1939 LaSalle w/mechanical and electric fuel pumps (resolved). Now a knock issue.

[Stude Light]

Car:

1939 LaSalle with rebuilt 322 (now 346) engine, all engine components rebuilt/new, fuel tank serviced by Gas Tank Renu, new sender, lines all flushed or replaced, I personally rebuilt the fuel/vacuum pump with a kit. Prior to my ownership someone had installed what looks like an Airtex pump forward of the left rear wheel along the frame where the fuel line runs (good location). It is powered through the ignition switch via a remote switch under dash. I use it for priming purposes. There is a tubular glass fuel filter that can be disassembled and cleaned back by the electric pump and there is a sediment bowl on the mechanical pump.

 

Issue:

The car runs great except when warmed up and you get on it upshifting through the gears, then it will just fall flat about the time you hit third gear, then surge on/off. If you let up on the throttle it will recover and drive normal unless you get on it hard again. If you just drive with low to moderate throttle, then no issue. Same thing happens when to accelerate onto the on ramp on the freeway and once in a great while during a cruise (perhaps crowding the throttle at the time). It feels to me like the float bowl runs dry. It's very driveable but in today's traffic there are times you want to get to speed quickly.

 

If you disconnect the fuel line from the carb and crank the engine or run the electric fuel pump there is lots of flow. If I run the fuel pump all the time I will get weepage out of the carb so my guess is it overpowers the needle. I surmise that once the underhood temps get to operating range and the mechanical pump starts trying to draw more fuel it gets some vapor lock while trying to pump. Not knowing the brand/model of the electric pump makes me just want to replace it with a new Airtex 8902. That should have the correct voltage, pressure and ability to draw through but I cannot find one. Everyone is sold out! 

 

Two questions:

1) Has anyone had similar experience and found something I'm overlooking?

2) Does anyone know where I can find a new Airtex 8902 or something similar?

 

Thanks

Scott

 

 

 

[Oldtech]

I Think you are correct that the float bowl is running low. It seems the fuel flow is inadequate but where? You have checked the fuel to the carb, as I read it. Could there be a restriction in the carb itself? Is there a screen in the inlet, or how about float too low and not allowing enough fuel in the bowl. In that case the fuel may "pile up" at the rear under acceleration and starve the jets.

[EmTee]

What is the condition of the gas tank?  Is there a 'sock' on the fuel pickup?  Initial flow may look OK, but can it be sustained?

[Rusty_OToole]

Could be vapor lock due to today's gas especially if it happens in hot weather or after a long drive. You could try adding some kerosene to the gas, or wrapping the fuel line with insulation under the hood.

 

The other possibility is a clogged fuel filter but I think you addressed that.

 

If you use an electric pump in series with the stock pump you are going to get a lot of pressure, if they each put out 4 PSI you are going to get 8 PSI at the carburetor. This is enough to overpower any needle valve.

[Grimy]

1.  Given modern fuel, you must insulate the fuel line between the pump and the carburetor because the line runs parallel to and just above the left exhaust manifold. Some people add a sheet metal deflector which attaches to the manifold studs.

2.  Check to be sure you have the 1943-era replacement fuel pump with the glass bowl on top of the pump--there were vapor issues even when the car was new, and the described pump was the solution at the time.

[Stude Light]

This is the setup I have. The line is already insulated from the pump to the carb. The exhaust is ceramic coated inside and out ( may help a little). The pump I have runs the vacuum wipers along with the fuel and is showered with radiator outlet air, which is hot but probably the coolest air underhood. I’ll start measuring fuel pressures and do some experimenting.

[Matt Harwood]

Does the full-throttle surging/stumbling stop if you turn on the electric pump?

[Bloo]

A few thoughts:

 

Is your Airtex pump a thumper? If so, it is probably fine for draw through. If not you may need check valves to let the fuel around it.

 

It is possible to check for restriction in the fuel pickup (or from the fuel pump in the rear) by teeing in to the suction side of the mechanical pump and connecting a vacuum gauge. Run a hose up and tape the vacuum gauge to the windshield. Clamp all hoses very securely. If you see the vacuum skyrocket just before the car starts to have trouble, the line is restricted.

 

Some 1930s GM fuel pickup sender combinations use a copper tube for the pickup. Crimped sheet metal holds the tube in place in a couple of spots. The copper cracks there, usually at the one on the outside of the tank, letting air in.

 

Does your mechanical fuel pump have phenolic check valves? They might leak, especially if new. If they have to be wet to seal, the car will have trouble recovering from vapor lock.

 

Is the fuel pump diaphragm getting enough stroke from the cam or whatever is pushing it?

 

 

 

Edited July 16, 2022 by Bloo (see edit history)

[Stude Light]

I decided I needed to do some better diagnostics. I made up a tee in the fuel line and routed it to a gauge set up I could watch while driving.

My mechanical pump runs 2.5 psi while idling (needle steady) The electric pump (only) runs about 3 psi. During my driving exercise, the highest pressure I got with both pumps running was 5 psi.

 

I started with the engine warm enough to open the choke and be driven normally and found when I accelerated hard up through the gears the mechanical pump would run down to 1.5 psi at the worst. If I kept the electrical pump on it would stay up around 3 psi. My test was wide open throttle and shift around 3000 rpm into the next gear back to wide open throttle.

 

Once the engine got up to operating temp and it was hot underhood, I could get the pressure to fall to zero. If I kept accelerating hard the engine would die out as described. Leaving the pump on during this test, the pressure would not drop below about 2.5 psi. I repeated this a couple times with the pump on and off and it acted the same each time.

 

One odd thing I noticed is that if I turned on the electric pump, did the hard accelerations, then shut it off I would get a slight smell of gas in the car. I immediately stopped and looked through the missing side panel with a light and the top of the carb was wet and maybe just a little down the side. The pressure gauge never went higher than about 4 psi when this occurred. I was able to hold 4 psi with both pumps running at an idle and could never get any wetness around the carb - only while doing the hard accelerations. Not sure why.

 

Then another issue after I was all done…there was a slight knocking noise coming from the engine at low idle ~600rpm. You can feel it in the mechanical fuel pump or at least the pulses in the pump are at the same interval and the pressure gauge needle pulses heavily. The needle was not fluctuating at the beginning of the test. I was thinking it was just the pump. I don’t like knocks, especially in my cars and on an engine that just got rebuilt about 4000 miles ago. I grounded each plug and the knock totally disappeared with number 2 plug and grew louder with number 6 grounded. No knock at higher idle and nothing you can hear while diving. Doesn’t seem like my hard driving should have damaged a rod bearing, piston or pin. I am sure the tank drivers didn’t baby their engines. No noise after letting the engine cool off for an hour and the pressure gauge is steady with no pulsing. You can still feel a slight pulsing in the pump but no knocking feel. 
 

I have no idea why grounding #2 plug would quiet a fuel pump knock. I’m hoping it’s just the pump and not something else. I will be buying a replacement pump and see how that goes.

Edited July 17, 2022 by Stude Light (see edit history)

[Bloo]

4 pounds might well be too much. It would be too much for many cars in that time frame. The shop manual should have that.

 

I would hook that gauge up on the supply side (vacuum) and see if you have a restriction at the tank.

 

Then I would probably take the pump apart, bolt part of it back on the engine, crank (no spark or fuel) and see how much the diaphragm moves. The manual might have that too. In any event I would not expect less than 1/4" or so. Some cars in that period used to wear the fuel pump lobe off resulting in insufficient stroke. If the pump or arm is incorrect, you might catch it this way as well.

 

While it is apart, rig up some way to test the check valves with a mityvac. Do whatever is necessary to make them not leak.

 

The knock that you can kill by shorting a plug sure sounds like engine bearings, wristpins, or piston slap, probably bearings. I wouldn't jump to conclusions though. Fuel pump knocks are not uncommon but I can't imagine the spark or heat really making any difference. Timing gear knocks can fool you into thinking you have bearing trouble. If it has timing gears (not a chain) and if the big gear is metal I would not rule it out.

 

The mechanical pump pressure will be pulsing and probably low when it is filling the carb. When the needle gets mostly steady the float valve is shut, the diaphragm is full, and the pump is not pumping more because the diaphragm has pulled away from whatever is pushing it. This is the true fuel pressure.

 

Edited July 17, 2022 by Bloo (see edit history)

[Stude Light]

Thanks for the reply. I’ll start poking around for specifications. It has a timing chain. Engine is now just warm (about 120 degrees) and there isn’t a sound to be heard or felt and the fuel pressure gauge is steady. I’m now thinking a check valve starts acting up when warm. That would explain the fuel pump behavior. As far as the little knock goes, maybe #2 fires and accelerates the crank at the same time the cam lobe engages or disengages the pump and that small difference makes the change. May also be opposite for #6. That’s what I’m hoping anyway. 
 

If you’ve not worked on a 1939 LaSalle … testing fuel pump stroking action is easier said that done but I can give it a try. I did look at the lobe that drives it from the cam when rebuilding and it had little wear.

 

[Stude Light]

On 7/13/2022 at 2:26 PM, Grimy said:

 

2.  Check to be sure you have the 1943-era replacement fuel pump with the glass bowl on top of the pump--there were vapor issues even when the car was new, and the described pump was the solution at the time.

Hey Grimy,

This is the fuel pump I have.

This one is listed for a 1940-1948. Is this what you are talking about? My biggest concern is line routing...I know its rather tight on my current setup, especially for the vacuum pump.

 

[Grimy]

10 minutes ago, Stude Light said:

Hey Grimy,

This is the fuel pump I have.

This one is listed for a 1940-1948. Is this what you are talking about? My biggest concern is line routing...I know its rather tight on my current setup, especially for the vacuum pump.

 

I owned a 1939 Cad 75 7p sedan for 42 years, but for the last 12 it has been a couple of blocks away with the perfect Next Custodian that we all seek.  Cliff, the current owner but not a forum participant, has all my Cad literature, including a reprint of a 1943 official Cad supplement for mechanics which states that the design of the fuel pump was changed (obviously post-1939 but pre-1943) to a narrower, "stacked-looking" (my term), fuel+vacuum unit which has the bowl on top, inverted.  So it appears that vaporization was an issue even then, pre-ethanol.  The two units can and do interchange 1937-1948.  I can try to get him to track down the number and photo for you.  The 75 used a Stromberg carb (AAV-26 I think) with the fuel entry at the rear of the carb, making even a longer run of pump-to-carb fuel line to insulate.

 

As I recall (subject to Geezer Memory Syndrome), the flex line entering the *other* pump is *much* closer to the block and therefore far less affected by the heat from the radiator.  You might try insulating that flex fuel line entering the pump.  I agree with @Bloo that if your electric pump is a "thumper" (i.e., rotary vane) you need a bypass with a check valve.

[Stude Light]

My testing continues....

 

I was checking flow from an open tee (out of the valve I installed) between the fuel pump and carburetor

 

Engine slightly warmed up. No knocking (good).

• With the engine idling, the mechanical pump would pump out a cup of fuel in about 12 seconds
• I turned off the engine and used just the electrical pump and got a cup of fuel in about 15 seconds
• This tells me there is no fuel restriction from the tank

 

Engine to full operating temp and a little hot soak for good measure - the knock was back (again, only at lowest idle).

Note: Pushing a screwdriver at different points on the engine, you could definitely feel the knock on the fuel pump the worst

• With the engine idling, the mechanical pump was pumping 90% fuel vapor and just a dribble of fuel
• I left the engine idle and turned on the mechanical pump and after a few seconds I had good flow (about a cup in 15 or 20 seconds) but also a lot of vapor coming out of the hose. I had the outlet of the hose below the fuel level so you could see all the bubbles. Given the temperature of the fuel going to atmospheric pressure, I kind of expected that.

The fuel pump was running about 135 degrees, the head was at 195, the oil pan was around 160. I figured I would just try lowering the fuel pump temperature so with the engine off, I turned on the electric pump and ran about a quart of fuel through the mechanical pump and out my tee. The pump housing was now reading 112 degrees. I started the engine back up and the knock was barely perceivable and the mechanical pump improved performance to about half of what it pumped with a cooler engine.

 

Unfortunately, I could find no specifications for the fuel pump in any of the 1937 - 1940 service manuals I have. It only says that they must be serviced by the manufacturer.  Based on all this testing, I guess my next step is to just get another pump, install it and see what problems get resolved. With that little knock tied so closely to the fuel pump performance and only heard at low idle, I'm not ready to pull the pan and head off to look at rod bearings and pistons. I wish I had not tried grounding plugs 😉

Scott

[Oldtech]

You are not supposed to use your fingers to ground plugs.😁

[EmTee]

20 hours ago, Stude Light said:

With that little knock tied so closely to the fuel pump performance and only heard at low idle, I'm not ready to pull the pan and head off to look at rod bearings and pistons.

Temporarily hook up a mechanical oil pressure gauge to monitor oil pressure.

[jvelde]

Have you checked the fuel tank cap for any blockage that would cause the tank to build up a vacuum, reducing fuel flow on hard acceleration?  

[Matt Harwood]

Remove the mechanical pump from the block to determine whether it's the pump making noise. I had the same problem and simply removed the bolts and pulled it out of the hole, then ran the engine on the electric pump. It told me instantly where the problem was--the pump was simply noisy.

 

 

 

[Bloo]

Regarding the noise, could your fuel pump be missing a spring?

 

The spring under the diaphragm is what sets the fuel pressure. When the float valve in the carb is closed, the diaphragm is all the way up (or down). It is full of fuel that has nowhere to go until the float valve opens again.

 

A lot of mechanical pumps have a second spring on the arm to keep the arm following the cam lobe when the pump is not pumping.

[Stude Light]

12 hours ago, EmTee said:

Temporarily hook up a mechanical oil pressure gauge to monitor oil pressure.

Thanks. I did that not long ago to verify my oil pressure gauge readings which are on the high side and found my dash gauge runs about right on the low end (25 gauge vs 25 dash)  and 10 psi high on the top end (50 gauge vs 60 dash). I have plenty of oil pressure.

 

6 hours ago, jvelde said:

Have you checked the fuel tank cap for any blockage that would cause the tank to build up a vacuum, reducing fuel flow on hard acceleration?  

Yes, cap does vent. Thanks.

 

6 hours ago, Matt Harwood said:

Remove the mechanical pump from the block to determine whether it's the pump making noise. I had the same problem and simply removed the bolts and pulled it out of the hole, then ran the engine on the electric pump. It told me instantly where the problem was--the pump was simply noisy.

Thanks Matt. I'll have to try that. The only issue is the engine has to be at full operating temp to get the noise....hot fingers!

4 hours ago, Bloo said:

Regarding the noise, could your fuel pump be missing a spring?

Hmmm. I rebuilt the pump and I'm usually pretty detail oriented (more like retentive) but you never know.

 

Thanks for all the suggestions.

Scott

 

[Stude Light]

I decided to go with the 1940-1948 style fuel pump which is supposed to be an improvement over the 1937-1939 design for vapor lock issues. The two vacuum lines for the wipers hooked up with no modifications and so did the fuel inlet line from the tank. I did have to make a new line that runs between the pump and carb which was relatively simple. That solved the issue of lack of flow at higher temperatures. It may just be that it has better check valves too so rebuilding my older pump would probably yielded the same results.

 

The problem is I still have that little knock at idle that only occurs when the engine is at full temp. It sounded like it was coming from the fuel pump so I took @Matt Harwoods advice and just eliminated the mechanical pump, make a blockoff plate and ran on the electric fuel pump only. Started the car, no knock, warmed it up - no knock, took it for a drive and got it up to operating temp and.....same damn little knock that goes away when shorting the #2 plug!!! Bummer!!!!

 

This engine was a complete rebuild (bores, pistons, valves, guides, bearings, etc.), done by a highly qualified shop and has about 4000 miles on it. I dropped the pan last night.

#1 & #2 rod to journal side play in the middle of the spec

Nuts on #2 rod cap were torqued properly

Removed cap - both bearings shells are in excellent shape

Verified the stamping on the bearing shells are 0.020" under - specified for the crank journals that were ground during the rebuilt

Crankshaft rod journal in excellent shape

Bore that I can see under the piston looks perfect with cross-hatch still nice

Cannot feel any play in piston pin

Crankshaft end play within spec

Oil pump pressure is actually a bit high of spec at rpm and in spec at idle

 

 

What am I missing here???

 

My next step is to reluctantly pull the nicely painted, sealed up, non-leaking head and pull the piston rod assembly and make sure no issue with the wrist pin or piston. Then I'll measure the crank journal and rod end bearing to ensure proper clearance. I really don't think I'm going to find anything and not sure else where to go after this.

 

Pictures of the bearings look worse than what they are to the naked eye the way the light reflects off them (if you've ever taken a picture of a black car up close with a flash, you know what I'm talking about).

 

This is the rod design (spare one I had laying around).

Scott

 

Edited August 2, 2022 by Stude Light (see edit history)

[Bloo]

Wow. Well...have the rod checked for straightness while it's out. I'd be tempted to plastigage that bearing before taking anything further apart.

 

[EmTee]

18 hours ago, Bloo said:

I'd be tempted to plastigage that bearing before taking anything further apart.

That was going to be my suggestion also.  It may still turn out to be a piston/wrist pin issue.  At least that would explain the light knock with good oil pressure.  How does the bore look?  Are there any suspicious marks (esp. in the pin area)?

[Stude Light]

5 hours ago, EmTee said:

That was going to be my suggestion also.  It may still turn out to be a piston/wrist pin issue.  At least that would explain the light knock with good oil pressure.  How does the bore look?  Are there any suspicious marks (esp. in the pin area)?

Bore looks perfect from the bottom side. I was going to just pull the head and pull the piston as soon as I get some time. Rather than plastigage, I figured I would mic the journal and use a bore gauge on the rod end bearing to get the most accurate measurement I can. With that all apart, I can verify no piston, pin, bore, rod, valve or bearing issue. Hoping I find something but that doesn't require removing the engine.

[Bloo]

Nothing wrong with that. The plastigage idea was to possibly save you pulling the piston if the problem did turn out to be lower end clearance. Small undersizes were made in bearings for cars of that age. Of course if the crank has been ground, that option probably doesn't exist.

 

[Stude Light]

Pulled the head and piston. All cylinders look the same from the top side. No marks from piston contact with anything. Piston pin to rod has no play you can feel and measurements  for clearance in spec. Piston bore is perfect and still has cross hatch. Piston itself has just a light witness mark on each side at bottom of skirt like you would expect- no scuffing and nothing you can feel. I did find a burr on the outer edge of the rod cap mating surface which I removed with a flat plate and 220 sand paper.

 

Crankshaft rod journal is in spec for 0.020” under (actually at the higher end - larger diameter) but has  0.0004” taper from edges to center. I reassembled the rod with bearing and bore gauge shows 0.0023 - 0.0028” clearance worst case (to center of journal). High end of the new spec but for a 2.5” diameter journal, not too bad. We’ll below the 0.0045” max worn.

 

There is just no witness marks saying anything is wrong so the only thing I can come up with is that burr maybe keeping the rod cap clearance a bit higher. I never torqued and measured the rod bearing with the burr in place.

 

I have new 0.020” under bearings coming and will probably pull all the rods and double check measurements with new bearings and reassemble.

 

 

[Bloo]

I'd do the piston skirt clearance test with a long feeler gauge. The trouble is I don't know if it would tell you anything. Since it's rebuilt, the skirt clearance in the shop manual probably doesn't apply. Whatever the piston manufacturer says applies, and it is highly likely to be different. If you don't know who made the pistons, I guess it isn't that useful. If it was my car I would still want to know what the skirt clearance is.

 

I would also get the rod checked for straightness.

 

[Stude Light]

Latest update….

 

Well, the rod is straight, piston to bore clearance is good. I took the parts to two different engine builders along with measurements and they had no answers. I reassembled with new bearing shells. I sent the oil pump out and had it run on an electric driven flathead engine spin stand as it was a bit high on pressure and they replaced the bypass spring to get it in spec which allowed me to change from the 10w30 I was using to SAE30 (a bit heavier weight). I put it all together and…

 

My oil pressure is a little lower at speed (where it should be) and a bit higher at idle. It runs and drives great but the slight little knock at low idle after the engine is at full temp is still there. As before, it’s hardly apparent but I hear it. After all that work at least I know what it isn’t so I’m just going to continue to enjoy the car and if things ever get louder, I’ll deal with it then. Thanks to all for the various tips and advice.
 

I’m amazed at just how quiet and smooth these flathead Cadillac engines are.

 

I’ll end this thread. Scott

[EmTee]

Did you rule-out fuel pump knock?  Maybe hunt with a stethoscope in order to try to localize or rule out other sources.  I've heard belts make noises that sound like knocking.  Have you tried removing the belt to eliminate it and the accessories?