3 Phase Frequency On Comms Cable

[Bruspark]

I think there are several blind alleys in the question here

You have 3 perfectly good 50Hz waveforms in 3 phase, but summed equally (or coupled to another cable) they come to close to zero (hence low Neutral current in 3PH).  In practice different load or different coupling means the induced sum is a mess of Harmonics - just like the Neutral.   Your measured frequency may tie in to a particular lighting fitting or PC power supply, or Ethernet frequency but as I said good luck deducing that from afar.

Most test gear costing less than £20k is not going to make sense of the harmonic mess . .  

Have you actually tried posting a listener by the intercom and going round turning things off ??

 

[Sidewinder]

No deliberate blind alleys Bru, just a confusing mess of interference, and a deliberate lack of some information from me.

There is IMHO a good reason for this retention of information.

As far as turning things off go, I did monitor the install during the lunch break.

There is no option of turning all the loads off.

BTW, I'm NOT shooting you down, I appreciate your input, it is just that this is a bit of a complex scenario.

I have hung not quite £20k's worth of kit off it, but quite a few £k's.

I'm not sure about the Zero of the 3ph when coupled especially in a defined time period wrt an fft, hence my comments & questions and no one has come back with the maths/science to show that a 3ph wave form over a fixed time period when superimposed onto such a comms line will give an fft of 50Hz or a multiple there of.

 

[Bruspark]

I seem to recall long time periods are needed for FFT analysis where the fundamentals you are looking for are down in the noise and distortion so it should be fine at 50Hz.  Really ugly switch mode powersupply type loads though are terribly difficult to analyse as frequency or vectors as they are distorted and often change in switching frequency.   One point about shielded Ethernet cable . .   if using Ethernet cable of other than its 'normal purpose'  you may need to ensure that the pairs of cores you select are indeed one or more of the twisted pairs and not one from one pair and one from another as the 'unscreened immunity' of the twisted pair will then be completely lost.  When connecting directly to equipment it is easy to see if your stripped cable is being connected in pairs but when it is already installed or untwisted /terminated in some other way this detail can get lost

 

[Marvo]

Have you proved the noise is being induced into the cabling and not being induced through the power supply units? I'd start by trying a couple of known-good PSU's and terminating to earth all the sheaths/shields on the cables that have them. Nuetral-Earth voltage on the supply at the PSU's might also be worth checking especially if they're switchmode.

 

[Sidewinder]

The kit I used for the fft claims only to require 2 full wave periods.

However I'lm not sure how accurate this is, hence one of the reasons fir my query on the maths, it has been years since I have done manual fft's, and not quite so long since I have used other kit that does fft's but this is all in the software.

Have you proved the noise is being induced into the cabling and not being induced through the power supply units? I'd start by trying a couple of known-good PSU's.

Yes, as I stated earlier I ran one of the vox boxes off a 12V battery and the noise was still there.

 

[Sidewinder]

Not yet, and for a good reason, I want to exhaust all ideas before I spill the beans, as it WILL get people jumping to conclusions, trust me.

 

[NozSpark]

I bet the cable's right next to a great big wocking AM/FM transmitter

What can you pick up on a coat hanger?????

 

[Sidewinder]

Not quite, but, you are getting close Noz!

 

[NozSpark]

I was serious about the coat hanger BTW

 

[Sidewinder]

Canoe, trust me, we NEED to exhaust all thoughts first because once I give the full details people will jump to conclusions & I have no evidence to prove the conclusions they will jump to, why, because I think I jumped to them all before doing proper investigations.

Please think outside the box and make some constructive suggestions based on the evidence provided, which is valid.

 

[Sidewinder]

I bet you will!

Mañana now TBH.

I have to drop something off @ a customers @ 07:15, then I will be back, a meeting mid morning, then a job mid afternoon, so I'll see what I can post in the way of detail tomorrow.

Trust me though it WILL have you all jumping to conclusions!

 

[Bruspark]

I just hope whoever was ringing the door intercom has been let in by now . . .  :innocent

 

[ProDave]

We are still awaiting the next drip feed of information.  ut the kettle on

 

[Sidewinder]

I'll try to put something together tomorrow, sorry guys, and thanks for the help so far.

 

[kerching]

I just hope whoever was ringing the door intercom has been let in by now . . .  :innocent

I think he was called ............Buzz!

Sorry

 

[Sidewinder]

I've started a write up on this, and I will finish it over the next few days, just been stupidly busy, and I now have to design & build an extruder panel (yes Canoe what you are thinking of) made from two existing panels as a breakdown, it should have been back on site to commission today.

I only got the job @ lunchtime today!!!