valve setting with tight cam followers

[Brandis]

It's all good.   Onward and upward.   I'll get back to you when I sort it out.

And thank you Chatterlea for the kind offer.    I'm hoping a dab of fine grinding compound and some gentle spinning will do it.
My own damn fault for being dumbass when it was there on the bench.

[Swarfcut]

 Dunno which bit yo' gonna spin.....Not these bits, if they are still in place. Off the engine, that's an altogether different ball game.

 Pictures show A7 Longstroke Exhaust centre tappet guide, along with well worn followers, and shown in situ.

 Swarfy.

[Brandis]

I thought I'd push them in upside down and spin. I think the entire machined surface can be polished. Not ideal but it will be easy to see if it works.  If the issue is in the flat surfaces, that's pretty easy to deal with.

But first, I have 2 other cam follower guide blocks. I'll see if they solve the problem.  And if anyone needs one . . .

[Brandis]

Hmmmmm . . . . .
So I took the engine down to the cam followers.  3 worked perfectly, one exhaust was a bit stiff.  I had hoped to loosen the adjacent nuts and ease the whole guide block out but that was not to be.  So I spent half an hour lubing and moving it up and down by hand until a firm thumb push moved it back down.  Much less force than 2 springs. Back together again.
Now, when I turn the crank with the plugs out, I get a loud 'sprong' sound when the valve springs in question relax.
Sticky valve, I thought.  It seemed unlikely because they were all new and had been well lapped in, so the stem had spun easily in the guide.  So, off comes the head and springs and the valve moves perfectly.  I had examined the springs to see if they were somehow askew, but things looked normal. No solution there.
So, It appears that something in the valve train is interfering with the closing motion until that something releases and it moves suddenly to catch up.  That delay in motion is major enough that sometimes, even at low speed the pushrod comes totally clear and disconnects from the rocker arm before the motion resumes.
Those are the push rods that I cut down from longer rods and they may be a bit shorter. So the valve adjuster shows perhaps 4 threads more in the exhaust side than on the intake side. But the other exhaust valve is the same and has no issues.  There's room for it to move at both ends of the rocking and I set the lash before spinning the crank.
The rocker arm moves freely.  I've switched push rods left to right with no effect.   
Someone in the Forum has been down this road and hasbeen smart enough to figure it out themselves.  Not me, unfortunately.

[Brandis]

Aha !!!

You will recall I said the exhaust push rods were from another BSA,  considering the length, probably a 500 single?  Well,  a close examination of the bottom, socket end revealed that the OD of the socket was a tiny bit larger than the OD of the proper A7 intake push rods.   Since they are so close together, it appears that they interfered with each other's motion, one popping past the other when the other moved enough to get out of the way.  That caused the spring to sprong every time.  It was the every time identical that got me thinking. A tight spot on the slide would be more variable? 
A quick grind to thin them up and all is well.
Sorry about all this hand wringing.   Hopefully this is the end of my travails, at least until the next one. 

[fido]

Apologies for adding to your post rather than create a new one but I have a question about cam follower orientation. It is obvious for the 2 centre followers but I don't know which way round the outer followers go. Does the convex side face inwards or outwards? My followers are a bit worn and there is an unworn bit at the edge, where the cam lobe didn't reach.

[Swarfcut]

 My thought would be for the curved sides to face outwards. No particular reason, just to keep things so-so and as the two inlet guides are symmetrical....  in the grand scheme of things, I doubt it matters.

 The two centre followers look happiest flat faces together, but on parts to hand, they will actually fit in the centre block curve to curve without fouling.

 The majority of used followers I have seen all exhibit the same wear pattern pictured above, it is obvious the cam has missed some of the follower face. Whether this is down to followers the wrong way round, camshaft end float or just the was it is, who knows. You can bet back in the day they were nailed together without too much thought to the finer detail we consider now.

 When I get a new cam cheap in the fire sale, I will take great care to get lobes and followers aligned as best I can, my guess is that the alignment can be seen via the sump plate aperture, before fitting the crank and by trial and error getting the best match. Then again,....Is it actually possible without moving bush positions or facings?

 Swarfy.