Voltage of an Inductor: Solutions to Get 200V

[iolantham]

Hello!

I have this circuit: a source connected to an amplifier, which is connected to a coil. I'm trying to get 200V through the coil at 40 MHz. The problem is that the amplifier is not calibrated at that frequency and now I can only extend measurements done at lower voltages, which seems to give wrong results. I can apply maximum 1 V from the source. I was thinking about connecting a smaller coil in series with the first and to measure the voltage on the second coil on the oscilloscope. I cannot exceed 10V on the oscilloscope, so I thought about making the inductance, L2, 10 times smaller.
Initially, I measured the voltage for L2 5 times smaller and the true power, P=1.115*10^-4 W. The apparent power, S=24.894 W and the power factor=P/S=4.479*10^-5, which is close to 0. I read that the power factor should be close to 1. Should I add a capacitance to get the power factor to 1? Maybe in this case I can get 200 V on the inductor 1.
Do you think there's a frequency dependent quantity, like the voltage applied by the amplifier, that I haven't considered?
Further, the amplifier cannot provide more than 200 W.
Any ideas will be highly appreciated.
Thank you.

 

[berkeman]

Welcome to the PF, iolantham. What you are attempting sounds a bit dangerous and non-physical (or at least on the fringe of what people would usually do). What in the world do you want to use a 200Vrms, 40MHz signal for? You will need some very special shielding to work with signals like that, or the FCC (or other agency in other countries) will come knocking at your door. Plus, those are downright dangerous power and voltage levels.

What are you attempting to build, and what is your educational and training level?

 

[iolantham]

Dear Bekerman,

We're trying to get a 100 A/m magnetic field strength H by applying a 200 V voltage to a coil. The heat generated by applying a magnetic field will heat nanoparticles inside the coil.
Regarding my educational level, I'm a senior undergraduate in biophysics and a research assistant in a Physics lab.
Do you have any ideas for the experiment?

 

[berkeman]

How big does the coil volume (diameter & length) need to be? Do you have a 40MHz RF signal generator? How about an RF amplifier? Will you be able to RF shield the apparatus so that you don't create harmful interference outside your lab?

Is this experiment based on published work that I could glance at, or is it an idea that somebody wants to try? Could you just use a 2.4GHz microwave oven to do the heating, or is there something special about 40MHz?

 

[iolantham]

paper attached

The signal generator and the amplifier can work in the MHz range. I read the user manuals for them and I don't need to insulate them.
The radius of the coil is 2.75 mm and it has 15 turns, with a turn density of 8000/3 turns/m.
Any ideas on how to obtain 200 V in the coil?

 

[berkeman]

Thanks for the paper -- that helps a lot. I'll need to look at this more tomorrow, but here are a couple basics.

** You'll need to try to ensure that the coil input impedance is reasonable over the operating frequency range. It looks like the paper shows a range of frequencies, so you may need to take that into consideration.

** The power amp will have an output impedance of 50 Ohms, and it will expect there to be a fairly stable 50 Ohm load that it is driving. To the extent that the load impedance is not a stable 50 Ohm load (which it won't be if you have a single coil that you are driving at several frequencies), you will need a 50 Ohm pad of at least 3dB and more likely 6dB between the amp output and the coil load. The pad will of course need to be able to handle the power that you will be pushing through it.

** We have a test setup in our lab where we routinely push about 75Vpp (150kHz to 80MHz) into a coil for an immunity test (EN 61000-4-6). The power amp is 75W, and can handle 50Vpp pretty easily, but is pushed a bit to go to 75Vpp. To go to 200Vpp, I'd guess that you will need at least a 250W RF power amp, but I'd need to do some better calculations to be sure. If you go with a coil that you wind for 50 Ohms at 40MHz, and only drive 40MHz for your experiment, you won't need the pad, which let's you go with a smaller amp. Alternately, you could wind one coil per test frequency, and go with discrete frequency tests and no pad.

Does that make sense so far? I'll try to look more at this tomorrow. BTW, here are some photos via google images of a typical test setup for the EN 61000-4-6 test:

http://www.arctechnical.com/Resources/Shortrac.jpg

http://www.sdpengineering.com/pictures/P04.jpg

 

[berkeman]

BTW, it looks like the power absorbed by the sample may vary with the experiment, which will change the input impedance of the coil versus what it would be with nothing conductive or lossy inside the coil. So based on that, you may still need at least a 3dB pad between the amp output and the coil, which increases the size (power output) of the required amp some.

 

[edmondng]

How do you find voltage drop produce by the coil/inductor? I know V = L(DeltaI/DeltaT),
so if i have 1A flowing and 2mH at 100kHz, does that mean i can generate drop upto 200Vp-p?

 

[berkeman]

V = L V f (another way of writing it), so yes. If you put 200Vpp across a 2mH inductance, you would get 1App through it.

But also keep in mind that real inductors have resistive losses, and if it's not air core, you will have some core losses as well.

BTW, this thread you posted in is about a year old.