[andyr]

Hi,

I'd like to get your advice ... the DC leads that go from a PS to a PCB - should they be twisted (to reduce inductance) or doesn't this matter for a DC power feed?  (I know that it is the right thing to do for AC wires.)


Thanks,

Andy

[cheap-Jack]

HI.

... the DC leads that go from a PS to a PCB - should they be twisted (to reduce inductance) or doesn't this matter for a DC power feed?  (I know that it is the right thing to do for AC wires.

Is it a Class A audio amp or what?

c-J

[Speedskater]

Yes, twisted.
Something like those super high capacitance speaker cables is great.

[Folsom]

Why twist them? Technically it'll add a little extra heat.

AC twisting has a minor noise rejection, DC doesn't work like that.

Do whatever looks nice and isn't a struggle.

[randytsuch]

Twist them, heat is not a problem

Reduces noise.

[Speedskater]

All (multi-conductor) cables should be twisted,  it reduces pick-up (and transmission) of interference.  That goes for AC power, DC power, interconnects, digital and speaker cables. The one exception is co-ax cables.
Also in DC supply cables, twisting reduces self-induction which is good. We want the DC supply source to appear as a zero Ohm impedance at both DC and to AC to as high a radio frequency as possible.

[Folsom]

You are going to have to explain how DC can reject noise with twisting. Considering there is no sine wave in the power section, I'm having trouble seeing it.

[*Scotty*]

Another vote for twist them. The DC rail is supplying power to a circuit that may operate to frequencies past 20kHz. The circuit is is limited by the AC impedance of the power supply, added inductance in the power supply wiring raises the AC impedance at high frequencies. In addition the combination of the wiring's inductance, resistance and the capacitance in the power supply can form a tank circuit that can resonate at some frequencies which will cause the power supply to ring. The best practice is to treat this aspect of the amplifier as though the amplifier was operating at RF frequencies. The more current there is in the rails the bigger the problems that may be encountered if steps are not take to keep power supply wiring short and minimize its inductance.
Scotty

[andyr]

Another vote for twist them. The DC rail is supplying power to a circuit that may operate to frequencies past 20kHz. The circuit is is limited by the AC impedance of the power supply, added inductance in the power supply wiring raises the AC impedance at high frequencies. In addition the combination of the wiring's inductance, resistance and the capacitance in the power supply can form a tank circuit that can resonate at some frequencies which will cause the power supply to ring. The best practice is to treat this aspect of the amplifier as though the amplifier was operating at RF frequencies. The more current there is in the rails the bigger the problems that may be encountered if steps are not take to keep power supply wiring short and minimize its inductance.
Scotty

Thanks, Scotty.  The device is actually a phono stage - so drawing a max of 40ma in each channel.  But yes, it does operate waaay past 20KHz! 

I'm surprised there's so much disparity in the advice given but the fact that twisting will reduce inductance in the wiring seems to me to be the key argument. 

So twist I will. 


Thanks,

Andy

[jtwrace]

Always twist

[*Scotty*]

There is a paucity of experience and no shortage mis-information online.
Scotty

[Folsom]

Scotty you are saying to twist them in order to cancel inducted AC signal?

[andyr]

Scotty you are saying to twist them in order to cancel inducted AC signal?

I took his statement to mean "Twist them to reduce inductance."  (As the wiring's inductance & resistance, and the capacitance in the power supply, can combine to form a tank circuit that can resonate at some frequencies.)


Regards,

Andy

[Folsom]

He said treat it like it operates in RF frequencies (AC).

Inductance goes up with more close proximity?

[andyr]

He said treat it like it operates in RF frequencies (AC).

Inductance goes up with more close proximity?

I don't believe so ... capacitance goes up with proximity, inductance goes down.

Regards,

Andy

[Folsom]

Capacitance goes up by electromagnetic fields passing through material that tries to act like a dialectic.

Inductors look like this.  They can hold capacitive charges. The more cross sectional area the more inductance. But reduction in length does mean less inductance.

[andyr]

Capacitance goes up by electromagnetic fields passing through material that tries to act like a dialectic.

Inductors look like this.  They can hold capacitive charges. The more cross sectional area the more inductance. But reduction in length does mean less inductance.

Sure, DoS - that is an inductor! 

But 2 wires twisted together don't look anything like that.  If I can explain my point about inductance going down with proximity of 2 wires vs. capacitance going up, a bit further:
* with speaker cables, there can be high currents flowing.
* inductance acts as a "brake" on current - which is obviously a bad thing.
* so extremely low inductance is the best for speaker cables.
* hence, some mfrs introduced braided cables a few decades ago.
* however, the resulting high capacitance can drive certain power amps to self-destructive oscillation - so some amps need to use cables with low capacitance ...  and, hence, higher-than-ideal inductance.

Old Naim amps were like this - they needed to have a certain amount of inductance on their output terminals - which was provided by 5m of Naim NAC-5 speaker cable.

Regards,

Andy