Thank you DDK,
Now you are talking !
You should not have any difficulties in setting and dressing those knots, if you start from the loose thief knot as shown in the attached picture.( Notice that I have placed the yellow-red marble into the 3 rd opening, for the A3. I have followed this simple notation for all those bends. For the B series, one has to pass the working end(s) from the other side.) May be I should have posted this picture from the beginning, to save us from exchanging meaningless bits and bytes.
Regarding the "jamming" issue, I am the least qualified person to speak about it...I do not understand if the "jamming" of a knot is a
quality that exists or not, or a
quantity that varies continuously, in relation to the loading and/or the slippage characteristics of the knotted material. ( There might also be a relation between the jamming of a knot, and the resistance of the rope material to
compression...) I, too, have made the vague observation that
pretzel shaped interlocked-overhand-knot bends have a greater jamming tendency than
"8" shaped ones. ( So, there is already a difference between the members of the class of the overhand-based bends, before the difference between the overhand-based, and the figure 8 -based bends.) I can only guess that the more effective compression of the encircled central nub, achieved by a pretzel shaped overhand knot, is what makes the difference.
Go on, try the rest of those bends. You could also compare them to the differently re tucked thief knots presented at (1), that are not so prove to jamming, I believe.
The main problem that made me post those bends of this thread remains : How, on earth, the "clogging" effect is cancelled so effectively, however we retuck the thief knot ? We pass the working end(s) from whatever opening, very different bends are produced, but anything we do makes this "clogging effect" disappear ! When I will understand why is this so, I would have understood something about this effect.
1) http://igkt.net/sm/index.php?topic=2085.0