simonplights
Member
Hello,
Sorry if this isn't the right forum - I had a look around and this seemed to be the closest match. Apologies, it's a bit of an essay but may be of use to others...maybe!
Haven't been here long - I posted more about me in the introduction forum, but in short I've been working as a touring lighting technician aka 'roadie' for 10-odd years and want to further improve my knowledge of electrical theory now that I've stopped touring and work at a rental company. Hoping to embark on the 2365 soon, so I have no actual qualifications but a lot of hands-on experience.
So, on to my question. I'm trying to get straight in my head the mechanisms behind which some loads can affect the power delivered to other loads, and where in the electrical system the waveforms are affected moreso and less so, and why.
A scenario - a supply (Vs) feeding a household with a SMPS laptop PSU upstairs (let's say voltage across input terminals 'V1') and a hifi audio amplifier in the lounge 'V2'. I often hear that things like SMPS power supplies 'send rubbish back down the mains' etc. The concept of this is intuitive on a simple level - the choppy switching operation of the power supply is vastly different to linear loads, and the strange harmonics and power factors that result aren't a surprise. What I can't fully understand is how these 'go back down the mains' and affect other devices, such as the hifi amplifier. It doesn't seem right that the power delivered to the system from the actual mains is affected by a small load nearby, especially when the affected loads are connected directly to the mains by (relatively) low impedance wiring. Because the voltage should come in from the supply as a (near enough) decent sine wave, because that point is electrically common with the mains input of our hifi amplifier - why should the voltage waveform look anything different? I understand that reactive power will transfer power back to the supply from reactive loads, but why would that affect other loads in the system? I.E. why should 'V2' be affected by 'V1'?
Quite often in our industry - venues will have a separate 'sound power' supply. There will be a 400A Powerlock supply on Stage Right for Lighting, and a 63A 1 PH CEE supply on Stage Left for Audio. However, both of these often go back to a single mains supply at the entrance to the venue. It is assumed that if the Audio power was to be taken from the same supply as the lighting dimmers, the Thyristor switching in the high power dimmers affecting the power factor would have an impact on the performance of the audio amplifiers. Moving the audio equipment on to it's own supply, a guess what you would call a radial circuit, is taken to remove this interaction. Similarly, I don't expect my next door neighbour's potentially inductive white goods to affect my audio amplifier in the lounge, even though we both take a supply from the substation behind the houses.
I understand in some of those examples that earthing is a big thing, particularly where audio and ground loops are involved, but the scope of my question solely relates to the actual voltage/current waveforms being distorted throughout a system and exactly how it happens.
Solution
So, rather than just ask for help straight away, I've drawn the attached picture and tried to have a think through this and come up with some logical idea to explain it to myself. I have this possible scenario, which may be partially or totally wrong:
Because the system isn't actually made up of 'common' connections between supply and loads, and instead made up of multiple resistive wire elements, the system acts like one big network of voltage dividers. Noisy loads draw irregular current waveforms, and ohms law dictates that these will drop a corresponding irregular voltage across the impedances they are in series with (the L and N lines feeding the device). IF an affected load has that wire impedance in common with the offending load, the irregular voltage drop over the wire impedance will subsequently affect the voltage delivered to the affected device.
Therefore, if I wanted a clean feed to my audio amplifier (or sensitive medical instrument etc etc) that is less affected by a SMPS in the same network, I could run a direct radial circuit from the supply. This means that of all the different impedances in the system supplying the SMPS (supply cable from substation, cable in to house, ring main etc), only the ones common to both the amplifier and the SMPS would drop a voltage that affect the amplifier.
Therefore also, because the mains wiring feeding the house must have some form of impedance, if there is a given load in the house with a poor power factor or inducing harmonics etc, because that distorted current is being drawn through the common impedance of the supply cable, some of that current waveform will drop a corresponding voltage over the cable supplying the house and thus affect every load in the house in a minor way.
Therefore again, because the wiring in modern installations is (hopefully) as low impedance as possible for safety and reduction of losses, interaction between loads is reduced because the voltage drop over the wiring in the system is smaller - therefore imparts less of an effect on the mains delivered to other loads, because the distorted current waveforms are 'dropping' less distorted voltage?
Any distorted current waveforms from noisy loads are dropping corresponding voltages over various system impedances, which are then superimposed on the mains waveform coming from the supply. Moving to a different house means that the only thing common is the cabling within the substation, which is much lower impedance compared to the smaller cable (dropping more voltage) in the walls of a different house.
Am I on the right track here?
Much appreciate any thoughts anyone has.
Thanks,
Simon
Another thought...reading up on power factor, I'm aware of lagging and leading current waveforms and the physical reasons why. Am I right in thinking there is no way a load could offset the phase of the voltage waveform? Voltage will always be 0deg and the current waveform will just shift back and forth depending on flavour and magnitude of reactance?
View attachment Multiple loads.tiff
Sorry if this isn't the right forum - I had a look around and this seemed to be the closest match. Apologies, it's a bit of an essay but may be of use to others...maybe!
Haven't been here long - I posted more about me in the introduction forum, but in short I've been working as a touring lighting technician aka 'roadie' for 10-odd years and want to further improve my knowledge of electrical theory now that I've stopped touring and work at a rental company. Hoping to embark on the 2365 soon, so I have no actual qualifications but a lot of hands-on experience.
So, on to my question. I'm trying to get straight in my head the mechanisms behind which some loads can affect the power delivered to other loads, and where in the electrical system the waveforms are affected moreso and less so, and why.
A scenario - a supply (Vs) feeding a household with a SMPS laptop PSU upstairs (let's say voltage across input terminals 'V1') and a hifi audio amplifier in the lounge 'V2'. I often hear that things like SMPS power supplies 'send rubbish back down the mains' etc. The concept of this is intuitive on a simple level - the choppy switching operation of the power supply is vastly different to linear loads, and the strange harmonics and power factors that result aren't a surprise. What I can't fully understand is how these 'go back down the mains' and affect other devices, such as the hifi amplifier. It doesn't seem right that the power delivered to the system from the actual mains is affected by a small load nearby, especially when the affected loads are connected directly to the mains by (relatively) low impedance wiring. Because the voltage should come in from the supply as a (near enough) decent sine wave, because that point is electrically common with the mains input of our hifi amplifier - why should the voltage waveform look anything different? I understand that reactive power will transfer power back to the supply from reactive loads, but why would that affect other loads in the system? I.E. why should 'V2' be affected by 'V1'?
Quite often in our industry - venues will have a separate 'sound power' supply. There will be a 400A Powerlock supply on Stage Right for Lighting, and a 63A 1 PH CEE supply on Stage Left for Audio. However, both of these often go back to a single mains supply at the entrance to the venue. It is assumed that if the Audio power was to be taken from the same supply as the lighting dimmers, the Thyristor switching in the high power dimmers affecting the power factor would have an impact on the performance of the audio amplifiers. Moving the audio equipment on to it's own supply, a guess what you would call a radial circuit, is taken to remove this interaction. Similarly, I don't expect my next door neighbour's potentially inductive white goods to affect my audio amplifier in the lounge, even though we both take a supply from the substation behind the houses.
I understand in some of those examples that earthing is a big thing, particularly where audio and ground loops are involved, but the scope of my question solely relates to the actual voltage/current waveforms being distorted throughout a system and exactly how it happens.
Solution
So, rather than just ask for help straight away, I've drawn the attached picture and tried to have a think through this and come up with some logical idea to explain it to myself. I have this possible scenario, which may be partially or totally wrong:
Because the system isn't actually made up of 'common' connections between supply and loads, and instead made up of multiple resistive wire elements, the system acts like one big network of voltage dividers. Noisy loads draw irregular current waveforms, and ohms law dictates that these will drop a corresponding irregular voltage across the impedances they are in series with (the L and N lines feeding the device). IF an affected load has that wire impedance in common with the offending load, the irregular voltage drop over the wire impedance will subsequently affect the voltage delivered to the affected device.
Therefore, if I wanted a clean feed to my audio amplifier (or sensitive medical instrument etc etc) that is less affected by a SMPS in the same network, I could run a direct radial circuit from the supply. This means that of all the different impedances in the system supplying the SMPS (supply cable from substation, cable in to house, ring main etc), only the ones common to both the amplifier and the SMPS would drop a voltage that affect the amplifier.
Therefore also, because the mains wiring feeding the house must have some form of impedance, if there is a given load in the house with a poor power factor or inducing harmonics etc, because that distorted current is being drawn through the common impedance of the supply cable, some of that current waveform will drop a corresponding voltage over the cable supplying the house and thus affect every load in the house in a minor way.
Therefore again, because the wiring in modern installations is (hopefully) as low impedance as possible for safety and reduction of losses, interaction between loads is reduced because the voltage drop over the wiring in the system is smaller - therefore imparts less of an effect on the mains delivered to other loads, because the distorted current waveforms are 'dropping' less distorted voltage?
Any distorted current waveforms from noisy loads are dropping corresponding voltages over various system impedances, which are then superimposed on the mains waveform coming from the supply. Moving to a different house means that the only thing common is the cabling within the substation, which is much lower impedance compared to the smaller cable (dropping more voltage) in the walls of a different house.
Am I on the right track here?
Much appreciate any thoughts anyone has.
Thanks,
Simon
Another thought...reading up on power factor, I'm aware of lagging and leading current waveforms and the physical reasons why. Am I right in thinking there is no way a load could offset the phase of the voltage waveform? Voltage will always be 0deg and the current waveform will just shift back and forth depending on flavour and magnitude of reactance?
View attachment Multiple loads.tiff
Last edited by a moderator:
