In general, surging is caused by a lean mixture and not related to crank seals. I found that If I richen it enough to eliminate surging my fuel economy is lower than I'll tolerate. My cure is... (see signature below) I average about 30 mph with my H2, with H2R porting, 34mm carbs, and chambers.

I'm confused by your center cylinder reading of essentially no sealing. If that's true, the cylinder next to it can't be sealing either, yet both outer cylinders are not terrible.

I suspect you had a leak in one of your plugs for the center cylinder when you did the test. As I said above, if one of its seals is bad, the cylinder next to it must also be just as bad, but your numbers disagree.

I don't have acceptable numbers for you. If I tested my crankshaft and one cylinder held 6 psi, both vacuum and pressure, for 5 minutes instead of 6 I would not get the crank rebuilt, but I can't tell you from experience when slight leaks become a big enough problem to make a rebuild necessary. Things like the RPM not returning to idle speed when you let off the throttle are signs it's a problem.

Have you measured the big end connecting rod radial bearing play? My crankshaft would have needed rebuilding for that reason alone, and its seals were shot on top of that.

I'll just post this once again, for those whom missed it all.

Testing is not only vacuum, but also pressure, as the seals must seal both directions.

The specification for our engines is 6 for 6. 6 in/hg of vacuum for 6 minutes, 6 psi pressure for 6 minutes.

The only issue is, earlier triples used an open labyrinth seal, one that was changing pressure/vacuum loaded onto the crank race, everything else uses a positive pressure seal, one that keeps seal lip pressure on the crank race all the time

BUT, testing all our triples should be all 3 cylinders tested, both ways, at once, not individually