The only thing we're trying to accomplish by twisting the heater wires together is to reduce the radiated magnetic field. Consider that the heater circuit is a loop (a single turn) and that one of the parameters that governs the magnetic flux is the area of the loop. By twisting the wires together, we're just reducing the area of the loop.
Twists per inch and crossover angles become significant at much higher frequencies- especially so when we look at a twisted pair as a transmission line (Radio Frequencies). It's really not critical for 60 Hz.- all we're trying to do is to keep the two heater wires close together as they make their way around the chassis. Using 18 gauge untwisted zip cord (lamp cord) would likely be as effective- but would look a bit strange...!
Twists per inch and crossover angles become significant at much higher frequencies- especially so when we look at a twisted pair as a transmission line (Radio Frequencies). It's really not critical for 60 Hz.- all we're trying to do is to keep the two heater wires close together as they make their way around the chassis. Using 18 gauge untwisted zip cord (lamp cord) would likely be as effective- but would look a bit strange...!
Is there anything to gain by twisting signal wires? eg. phono input, tape out, main out etc....
Or perhaps twisting B+ in a tube pre?
Is there any advantage to case grounding a shield at one end for any of these signal?
Or perhaps twisting B+ in a tube pre?
Is there any advantage to case grounding a shield at one end for any of these signal?
In the Paragon pre I have, the heater supply and main B+ come from the same transformer. I take it that this is a no-no? What problems will I encounter by not separating these?
Actually, I won't use the same transformer for B+ and heaters, and most of my amps are AC heated. Separate transformers for heaters are very effective for keeping rectifier hash out of heater circuits. By using a cheap Hammond split bobbin, you greatly minimize the chance of RF creeping into AC heated circuitry. Ask me how I know.
Additionally, the typical ceramic caps right at the heater pins to chassis.
Most of the designs where I have had hum pickup has not been from AC heating (DHT and IDHT), but from magnetics. The choke seems to be the biggest culprit, as it has to deal with nonlinear voltages and spits out a bunch of B field that gets picked up by either a tube or signal transformer.
Agreed that TPI is less of a concern than just getting the two wires tightly bound together. However, there have been tests made (I'd have to track down the source) that confirm zip cord as parallel conductors is not as effective against RF pickup as a twisted pair. Has to do with the continously changing position of each conductor relative to the direction of the source of noise. With parallel conductors, one conductor will consistently be arranged closer to the noise source from the other, so there is less attenuation. Might only be a few dB, but the difference is measureable. Go for twisted if at all possible.
Please note,
You can use the same power Tx for B+ and heater. It depends why you are splitting the heater and what tubes are in use!
The problem is with the rectification if you use DC heaters and volt drop!
So you may need a higher voltage heater supply. "Not always". This depends on the loading of supply and smoothing capacitance.
Twisted pair signal cables can be more effective than screen. This depends on impedance of O/P or input that it connects to.
Screens can be grounded at one end to prevent earth loops.
Regards
M. Gregg
Last edited:
This is only half of the story. In addition, twisting ensures that induction from one piece of wire is cancelled by the opposite magnetic field from an inch further along. The cancellation is not perfect, but it helps. This is why close parallel wires are not as good as twisted wires, but both are better than random wires.
Just for interest,
I think that the insulation "dielectric" is very important.
Just my experience rubber mic cable is very poor. "twisted or not".(As a signal cable)
Regards
M. Gregg
I think that the insulation "dielectric" is very important.
Just my experience rubber mic cable is very poor. "twisted or not".(As a signal cable)
Regards
M. Gregg
It depends on cost!
PTFE silver coated copper either flex or single core. The flex tends to give more detail in the music. If you do a DIY twist this is better than screen, however if you are using "very" low level signals you may need to cover the twist with tin foil use a drain wire and cover with heat shrink!
Regards
M. Gregg
It's the supply voltage that was the reason I used separate transformers in my own builds. The transformers I had to hand couldn't give me enough voltage for the job from their 6.3vac windings.
So why did I go DC for these projects? Just to see whether I could do it and whether it was worth the hassle. 😀 😉
To be honest, AC heaters with a humdinger are just as quiet in my own builds as the regulated DC builds. With the DC builds, I had to think carefully about the layouts and where I mounted the regulators, both in terms of noise and heatsinking. With the AC builds, I had to pay more attention to lead dress and cable routing. But the AC builds were "easier".
- Status
- Not open for further replies.