| On 3 Aug 2004 16:05:58 GMT, snipped-for-privacy@ipal.net put forth the | notion that... | | |> On Tue, 3 Aug 2004 01:09:49 -0700 Checkmate wrote: |> |> | Sounds do-able. So you're saying the wire gauge isn't that important, |> | as long as long as there's room for 10 times as many turns as the 24 |> | volt coil has? |> |> Just guessing, here, but I'd say that the total _volume_ of wire would |> remain a constant to get the same amount of "kick" for the same amount |> of power at the same frequency with the same surface for dissipation of |> heat. Physics often has a nice way of doing things like that. You're |> just changing the voltage to current ratio (something a transformer |> does with two windings). | | This is what's got me temporarily fascinated. It seems like a | relatively simple problem, but there are several factors that have to be | taken into consideration... coil resistance, inductance, magnetic flux, | physical space, etc. I don't want to do this for a living, but I'd like | to know how to calculate what the physical properties of the coil need | to be... partially because I have several contactors I want to convert, | and partially because it's a good learning exercise. I suppose I could | buy one and dissect it, since I have three contactors I want to use at | 240 volts, but I'm betting someone in here knows an easier way.
Sure, resistance will go up. And inductance will go up. They should since the voltage is going up and the current is going down. For 10 times the voltage, impedance needs to be 100 times as much to get 1/10 the current to keep the power level constant.
Here are some example figures for a 6 watt load at several voltages common in North America:
amps: 0.50000000 ohms: 24.0000000 volts: 12.0000000 watts: 6.00000000 amps: 0.25000000 ohms: 96.0000000 volts: 24.0000000 watts: 6.00000000 amps: 0.12500000 ohms: 384.000000 volts: 48.0000000 watts: 6.00000000 amps: 0.05000000 ohms: 2400.00000 volts: 120.000000 watts: 6.00000000 amps: 0.02886751 ohms: 7200.00000 volts: 207.846096 watts: 6.00000000 amps: 0.02500000 ohms: 9600.00000 volts: 240.000000 watts: 6.00000000 amps: 0.02165063 ohms: 12800.0000 volts: 277.128129 watts: 6.00000000 amps: 0.01732050 ohms: 20000.0000 volts: 346.410161 watts: 6.00000000 amps: 0.01443375 ohms: 28800.0000 volts: 415.692193 watts: 6.00000000 amps: 0.01250000 ohms: 38400.0000 volts: 480.000000 watts: 6.00000000 amps: 0.01000000 ohms: 60000.0000 volts: 600.000000 watts: 6.00000000
So if you have 10 times the wire length, at 1/10 the cross sectional area (1/3.16227766 the width) you'll end up with 100 times resistance. But you'll also have 100 times the inductance, so the resistance will still be about the same proportion. Model the wire as 10 resistors in parallel that you are changing over to 10 resistors in series.
Loss is proportional to I^2*R. Since I is 1/10th as much and R is 100 times as much, it comes out the same. It should get just about as hot.
The issue you could run into is that the wire insulation might have to be thicker to handle the higher voltage. That can end up making the whole coil assembly larger, requiring longer wire, introducing more resistance. You can compensate with larger wire, but now the assembly gets even larger.