Wire types, uses and sources...

[romberg]

As R.G. already mentioned, the gauge of wire is mostly dependent on the current it is going to carry. And the only substantial currents in most guitar amplifiers are in the filament and OT secondary circuits. In both of these cases, presumably, the transformer leads have already been correctly selected with the proper gauge wire to handle the maximum current the transformer can deal with anyway. And in most cases there is enough lead length to just use that wire for all parts of the amp where high currents matter.

In the case of the output transformer secondary it is just wired up straight to an impedance switch and or jack. And you are good to go.

The filament current is almost highest (parallel wired) on the runs from the power transformer to the output tubes. So, again you can often just use the leads from the transformer to take care of a pair of output tubes. When wired in parallel the current drops as you go down the string. For example using an amp with 2 el34s and three 12ax7s the current on the first run would be about 1.5 * 2 + 3 * 0.3 = 3.9A. And the current between the 1st and 2nd power tube would be 2.4. And to the preamp tubes 0.9A. So, you can easily get away with using 20 gauge for everything after the power tubes.

It is also interesting to note that the number of conductors does have an effect on the amount of current a wire can safely carry. Charts like this will point that out:

https://www.engineeringtoolbox.com/wire ... d_419.html

For example note that 20 gauge single core wire is good for 6A. But 20 gauge 7 strand wire is only good for 3.5A. Again this only matters if you intend to be pushing currents around these limits. Also note that underneath the table there is a mention of correction factors for temperatures over 30 degrees C (86 F). The insides of an amp chassis more often than not are going to be in these temperature ranges. In the range (88F - 104F) which is barely above body temperature, the max currents are de-rated by 0.82. So, our 20 gauge single core is only good for 4.9A and the 7 strand 2.9A. At 113F (which does not seem out of the question in a tube amp in a hot environment) the rating is down to 0.58. Which drops the max for 20 gauge to 3.5A single and 2A 7 strand.

I mention the temperature stuff because I have seen some amps wired with like 22 gauge stranded on some power tube heater runs and the insulation became extra crispy . In general I've always been fine with using transformer leads on the high current stuff already mentioned (probably 16 gauge) and 20 gauge single core everywhere else.

Mike

[martin manning]

I've examined several tables like the one linked above, and it's unclear as to what is meant by multi-conductor. This one mentions that the diameter of a stranded wire will be larger than a solid wire of the same gauge, implying that gauge is representative of the conductor cross section, and therefore the I^2R heating and current capacity. In others it seems this correction is referring to a multi-conductor cable in which several insulated wires are bundled together. That would surely reduce the current capacity for a given insulation temperature limit, and the derate of ~50% shown for a four- to six-conductor cable seems reasonable. It does not seem reasonable that a single wire having the same conductor area made up of six strands would have half the capacity of a solid wire.

[M Fowler]

Your going to fine that many different types of wire will be needed on your electronics work bench if you start and continue to build tube amps.

You will want cloth covered solid wire 18awg for Fender type amps on filaments and 20awg solid cloth covered for rest of amp. Many vendors like Mojotone. com and etc.

Marshall you'll find yourself wanting some pink wire so your amp looks like a real old tube Marshall amp so I then go to ValveStorm.com for a wire pack.

Can't deny the need for wire from Steve at Apexjr.com for coaxial cable and silicone covered wire and he has a vast supply of NOS carbon comp resistors.

Also, CEDist.com for dealers and AES.com for non dealers has vast supply of wire and all else.

Wattstubeaudio. com has wire and parts but shipping is really slow.

So many wire sources of wire I have not mentioned.

Mark

[labb]

Right or wrong when I look for wire size amp rating for wiring in chassis I go to the ARRL Radio Hand book..It rates 20 ga. as being good for 11 amps.

[martin manning]

Right or wrong when I look for wire size amp rating for wiring in chassis I go to the ARRL Radio Hand book..It rates 20 ga. as being good for 11 amps.

That's consistent with this table: https://www.powerstream.com/Wire_Size.htm It includes the following note (but does not specify what ambient temperature and insulation temperature rating is assumed):

"The following chart is a guideline of ampacity or copper wire current carrying capacity following the Handbook of Electronic Tables and Formulas for American Wire Gauge. As you might guess, the rated ampacities are just a rule of thumb. In careful engineering the voltage drop, insulation temperature limit, thickness, thermal conductivity, and air convection and temperature should all be taken into account. The Maximum Amps for Power Transmission uses the 700 circular mils per amp rule, which is very very conservative. The Maximum Amps for Chassis Wiring is also a conservative rating, but is meant for wiring in air, and not in a bundle."

Emphasis on "the rated ampacities are just a rule of thumb." The actual temperature at the conductor-insulation interface, which is critical to insulation breakdown, is a complex heat transfer problem with widely varying boundary conditions. As a side note, don't call this Thermodynamics; that is something else again ;^)

As pointed out above, in filament wiring the first power tube doesn't count (only the transformer leads carry that current), and it's only the next few inches of wire that carry the whole of the remaining tube filament current. In a 100W marshal with 4x EL34's that will be 5.4A, and in a Twin Reverb with 4x 6L6 it will be 4.5A. I've been fine using AWG22 UL 1015 appliance wire (like Valvestorm sells) for everything. This is top-coated wire with PVC insulated rated at 105C. Like R.G., I don't care what it looks like. I'm more interested in how it strips, routes, and solders, in addition to having a full range of colors available for ease of tracing.

Twisting filament wiring is a good idea to reduce hum, but twisting it very tightly will increase the length of the filament string (increasing resistance) and restrict the exposed area of insulation available to dissipate heat (not to mention over-stressing the wire). I think a good (faster, easier) plan might be to use closely-paralleled (untwisted) solid conductor wire running just above the power tube sockets (in a variation on the Fender scheme), and a twisted pair for the rest of the string.

[romberg]

I've examined several tables like the one linked above, and it's unclear as to what is meant by multi-conductor. This one mentions that the diameter of a stranded wire will be larger than a solid wire of the same gauge, implying that gauge is representative of the conductor cross section, and therefore the I^2R heating and current capacity. In others it seems this correction is referring to a multi-conductor cable in which several insulated wires are bundled together. That would surely reduce the current capacity for a given insulation temperature limit, and the derate of ~50% shown for a four- to six-conductor cable seems reasonable. It does not seem reasonable that a single wire having the same conductor area made up of six strands would have half the capacity of a solid wire.

Yea. Thinking about it, I agree with you. I was interpreting that table incorrectly. The number of conductors it refers to must be multiple bound together wires rather than strands in a single conductor.

[johnnyreece]

Bob S turned me on to Belden pre-tinned stranded wire (He thought the name was J TEW). It's PVC, but rated up to 105 degrees C. Stays put like solid core without the breakage issues. Unfortunately, I didn't get to talk to him about his source before he passed...Miss you, Bob!

[martin manning]

That is the top coated UL 1015 wire (600V, 105C) mentioned above, which is available from Valvestorm.

[johnnyreece]

That is the top coated UL 1015 wire (600V, 105C) mentioned above, which is available from Valvestorm.

Thanks, Martin! Well, toss my vote in for that stuff. It's really great!