point to point

[amo]

Hi-

a-msystems.com sells teflon coated silver wire in the following guages:

41
36
32
30
27
22

Although there is quite a price difference, it will not be great for a project requiring several feet or less. So based purely on performance, what is the best thickness for point to point work, within the signal carrying portion of the circuit? Should I just go with 22 or are there side effects with going too thick?

For the power supply portion, I am assuming copper is best . What guage should I use here?

And while we are on the toppic of point to point: Can someone recommend an adheseve to use for attaching parts to the perfboard? I know that resistors can run rather hot in some PASS designs, and perhaps regular hot-glue will just melt?

Thank you for your advice!

[Nelson Pass]

Hot glue will melt. Epoxy ought to be OK most of the time.

The gauge of the wire depends on the current. If I have only
a few mA going through the wire, then 22 is fine for me, but
otherwise for more current 14 or 16 gauge is more appropriate,
as with a power supply.

[amo]

Thank you!

[planet10]

In general the thinnest stuff you can work with. I use 30 guage (because that is the thinnest i have around). I work mostly in tubes which typically are lower current (and higher voltage).

dave

[Brett]

Originally posted by Nelson Pass
The gauge of the wire depends on the current. If I have only
a few mA going through the wire, then 22 is fine for me, but
otherwise for more current 14 or 16 gauge is more appropriate,
as with a power supply.

Originally posted by planet10
In general the thinnest stuff you can work with. I use 30 guage (because that is the thinnest i have around). I work mostly in tubes which typically are lower current (and higher voltage).

dave

I agree with both of you gents.

For signal I sue the smallest I can, and for PS, just what's neccessary, but again I work with tubes and the currents are quite low, usually <100mA.

Maybe try some of the smaller guage for signal, and just some decent copper for the PS.

[amo]

So what are the benefits of working with "as thin as possible" besides the cost? I am just trying to understand if there is a mathematical way to calculate the best wire thickness. By the way I am currently working on the BOSOZ, where current is rather high.

[planet10]

Originally posted by amo
So what are the benefits of working with "as thin as possible" besides the cost?

There are sonic benefits. Malcolm Hawksford showed this mathematically and listening tends to back him up.

The dialectric also plays a role -- air being the best practical (a vacuum is probably better yet).

I am just trying to understand if there is a mathematical way to calculate the best wire thickness.

Consider the length of wire to be a resistor... if the voltage drop becomes too high (or the power dissapation of the voltage drop threatens to turn your wire into a fuse) then it's too thin.

dave

[wuffwaff]

Amo,

current in BOSOZ is not very high. It´s around 40mA per Fet or 80mA for one channel. This can be done with any wire you can see with your bare eyes.

william

[Brett]

Originally posted by wuffwaff
Amo,

current in BOSOZ is not very high. It´s around 40mA per Fet or 80mA for one channel. This can be done with any wire you can see with your bare eyes.

william

Yep, just don't get silly small guages as they're a pig to work with.

[grataku]

Originally posted by Brett


Yep, just don't get silly small guages as they're a pig to work with.

So much so that even AW uses PCBs for his ultimate pre-amp after getting off on the thin wire for 10 pages in his book.
There are other things to consider when thinking about point to point, first and foremost: can you do it well enough to be worth doing?
This small wire business may be a good idea for interconnects, as far as the circuit a carefully laid out PBC in which the signal path is kept AShortAP is hard to beat.

W/O getting into details what does the mathematical threatment of thin wire show? Better S/N? Extended frequency response?