Thank you fatmobile. The engine is an AAZ.
I like your vacuum gauge idea. What amount of vacuum do you normally see on that gauge, as a benchmark? Any specific recommended gauge that stands up to diesel fuel contact?
I did remove the gas cap to see if it ran better that way, but it did not. So I don't think it is a clogged vent.
Only when fed from the jerry can does it run properly.
Maybe the screen in the tank is clogged at the pickup, irrespective of any issues with the fuel filter or check valve.
I'm thinking that pulling the tank out is the only way to see for sure.
Even if I backflush the clog on the supply line, I expect that the crap will stay in the tank, and it will only be a matter of time before a clog happens again.
I have been googling around for diesel tank algae cleaning, and there are plenty of "mechanic in a can" products out there, including "fuel polishing services" and such.
At the moment I am leaning towards pulling the tank out to get a proper look and deal with whatever is in there, be it algae or rust.
In order to access the fuel gauge sender on the van, you have to remove the tank as far as I can tell, so at that point, "in for a penny, in for a pound".
I'm seeing some suggestions to pressure wash the inside of the tank and call it good. We'll see how it looks inside once it's out, I guess.
I just saw a youtube video where an old Benz guy was cleaning his diesel fuel tank at a coin-op car wash with a pressure wand, which I think might be a little shady.
No idea how they handle recycled water at those things, but that's a little too close to "straight to dolphin mouth" for my comfort level, without knowing more. Pretty sure they weren't expecting people to wash out the inside of diesel fuel tanks in the coin-op car wash bays...
Interesting that what comes from the fuel filter drain is unfiltered, because it was pretty clean. Thanks. That suggests the main clog is more likely at the pickup, than the filter.
Once I pulled the filter right off, it was clear there was a mess inside it though, as you can see from the photo.
Edit - I just slid a pretty powerful magnet around the bottom and sides of that plastic tub in the photo, and none of the red/brown particles suspended in the diesel fuel reacted to it. If they were ferrous, then I would have expected them to follow the magnet. Maybe that's not a fully determinative test, but certainly it more supports the algae hypothesis.