1950s Fuel filter bubbles bowl not full

[Bob Kielar]

My mechanic friend who rebuilt my 331 engine was over troubleshooting for why the car was running poorly when cold and not performing well when warm. He remembered he forgot to connect the vacuum advance line. After connecting the vacuum advance we were admiring our work and saw the glass bowel fuel filter was bubbling and running low on fuel in the bowl. He said looks like the fuel pump is bad so I changed it with a rebuilt one from one of the reputable vendors. Had the same results after installing the rebuilt one. So then I disconnected it from the fuel tank and ran it out of a gas can same results. Then I hooked the tank back up disconnected the fuel filter and ran a line from the outlet of the fuel pump into my gas can and fuel came out with a good amount of pressure and some bubbles until the car quit because the carb ran out of fuel. The Cadillac idles fine and doesn't quit running even when you increase the rpm. It's only 50 degrees here so it's not vapor lock. Is this common to have these bubbles and is the fuel filter bowl supposed to be full at all times?

Thanks,
Bob Kielar

[James Landi]

Somewhere in the input line or in the vacuum side of the fuel pump, you have a pin hole leak. They are challenging to find because the most frustrating ones don't drip gasoline, they only suck air especially when your car is under load and requires a large volume of gas.  I've owned several 50's Cadillacs, and simply switched over to electric pumps out of frustration.  If you go this route, make certain to wire and plumb the pump for safety.  James

[Jon S]

What you have is completely normal. The glass bowl is usually 1/2 full. Occasionally it does go to completely full but normal is 1/2 full. The "bubbles" is gas refilling the filter.

[Clewisiii]

With his fuel bowl upside down I would expect trapped air. Unless you had a lot of turbulence it would not escape easily.

[Bob Kielar]

Thanks for the replies. The mechanic is coming over today to do some vacuum and pressure checks. Will report back.
Regards,
Bob

[Lexi]

Thanks for the replies. The mechanic is coming over today to do some vacuum and pressure checks. Will report back.
Regards,
Bob

All good comments. I have owned several 1956 Caddies and they always ran around town with the fuel filter bowl basically near empty. My current '56 has most everything rebuilt under the hood including engine, carb, new fuel pump. The fuel filter on the '56 is mounted right side up, (so glass bowl facing upwards). This "running on empty" fuel filter issue always bothered me. I could never get any of my cars to drive around with the filter anywhere near full. Usually looks empty, (or near so), with lots of bubbling action. Like the cylindrical glass sight glass on the top of an old time coffee pot when percolating. I have installed a separate electric fuel pump on a separate line, just in case. I can turn it on and off as required. A few years back I wrote a piece either for this or the old Mid-Century Cadillac Forum on these matters. Pasted below are my comments, albeit for a 1956 Cadillac. As I recall the specifications given on the push rod length also apply to your 1955 engine. Let us know how you make out! Article attachement begins here:

Fuel pump efficiency tests, (done with an assistant), on the mechanical fuel pump for a 1956 Cadillac include:

1) Volume output - disconnect OUTLET line, (carb side of pump) and place end of line in a container. Run car for 45 seconds to 1 minute at idle. You should get 1 - 1.5 pints of gas if output OK.

2) Crank only test with coil wire off (no engine starting wanted here) for about 20 sec. You should get about 1/2 pint of gas if OK, OUTLET side.

3) Vacuum test at idle with a vac gauge at pump INLET side. Disconnect line here and fit gauge to pump. Start car. A normally filled carb bowl will permit your car to run for at least a minute or more with the fuel pump disengaged. Run car and observe gauge. Normal minimum vacuum should be 10 - 13 inches. I believe the gauge needle indicator should be steady. If it fluctuates, especially wildly, again, I believe that is indicative of an air leak. If this asspect of my comments require additional comments, please chime in.

4) Pressure test at OUTLET side of fuel pump. The reading on your vacuum gauge should be 5.5 psi with engine running at idle, (4 - 6.5 as per shop manual). I believe there is a separate section to read this on your vacuum gauge dial.

5) Check torque for the approximate 10 hold down screws around the perimeter of your pump (unless yours is a crimped unit w/o screws).

6) Check all connections as well as strainer nut seal, (see my earlier post) and strainer screen for debris. There should be no air leaks here. Note: A strainer nut air leak will not reveal itself with leaky fuel. It will suck air and seriously affect engine operation. This is often a serious culprit in fuel delivery problems with these cars. To be sure I wrapped my strainer nut threads with gas fitters tape to be sure seal is tight. There is also a sealing washer there.

7) Run car from a gas can if you think there is a problem somewhere in the fuel line before the pump or at the car's gas tank, as a test.

Check lines for kinks & blow out with compressed air both directions, gently

9) If all fails consider checking push rod length as might be worn, manual says should be 7.1425" - 7.1475" long. If OK but you suspect still a problem check manual for specs on push rod travel and possible causes. Tests done with car at room temperature as per manual. If you suspect a leaking well plug in your carburetor inspect your engine oil for odour of gasoline. I also believe a leaky pump diaphram may cause gas to also leak into your engine. If you remove the pump and push rod, you may have to slightly 'bump' the engine to make it easier to fit the pump back on if push rod on high end of stroke and interferes with pump lever when re-installing. You may have to tear down the pump to inspect the internals.

Clay/Lexi

Edit 1: 1956 Cadillacs were known for generating lots of heat under the hood and "running hot", a recipe for vapour lock.
Edit 2: Cannot emphazise enough the BS that these strainer nut leaks can cause. I believe your 1955 pump should have one. My new, reproduction fuel pump back in 2019 leaked air through this strainer nut right out of the box. This nut is the large one you see on top of the fuel pump. I think its size is 7/8, like the oil filter cartridge size top nut. You can't miss it.
Edit 3: James Landi's comment on pin hole leaks might be your culprit. It got me thinking to search for the above old text to re-post with an added bit on those pesky strainer nuts. Thank you James!

[Bob Kielar]

First of all thank you so much Clay for the detailed procedure on how to check the fuel system. Also everyone else who offered advice.

We checked the fuel pressure and it is 5.5 gave the big filter nut an extra tweak. Still getting the percolator effect with bubbles in the filter bowl. The Cadillac idles fine and when we gave it more gas it runs fine and doesn't sound at all like it's running out of gas. The reason for the concern is that he has been rebuilding engines for 30 years and has never seen this before but I may ad this is his first 331 Cadillac engine. So hopefully tomorrow I will be able to take Annie out for a test drive being followed by my better half just in case.

Regards,
Bob Kielar

[Jeff Rosansky CLC #28373]

That always bothered me too. Our 55 has an electric and does the same thing. It seems to fill up more after longer drives. We tend to forget that they run off the fuel in the carb bowl, not the fuel filter. As long as there's gas in the carb, she'll run.
However, I admit, I did a lot of troubleshooting on it and finally gave up because it ran so well.

One test you can do is to pull the line at the tank and plug it. Then fill the line with mineral spirits at the pump side. Then pressurize the line with shop air. Any leaks in your hard line will show up that way. In my case it was bad under the clips. Small leaks won't show normally because the pump is sucking fuel. This test pressurizes the line and makes it easier to see.

[Lexi]

That always bothered me too. Our 55 has an electric and does the same thing. It seems to fill up more after longer drives. We tend to forget that they run off the fuel in the carb bowl, not the fuel filter. As long as there's gas in the carb, she'll run.
However, I admit, I did a lot of troubleshooting on it and finally gave up because it ran so well.

One test you can do is to pull the line at the tank and plug it. Then fill the line with mineral spirits at the pump side. Then pressurize the line with shop air. Any leaks in your hard line will show up that way. In my case it was bad under the clips. Small leaks won't show normally because the pump is sucking fuel. This test pressurizes the line and makes it easier to see.

Good advice. A buddy of mine has owned four 1956 Cadillacs and all ran with a near empty glass fuel bowl. His cars performed well. That said, like Jeff and I, he could never identify the issue causing the low gas level in the fuel filter bowl. Wonder if now it is partially due to the new gasoline blends? Also, in our cars the fuel filter is situated in what is probably one of the most hottest spots in the engine bay. Perhaps why Cadillac literature at least for '56 shows the glass bowl mounted "up", in an attempt to keep rising heat away from the bowl. My buddy wrapped his fuel lines to insulate them from the heat but he still wound up with this near empty glass fuel filter bowl issue. Clay/Lexi

[Lexi]

As an after thought, on two occassions I have experienced leaky strainer nuts that while physically tight and properly torqued-they still leaked air. In both cases I had to wrap the threads with gas fitter's tape (not plumber's teflon tape). In one case there was also a missing washer/seal. In the other case the seal was there but I had to replace it. Both of these were with new pumps. Clay/Lexi

[Jon S]

If it's any help, as you know my dad purchased the 1958 brand new and the fuel bowl level has always varied from full to half full. As the pump adds fuel it will "bubble" but normally stays at half full. As others have mentioned,it really doesn't matter as the carburetor fuel bowl is what keeps the engine running smooth. I have never had a lack of fuel problem even at sustained WOT.

[Lexi]

If it's any help, as you know my dad purchased the 1958 brand new and the fuel bowl level has always varied from full to half full. As the pump adds fuel it will "bubble" but normally stays at half full. As others have mentioned,it really doesn't matter as the carburetor fuel bowl is what keeps the engine running smooth. I have never had a lack of fuel problem even at sustained WOT.

That is amazing that your '58 is still in the family after all these years. Your car is a wonderful reference vehicle. I am not aware of these fuel bowl issues after 1956. Just curious, but was your fuel bowl originally installed glass side up or glass side down, in 1958, when new? Also, does your car have a period pumice stone filter installed in the fuel filter bowl or is it a new paper replacement? Clay/Lexi

[Bob Kielar]

Had Annie (55 Fleetwood) out for a ride yesterday and she ran well. No stumbling from stopping and going also varied speeds up to 65 mph ran well. When going about 45 I put it to the floor and she downshifted kicked in the other two barrels and took off. The only thing I didn't do was a hole shot save that for another day or not. I watched the fuel filter when warming up and the filter glass stays full. Got home parked shut Annie off opened hood restarted her and filter glass was empty then started percolating. Maybe someday someone will read this thread and explain what's happening with the bubbles and low level. As for now I'm just going to Keep Cruzin.

Thanks for all the replies,
Regards,
Bob Kielar

[Chopper1942]

When I am checking for fuel delivery, after Lexi's post's step #3, I always remove the line at the fuel tank and verify that the vacuum there is exactly the same as at the pump inlet and that it will hold a vacuum.

[Jon S]

That is amazing that your '58 is still in the family after all these years. Your car is a wonderful reference vehicle. I am not aware of these fuel bowl issues after 1956. Just curious, but was your fuel bowl originally installed glass side up or glass side down, in 1958, when new? Also, does your car have a period pumice stone filter installed in the fuel filter bowl or is it a new paper replacement? Clay/Lexi

The 1958's had the glass side down from the factory and the slant top AC fuel pump as displayed in my photo several posts above. Also I run the stone filter as was originally installed. Lastly, the 1958 fuel pump was raw aluminum; not painted.

[Lexi]

The 1958's had the glass side down from the factory and the slant top AC fuel pump as displayed in my photo several posts above. Also I run the stone filter as was originally installed. Lastly, the 1958 fuel pump was raw aluminum; not painted.

Great info. Thanks Jon. Clay/Lexi