I came across the Resonance Induction book that comes with the RICK kit from Rick Friedrich, at some point he shows a series-parallel circuit that contains resistors to measure power going through them. His style is such that he gives more questions than answers. I had some coils that I used for some previous experiment and some capacitors, so I gave it a try. At first I didn't measure anything unexpected and wanted to ask here if you guys knew how to do this measurement properly, but then I turned my probes from 10x to 1x and measured the following.
To describe the setup. I have a signal generator that has sine wave at output, 12 Vpkpk, tuned to about 801 kHz frequency to resonance (the resonance changes a bit when I move stuff around). The capacitors are 100 pF, 3 kV, china made, ESR unknown. The coils are 61 turns, .3 mm wire, 25 mm radius, should have about 150 uH. The resistors are 2 W 1 ohm ceramic resistors.
The setup looks like this:
In the book Rick says to measure the power on the resistor on input and the power on the resistor in the resonance circuit. I guess the best way would be to use a differential probe, but I don't have those. So I decided to use ungrounded probes on each side of the resistor, square the voltage and divide by resistance to get power.
And this is what I got on the scope.
The M1 trace should be the voltage over the input resistor, and the M3 trace should be voltage over the resistor in the resonant circuit. The trace shows that the voltage over the resistor in the resonant circuit is about twice the voltage over the input resistor.
What do you think? Is this the way to go about measuring the voltage over a resistor? If so, why is the power in the resonant circuit double of the power over the resistor on the input? If it is not, how to improve this measurement?
I came across the Resonance Induction book that comes with the RICK kit from Rick Friedrich, at some point he shows a series-parallel circuit that contains resistors to measure power going through them. His style is such that he gives more questions than answers. I had some coils that I used for some previous experiment and some capacitors, so I gave it a try. At first I didn't measure anything unexpected and wanted to ask here if you guys knew how to do this measurement properly, but then I turned my probes from 10x to 1x and measured the following.
To describe the setup. I have a signal generator that has sine wave at output, 12 Vpkpk, tuned to about 801 kHz frequency to resonance (the resonance changes a bit when I move stuff around). The capacitors are 100 pF, 3 kV, china made, ESR unknown. The coils are 61 turns, .3 mm wire, 25 mm radius, should have about 150 uH. The resistors are 2 W 1 ohm ceramic resistors.
The setup looks like this:
In the book Rick says to measure the power on the resistor on input and the power on the resistor in the resonance circuit. I guess the best way would be to use a differential probe, but I don't have those. So I decided to use ungrounded probes on each side of the resistor, square the voltage and divide by resistance to get power.
And this is what I got on the scope.
The M1 trace should be the voltage over the input resistor, and the M3 trace should be voltage over the resistor in the resonant circuit. The trace shows that the voltage over the resistor in the resonant circuit is about twice the voltage over the input resistor.
What do you think? Is this the way to go about measuring the voltage over a resistor? If so, why is the power in the resonant circuit double of the power over the resistor on the input? If it is not, how to improve this measurement?
Hi Kloakez.
It's a bit hard to help without a schematic. If the coils are only inductively coupled, then use the ground clips. Also might help to multiply your voltage by 0.7 to get approx RMS voltage for a sine wave before squaring when doing power calculations.
Kind regards, Sandy
Fair point, thanks Sandy. I wasn't sure if I can write legibly, but I hope it helps.
I think the input voltage is not that important in this case, it will change the voltage on the resistors, but the relative difference stays the same. The signal generator can output about 0.5 W. What I find most interesting is that on input there seems to be less power than in the resonant circuit. It would mean for example that I can get twice the heating for the same input power.
(10-18-2024, 06:17 AM)kloakez Wrote: Fair point, thanks Sandy. I wasn't sure if I can write legibly, but I hope it helps.
I think the input voltage is not that important in this case, it will change the voltage on the resistors, but the relative difference stays the same. The signal generator can output about 0.5 W. What I find most interesting is that on input there seems to be less power than in the resonant circuit. It would mean for example that I can get twice the heating for the same input power.
Hi Kloakez. Thanks for schematic. I'm not EE. But if you swap positions of R1 and C1, then swap positions of L1 and R2, then you can take power measurement on both resistors with two probes and with ground at the node between R1 and R2. Otherwise just measure one resistor at a time with probe on one side of resistor and ground on other side. Also make sure you don't have ground loop between sig gen and oscilloscope. Scope probes must reference ground, AFAIK.
Kind regards, Sandy
10-18-2024, 11:14 AM (This post was last modified: 10-18-2024, 11:18 AM by neomagnet.
Edit Reason: addition
)
(10-18-2024, 04:55 AM)Sandy Wrote: Also might help to multiply your voltage by 0.7 to get approx RMS voltage for a sine wave before squaring when doing power calculations.
Hi Sandy, would like to add this to your above suggestion: it is ok for the peak amplitude (Vp) of a sine wave.
And when we use an oscilloscope and measure peak to peak amplitude of a sine wave, (Vpp), then the RMS value is received by dividing the peak to peak value by 2.82
or multiplying the Vpp by 0.354. (This latter number is the reciprocal value of 2.82 and the 0.7 is the reciprocal value of 1.41)
Thank you neomagnet, the document you shared contains the same circuit that I tested (with different values) and presents it as a novel resonant circuit that amplifies current 2x for free. I guess the values I measured are then expected results.
10-22-2024, 03:33 PM (This post was last modified: 10-22-2024, 03:34 PM by kloakez.)
So with the help of Ansys, I recreated the simulation. The electronic parts parameters are the same. Signal generator has 12 V pkpk again and can supply 1 W of power. The scheme:
Let's look for the resonant frequency.
The signal generator will be set to 816488 Hz. Transient analysis should now show the power at the resistors here inconveniently called R36 and R37. It took some tries to assemble the circuit. I skipped the transient part of the signal and zoomed on time interval of 4.99-5 ms where the signal is steady.
The voltage drop on the input resistor R36 is 154.6385 mVpkpk, and R37 330.3649 mVpkpk which is 54.68 mV RMS and 116.82 mV RMS. Then current at I_R36 = 54.68/1 = 54.68 mA, and I_R37 = 116.86 mA, which is about double of I_R36. Power at R36 is P_R36 = 3 mW, and power at R37 is P_R37 = 13.65 mW.
From the values I measured the voltage drop on... R36 is U_R36 = 64.299 mV RMS, and U_R37 = 121.76 mV RMS giving P_R36 = 4.13 mW, and P_R37 = 14.8 mW.
Pretty close. So there a good reason to think that this circuit doubles the current while the power increases in ratio from 3.58:1 (measured) to 4.55:1 (theoretical)
I will try to perform these tasks:
Make a theoretical computation about what voltage should be in this circuit at different places, the document from neomagnet contains only current calculation.
Make the same for power. If there are doubts.
Try to design a circuit, that would do this for 50 Hz. Who knows, maybe there can be a passive device that will half the current demands of circuits.
When I finish soldering a gate driver, I can put more power into the circuit which I cannot do with signal generator. That should help make more interesting power measurements.
(10-22-2024, 03:33 PM)kloakez Wrote: So with the help of Ansys, I recreated the simulation. The electronic parts parameters are the same. Signal generator has 12 V pkpk again and can supply 1 W of power. The scheme:
Let's look for the resonant frequency.
The signal generator will be set to 816488 Hz. Transient analysis should now show the power at the resistors here inconveniently called R36 and R37. It took some tries to assemble the circuit. I skipped the transient part of the signal and zoomed on time interval of 4.99-5 ms where the signal is steady.
The voltage drop on the input resistor R36 is 154.6385 mVpkpk, and R37 330.3649 mVpkpk which is 54.68 mV RMS and 116.82 mV RMS. Then current at I_R36 = 54.68/1 = 54.68 mA, and I_R37 = 116.86 mA, which is about double of I_R36. Power at R36 is P_R36 = 3 mW, and power at R37 is P_R37 = 13.65 mW.
From the values I measured the voltage drop on... R36 is U_R36 = 64.299 mV RMS, and U_R37 = 121.76 mV RMS giving P_R36 = 4.13 mW, and P_R37 = 14.8 mW.
Pretty close. So there a good reason to think that this circuit doubles the current while the power increases in ratio from 3.58:1 (measured) to 4.55:1 (theoretical)
I will try to perform these tasks:
Make a theoretical computation about what voltage should be in this circuit at different places, the document from neomagnet contains only current calculation.
Make the same for power. If there are doubts.
Try to design a circuit, that would do this for 50 Hz. Who knows, maybe there can be a passive device that will half the current demands of circuits.
When I finish soldering a gate driver, I can put more power into the circuit which I cannot do with signal generator. That should help make more interesting power measurements.
Great work, thanks Kloakez. Please keep us informed.
Kind regards, Sandy
10-24-2024, 05:02 PM (This post was last modified: 10-24-2024, 05:08 PM by kloakez.)
I guess I should not be surprised, this circuit has been used by Tesla and it is mentioned in his Colorado Springs notes. And it makes a lot of sense for a Don Smith system that uses L1 and L2 coils for powering the L1. It is already a gain mechanism by itself. The reason why it is great for a Don Smith system is because it is very easy to get out of resonance, but if there is a transformer (or a L1 and L2 coils with an air core) then the circuit is isolated and it will not get detuned by stuff beyond the transformer.
Something like this makes a lot of sense:
There is one version with three capacitors in the notes by Tesla too and I have seen it in circuits by Zilano. The generator has to be set to the resonant frequency of the circuit, both capacitors have to have the same value and the transformer in the end is the L1 of a Don Smith system. Of course, usually in a Don Smith system there would be a spark gap in parallel too, but to learn what it will do I need to do more experiments.
For those interested in this subject, try to get information on series-parallel resonant circuits. I tried to google, tried the GPT models, there is not really much information, if any. How come?
I tried to use a gate driver today instead of signal generator so I could choose the input voltage and current and it was not possible to do measurements like before, the sudden changes in the square wave signal cause too much interference.
