Degausing/Demagnetizer Coil Driver Design 1

[Hyteck]

Hi

Can anyone sugguest a suitable method for the design of a suitable circuit for a demagnetizer coil. I currently use a HV 470mfd @ 300V Capacitor that is discharged via an SCR into a coil for magnetizing function which works fine but I also need a demagnetizing function?

Normally I would use a Phase controlled Traic to apply an AC current which is progressively reduced to provide a decaying field. However Due to the Inductance of the coil an AC power line frequency I cannot get a high enoungh AC current in the coil to reach sufficent magnetic field strength required.
Some professional demagnetizer's use a "Ringing Circuit" to achieve this demagnetizing function - but I do not know how to go about making this circuit - any idea's ?

Ideally one would need to somehow discharge stored energy in a capacitor and make it "ring" to provide the decaying and reversing current needed for demagnetizing ?

thanks

 

[BrianG]

Some points that come to mind: What are you trying to magnetise/demagnetise? You don't state what the inductance of your coil is or the field strength you require for demagnetising. If you can't get enough current through the coil at supply line frequency to get the required field strength, then why not supply the coil from via a step-up transformer?

 

[Hyteck]

Brian, Coil inductance is 0.25H, Wire Resistance is 20 ohm, operating from 240Vac 50Hz supply. In magnetizing mode I need approx 15A DC @ 300VDC to get the required field strength which is the maximum the circuit will allow due to wire resistance. In Demagnetizing mode the inductance presents an impedance of approx (78.5 + 20) = ~100 ohms @ 50Hz which will only give me around 2.4Amps AC from a 240Vac supply. This means I would need to raise the AC voltage to around 1500Vac (and power) to get the necessary initial field strength as per the dc magnetization mode.
Unfortunately the voltages and Power requirements are too excessive so I need a different approach?

Another way would be to play around with the coil design and use heavy guage wire with fewer turns and hence smaller inductance but then the current requirements will be correspondingly larger. Prefer to store the energy in capacitors and discharge through the coil - this is fine for magnetizing case however demagnetizing is the problem?

thanks

 

[dspDad]

All r/c circuits are 'ringing' circuits with a mix of oscillation and damping.

With the values described, you have a highly damped circuit resonant at about 14 Hz with a Q of about 1. Your peak currents are probably closer to 10 Amps than 20.

f = 1/ (2 * pi * sqrt(L * C))
Q = w * L / R or 2 * pi * f * L / R


To get a ringing circuit, you need to decrease the resistance and increase the frequency. Fewer turns of larger wire will do both of these. Shoot for a Q of 5 or more to get a nice decaying sinusoid.

Electrolytics have high internal resistance. It may be small enough to ignore, but maybe not. In addition, they don't tolerate reverse voltage which will happen with a ringing circuit. So you need two electrolytics back to back with shunting diodes across them, or better, a paper or mylar capacitor which has much lower resistance and is non-polar.

Third, you need to provide a reverse current shunt (an antiparallel diode) across the scr to allow the current to flow backwards.

And last, you need to keep the scr drive turned on for the entire demag interval, because the scr would normally turn off when its current drops to zero. A triac would be another option and eliminates the shunting diode, but would still need to be gated on for the duration.

Sometimes an scr will retrigger without being gated on because the reverse recovery characteristics of your shunting diode allow an inductive kick to occur. It appears as though the scr is continuously enabled. This inductive kick can be destructive, and at the power levels you are describing (21 joules per cycle) you might destroy the scr. Then again, you might not.

When I was much younger, they had focussing coils around the yokes of tv's that had a 1.5 inch or so aperature. These could be hooked up with a diode bridge to the 110 mains for magnetizing, and hooked up to 60 Hz directly for demagnetizing. Plug it in, run a screwdriver through it, and voila. I built one, and at one time every tool in our workshop was magnetized. I had a relatively tolerant father so I'm here to tell about it.

DspDad

 

[Hyteck]

DspDad - many thanks for the useful advise.

I tried a preliminary setup today and it works as you say, but I now need to work on getting a higher Q value by playing around with coil design.
One problem is I need fairly large values of Capacitance to store sufficient energy and this means lower Q. Also the fewer turns (for low Resistance) means much smaller Inductance and hence lower Q.

How does one choose the correct/optimun amount of capacitance needed for job? If my coil needs say 100A to achieve the required amount of magnetic field strength for magnetization and if its inductance is say 5mH?

Do I simply use U=0.5*L*I^2 = 25 Joules to be the amount of Stored Energy required in the Capacitor or do I need more for wire losses etc?

thanks again,