Soldering question

[catalin gramada]

Someone know a trick how to get decent "T" conections ? l tried all my best but the results are random. Someone told me to cool the proximity of the hot point keeping the wire in wet sponges...does it work ? Any useful tricks please ? I intend to isolate at the end the spots with an epoxy resin thin film. The conductor I used is pvc type...

[Greg Robinson]

Not sure what the application is here, but you'd have better luck soldering before twisting, as the twisted wire will still have inbuilt strain (tension/compression), especially in the insulation which will still be in the elastic deformation rather than plastic deformation region like the copper (the copper will still be in a transitional stage between elastic and plastic deformation too). The added heat on top of the strain trying to relieve itself is a recipe for the insulation to melt.

[catalin gramada]

Thanks. There is interesting point here: I used and use heating with a heater gun the twisted stranded wire before to get rid of insulation plastic memory and get a rigid twisted pair. Now I thought the outside heating is not necessary as time the wires are heated from inside. As it seen from pics the outside part of insulation still peeling off from remaining tension whilst inside of insulation remains melted stucked to the wire. Now is clear may get better results forming insulation from outside with heat first then soldering. When the wire get hot the melted plastic will not go anywhere just remaining to the wire into original position till cooling down . Is like forming insulation in two stages: outside first and inside then. As second benefit it will shrink less once the plastic will be formed before the cut. I think it is more easy to deal with the plastic insulation shaping then to try to keep the extremities cooled out of its plastic deformation. Thanks.

[mozz]

Teflon wire.

[dai h.]

hi, (I haven't done much 'hands-on' stuff in a long time, but) if I was trying to do this, I would try things like heatsinking the wire on both sides of the joint with those commercially available ones or maybe metal tweezers with a rubber band around it (the wet sponges idea to me doesn't seem to make much sense since I think you want to cool the wire because that is what you're heating--the heatsinking is to keep the heat from traveling but placed away from the joint (which of course needs to be heated)). So somehow heatsinking (maybe even cutting a bit of the insulation trying to keep the exposed wire area minimal), then tinning the spot where you're going to make the joint (maybe use thin gauge solder). Also the same with the (non-twisted) wires being connected. I don't think you can tin then bend (it breaks IME), so bend then tin. (I have no idea if this will work for you, but) I have tinned an end, then wicked it off (with solder wick)--this way all the strands barely stay together--then bend into a 'J', then tin. (Basically, my thinking is that the solder joint here has to be made as quickly as possible (about the time you can say, "poof!"--maybe one second). Also I might use a bit of kitchen alu foil to shield the insulation near the joint just in case the soldering iron tip 'wanders' as a safety.

oh, and maybe the Teflon idea sounds like it'll work (unless there is some of issue being too stiff or too pricey or whatever).

Also, maybe there is some other way to do it more easily with a small perfboard or something?

(what is this for anyway--some stringent noise issue?)

[catalin gramada]

Thanks. ... well the preforming insulation by heat it works. That means twist the wires. Spread the loops apart with a needle. Heat the loops insulation till slightly melt. Strip insulation . Soldering. Done.
I used to heat the wires with a lighter so get a little messy and not to proud to post a pic, but get my conclusion and my heat gun for precise job is on the way, show you later the result please. Just need to insulate with a drop of epoxy resin and colour with a permanent marker.

[catalin gramada]

...and sort of final result. It don't looks too bad even for a sloppy job...

[catalin gramada]

(what is this for anyway--some stringent noise issue?)

There is nothing more than heaters supply rail. If doing so the heavy currents are no more beared from socket to socket. You may keep the heavy currents on a twisted wires rail far away and only run small currents for each tube due to derivation wires. Why to make sloppy wiring conection when can be made almost perfect one with same effort ? When know how to deal with wire insulation is as simple as any hook conection. More than that is even simpler to deal at the socket with two wires instead of four specially for heavy awg.

[dai h.]

>>for heavy awg.

for preamp tube heaters, you don't need really heavy wires, do you? And for power tubes, the signals are bigger, so would seem less sensitive (and maybe further still with push-pull configuration).

The heater wiring in Suhr ('68?--John Suhr's version of a Marshall plexi) looks pretty ideal to me (seems to follow the recommendations in merlin's document).



Soldano SLO100 doesn't even use twisted wires (just a pair of thick solid core wires--though in a relatively recent interview he said he had to go to DC heaters due to issues with recent tubes). Basically (what I gather) you need to do the same sort of thing outside the tube as inside the tube (where a noise sensitive grid and noise source (heater) are going in (x2 in a 12A_7). That is use distance (try to keep it away from) and relative orientation (if you have to have them close try to cross at a right angle). Also, having the heater wires close (or hugging) the chassis can provide a bit of shielding (even if it isn't entirely enclosing the wires) which can provide a bit more isolation from the grid(s) with a bit of capacitive coupling.

What you did looks pretty cool, but if an easier way can be found I think that would be better, no? I guess I just have a bit of a concern of overelaboration without a worthwhile enough benefit. (I think about this for my own layouts and layout ideas (mostly with paper and pencil or a drawing program these days) also.)

[Helmholtz]

... well the preforming insulation by heat it works. That means twist the wires. Spread the loops apart with a needle. Heat the loops insulation till slightly melt. Strip insulation . Soldering. Done.
I used to heat the wires with a lighter so get a little messy and not to proud to post a pic, but get my conclusion and my heat gun for precise job is on the way, show you later the result please. Just need to insulate with a drop of epoxy resin and colour with a permanent marker.

I like that method.
It probably works because the pre-heating makes the PVC insulation separate from the wire, so less heat is transferred when soldering.

One problem with Teflon insulation is that it tends to "flow" under pressure (like with tighly twisted wires).

There might be less problematic wire insulation types, maybe polyurethane.

[catalin gramada]

I guess I just have a bit of a concern of overelaboration without a worthwhile enough benefit. (I think about this for my own layouts and layout ideas (mostly with paper and pencil or a drawing program these days) also.)

Hey. There is a benefit as time you're not forced to design the layout in a particular order just because to avoid large currents from sensitive areas. We may cross the wires as much we want, the code of good practice said to separate large by small currents. Doing a rail you may supply the first high gain stage than directly the power tubes, it doesn't matter, the currents at the sockets are isolated. You get the liberty to configure the heater supply scheme in any mode you wish. Not the case of supply in a string.
I particular I.m also curious to know how it make a difference between a good wired socket running couple of Amps , or a mediocre wired spcket running 1Amp or less.

[dai h.]

Hey. There is a benefit as time you're not forced to design the layout in a particular order just because to avoid large currents from sensitive areas. We may cross the wires as much we want, the code of good practice said to separate large by small currents. Doing a rail you may supply the first high gain stage than directly the power tubes, it doesn't matter, the currents at the sockets are isolated. You get the liberty to configure the heater supply scheme in any mode you wish. Not the case of supply in a string.

so I gather you are talking about separating the higher current and lower current impedances (paths) for the pre and power tubes? (IIRC) I've seen some amp modder do that here (some Japanese site) and claiming some sort of improvement, but I have no idea if that is the case (or at least I can't think of one). Hmm... more consistent heater voltage for the preamp? Wonder what that does exactly.

I particular I.m also curious to know how it make a difference between a good wired socket running couple of Amps , or a mediocre wired spcket running 1Amp or less.

do you mean induced noise (so I guess more if higher current) from the heater (to the grid(s))?