Interaction between loads - power quality and noise.

[Simon Pettitt]

Hello,

Sorry if this isn't the right forum for this. Didn't look to be far off...

So I'm looking for some advice please with regards to interaction between loads sharing a circuit. I understand (or think I do!) a lot of the composite concepts and I think I have an idea of what my answer might be - but I want to bounce this off those with more experience and see where I'm wrong. I work in the entertainment industry with lighting and audio systems. I have a fairly good understanding of electrical principles but have no actual electrical qualifications (although I've considered training soon, that's another story).

So...

In several instances in our industry and outside it, I often hear people exclaim that a given device on a supply is affecting the power delivered to others on that supply. Take the scenario of a TRIAC-based lighting dimmer and an audio amplifier, both taking power in parallel from a given supply, with the amplifier buzzing when the dimmer is at certain dimming values. Often someone will say something to the tone of 'that dimmer is sending c**p back down the line'. SMPS supplies are said to do the same.

Now, on a conceptual level this makes sense - switching noise generated in the TRIAC on each dimmer channel is affecting the amplification stages in the amplifier and becoming part of the output signal. However, I can't seem to fully understand the electrical details behind this.

(For the moment, I'm discussing noise/distortion transmitted through mains wiring - not signal-level interconnection between two devices - I know Ground Loops are another story)

Both devices are connected to earth and the supply via suitably low impedance wiring, so why would any noise generated in load 1 (an SMPS/dimmer/other noisy supply) affect the power supplied to load 2? How can load 1 actually distort the waveform delivered to load 2 if load 1 is directly connected to the mains?

(I guess part of the confusion also lies in the difference between noise on the earth and noise/reactance on the circuit conductors superimposed on the mains waveform, maybe there are two different answers...)

The answer that I've came up with in my head is down to the natural impedance of the wiring in the system forming a voltage divider - so although load 2 is connected straight to the mains, it is done so through wiring with a finite impedance that will 'drop' a voltage, meaning the supply voltage and the voltage measured at the input to the device will be different - they are either side of a given impedance of wiring, thus different nodes in a circuit.

Or put another way, the voltage at the socket outlet feeding the affected device will be a summation of 1) mains voltage (through the impedance of the supply wiring) and 2) noise/distorted power from another load (through the impedance of whatever wiring affects the two) - with the affected load sitting at a given node in a voltage divider.

One last point - a lot of venues have 'Sound Power' supplies, Technical Earth etc. Again while I broadly understand the concept of separating noisy dimmers/SMPS from audio equipment, surely the fact that they all common back to the same earth would introduce interference? ...or again, is it the fact that because you're lowering the impedance between earth and the affected load (giving it a good direct connection) and increasing the impedance between noisy load and affected load (putting it on a different radial earth branch) then it naturally puts the affected load at a better point in the theoretical voltage divider?

Would appreciate if someone could steer me in the right direction, not sure how many other people end up thinking the same.

Thanks,

Simon

 

[The Ghost]

Just saying hello do you don't feel lonely, but as far as I am concerned you're on your own with that one. Not to say there is no-one on here who can't grapple with what you are looking at. I am aware of noise from slightly different perspective which can cause all sorts of different anomalies in electrical circuits but must admit to not really knowing much about sound engineering. Anyway just saying hello mainly.

 

[haptism]

Im no expert on this atall, but would it not be an idea to have each device on its own separate supply circuit (cable), this might eliminate any volt drop (even if they are wired in parallel) ! Without getting into the physics of it, stage lighting with dimmers and audio equipment on the same circuit sounds wrong !

 

[static zap]

Already thinking put audio amp on a chunky wound up extension lead , if they are fed from the same distibution point .
( and wander around with an AM radio -Ghost hunting )
But i'm probably breaking a rule ! How many watts is the amp ?

 

[Simon Pettitt]

Thanks for your replies. No worries if it's not your specialist subject. I don't desperately need help, it's just a curious question in case any experts in this kind of area hang out here. I'm aware that sound engineering isn't really the point of this forum, but this could apply to anything - a noisy computer PSU or discharge lighting etc.

I should clarify that I don't have an actual specific issue here - this is a theoretical post to gain understanding for when someone next says that a specific piece of equipment is corrupting a power supply. The audio amplifier is just a very convenient way of seeing (or hearing) the manifestation of corrupted power from other loads. You could always substitute this example for a computer with a noisy SMPS PSU corrupting the mains supply for some medical sensor equipment etc.

I did make one error in my post - at the top I put "interaction between loads sharing a circuit". I would never put an amplifier and dimmer on the same actual circuit in parallel, of course, that would intensify any issues. However, although not on the same final circuit, they would likely be fed from the same supply, which still goes back to my point.

A simple example:

400A Powerlock supply in a music venue.
Feeds a portable mains distro unit, which feeds a lighting dimmer with a 125A 3PH CEE and then feeds a second mains distro unit a 63A 1PH CEE with all sound equipment connected.

Once the dimmer is on and working, hums, buzzes etc are heard on the sound equipment.

So, often the second distro with all of the sound equipment is deliberately fed from it's own supply on the wall, which is generally regarded as helping/solving the issue - but why?

My point is, the main incomer and earthing arrangement of the building must be feeding both supplies - so all we've done is move the point at which they become common with each other further and further upstream. The earths of both systems still go back to the earth at the supply - so why does it make a difference? If all devices are fed directly from the supply and have a suitably low impedance connection to that supply - how are other devices corrupting that? I can only surmise that because moving affected devices on to a different branch increases the wire resistance between the noisy and affected loads and decreases the wire resistance between the affected load and the supply that it must have some affect similar to changing the ratio in a voltage divider.

 

[haptism]

Dimmer switches (because they operate by alteration of the common voltage waveform; (at least the old ones do) may be the culprit, causing harmonic distortion in the amp. The operation of the dimmer switches is obviously causing noise, enough to be picked up by the amp. You can get dimmers with inbuilt filters (of various specifications I expect) to reduce noise. Probably can get specialist amps that deal with that aswell, although which ones you need would be a mystery tome.

Im guessing dimmer switch noise is more prevalent in the wire in which its situated, that may explain why its not such an issue if you put the amp on a different (sub) circuit of the distro unit !

 

[static zap]

All wires are not perfect conductors , they also have inductance so in effect they like becoming transmitter masts .
Star earthing (like spider legs/body)
is a similar stategy wires are kept seperate until a known good point.
but if that has its own associated wiring it may not be good enough to not re radiate. back up all the seperate legs again .
Distance / parallel wiring , but you case sounds simple . Keep lighting and audio fed seperatly , or audio a long enough path to filter out some of interference .
(RF interference / capacitive pickup / inductive effects -may be breaching atempts to seperate )

 

[haptism]

Just reading up on this topic; anyhow a thrash-metal forum poster had the same problem with his small guitar amp and a dimmer. And that was in a house (on two separate circuits buried in the walls); had to play with the lights off for fear of mass distortion (although not sure what difference it would make to the sound). Anyhow the point is this is a common issue by the looks of it. I gonna make a fairly confident assumption you need filters on the dimmer switches

 

[b-spark]

I've got a book on this at work. Without getting lynch on here. I can read up on it for you. Its to do with the waveform and the triac or diac not being a gate like a transistor. Your basically putting to semi conductors together to able the ac to flow.

If your in no rush I can pick through it. Its a really good book from the (teach yourself) range of bookmakers. Has helped me no end. One on ebay £3.00 "teach yourself electronics"

 

[Simon Pettitt]

Thanks b-spark, I'll pick that book up anyway and have a look. Cheers

 

[Simon Pettitt]

In case anyone is interested, this document proved very helpful - and on page 10 my suspicion on moving loads so as to reduce common mains line impedance between them (I.E. moving the point of common connection to a point with much lower mains impedance) appears to be true.

http://www.powerlines.com/clips/pq2kdoc.pdf

Cheers

 

[davesparks]

From the point of view of a lighting tech the buzzing which sometimes appears in speakers when the dimmers are operated is great for winding up sound techs, if youre feeling particularly vindictive you can have them chasing their tails for hours looking for this hum which seems to disappear every time they walk up to the speaker then reappears when they walk away.

 

[static zap]

I remember a hall setup , Dimmers fed stage end , audio gear other end of the hall.
(great until the audience started carrying Bir..bir..bi Mobile phones -Next to audio cables )

 

[davesparks]

I think the vast majority of the time these issues are going to be as a result of incorrect earthing/grounding in the audio kit.

If the installed audio cables are picking up noise from mobile phones the there's definitely something wrong with the audio setup.