Hi, this is Kurt from Blue Jeans Cable. Let me flesh this out just a bit.
In principle there's nothing wrong with the general idea. However, ultrasonic welding has some odd wrinkles and that's where it becomes troublesome. I'll go into some detail here to explain.
First, the reason we really like ultrasonic welding for speaker connectors is that it gives the cable really high pull strength and very low contact resistance. Since a lot of people seem to want to pull terminated speaker cables through conduit, and since speaker wire is a high-current, low-impedance application where low resistance is an important consideration, it seemed to us like a great way to proceed.
Considerations are different at line-level. First, the inherent pull strength of the cable isn't as high, so enhancement to pull strength isn't as big an advantage. Second, resistance is no longer a major factor -- you've got a high-impedance circuit with very little current flowing.
Any ultrasonic welding application has to work off of custom-designed tooling, because otherwise what happens is that you're pumping a lot of energy into the weld and the pieces being welded are liable to fly apart (and if they slip out of the welder, the weld head welds itself to the anvil and you've got some serious downtime and expense). To get good, reliable welding, you need to be able to thoroughly immobilize the terminal you're welding to, AND you've got to be able to get an unyielding anvil underneath it (and that means under the terminal itself, not under the part as a whole) so that as the weld head closes on the piece, the piece can't recede away from the weld head. Speaker terminals are easy to deal with in these terms, because there aren't multiple contact points and because you don't have to worry about what all that ultrasonic energy load is going to do to a dielectric (things can shake apart -- indeed, they can literally explode -- when you pump all that ultrasonic energy into them).
Changing out tooling can be slow work, and alignment can be critical, so one wouldn't want to change it out on a regular basis. But if you don't, you need a new welder. The up-front cost of the welder is around $30K, so while this isn't a huge-industry-only kind of affair, it's not something one invests in without being pretty sure the yield will justify it.
RCA plugs are really difficult to weld. You'd need two completely separate sets of tooling, one for the center conductor and one for the shield, and that'd mean, in a small-batch environment like ours, two welders instead of one, so $60K up front. But getting an anvil underneath the center-pin contact on an RCA is itself a real mess-- there's just not a lot of clearance, in most designs, to work with, and you're talking about a very small weld head so small variance in dimensions of parts is liable to drive you a bit crazy. Meanwhile, you'd need to get clearance AROUND the center conductor for the shield weld. These things could probably be done, with some difficulty, but as with our speaker connectors, we'd have to have the plugs custom-manufactured for welding -- no existing connector would work. As a practical matter, when customers have unorthodox requests such as silver plating on connectors, that means we couldn't accommodate them -- we'd have to go to one substrate and one plating material and stick to it. That'd probably be gold-plated brass; I don't really know of anyone making copper connectors, and silver is not generally regarded as a good plating material. Honestly, if it were entirely up to me and not dependent on the market, I'd go to nickel plate instead, but we live in a world where nickel -- a lovely non-corroding plating material with exceptional tolerance for wear -- is looked down upon, for reasons which are entirely obscure to me other than the obvious "gold looks expensive" reason.
Star quad is not a good choice, in general, for RCA plugs. We will solder it into RCAs when people place custom requests for it, but the whole raison d'etre of star quad is common mode noise rejection, which you cannot get in unbalanced circuits. Capacitance is very high in star quad, and while that's a reasonable tradeoff in a low-signal-level, balanced, high-noise-rejection-required application like, say, a microphone cable, there's no corresponding benefit in an RCA cable because of the lack of CMNR. So while this is the sort of thing we build on request, we recommend against it and we wouldn't therefore ever offer it as a standard configuration -- similar considerations apply to other uses of mic cable in RCAs, though with conventional two-wire designs the capacitance is lower. If we don't offer it as a standard configuration, we're talking about a lot of setup cost in order to be able to do a very, very small volume of work.
Good old crimping works really quite well on RCAs. We're able to deliver 3 tons of force to our dies in our pneumatic presses; we get crimps that are hard to distinguish from welds. If you get out your dissecting microscope, tweezers and pliers and try to dissect a crimp pin and peel its body free after it's gone onto one of these center conductors, you can't do it -- it's effectively fused together. Indeed, people do welding of just that sort -- cold pressure welding -- on small parts, and the sole ingredient there is pressure. So, we've got a very high reliability termination method there, and very little reason to ditch it. Contrast that with speaker cable, where you've either got setscrews, compression screws, axial-compression style connectors, or soldering as competing methods. Soldering's arguably the best of those, but the sheer amount of heat required to solder large wire to high-mass connectors means insulation meltback (unless you go to Teflon just to avoid that, which is costly).
We very well might add ultrasonic welding on other products, at some point. We've had a lot of people asking us to make IEC power cords, for example, and if we were to do that, we very well might go to welding terminals on and then overmolding the connectors. But ultrasonic tends to be well-suited for free terminals without a lot of plastic parts or complicated angles, being attached to discrete wires. It's much more troublesome to do it with more complicated things like two- or multi-contact assemblies, and coaxes.
Kurt
Blue Jeans Cable
