R6T7

I have been trying to get this carburetor running right and I've got the driveability issues solved but it continues to dribble a very small about of fuel from the primary transfer slots down into the intake, and some of the fuel seeps out around the throttle plates when the engine is shut down. And of course, when you restart it when hot it takes a few seconds of cranking because there is so much fuel in the intake.

The car is a "66 350HP coupe with A.I.R. (fully functional) - the carb is a correct list 3605 with a 4920 metering plate. The float levels are correct, the power valve is a 6.5 and has been replaced to isolate that as a cause.
What I don't understand is what would cause fuel to seep out from the transfer slots?

I just rebuilt a friends' list 3367 carb off his "66 300HP engine and when I dialed it in on my car it runs perfectly - no leaks or seepage, and the engine starts immediately when either hot or cold.

Any suggestions?

Thanks,

Bob

TCracingCA

Ok first I assume that the car is on level ground? any rake front to rear because of giving it some attitude with ride height. First I adjust my secondaey float level with the car running, and it sounds like the level is still too high, or you have a gasket sealing problem and maybe some flaw in your metering plate etc.. I would take those back off and see if they are wet were they shouldn't be, etc..

I generally do things that are a little different, but i am one of those fancy gasket type of guys, but I have always set up a carb and got it tuned running on old paper gaskets. If everything is looking good, then i change those over after the set-up of a new carb or when rebuilding one! So far i am not sold on teflon anywhere gaskets!

856666

Hi Bob,

What are the conditions when you have the problem?

- Engine hot
- Hot day
- When you set up your friends Holley were the same conditions met to compare the problem?
- I am thinking fuel percolation as a potential cause

What is the primary fuel bowl temperature?

Phil

R6T7

Thanks for the feedback - the conditions were exactly the same for both carbs - ambient temps here are in the 50's. Started the car in the garage, checked for leaks with both carbs, adjusted float levels, set idle mixtures, set idle at 750, and then took the car out for a short drive - enough to get coolant temperature up to normal. Pulled it back into the garage, let it idle for a few minutes while I rechecked for leaks then shut off the engine. Within a few minutes after shutting off the engine my carb started dribbling fuel, and I could see vapor developing in the primarys. The other carb remained bone dry. After 10 minutes or so I observed fuel leaking out through the throttle shaft on both sides.

I'm thinking that I'll repeat the procedure, except that before I shut off the engine, I'll remove the sight hole plug, recheck fuel level in the primary bowl, and then shut off the engine. If the fuel rises in the bowl it will run out, indicating that it could be percolation, or a bad needle valve or O-ring. If it does not rise, and still seeps out through the idle transfer slots, it has to be an internal leak in the metering block.

Bob

TCracingCA

It is percolation! Therefore now for the cure to that!

R6T7

I don't think it's caused by percolation. For one thing, it's not hot enough, and when I just ran the test I suggested, the fuel levels in the float bowls did not change. That rules out the fuel pump, needle valves and o-rings. But it starts leaking within 60 seconds after the engine is shut off. I immediately pulled the carb off the car and found a lot of fuel in the manifold. Once I got it up on my bench on legs I could see fuel dripping freely. I suppose it's possible that the power valve I just put in is leaking, so I'll change that again. Other than that, I'm not sure where to look next!

Bob

Kerouac

You may not find this helpful, but if I were you I would steal my friend's carb.

R6T7

Well, that's tempting, and I've suggested to him that he might lose it, but unfortunately, it's not the right carb for my car which is being judged in March!

TCracingCA

Your follow-up paragraph sounded more like this was just happening (heat) after taking it around the block. Therefore then I suggest what I said as far as the seal on the meter block and/or the meter block has a surface issue that is causing fuel to travel where it shouldn't at just idle or under rpms, if you are absolute on the fuel level.

Do you own a good Holley book on this carb. There would be drawings as to where the fuel comes from to get to the transfer slot, where the passages are etc.. Now if fuel is leaking in your general shafts/throttle plate area, is that fuel from the transfer slot leaking in that general area or does your unit have two problems? I don't know your model carb, but the gasket seals are very very important to control where the fuel is held in the carb and for what purpose and where fuel has to be available for this run condition or transition from state to state.

Someone should be able to find a diagram or picture of the inner passages on your particular unit for us to nail this down.

Darn I should have gotten in the line for Xray vision! But if fuel is leaking with nothing being moved, then something from a higher point were the fuel is stored internally is getting down low to the transfer slot-- and that sounds like a gasket leak to me! The only way fuel moves in a Carb is by pressure- such as the fuel being made available at the inlet, through mechanical motion like turning the throttle, and vacuum under run condition that pulls fuel somewhere! Unless I missed one- oh I did, you also have the guys that spray or pour fuel down the throats! Now that was a joke!

vetrod62

Air bleeds are probably clogged and gas is being siphoned through the carb.

TCracingCA

But that would be easy to blow some gumout down those top air bleeds, followed by compressed air to ensure they are clear. With the car running the shot of gumout down each should cause the rpms to dive toward stalling, so give it a little throttle as you hit them with a shot. you will quickly know if each is drawing air flow by doing this.

Also it would be good to show where if any vacuum/pcv lines are connected.

R6T7

Thanks for the feedback guys, and it helps to have some direction to isolate this problem.

I decided to bench test the carb to see if I could replicate the problem with no external factors such as manifold vacuum, heat, etc.

So I decided to reinstall the primary float bowl without making any other changes, and with the carb on the bench, filled the fuel bowl to see if it would leak. It did. When it was a little over half full fuel was seeping out in the area generally between the two primary throttle plates.

So I removed the bowl, replaced the power valve and the accelerator pump and repeated the test. It leaked as before. That eliminated the power and accelerator pump diaphragm as possible problems.

Since I had another 4920 metering block off another carb I installed that block and repeated the test. It leaked just like the other one.

I then decided to remove the primary bowl from my friends carb which is a list 3367 with a correct 4743 metering block, and inspected both the metering blocks and the main body of the carburetors to see if there were any differences. As near as I can determine they are identical - both the metering blocks and the main body.

I then checked the original 4920 block for straightness with a straightedge, and it is in excellent condition - shows minor warpage but not enough that the gaskets would not seal. Besides, during each of these tests, there was absolutely no external leakage.

I then installed the 4743 metering block from his carb on my carb, filled the primary fuel bowl and checked for leaks. It has been sitting on my bench now for over an hour, and the throttle plates and entire underside is dry as a bone - no leakage, seepage of any kind.

This makes be believe that there is something different in the way a 4920 is drilled, and I'm now wondering if it really is the correct metering block for my carb, even though the judging guide specifies it as being correct for a '66 350HP with K19.

Am I missing something here?

Thanks,

Bob

Mossy66

Hi Bob,

I remember when I rebuilt one of my carbs that the rebuild kit had a bunch of different gaskets for the metering plates, and a few were very similar, but different. I wonder if maybe you have the wrong gasket installed?


Gerry

R6T7

Hi Gerry -

You may be right about that, but how would you determine which one would be correct. I've always selected the gasket that has holes where the block is drilled.

I guess I could do a trial and error approach!

Bob

Mossy66

I would compare the gasket with both the holes in the block and the holes in the carb body. There might be a hole in the metering block that is meant to be covered by the gasket.

Also, I might squirt a little carb cleaner into the holes in the two metering blocks you have (the correct one and your friends) and see if it sprays out in the same places

TCracingCA

Good Job!!!! You got it!

VintageMusclecar

A straightedge is not accurate enough to check a metering block. I've seen innumerable blocks that "looked" OK just eyeballing them, but once they were accurately checked were out up to .030" or worse.

It takes surprisingly little warp to cause the issues you're having.

The 4920 is the correct block for that carb, and unless it's damaged in some manner or severely warped, chances are it can be straightened.

Eric

R6T7

It turns out that it is warpage that caused the problem. Adding a second gasket between the metering block and the main body stopped the leak completely. So do I leave the second gasket in place, or try to straighten the metering block? Is there any downside to having an additional gasket there? And how does one go about straightening a metering block?

Bob

856666

Bob do you have access to a surface plate and feeler gauges?

In order to straighten it you need to identify the location and amount of deflection.

If its a smooth surface you could use very fine sand paper on the surface plate - IMO

Phil

R6T7

I don't have one here but I could get to one. Does one just sand/grind it flat or attempt to bend it?

And how do you work around the locating tabs?