowing in through the fuel port and manually raising them into place. As far as I can tell they are performing perfectly. I’ll install the new top gaskets next week. My current hope is that the needle valves were simply sticking (they did seem gritty before blowing them out).
I’m finally realizing that there is essentially no pressure on the gas inside the float bowl beyond gravity pushing down as there is a vent to daylight from the top and a vented washer under the top banjo.
This for me makes this more mystifying.
There is no pressure inside the float bowl, Bill. There is a little pressure - 3 or so psi - applied by the fuel pump to fill the float bowl to the prescribed level but that pressure is cut off by the closing of the needle valve and from that point gravity does indeed take over such that the level of gasoline in the float is at the same level as the gasoline in the carburetor jet. If you were having problems with fuel level in the float (from a sunken float, maladjusted fork or sticking needle valve) you would see flooding in the carburetor and the engine would run poorly, lots of black smoke and soot in the tailpipes and spark plugs.
You have either a sealing problem (bad gasket, misaligned sealing surfaces) or you have a crack somewhere.
Correct, the pressure is just there to push fuel into the bowls. As for getting from the bowl to the carb, I don’t think it’s gravity or it would dump in and the carb would flood. It’s some other form of magic.
Did you verify the float height? It’s critical and simple to check. You just stick a certain diameter rod between the lever fork and the bowl lid (7/16 IIRC). A drill bit will suffice. If at that distance, the lever arm is contacting the needle valve, all is good. If not, you bend the lever until it does contact it.
Pressure differential. Bernoulli principle, venture effect.
That’s too many things to understand at once. I prefer magic.
3 Likes
They’re connected and the fuel in both “seeks the same level.” Very useful if you need to level things over large distance–use a “water level.”
Regarding the float level adjustment–I was helped by info in the archives stressing that the fork blades need to be exactly aligned and horizontal, or else the float can be cocked (and stick) by hitting one side first, which can generate a sideways force.
As I understand the dynamic, air is drawn across the top of the fuel and causes a thin layer of it to be drawn up and vaporised.
Where gravity comes into effect, as I also understand it, is establishing the level of the gasoline in the jet. It will be at the same level as the level in the bowl. The bowl is vented so there is no differential effect of pressure on the level of gasoline, only gravity.
Edit. Discussing steady state now. The OP’s problem is leakage without the complication of a running engine.
I found my dad’s old Bentley manual (s) and have been working through them - including reading about setting float levels. As far as I can tell the float levels are set correctly. Sadly one of the likely culprits might be a crack somewhere. I’ll know more when I reassemble the floats using new gaskets later this week.
Right, but only with the fuel pump running. That would indicate a sticking float valve. But the OP (Bill) says the engine runs fine. That would not mean flooding-rich. So maybe the vibration of the engine whilst running “loosens up” the float valve, but it resumes sticking when the engine is shut down? Or some sort of capillary action–a wick?
A likely place to look for a casting crack would be the banjo bolt threads in the bowl lid due to over zealous tightening. That would likely manifest as appearing to leak from the top. Fortunately they’re readily available new and used for not too much money.
The strange thing is that I probably have 500 miles on this after it was first restored last fall with no leaks. First observed on the float bowl nearest the rear and then progressing to the other two. At no point during this time was anything touched under the hood. I also consistently used ethanol free fuel. Car sat in cold storage for the winter (very cold) but started fairly easily.
1 Like
Cleaning out those needle valves likely did the trick. That would be in keeping with a car that sat up if sediment got past the filter. Ever cleaned out the tank sump?
When you reassemble with new fiber washers expect to see a few drops. They seem to sweat fuel a bit until it seeps into them and makes them swell. Very gradually tighten down until that stops. I like using a socket as it provides better grip and control than an adjustable spanner I forget which size works best. It’s probably metric. These are British Standard thread I believe.
Thanks!
The slightly leak as the fiber gaskets expands makes sense.
I’ve been reading the Bentley and can retrace some of the work my dad did on the engine by looking for carefully written notes in the margins in pencil.
PS. One of the mechanics used some sort of silicone goop to seal the cork gasket at the top of the new fuel tank. I have plunged out this glop in addition to sediment after getting the car on the road (for the first time in 17 years) last fall. As far as I know there should be no new sidement in the tank.
The hexs are, IIRC, 3/4W.
That’s my attitude on computers. I don’t care what’s inside the box, could be a tiny guy with an abacus for all I care, as long as it (he) gets the job done. 
2 Likes
Engine not running, fuel pump off, yes, I agree.
Bill could clean everything up and dust the suspected area with talcum powder to help pin point the leak.
…it’s not??
One of the best things I did during my restoration was to send them out for new bushing and overhaul. I also installs a second in line filter and was amazed at what it caught before clogging the carb
1 Like
I think you’re right.
1 Like