Nils -- First, thanks for the puff, but I'm by no means an "expert" on any of this. I'm just, like a lot of the rest of us, a long-time CBX'er (and some other bikes before), and an even longer-time amateur electronics diddler (ha) -- since I was a kid. Just a whole lot of curiosity that drives me to try to learn about the things I diddle with. Lots of mistakes involved in those learning processes.
I do agree with you about the tapes -- I wrap bare splices with electrical (insulating) tape -- 3M Super 33 is a great insulator, 3MSuper 88 also, for extra scuff protection (it's thicker). But PVC tape and/or tubing is as you note is tougher against abrasion and also add great heat insulation. Electrical tape alone can lead to problems.
On the soldering. Again, I'm not an expert, so here's my amateur 2 cents worth. A lot of controversy here. I've read academic papers where they lab-tested soldered vs crimped electrical joints, primarily for electrical conduction in low voltage applications (this had more to do with electronic apps than anything else), and others on twisted vs bunched wire conduction, etc etc. In the end there was no consensus on "do" or "don't do" -- just best practices in particular applications. What I gleaned from this is that properly soldered joints do increase conduction, but just a bit. In general, the main thing I've learned about it is that one solders a joint to increase strength more than conduction. But as with anything, here's the fly in the ointment.
Soldering is a learned skill, and it takes lots of practice (and failures) to get reasonably good at it. A poorly soldered joint can reduce conduction, by being either too cold or too hot, frying or hardening insulation, oxidizing the wire strands (introducing resistance at the joint) and so on. A properly soldered joint is made by knowing/using the right heat with the soldering tool (an adjustable tool is helpful, especially for novices). The heat range is related to both wire size (here all copper wire of course) and heat tolerance of the copper connector (mostly wire terminals on our bikes). For most of us, this means experimenting with the tool and the materials before tackling the actual job. As with any soldering, heat the wire and terminal to the point where the solder will flow freely (thus the experimenting) when just touched to the copper, remove the heat, flow the solder -- just a very small bit -- too much makes a poor bulky joint. The objective is to get the solder to coat -- "tin" -- the wire strands and the surface of the terminal in contact with the strands -- no more. This adds both small conduction improvement through the joint and some strength to the joint. By strength, I mean it helps prevent the wire strands from loosening in the crimp over time, something that can happen if the mechanical crimp is not quite as well made as it should be.
Solder or not solder? I'm in the middle on this (sorry -- no controversy). Properly crimped joints are, in my opinion, just as effective as soldered joints, if we're talking about the terminals on a CBX harness. If properly crimped, the wire bunch is compressed so that the strands are all in optimum contact with one another for good conduction and in good circumferential contact with the terminal itself. A properly crimped joint also will not likely see the wires loosen in the joint over time. A good example -- the original CBX harnesses. Beautiful work there, in general. We mostly have problems with poor wire maintenance -- some physical abuse too -- where the terminals corrode, introducing resistance.
The other, more prominent culprit in volt drop at terminals, is the connection between the male and female sections joining within the couplers. There, the contact points for conduction are very small surfaces between the "rolls," as they're technically called -- the little semi-round parts on the female terminals -- and the blades on the male terminals. The very thin bottom of the tips of the rolls and the flats of the blades are the primary points where conduction occurs. Those conduction points are visible as two short score marks on the blades. Over time, and with numerous disconnects, the physical orientation of the roll tips with the blades can and does loosen up, sometimes to the point where only one roll is making contact (typically), or poor maintenance leads to oxidation on the blades and rolls, reducing contact and introducing resistance, or both rolls are only minimally in contact with the blade. I've even had brand new terminals with miss-aligned rolls, so that there is poor conduction contact from the get-go.
So the bottom line, for me: I agree with your sage, with some caveats (no "rules" are ever absolute). IF you know how to properly crimp a terminal, and use a good crimping tool, no need to solder. IF you don't know much about soldering, either learn, and practice, or do not solder. If you can crimp pretty confidently and solder -- then why not? It's extra insurance. A good solder job will not degrade the joint. As for flexing the whole thing, a good solder job should never see solder running up the wire strands past the point where insulation is crimped into the retainer at the end of the terminal, and so the wire remains flexible all the way up into that point, within the insulation. A bad solder job will, yes, push too much heat up into that insulation, drawing solder with it, likely damaging it and stiffening the soldered wire inside and beyond (however far it travels). On a good job, though, with our terminals, no wire strands should ever BE exposed (out of insulation) beyond the terminal. If it is, it can be damaged, yes -- soldered or not.
The photo below is something I did for an article on switches, using an EBay junk harness for one of the illustrations. It might help visualize some of the points here.
You do not have the required permissions to view the files attached to this post.
