Tech Square

Mmmm ...

157.128.218.206

andyr

In Reply to: RE: Twist. posted by Ugly on December 10, 2008 at 14:12:51

If the wires you are twisting are input or output signal wires ('hot' & 'ground') then yes. :-))

However, my OP was about what to do with the '+ve' and '-ve' DC rail wires, from PS to PCB. I can't see why twisting these would provide any benefit.

Regards,

Andy

Follow Ups:

RE: Same reason.... - Ugly 11:08:27 12/11/08 (3)
Picture the scenario: a power supply putting out two rails, two conductors stretching to some finite impedance load. Put all of these items into the shape of a circle and you have just created a great little loop antena. Note that a circle will be the most efficient antenna (bad), but any loop area results in some antenna gain. Twisting in this case is merely a method of reducing antenna loop area. The antena loop in this case is the loop from the power supply rails through the load (amp?) itself. Generally, the more loop area there is the more noise power recieved by this loop. Twisting the wires reduces this loop area to almost nothing ensuring the poorest reception of ambient noise possible.

Just as you don't want noise on your I/O you do not want it in your load (amp?) rail voltages. You can bet that ambient noise fields will not discriminate between antennae intended to be I/O lines vs antennae intended to be power supply rails.

Presumably these rail voltage wires feed something such as an amplifier circuit, no? All amps will have some finite noise rejection capacity, some better than others. Key word here being finite. Any noise that can be removed from amp supply WILL result in a better amp SNR.
- Thank you, Ugly ... - andyr 11:36:48 12/11/08 (2)
  great explanation - I can understand your proposition. :-))
  
  However, the only thing still unclear to me is that while I can see how twisting, say, input signal wires together ('hot' & 'ground') will reduce noise pickup ... this is because the 'hot' wire carrying the ac music signal is twisted against ground.
  
  Yes, I can see (as you explained) that certainly the PS rail wires plus the amp PCB form a loop antenna but the wires being twisted are not a 'hot' & a 'ground' ... they're a '+' & '-' Xv pair?
  
  Or is that irrelevant - whatever "signal" they are carrying, twisting the wires together reduces the antenna loop area?
  
  Is it better in fact to use the 0v wire as well and braid the 3 together?
  
  Thanks,
  
  Andy
  - RE: You're welcome andyr. IIRC I owed you one for my phono inner tube. - Ugly 15:00:48 12/11/08 (1)
    In the case of bipolar amplifier rails the signals could be treated as differential. An example: consider a + and - rail amp loop as we have described with fairly large loop area. Say a differential noise spike is induced on these rails. Being differential means that the noise does the oposite on each rail...in other words differential noise is essentially reducing or raising the rail voltage from nominal whenever it is present. For a 10V rail to rail system 2V of diff noise induced to the rails results in not 10V rail to rail but either 6V r-r or 14V r-r depending on the direction of the noise. Depending on amp topology this noise will be felt at the output on at least some level.
    
    "Or is that irrelevant - whatever "signal" they are carrying, twisting the wires together reduces the antenna loop area?"
    
    Well, you have to be careful a bit careful about what you twist. Unrelated AC signals twisted together could be bad due to crosstalk (capacitive/inductive coupling). I would say always twist differential pair signals to each other or single ended signals with ground. Twisting is in lieu of and kind of the poor man's version of the more technically proper controlled impedance design which is probably overkill and too expensive for an audio amp.
    
    While it does always reduce the antenna efficiency to reduce the loop area, one must consider potential tradeoffs such as crosstalk. With diff pairs and single ended with ground the close coupling is actually beneficial where other types of signals may suffer signal integrity issues from twisting.
    
    "Is it better in fact to use the 0v wire as well and braid the 3 together?"
    
    It really depends how the amp circuit is realized, ie fully diff, single ended, or both. It's pretty difficult to come up with a generalization here since the way products are implemented varies so much for example in how grounding and shielding is accomplished. I think it's fairly safe to say anytime the 0 volt line is actually used as part of the amp circuit ie as a reference/current return path it should be in there too. However, for example, a fully diff circuit probably won't even use it for anything related to the amp circuit and it could probably be left out if no shielding is necessary.
    
    I hope this helps. It's kinda to try and talk about this stuff in a very general manner with someone where you really don't know their technical background.
    - What goes around, comes around! :-)) ... - andyr 19:00:06 12/11/08 (0)
      Hi Ugly,
      
      Thanks for your detailed response - and just FYI, I don't have an EE background, although I can understand basic concepts (when explained like you have done! :-)) ) like a larger loop antenna picking up more noise then a smaller one.
      
      My amps do use the 0v line as the current return path ... so I will proceed to braid the 0v wire in with the '+' & '-' wires.
      
      Re. the bicycle inner tube ... I remember posting about that but I had forgotten who it was to! :-)) If I can suggest one improvement - it may not isolate any better but it should have higher WAF (and it's what I have under my own phono stage and preamp).
      
      Get a wood-turner (or a mate with a lathe) to create 4 annuli out of 1" thick MDF. ID of these annuli should be what just encloses 3 squash balls (racquet balls to you) and stops them moving around. OD is 1" bigger. Spray them gloss black.
      
      I'm not sure if I suggested putting a stone slab on top of the inner tube (with the component on top of this slab) or you simply have the tube under your phono stage? Anyway, I position the 4 annuli just in from the 4 corners of a 1" thick slab of slate (which weighs about 20lb), put 3 squash balls in each annulus, put the slate slab on the squash balls and then put my phono stage on top of that. :-))
      
      Regards,
      
      Andy

Post a Followup:

FAQ

Post a Message!

Forgot Password?

Moniker (Username):
Password (Optional):
	Remember my Moniker & Password (What's this?) Eat Me
E-Mail (Optional):
Subject:

Message:	(Posts are subject to Content Rules)

Optional Link URL:
Optional Link Title:
Optional Image URL:
Upload Image:
E-mail Replies:	Automagically notify you when someone responds.

To view your new posting or follow-up, click on the RELOAD or REFRESH button on your browser.

This post is made possible by the generous support of people like you and our sponsors:

Schiit Audio