Fuel pump and air filter (09)
Drawing of Mercedes-Benz type 219, © Daimler AG
As with the carburetor you need to level all surfaces where gaskets etc sit. I did that with hard, fine sandpaper (silicon carbide paper?) on a leveled surface, a cutting board made out of marble. I also leveled the surfaces for the intake and exhaust valves' valve plates.
The fuel pump's intermediary flange is made of aluminium. After glass beading it, the surface gets a little bit too rough. This is taken care of with tumbling (as with for example the cylinder head cover). Glass/bead tumbling gives beaded aluminium parts a slightly more glossy finish. In Stockholm you can find the shop Svenska Trumlingsaktiebolaget that does that type of work.
I re-used the diaphragm, it wasn't too old at the time for this restoration, and I thought that if it break it's easy to change. Thought I had to spare a few weak Swedish Kronor at least... In the right-hand corner of the "parts picture, you can almost (!) see the tapper for the fuel pump drive with its bushing and sleeve. Check the tapper for wearing. Using an ordinary puller it was an easy job to extract the bushing and insert the new sleeve. But remember to measure the bushings overhang, outside the intermediary flange - it's difficult to see when it's fully pressed in. I used a vise to press it in. The last picture shows the rubber sleeve behind the bushing in the intermediary flange, the one that is attached to the cylinder crankcase. See the thorough article written by Doug on the Ponton pages, Mercedes-Benz Ponton Fuel Pumps (lever-type, priming, gasoline).
In April 2017 I replaced the old diaphragms, all other gaskets and valve plates and most of the springs. Cheap and worth it... I don't know what I was thinking of the first time I went through it.
And once again...
While testing the engine (see Running the engine on test bed) I had some intermittent problems which pointed towards the fuel supply. After having run it on high rev (in this case 4500rpm) it started to go bad and finally stop totally. If I engaged the lever on the fuel pump I got the familiar "pump sound" and I could hear fuel coming into the carburetor bowl. If I restarted the engine it normally run well until I did a high rev again. So, no fuel but the pump was working!? Took the pump off and apart. Everything looked good. Assembled it and holding my lip or tongue against the intake nipple I could feel that it had suction, created a vacuum. Stupid me thought that was enough, I didn't realize that it was a problem that the vacuum didn't hold more than a couple of seconds. Kind people informed me that it was. So where did I have the leak? Everything was newly restored and looked good. Henry M on the Ponton list, kindly asked me to see to that the top of the fuel pump cap, where you have the screw and a fiber washer, was flat. Mine was not, it was pressed in by years of tightening the screw. The drawing of the cap in the workshop manual shows that it originally is a bit pressed in (mine was more than that) but I did as Henry told me - and voila, the pump could hold the vacuum for at least 2h.
So, I put it back on the engine and still had the same problem. Tore J on the Ponton list then kindly informed me that you have to assure that you have no air leaks (can hold vacuum) all the way over the fuel filter back to the fuel tank (in my case a fuel canister since I was testing the engine on a test bed). I didn't have that. Tightened everything and verified that I had vacuum 2h (had other things to fiddle with in the meantime). After that the engine run as it should without stalling.
Lesson learned; see to that you have NO air leaks on the intake side of the fuel pump back to the tank. I used "holding vacuum for 2h" as a measurement of that and it was obviously enough. If you have leaks on the fuel pump's pressure side that will be more visible, fuel leaking out.
Measuring that the fuel pump delivers enough fuel
During the search for reasons for the stalling engine during the test runs, as mentioned above, I also decided to test the fuel flow according to the workshop manual. According to the workshop manual the fuel pump should deliver 35-40l/h with camshaft 2500rpm (equals crankshaft 1250rpm) through a nozzle with 2mm diameter. My starter runs with approximately 400rpm and I tested during 20sec. Given this I should get around 0,08l. I got 0,17l.
The test had a very simple setup, on the outlet nipple I attached a hose in which I had inserted a nozzle with a 2mm opening and then I run the starter, holding the time - 20 sec - using the timer in my cell phone.
Video on YouTube showing how I tested the fuel pump flow
The restoration of the Ponton's air filter I'm actually quite proud of too. The pictures are too bad to show the end result. Everything was glass beaded or sandblasted; if necessary buffed and then white cad'ed if not supposed to be painted to 100%. I had the control arm for the linkage pull rod to carburetor partly white cad'ed and partly semi gloss black. Can be noted that the control arm for the linkage pull rod actually belongs to Carburetor but I decided to have it viewed here together with the air filter support. And yes, the air filter itself was more than thoroughly covered with 2-5 layers of tape and paper etc during the process.
Before installing the engine, with its auxiliary systems mounted, I wanted to check that everything worked as it should, do necessary adjustments, re-tighten all screws and see to that I had no leaks, see Running the engine on test bed.