Wiring size for compressor

[andeepdhanjal]

Hi there,

I am looking to order some cable to wire in an air compressor. The compressor motor is 3kW which is around 13amps, but on startup it can surge up to 40amps.

My question is what amperage do I use in my calculations for sizing the cable? Since its only a 40amp startup surge, do I base my cable on a maximum load of 13amps?

Many thanks

 

[bernardgreen]

Two things

[1] the cable has to have an impedance that is low enough such that the voltage drop at 40 amps is low enough not to compromise the motor during start up.

[2] As has been discussed in other threads you will need to notify the DNO ( the people who maintain the supply network ) about this load. It is likely that the 40 amps will dip the local supply voltage and affect your neighbour's electrical supply. If the affect is noticeable they may report it to the DNO who may then request you cease using the compressor.

 

[JohnW2]

[2] As has been discussed in other threads you will need to notify the DNO ( the people who maintain the supply network ) about this load.

Indeed - that appears to be the case.

It is likely that the 40 amps will dip the local supply voltage and affect your neighbour's electrical supply. If the affect is noticeable they may report it to the DNO who may then request you cease using the compressor.

How does switching on my 10.5kW shower (nearly 45A at my usual 245V supply) fit with that?

Kind Regards, John

 

[westie101]

How does switching on my 10.5kW shower (nearly 45A at my usual 245V supply) fit with that?

It, er, should have been notified and load approval gained!

In reality the odd flicker now and then is rarely noticed so the use of a shower once or twice a day wouldn't be an issue. A motor start for a compressor if used heavily could be more noticeable

 

[JohnW2]

How does switching on my 10.5kW shower (nearly 45A at my usual 245V supply) fit with that?

It, er, should have been notified and load approval gained!

I just knew you would pop up and say that We've discussed before that publicity is so poor that only an insignificant number of people even know that they are meant to get permission from the DNO before using welders, compressors etc., and that probably only a tiny number of domestic users of such equipment ever do. However, when it comes to showers, I suspect that even fewer people would even dream that they needed permission to install them, so I suspect that virtually none do. Is that correct?

In reality the odd flicker now and then is rarely noticed so the use of a shower once or twice a day wouldn't be an issue. A motor start for a compressor if used heavily could be more noticeable

Yes, possibly - although I would imagine that welders are far more of a potential problem.

Kind Regards, John

 

[westie101]

I suspect that even fewer people would even dream that they needed permission to install them, so I suspect that virtually none do. Is that correct?

Correct

Interestingly before privatisation there never was a problem with all those connecting large loads checking first, be they DIYers or electricians, now I can understand DIYers not knowing but long serving electricians who used to ask seem to have stopped - why? The newer breed I can possibly understand as I guess they are never told at college or by whatever scheme, if any, they are in.

 

[SimonH2]

My question is what amperage do I use in my calculations for sizing the cable? Since its only a 40amp startup surge, do I base my cable on a maximum load of 13amps?

Wrong way round !

You need to determine what supply you will need - specifically what rating and trip curve breaker you need. Once you've determined that, then you can calculate the minimum cable size allowable. I believe (up for correction) that there are 3 requirements to meet :

1) The current rating of the cable, allowing for installation method and any correction factors, is not less than the rating of the breaker. That's possibly the only aspect you've considered !
2) The circuit impedance is low enough to guarantee operation of the breaker under fault conditions. There's an element of this determined by the trip curve - a type C will let through much more energy before it trips than a type B, I think this is where the "adiabatic equation" you may have heard mentioned comes in.
3) (Already mentioned) The volt drop under startup conditions isn't so much as to affect the ability of the motor to start. I've seen this at my mate's - compressor with 2x3hp motors, delayed start on one of them. In cold weather, the motors can be slow enough to start that the second one fires up, and doesn't get going before tripping it's overload (or occasionally blowing the supply fuse). The supply, while adequate for the running load, "could be better" in terms of startup. Doesn't help that the motors are 2 pole and permanent capacitor - both of which makes them have poor startup characteristics compared to w 4 pole with switch-start - which means that with the compressor cold (thick oil etc) it can be quite slow to get spinning until it's warmed up.

 

[JohnW2]

2) The circuit impedance is low enough to guarantee operation of the breaker under fault conditions. There's an element of this determined by the trip curve - a type C will let through much more energy before it trips than a type B, I think this is where the "adiabatic equation" you may have heard mentioned comes in.

Let-through energy and adiabatic calculations do not come into determining whether the MCB will operate satisfactorily (i.e. trip magnetically) under fault conditions - that is simply a function of the current required for a magnetic trip (5xIn for a Type B, 10xIn for a type C) and, as you say, the loop impedance of the circuit.

Kind Regards, John

 

[stillp]

If the voltage dips on starting, it is likely that the inrush current of the motor will continue for too long a time, possibly damaging the starter, the motor, or both.

 

[SimonH2]

Let-through energy and adiabatic calculations do not come into determining whether the MCB will operate satisfactorily (i.e. trip magnetically) under fault conditions - that is simply a function of the current required for a magnetic trip (5xIn for a Type B, 10xIn for a type C) and, as you say, the loop impedance of the circuit.

OK, so that should have been item 4 in the list

 

[JohnW2]

Let-through energy and adiabatic calculations do not come into determining whether the MCB will operate satisfactorily (i.e. trip magnetically) under fault conditions - .....

OK, so that should have been item 4 in the list

Maybe, but you would then have to reverse the language as compared with your item (2). Whereas with (2), you rightly say that one has to ensure that the loop impedance is low enough to guarantee operation of the MCB under fault conditions, in terms of the adequacy of the the CSA of the CPC (determined by an adiabatic calculation), the need is to ensure that the loop impedance is high enough for the CPC's CSA to be adequate. Combining those two, one therefore has to ensure that the loop impedance is between the minimum (for CPC) and the maximum (for MCB operation).

I may be very wrong, but I get the impression that these adiabatic calculations are probably not often routinely undertaken.

Kind Regards, John

 

[EFLImpudence]

I may be very wrong, but I get the impression that these adiabatic calculations are probably not often routinely undertaken.

I suspect that you are probably correct in what you meant.
However, 'often routinely' is a bit confusing.

With routine domestic circuits, it is unlikely that it would not comply with the requirements.

In a special circuit such as this thread it may be necessary.
Having said that, increased csa if employed would make the likelihood of non-compliance less likely.


I once blew a supply fuse - only 60A - with a 1mm² cpc at the CU resulting in no damage to the conductor at all, other than a little blob on the end of it.
A very nice man came from SSE very quickly.

 

[SimonH2]

Whereas with (2), you rightly say that one has to ensure that the loop impedance is low enough to guarantee operation of the MCB under fault conditions, in terms of the adequacy of the the CSA of the CPC (determined by an adiabatic calculation), the need is to ensure that the loop impedance is high enough for the CPC's CSA to be adequate. Combining those two, one therefore has to ensure that the loop impedance is between the minimum (for CPC) and the maximum (for MCB operation).

Not sure I follow.
As I see it, there is a need for the loop impedance to be low enough that the OPD will operate before the cable overheats.

As EFLImpudence points out, with a short cable, the impedance is low even for a small cable, and the OPD will operate quickly. As the cable gets longer, so the impedance increases, the fault current reduces, and the OPD may take longer to operate. There may come a point where the OPD takes long enough to operate that the cable may overheat.
I strongly suspect that for most domestic circuits, the loop impedance required for reliable OPD operation is a limiting factor before the heat issue comes into play - this may be a case where it matters.

Eg, since the running current of the compressor is only 13A, one may be tempted to fit a 16A breaker and use (say) 1.5mm^2 cable. A B curve will trip pretty well every time the compressor starts which would be "inconvenient". So the user then fits a C curve breaker, and possibly even a D curve breaker. All will trip "quickly" if the cable is short enough and it's a dead short near the supply end - but I can't help thinking that with enough length of cable, a D curve in particular could let an awful lot of energy through before it trips.
One case where the user might swap the breaker would be if the supply was originally for a different load, with a low start up surge. The supply then gets re-used for a different purpose - user things "16A is 16A" but perhaps has heard "down the pub" that C or D curve breakers don't trip as fast.

It would also (as I think has already been said) depend on the load. If the compressor has a proper starter with overload then a stalled motor would trip the overload. Otherwise, it would remain stalled and reply on the breaker tripping - eventually.
But can you rely on the load when designing such a circuit ? Suppose the user has a "proper" compressor now - so the circuit is designed on the basis that there's an overload in the motor controls. The user then replaces the compressor, and the new one has no overload. Most users would have no idea - but now there is no load side overload protection.

As an aside, a mate did replace his compressor a few years ago - I bought his old (broken) one. The old one was 3 hp and plugged into a 13A socket. The new one was 4hp and was also plugged into a 13A socket He found, by trial, that if he plugged it into an extension lead (say around 10m of 1.5mm^2 flex) that it didn't blow fuses. He asked me for help when he realised he'd "burned out" half of the sockets in the unit (switch failed internally due to overheating, when one went, he'd move the plug to another) and I arranged a hardwired supply for him.

 

[ericmark]

As already said compressors are a real problem the same applies to many pumps be it gas or liquid in that one they start the load starts to increase so the time taken to start is rather important. In the main the de-load valve is worked by the delta connection coming in with larger units but single phase it's pure time delay in the main.

The use of resistor start I suppose could work as well with single phase as three phase but never seen it used idea of an extension lead is interesting.

In real terms 3kW motors are not domestic and really not a DIY project but there are ways to reduce the load from 3 phase with inverters to type of motor starter. Some single phase motors have a capacitor in circuit all the time and some have it switched by centrifugal switches and some have a timer. The windings also vary and so it is near impossible to work out the supply required.

Even when you have all the details it is often a suck it and see situation rather than some calculations and I would normally consider 6mm² cable as a start point. I would consider it would likely need a C16 MCB and again start point I would be looking at the loop impedance when using a C16 MCB.

If you use your figures then 40A start would only need a B16 MCB (80A on magnetic part) but it only needs to exceed 80A for 0.01 of a second to trip so I would be looking for 160A so the loop impedance must be better than 1.44Ω.

It is clearly over 2kW so dedicated supply of course. Over the watts for rotating machines as well so must not auto start after a power cut. Likely an Active RCD will ensure that.

 

[SimonH2]

As already said compressors are a real problem the same applies to many pumps be it gas or liquid in that one they start the load starts to increase so the time taken to start is rather important. In the main the de-load valve is worked by the delta connection coming in with larger units but single phase it's pure time delay in the main.

On many compressors, there is a start valve in the outlet piping. A small bush with a nipple in the middle which passes air and relieves the pressure until there is enough flow/pressure to push the nipple out and close it. When my mate got his compressor and we had trouble making it start, the supplier came out and changed these for ones with a higher closing pressure.
"Hiss like hell" for a couple of seconds (more so if the tank is empty), then eventually close.

Some single phase motors have a capacitor in circuit all the time and some have it switched by centrifugal switches and some have a timer.

At the time I asked dad for advice - many years ago he designed motors etc for a living. permanent capacitor motors have poorer starting torque than switched start motors - less current in the start winding for permanent cap, while switched cap can have a much more powerful start winding.
So slower startup - which means the motor is still running up as the back-pressure builds, hence the start valves fitted to this unit. IME (limited) experience, these are the only load dump valves fitted on smaller systems - other than teh dump valve that's part of the pressure switch (dumps the pressure in the line between comp and tank when the switch turns off).

If you use your figures then 40A start would ...

IME I'd say 40A is wildly optimistic for a 3hp motor.

It is clearly over 2kW so dedicated supply of course. Over the watts for rotating machines as well so must not auto start after a power cut.

Interesting. None of the compressors I've ever looked at need manual intervention after a power cut - they all just start up if the pressure switch calls for it. Mind you, depending on the circumstances, many will fail to start as the pressure switch won't have unloaded the compressor outlet.