Record for most underrated cable?

[Lectrician]

Turkey has to be the worst for wiring - has to be seen to be beleived.

I have worked with a sprinkler system in a factory. The glass in the sprinlker heads explode at a set temperature, allowing the water to 'sprinkle' out.

The water is stored in a huge pressurised vestle. This vestle is kept under constant pressure by a small pump which turns on and off to keep the tank 'ready'. This pump is typically only a couple KW.

There is also a huge pump (the one I have seen was fed with 120mm cable!). This pump can keep the pressure in the system up when the sprinkler heads are open. It takes water in from a huge resting tank.

Dotted in the sprinkler pipework are flow switches which are used to signal to a central display, and interfaced to the firealarm.

There are also very often fire hoses attached to the system, and operating these will bring in the huge pump and the firealarm.

 

[kendor]

21KW total?, so 7kw per phase, somewhere around 30AMP, 2.5mm is rated at about 20A, ins't it, I would expect it to just heat up a little too much, maybe brown a bit, but not set fire to wood (well not unless it had been running for hours)?

Unless I've mis-understood, and the cables were carrying 90A (21kw), then I could imagine it causing that sort of damage.

Most certainly undersized whatever, I would imagine for the first way, it would be 6mm minimum and the second 25mm minimum?

Adam, 21kW PER PHASE (mistype sorry)

Also, 21kW total on 3 phase would not be 7kW per phase either, 7kW per phase would be 12.1kW total.

21kW total would be 12.138kW per phase.

The root 3 calcs always catches em out

 

[ban-all-sheds]

Even those who, it seems, are gainfully employed as electricians in commercial environments working on 3-phase equipment:

http://www.screwfix.com/talk/thread.jspa?threadID=17787

 

[plugwash]

hehe some things can be very unintuitive

for example the voltage at the load end of a cable can be higher than the voltage at the source end. can anyone here figure out why?

p.s. fwl i presume you know the answer keep quiet for a bit and let the others have a guess

 

[andy]

hehe some things can be very unintuitive

for example the voltage at the load end of a cable can be higher than the voltage at the source end. can anyone here figure out why?

p.s. fwl i presume you know the answer keep quiet for a bit and let the others have a guess

er... i dunno


someone got the feed and load end of the cable mixed up?!

 

[mapj1]

Noo. Think about conditions where current and voltage may not be in step... (or why we add capacitors to lamps with ballast, power factors etc) those sort of thoughts will send you in the right direction..
Actually you can get this effect on a single phase supply.

Next trick question, aimed squarely at our university men.

2 lamps in series with 2 switches, all accross the AC supply in one long loop, no branches bypasses.
What component do I add at each lamp holder and each switch, to make switch 1 only affect lamp 1, and switch 2 only affect lamp 2, and the two to be independant of each other.
Clue, people with good eyesight may notice increased flicker.
5 points for an answer in 24hrs.
1 more point for the other symptom.
You may not confer.
Oops my other job showing through...

 

[ban-all-sheds]

2 lamps in series with 2 switches, all accross the AC supply in one long loop, no branches bypasses.

 

[plugwash]

Noo. Think about conditions where current and voltage may not be in step... (or why we add capacitors to lamps with ballast, power factors etc) those sort of thoughts will send you in the right direction..
Actually you can get this effect on a single phase supply.

Next trick question, aimed squarely at our university men.

2 lamps in series with 2 switches, all accross the AC supply in one long loop, no branches bypasses.
What component do I add at each lamp holder and each switch, to make switch 1 only affect lamp 1, and switch 2 only affect lamp 2, and the two to be independant of each other.
Clue, people with good eyesight may notice increased flicker.
5 points for an answer in 24hrs.
1 more point for the other symptom.
You may not confer.
Oops my other job showing through...

you add a diode accross each lamp and each switch

you set up the diodes so that on each half cycle one lamp and one switch are shorted by the diodes.

the average power to the lamps assuming the diodes are perfect and the lamps purely resistive will be half what it would be if they were wired in parallell accross the supply or twice what it would be if they were just wired in series accross the supply.

its an interesting idea for upgrading a single switch to a double switch with no extra wiring although you would need special bulbs to do it well (you would wan't bulbs rated at 170V rms).

 

[ban-all-sheds]

you add a diode accross each lamp and each switch

you set up the diodes so that on each half cycle one lamp and one switch are shorted by the diodes.

Like this?:

My first thought too, but it doesn't work....

 

[plugwash]

that cuircuit DOES work but it involves a bypass which isn't allowed

however it should also work if you use 4 seperate diodes on the lamp and the switches

 

[Damocles]

Perhaps a couple more diodes. One in series with each lamp and switch, with opposite polarity to the bypass diode. The bypass diodes must be as you have drawn them, in opposite directions. You now have two separate current paths for the negative and positive half cycles. The lamps/switches would not interfere with each other. Each lamp would lose power for half a cycle so it would flicker more. But they would catch up a bit and would be using more than half power.

 

[Damocles]

Oh, and I meant to say. Any electrical engineer who manages to create a device which pushes 12Kw down three wires and gets out 7Kw on each of three pairs of wires is going to get very very rich.

 

[ban-all-sheds]

that cuircuit DOES work

It works in the sense that the lights can be turned on and off independently, but it does not work in the sense of meeting the requirement for S1 to only affect L1 etc.

but it involves a bypass which isn't allowed

Who says it's not allowed?

however it should also work if you use 4 seperate diodes on the lamp and the switches

Perhaps a couple more diodes. One in series with each lamp and switch, with opposite polarity to the bypass diode.

Nice idea. I was about to say that it wasn't allowed because the spec was "what component do I add at each lamp holder and each switch", which I had been interpreting as only 1 thing at each (lampholder and switch) and which I had been using as a bound for my solution, but I now suspect that that interpretation was wrong, and that you do add 1 diode per switch and 1 per lamp. I could argue that inserting the series diode is not "adding to the lamp" in the way that the parallel one is "added to the switch", (i.e. to fit it you have to remove something that is already there, namely the wire between the switch and the lamp, and replace it with the diode), but that would be pretty desperate..

 

[mapj1]

The one I was thinking of was 4 diodes 800v 3A or similar will do, one in each switch, one in accross each lampholder - or at least that was the one we used to show the students on open day, all screwed firmly to a pywood demo board, with the diodes not visible. Done with 15w nightlight bulbs no-one notices they are extra dim, but it certainly gets some quizzical looks.
OK, lots of points all round - I really wasn't expecting people to rise to the challange
top o the morning to you
- best regards Mike.

 

[kendor]

Half wave rectification?