Mysterious Resonant Circuit

[aleks]

Since heat rise is a good relative indication of power level, I have run some temperature rise tests on a 1/2 watt 150 Ohm (approximate)  carbon resistor using two tiny thermocouples, one taped to the resistor, the other sensing ambient 77 F  (which did not move during the test)

Nice test, but DC is known for high thermic losses. Do you think at 20 MHz losses will be similar? After all, we have 50/60 Hz in a wall outlet because previously at DC losses were unbearable. For even better test you could attach a function generator.

[Vortex1]

Aleks

The RMS value of an AC waveform is defined as "the heating effect of an AC waveform when applied to a non-inductive resistor", in other words, the DC equivalent. So, if the instrument is a 'true RMS' meter, it should read the RMS value regardless of the shape of the waveform.

Most RMS measurements relate to a sinusoidal (i.e.undistorted) waveform. Simpler instruments actually respond to the peak value of the AC waveform and apply a standard calibration factor to display the RMS value on the scale, but in such cases it is only accurate for pure sine waves.

AC voltages sitting on a DC bias will give incorrect readings. You should use a DC voltmeter to measure the average DC, then interpose a capacitor to isolate the DC and measure the 'ripple' or AC component using your RMS voltmeter.

But some meters don't like excessively 'peaky' waveforms. Generally, the solid-state digital meters are accurate as long as the peak value is less than 6 times the RMS value, more than that, all bets are off.

The most accurate true-RMS meters are hot-wire ammeters. Moving-iron meters are inhrently true-RMS reading too, but their accuracy diminishes with frequency.

Analog Devices and Linear Technology make several RMS-DC converters as integrated circuits. One of the Linear Technology devices is actually a dual heater-on-a chip. It works very well.

There are True RMS meters and True RMS meters.

An average responding meter, such as a Fluke 77 will assume a sinewave. If measuring anything else, the reading will be subject to error.

Whereas a Fluke 87 TRMS meter will measure the TRMS content of an AC waveform only. If you are measuring mixed AC/DC, if I remember correctly, the manual says you should measure the AC voltage and then the DC voltage.

The next step up is say a Fluke 189. This can measure the combined AC+DC TRMS content.

At the top of the scale, so to speak, you've got a meter with a thermal converter, such as a Fluke 892X, or a Fluke 8506. This will give the most accurate results for the most distorted waveform, and/or an AC/DC mix.

Additionally, TRMS meters have a wider frequency response, an average responding meter such as a Fluke 77 will start to roll-off at 1kHz at best; possibly more likely nearer 400Hz.

A TRMS meter such as a Fluke 87 will be good to 20kHz or so.

Thermal TRMS meters are often good to MHz.

source:vintage radio UK

I've spent half a lifetime making low level power measurements. Thermal conversion is still known as the most accurate method when high frequencies and distorted waveforms are involved. IfI you can't afford the fancy equipment you can make a differential thermocouple circuit that compares an unknown temperature rise with a known temperature rise.  Cost to you: two matched resistors, two thermocouples, any cheap meter that can read millivolts since you will be adjusting the power input to the "reference" power resistor for a "null". The thermocouples are taped to the resistors and connected inversely. Need I explain more?

After all, we have 50/60 Hz in a wall outlet because previously at DC losses were unbearable.

This is another "Tesla" type urban legends that annoy me. Actually DC is not practical because of corrosion effects at dissimilar metal junctions. It is actually more efficient since it does not radiate, hence is used for the newest high voltage transmission in parts of the US grid. DC is still used throughout your automobile with few negative effects except at the battery terminals.

......V

[aleks]

The RMS value of an AC waveform is defined as "the heating effect of an AC waveform when applied to a non-inductive resistor", in other words, the DC equivalent.

Well, no argue here, but heating effect does always depend on current's frequency and characteristics of the medium. Does it mean that "non-inductive resistor" heats equally well on any frequency, and that resistor basically works as a heater? I always assumed that resistor is more of a "current blocker" than direct current to heat converter (otherwise energy coming from the wall outlet could not be limited). Your Power/temperature graph does not tell much since average surface temperature depends on the thermal capacity and thermal conductance. I mean, from temperature measurement you can't tell how much energy is dissipated into surrounding air. OK, I may be wrong, but then it's good since it means I need to rework my understanding.

[aleks]

This is another "Tesla" type urban legends that annoy me. Actually DC is not practical because of corrosion effects at dissimilar metal junctions. It is actually more efficient since it does not radiate, hence is used for the newest high voltage transmission in parts of the US grid. DC is still used throughout your automobile with few negative effects except at the battery terminals.

You probably wanted to say DC does not "inductively couple". DC is not always used in automobiles, and DC motor driver circuit is subject to heavy overheating due to excessive quantity of transients. More advanced autos use 3phase AC motor/generator. Please if you can, give me an URL that compares transmission efficiency of DC and AC power. From what I've read, AC always won. In fact, for more advanced power lines a much higher frequency (in kilohertz range) was suggested.

[aleks]

Well, I should have rememberd Joule's law: http://en.wikipedia.org/wiki/Joule%27s_Law Sorry, need to review my understanding.. (I assumed resistor works as a voltage/current blocker, but not as a result of heating which is so much inefficient).

However, why not measure voltage across all other resistors? and then sum the dissipated power up? Won't it be funny? ah, not in this scheme as another resistor is 100k which will give a low dissipation value per formula.

To sum up, I'm sorry for my rambling, I have to work on my understanding better. You probably did achieve overunity. I personally just have a problem with physical formulas - so much reality packed into so little formulas. Probably that's why physicists tend to be pretentious (I do not want to insult anyone, sorry).