Temporarily fitting a small chainring to an old Stronglight single crank with a swaged arm. The single chainring is also effectively the spider (it's just machined from a flat piece of alloy), and has a currently unused 86BCD set of holes in it.

So I have big ring - 3.5mm spacer - small ring.

The chainring bolts I have to hand are 7mm long, which is just over 2mm short of the actual measured stack depth. The bolt is getting less than half way through the chainring. At present I can get about three threads engaged, which is probably suboptimal. I think a 9mm long bolt would be perfect.

Not sure if this all works by the clamping force of chainring against 'spider' (in which case I'm probably OK) or whether the chainring bolt body actually bears load in the direction of rotation (seems that getting the bolt almost all of the way through would be more important then).

Can you engage Half or more of the threads of the sleeve nut with it's Bolt?

It's a little hard to count threads inside the sleeve, but I would guess the answer to that is "not quite".

Is that a good rule of thumb?

I was more concerned that the rotation of the chainring will be bearing on just the last 1.5mm of that slightly-too-short sleeve, where a longer set would spread over 3.5mm (although as I said, I'm not sure whether the whole assembly transmits loads that way or just via clamping force on the faces of the rings / spider).

you can get longer bolts.

a triple , for a crank with a spacer , uses a longer bolt than a double.. or the outer 2.

then you use the double nut , to get the compression of all the parts

its maybe more comforting to you to have the bolt stick out past the nut, a bit.

the classic little spanner to grab the slots in the Nut edge should allow that .

Thanks - yeah, I think I'll just accept that I can't complete this today and wait on getting some longer bolts.