I would like to know what kind of internal wiring for power distribution I can use inside my tube pre-amplifier and DAC.
I've seen many amplifiers with twisted pair wires between powersupply and the amplifier PCB.
But is this the best way?
Is it better to use shielded twisted pair cables or shielded cables with 2 or more wires inside? (ground and +)
For (tube) amplifiers a wooden case is the best because there is no influence on current (signal). But is there a risk, when you use shielded power cables, that you introduce current loops inside the shielding? Does a shielding around cables act like a metal housing?
I've seen many amplifiers with twisted pair wires between powersupply and the amplifier PCB.
But is this the best way?
Is it better to use shielded twisted pair cables or shielded cables with 2 or more wires inside? (ground and +)
For (tube) amplifiers a wooden case is the best because there is no influence on current (signal). But is there a risk, when you use shielded power cables, that you introduce current loops inside the shielding? Does a shielding around cables act like a metal housing?
Jonasa,
I think that for power-wiring, layout of wiring is more important than type of shielding on the wire. Most folk (I believe) simply use single-core mutli-strand wire that has a simple plastic insulation. 18 gauge wire seems to work pretty well as you want it thin enough to be flexible but retain some strength while it is worked. The mA power-load of most preamps means that light gauge wire can likely be used. One place where the wiring can be critical is for the heater/filaments. These tend to carry higher currents even though the voltage is low (e.g. 0.6A at 6.3VAC for a 6SN7). Furthermore, this is AC and should be placed away from sensitive components. I used solid-core 16 gauge wiring, taking two long strands and winding them together using an electric drill. I then route this between trafo and filaments along the edge of my chassis.
Between PSU and PCB, I use two 18gauge wires that are twisted to form a pair. I don't know whether this has an effect, but I expect that the twisting may result in some common-mode noise rejection. Maybe someone else can verify this.
As for the chassis, I use wood for the sides with a steel top-plate. Steel is pretty easy to work with and you only require a fairly thin steel to get sufficient strength. However, chokes and trafo mounted to the steel can cause magnetic interactions, which may add hum. When I rebuild, I'll use 3mm thick aluminum.
You can use a shielded wiring for your signal wiring and this is quite trivial if you are only using a single input and single output. I use Cardas microphone cable with a twisted pair inside a foil shield. The twisted pair work for the signal +ve and gnd, while I use the foil and connect only at one point. If I remember, the foil is connected only at the input signal GND and output signal GND for the input and output respectively. The shield acts to pick-up noise and carries it to GND. Connecting the shield at both ends results in an antenna (I think) and should be avoided.
Charlie
I think that for power-wiring, layout of wiring is more important than type of shielding on the wire. Most folk (I believe) simply use single-core mutli-strand wire that has a simple plastic insulation. 18 gauge wire seems to work pretty well as you want it thin enough to be flexible but retain some strength while it is worked. The mA power-load of most preamps means that light gauge wire can likely be used. One place where the wiring can be critical is for the heater/filaments. These tend to carry higher currents even though the voltage is low (e.g. 0.6A at 6.3VAC for a 6SN7). Furthermore, this is AC and should be placed away from sensitive components. I used solid-core 16 gauge wiring, taking two long strands and winding them together using an electric drill. I then route this between trafo and filaments along the edge of my chassis.
Between PSU and PCB, I use two 18gauge wires that are twisted to form a pair. I don't know whether this has an effect, but I expect that the twisting may result in some common-mode noise rejection. Maybe someone else can verify this.
As for the chassis, I use wood for the sides with a steel top-plate. Steel is pretty easy to work with and you only require a fairly thin steel to get sufficient strength. However, chokes and trafo mounted to the steel can cause magnetic interactions, which may add hum. When I rebuild, I'll use 3mm thick aluminum.
You can use a shielded wiring for your signal wiring and this is quite trivial if you are only using a single input and single output. I use Cardas microphone cable with a twisted pair inside a foil shield. The twisted pair work for the signal +ve and gnd, while I use the foil and connect only at one point. If I remember, the foil is connected only at the input signal GND and output signal GND for the input and output respectively. The shield acts to pick-up noise and carries it to GND. Connecting the shield at both ends results in an antenna (I think) and should be avoided.
Charlie
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.