"COOL JOULE" makes a claim

[Matt Watts]

You wanted to see power dissipation that makes no sense,then watch the 3 video's below.

The only thing I can suggest Brad, is for these types of signals, one really needs a measurement tool capable of true signal integration--much different than average voltage.

My exploration in this topic has led me to three choices:
  1.  If you have a nice pure sine wave, you can use RMS values for power calculation.
  2.  If you have a nice clean square wave, you can use average or mean values for power calculation.
  3.  Anything else requires you somehow get back to the source power and make it flat-line DC and measure there.

Also, attached is the spec sheet for a device I have found rather inexpensive and very valuable for current measurements.  I have put it side-by-side against precision current shunts and the accuracy within its frequency range exceeds that of my scope.  In certain situations this device really reveals what adding any resistance to a circuit actually does.

The old me is still here ,alive and well Erfinder .

Good to have you back Brad.  I was starting to worry a little you might be a bit bipolar--all seems to matter what mood you're in when asked a question.

This stuff is more complicated than it looks and having two perspectives active while analyzing a device is helpful--so please try to kick-in both the new and old Brad at the same time if possible.  And always remember you have other sets of eyes trying to help and understand with you.  I'm pretty sure we all are seeking and hoping for the same goal.  It only takes one of us to move us all in the right direction.  You have done many experiments I go back and refer to when I'm not sure what I'm looking at.  So yes Brad, we do appreciate what you do and together we can get there.

[Matt Watts]

Below is an example where power measurements become non-trivial.

I have a 92uF cap connected to an air-core coil, approximately 17 meters of wire.

Scope channel A (yellow) is monitoring the amperage circulating in the tank circuit.

Scope channel B (blue) is monitoring the current driving the tank circuit from my PA amp.

I have tuned the frequency for the lowest possible amperage required from the PA amp.  Notice the signal shape.  The actual amperage here is something less than the peak of 2.88 amps and more than 720mA RMS, but by how much...?   Very difficult to say unless you integrate across one complete cycle.  I seem to recall some high-end scopes can actually do this calculation, but for the rest of us...

However, the amperage in the tank circuit shows a nice sine wave and we can be pretty sure the 10 amps RMS is quite accurate.


Of course the trick now is, can I extract this cyclic 10 amps in the tank circuit without adding any more amps to the driver...?   Supposedly Dally did it and Ruslan, but can we do it?

[Smudge]

The rms function on a modern sampling 'scope should not be limited to sine waves but should be accurate for any wave shape.  Doesn't the SigLent manual make this clear?  For a series of sampled values the math rms function should square each value, take the mean value of that series of squared values then take the square root of that mean value.  The important thing is to ensure that the samples are taken over a full cycle or a number of full cycles, else there is an error due to incomplete sampling.  Alternatively the scope can take a running mean value over many cycles where the error gradually reduces to insignificant proportions.  I would guess that the SigLent 'scope does just that.  However if the math channel just takes the peak to peak values then uses the sine wave calculation we are all familiar with then of course it is limited to sine waves, but I can't imagine a scope manufacturer doing that.

Smudge

[wattsup]

Wow so Camelot is back to its peaceful mode. Great.

@TinMan

In the videos when the cap is connected the scope frequency reading goes blank, so there may be a relation to what you are seeing because of that. I know when the waveform goes off the screen because of a wrong volt/div setting, the readings can go blank, but is it the same when the frequency is not shown or read properly.

I get what you are trying to show there but there are so many variables in this set-up that it would require way to much analysis for me right now as I am busy on some other stuff.

But I think the 1st 25v bulb always had enough amperage already, more then it needs so it is passing some to the motor stage when that cap is shorting out the 1st bulb and battery.

Reading the light bulb through a solar panel is another quandary because it wants light which is really only available via voltage and not a function of amperage when amperage is always sufficient more will not produce more light. I don't think you can equate brighter light with more amperage. Hmmmmmm. It's hard to put in words.

I personally do not work my bench like that. I rarely use the ground clip of the scope except when I know what I am trying to show but in general just the probe is enough to show what is going on at that one point, which I am more interested in then a differential reading across two points.

Guys will say, yes but what you see is not a differential and how do you make due without a reference. Well my reference is the open ground to atmosphere which is the same everywhere around. But more to the point, with just the probe poking around and seeing the different waveforms is the point of reference as a fair comparison by itself.

I think if there was a formal method to use only the scope probe and be able to rationalize those waveforms, because sometimes they show some very crazy results, that guys will get a better heartbeat of their devices then by simply reading differentials. Example: What would you see if the CH1 probe was on that second bulb without ground compared to a CH2 referenced reading across the CVR. hahaha

wattsup

[Matt Watts]

@ Smudge,

To be honest, I'm not real confident in the numbers my scope is showing.

Based on this explanation:
http://www.electronics-tutorials.ws/accircuits/rms-voltage.html

It seems to be using the graphical method vice the analytical method.

The only way I would know for sure though is to dump the sampling points to a file and calculate them by hand.

Also, this being an AC signal, I would really need to split the numbers into two segments, each showing the direction of current flow.  In the case of my PA amp, current that flows back into it is likely just lost and cannot be assumed as power generated by the DUT.  Unfortunately, with the current and voltage in-phase, this is real power that could do real work if properly accounted for and distributed as appropriate.  This is something Jim Murray talks about when passing source power through a load and to some device.  If the device sends power back to the source, in effect the load gets access to this power twice even though it was only generated once.

I guess in summary and to my point, power calculations cannot be assumed see-spot-run based on a few screen shots.  Real number crunching and verification is required.

[Smudge]

@ Matt,

That RMS Voltage Tutorial treats the "graphical" method as though it only applies to sine waves, and that is not true.  That "graphical" method (which is not really graphical but is classical sampling method) applies to square waves, triangular waves, pulses, in fact any waveform.  And since your scope performs that "graphical" method you can have confidence that it gives correct results if done over full cycles or as a running value over many cycles, and at sufficiently high sampling rate.

Smudge

[Matt Watts]

I think I'll do some data dumps and do a crosscheck on a spreadsheet.

You've got me very curious now.  I have always taken those readings with a grain of salt, but it's looking like I was incorrect in my belief about these modern test instruments.

Thank you Smudge for jumping in here and setting me straight.

[Matt Watts]

Also, this being an AC signal, I would really need to split the numbers into two segments, each showing the direction of current flow.  In the case of my PA amp, current that flows back into it is likely just lost and cannot be assumed as power generated by the DUT.

I got thinking about this some more...

Would it be valid to say that ALL power in AC form flows from the source, like a PA amp?  I'm thinking so.  Power does not flow from the DUT to the source.  All that happens is the source changes direction.  So if you have in-phase (voltage & current) sine wave signals, lets say 10 watts peak-to-peak, the actual power delivered by the source can never be zero or negative.  You have the positive side PLUS the negative side.  So in my example of 10 watts PtP, that's 5 watts (positive) * 0.7071  PLUS  5 watts (negative) * 0.7071 or more simply just 10 watts * 0.7071 = 7.07 watts.

For the case when voltage and current is out-of-phase, I still think the same is true.  The only exception (and I'll have to do some serious thinking about this) is if by some means the source is actually working as a sink for part of the cycle, meaning the voltage kicked-back by the DUT exceeds the voltage of the source, in which case the voltage potential dominates the direction of current flow.  I'm pretty sure this is what was happening with Bill Alec's transformer device--he wasn't taking into consideration the direction of current flow due to the difference in voltage potential.  Likely he just added the whole cycle to the wrong side of the equation instead of just the small part where the source & sink trade spots.

To me, the part I underlined would be a very difficult measurement to make, because in all likelihood the source and DUT are connected together; you can only get one voltage reading for both, so you have no good way to see the voltage on each the source and DUT.  If you could get these two measurements somehow, then you would know exactly when and where voltage domination takes place and thusly true current flow direction.

Just a thought and some ramblings to consider.

M@

[Allcanadian]

@Erfinder

I don't think you or anyone for that matter can or even should care about what I think is missing Brad.  I once thought that this was a community filled with renegades.  I reread the forum creed the other day and cried (figuratively speaking....)  When did "free energy research" become methods for proper measurement and measurement practices?  As I am a thinking person, I double checked the definition for "energy" and was reminded of the fact that the term itself has no direct relation to measurement or measurement practices....go figure....  It's rather fascinating seeing how the separatists escape the box only to take with them to their camp, tools and practices which are used to build a more ridged box.....(....face palm slap smiley.....) Maybe a name change is in order, or, and this is probably more accurate, I don't belong here.  See, I am looking at how one can do work for free, plain and simple.  I thought this was the magic frequency, or, the common thought that brought us all together.  Equating the idea of "free energy" with "overunity" derailed the runaway train, the train being symbolic of the "free energy" movement. 

Well said and I couldn't agree with you more. I also like your other comments concerning the phrase "free work" which seems more fitting in light of the quagmire "free energy" has become.

AC

[Smudge]

I got thinking about this some more...

Would it be valid to say that ALL power in AC form flows from the source, like a PA amp?  I'm thinking so.  Power does not flow from the DUT to the source.  All that happens is the source changes direction.  So if you have in-phase (voltage & current) sine wave signals, lets say 10 watts peak-to-peak

You shouldn't talk of power being peak-to-peak, that is reserved for voltage.  For your in-phase example the power is delivered in positive bursts, so only peak power is relevant.

the actual power delivered by the source can never be zero or negative.

Of course it can.  When both voltage and current are passing through zero there is no power being delivered.  Power is energy flow, and with sine waves the energy is delivered in bursts, each half cycle delivers a known quantity of energy.  The rms function is merely a way to create the equivalent averaged power as a continuous energy flow that we know as DC.

You have the positive side PLUS the negative side.  So in my example of 10 watts PtP, that's 5 watts (positive) * 0.7071

No, you are confusing PtP voltage with PtP power.  The O.7071 applies to the peak voltage.  That gives you the effective DC voltage that would supply the same average power as the peaky sine wave.  If you have a 10 watt peak power (that repeats every half cycle) its average is 5 watts. 

PLUS  5 watts (negative) * 0.7071 or more simply just 10 watts * 0.7071 = 7.07 watts.

No. no, no.  You can't add power for successive half cycles.  If you follow that road you would add another half cycle power and get an even bigger number, and then another and so on.  If you want to add things then you add the energy delivered each half cycle.  And in the 10 watt peak power example the energy (joules or watt-seconds) in each half cycle is 5 watts multiplied by the half-cycle time .  The next half cycle delivers the same amount of energy so the total energy in 1 full cycle is now 5 watts multiplied by the full cycle time.  That quantity of energy is delivered at the cycle rate which is the frequency, so the average power becomes 5*cycle-time*frequency and since cycle-time=1/frequency you end up with 5 watts (joules/sec).

For the case when voltage and current is out-of-phase, I still think the same is true.   

If you take 90 degree phase shift as an example there are points in time where either voltage or current is zero, hence power flow is zero.  There are regions where both voltage and current are positive where you get power delivered from source to load.  There are regions where both voltage and current are negative where you also get power delivered from source to load.   And there are regions where one is positive while the other is negative where power flows from the load back to the source.  This is quite often referred to as "reactive power flow" because that 90 degree situation applies to inductors and capacitors that store energy on the forward flows, then give that energy back on the reverse flows.

The only exception (and I'll have to do some serious thinking about this) is if by some means the source is actually working as a sink for part of the cycle, meaning the voltage kicked-back by the DUT exceeds the voltage of the source, in which case the voltage potential dominates the direction of current flow.  I'm pretty sure this is what was happening with Bill Alec's transformer device--he wasn't taking into consideration the direction of current flow due to the difference in voltage potential.  Likely he just added the whole cycle to the wrong side of the equation instead of just the small part where the source & sink trade spots.

To me, the part I underlined would be a very difficult measurement to make, because in all likelihood the source and DUT are connected together; you can only get one voltage reading for both, so you have no good way to see the voltage on each the source and DUT.  If you could get these two measurements somehow, then you would know exactly when and where voltage domination takes place and thusly true current flow direction.

But the current direction is easily monitored so you then know in which way the power is flowing.

Smudge

[Matt Watts]

I agree Smudge.  I should have stated "per cycle".  The chunk of which is transferred in a specific time region/sector.


The last part still has me thinking though.  Sure I can collect data that gives me the direction of current flow, but still sitting in my mind is the voltage potential between source and DUT.  I also cannot put my finger on why I see AC power differently, I just do.  It's not a zero referenced thing.  From a single vantage point, it is a push/pull phenomena which takes energy to make happen either way.  If we color code two leads from the source as black and white, the source may be pushing on the white lead and pulling on the black lead.  Then it changes direction and pushes on the black lead and pulls on the white.  Or you could say it does nothing on white lead and only pushes or pulls on the black lead.  My gut feeling is the frame of reference is important even though the abstract concept seems the same.  My thought on this has to do with charge which we don't normally reference to the environment; instead we just connect the white or neutral lead to ground and call it good.  There is no mandate to do this however, giving us the concept of "floating", which we can live with using a DMM, but not so easy with a scope which typically forces us to ground one side.  So suppose for a moment we have a source and DUT that is floating and after running for a few minutes we detect the whole DUT is now sitting at a voltage potential of 50,000 volts referenced to earth.  How did that happen and where did the charge imbalance come from?


I understand the concept of reactive power, but at the same time I'm not completely convinced.  If the voltage at the DUT is lower than the voltage of the source, but the measurement of the current is negative, I'm compelled to think this is not reactive power, i.e. the DUT is not pushing power back to the source.  Again, we can't know what those two voltages are exactly at any instant in time because they are hooked together and there is only one voltage to measure.  We are kind of stuck.  The only way to truly measure the two voltages is to break the circuit and by doing so we completely destroy the natural dynamics of the system, intended or otherwise.

I guess what I'm getting at is a means to validate how systems like Ruslan's, Akula, Dally and many others could possibly operate.  Like a lawyer looking for a loophole, I'm trying to see if there is something we missed.  A case where if you did standard measuring practices, you would actually kill any OU that might exists, but when simply connected to a light bulb for example, clearly shows something anomalous.  If I could actually spot where we go wrong in taking measurements, maybe this could provide a means to build similar devices as well as explain how to properly test and tune an OU device.  I know there are several very qualified people (like yourself) that test systems with high precision, known principals and prove beyond any reasonable doubt they operate as designed in the under-unity realm.  I ask myself though, could this be the very reason we don't have a flurry of OU devices.  Is it possible the very reason we don't is because we don't test and measure them with a framework that allows for the very phenomena we seek to expose itself.

[Allcanadian]

Hey Matt

I enjoyed your last post and your moving in the right direction and I started writing a response and wanted to explain many of the things I know may be going through your mind then deleted it. Then I thought about sending a PM for a chat but that's not going to fly. My only advice would be to stop listening to most of the people here, I know you want to and your looking for direction but they cannot help you only you can... it's that simple.

In fact I was where your at a very long time ago, stuck between a rock and a hard place. Knowing something was seriously flawed in other's logic but not knowing how to proceed and it is not my place to tell you. So here is my answer, your subconscious mind has most likely already put most of the pieces together but your not listening. Your trying to force a square peg into a round hole and it just ain't working because you want to comply and get along with others. The problem here is others cannot help you and these thoughts your having are not theirs because your not them.

So I will put this to you the only way I know how, you need to discard this dependence on others and find your own way. You need to stand on your own two feet and let your own good judgement guide you despite what others may think. Lose the measures and equations and focus on concepts, discard the terminology and semantics and embrace simplicity, do what you do not for yourself but something greater than yourself.

Myself I'm a farm boy, born and raised, but that doesn't mean I'm stupid and every time I look up to the stars I know we were meant for something better than the fucking stupidity I see down here. We are given a short time here and what we do matters, do your own thing, make a few waves, follow your own arrow, have some fun and never follow the beaten path... life is but a brief moment in time my friend...be great in your own time whenever you can.

Merry Christmas and a awesome new year.

AC

[Farmhand]

I agree Smudge.  I should have stated "per cycle".  The chunk of which is transferred in a specific time region/sector.


The last part still has me thinking though.  Sure I can collect data that gives me the direction of current flow, but still sitting in my mind is the voltage potential between source and DUT.  I also cannot put my finger on why I see AC power differently, I just do.  It's not a zero referenced thing.  From a single vantage point, it is a push/pull phenomena which takes energy to make happen either way.  If we color code two leads from the source as black and white, the source may be pushing on the white lead and pulling on the black lead.  Then it changes direction and pushes on the black lead and pulls on the white.  Or you could say it does nothing on white lead and only pushes or pulls on the black lead.  My gut feeling is the frame of reference is important even though the abstract concept seems the same.  My thought on this has to do with charge which we don't normally reference to the environment; instead we just connect the white or neutral lead to ground and call it good.  There is no mandate to do this however, giving us the concept of "floating", which we can live with using a DMM, but not so easy with a scope which typically forces us to ground one side.  So suppose for a moment we have a source and DUT that is floating and after running for a few minutes we detect the whole DUT is now sitting at a voltage potential of 50,000 volts referenced to earth.  How did that happen and where did the charge imbalance come from?


I understand the concept of reactive power, but at the same time I'm not completely convinced.  If the voltage at the DUT is lower than the voltage of the source, but the measurement of the current is negative, I'm compelled to think this is not reactive power, i.e. the DUT is not pushing power back to the source.  Again, we can't know what those two voltages are exactly at any instant in time because they are hooked together and there is only one voltage to measure.  We are kind of stuck.  The only way to truly measure the two voltages is to break the circuit and by doing so we completely destroy the natural dynamics of the system, intended or otherwise.

I guess what I'm getting at is a means to validate how systems like Ruslan's, Akula, Dally and many others could possibly operate.  Like a lawyer looking for a loophole, I'm trying to see if there is something we missed.  A case where if you did standard measuring practices, you would actually kill any OU that might exists, but when simply connected to a light bulb for example, clearly shows something anomalous.  If I could actually spot where we go wrong in taking measurements, maybe this could provide a means to build similar devices as well as explain how to properly test and tune an OU device.  I know there are several very qualified people (like yourself) that test systems with high precision, known principals and prove beyond any reasonable doubt they operate as designed in the under-unity realm.  I ask myself though, could this be the very reason we don't have a flurry of OU devices.  Is it possible the very reason we don't is because we don't test and measure them with a framework that allows for the very phenomena we seek to expose itself.

TK does floating measurements with his scope, he has a scope with a switch installed so he can have the scope floating or grounded just from memory, I think he may have installed it himself or the scope already had it, but I am fairly sure he does do floating measurements when he feels it is proper.

Speaking of incandescent light bulbs and current, I think I remember showing at OU dot com a HF circuit I made up which appeared to pass way more current through a bulb than it was rated for but the phase difference was there and with the power factor considered the bulb was apparently using about the right amount of power for the calculations and a lot of current went "unused" as it were. The bulb itself had a fair bit of inductance considering the frequency I used and that also figured in with the phase difference ect.

I still have the setup in working order, it produced some crazy tank voltages for the turns and input voltage, and perfect sine waves with a DC pulsed input, though they deformed under some loads, lots of resonant rise, the setup was able to hold a resonant rise while powering a fluro tube with real current spikes through the tube but the power the tube consumed was difficult to define for me at the time. I could do better. I think it would all work out considering all energies ect. but that's neither here nor there. The incandescent bulb passing more current than it consumed due to phase difference at HF is noticeable, the inductance of a regular incandescent bulb shows itself, such a bulb can be made to produce spikes just like a coil as well, if it has inductance of course.

[TinselKoala]

Just to clear things up.... as far as "floating" the scope goes.... this can be a dangerous practice. It is much better to float the DUT than to float the scope, from a safety viewpoint. The fact that the scope probe reference leads are connected by the Mains connection to an Earth ground can be a valid concern when testing devices that have unknown, maybe "OU" characteristics. In cases where that issue has been raised, I have sometimes used an isolation transformer that doesn't have a straight-through ground connections as many do, or a "ground-lift" adapter that simply removes the ground prong from the mains connection, or I've used scopes that are actually ground-isolated themselves such as the Fluke 123 or 199 ScopeMeters. This is not my normal practice though.

The "switch" that Farmhand remembers is probably the one on my old Interstate (Racal-Dana) F43 FG, which has a built-in ground isolation switch that floats the signal "ground" or Black, BNC shield, output connection. This switch is normally on the rear panel of the instrument which is hard to get to in my "stack" so I installed a parallel switch on the front panel. Not too many more modern FGs seem to have this feature, I think. Their "black" or BNC Shield outputs are connected permanently to chassis and Mains ground. For instruments that can be isolated, of course whenever the BNC connections are used to connect the FG to another, grounded instrument, this isolation goes away, so one has to be careful using it. The same is true when "floating" the scope itself: as soon as a probe reference lead is connected to something else like another instrument's ground that isn't floated, the isolation is removed.
I've taught myself to be very conscious of inadvertent groundloops of this kind; that is one reason why I'm an advocate of clean, simple layouts so that the chances of making wiring mistakes are minimized.  And I do this because of costly mistakes I've made, myself! Fortunately only instruments have suffered, I haven't electrocuted myself (yet)! 

@Smudge: Unfortunately TinMan's scope has a little weirdness concerning the Math trace. While it does the usual Math functions, like trace multiplication of a voltage and a current trace to obtain an instantaneous power curve, it _does not_ do any automatic Measurements on the resulting Math trace. SO it can't even take the "mean" or average value of the Math trace or p-p or any other measurements of it. This may be a flaw in the old firmware or it may just be a peculiarity of the particular model of the scope. The User's Manual for the 1000 series Atten scopes gives instructions for getting measurements from the Math trace but TinMan's scope doesn't seem to have the necessary options in its Menus, so he can't do it. We've gone over this several times and it doesn't seem to be "user error", it's a flaw in the scope itself. It may be possible to download the channel trace data to a CSV file for a spreadsheet and use the spreadsheet to generate and then get measurements on the Math trace but as far as I can tell he hasn't tried to do this yet. It seems very strange to me that a scope with Math capability can't do measurements on the resulting Math trace, but there it is.


As far as TinMan's experimental setup where the bulb got brighter but the "average" or mean current through it was less... I did repeat and reproduce that effect using a simple pulsed circuit and did some measurements on it. We had a big discussion on OU about it, where we showed that using "average" values of voltage and current gave wrong results when computing average power. My scope, fortunately, does do Measurements on its Math trace, so I was able to show that the bulb brightness in the experiment follows the power, as it should. Less brightness, less power dissipated in the bulb according to the properly obtained and computed measurements, and more brightness = more power dissipated in the bulb. The Red Herring was only using the mean current instead of doing the instantaneous multiplication of current x voltage and then taking the mean of the resulting Instantaneous Power trace. 

[Farmhand]

My bad Tinsel, It was the Function Generator then and not a scope, I remembered falsely. Concerning the Math trace and no calculations my RIGOL DS1052E scope does produce a math trace but there is no way to get any Math measurement from it directly, as a readout. I remember someone urged me to produce a CSV file to get a calculation from it but I had no idea what to do with it. My bad again. I should learn. The limited Math function of my scope is a sore point with me as I thought it would do it considering it offered the Math trace/function. But no dice. I sure cannot afford to buy another better one.

Thanks for clearing that up Tinsel, appreciated muchly. Still no Cheeseburgers ?

[TinselKoala]

Below is an example where power measurements become non-trivial.

I have a 92uF cap connected to an air-core coil, approximately 17 meters of wire.

Scope channel A (yellow) is monitoring the amperage circulating in the tank circuit.

Scope channel B (blue) is monitoring the current driving the tank circuit from my PA amp.

I have tuned the frequency for the lowest possible amperage required from the PA amp.  Notice the signal shape.  The actual amperage here is something less than the peak of 2.88 amps and more than 720mA RMS, but by how much...?   Very difficult to say unless you integrate across one complete cycle.  I seem to recall some high-end scopes can actually do this calculation, but for the rest of us...

However, the amperage in the tank circuit shows a nice sine wave and we can be pretty sure the 10 amps RMS is quite accurate.


Of course the trick now is, can I extract this cyclic 10 amps in the tank circuit without adding any more amps to the driver...?   Supposedly Dally did it and Ruslan, but can we do it?

Matt, in your case I have a couple of comments. First is that I think your scope does have Math capability, so you should be able to generate an instantaneous power curve by multiplying a Current Trace by a Voltage(drop) trace. I don't know if your scope will have the same problem doing measurements on the Math as Brad's scope does (since Atten and Siglent scopes are very similar). I'd suggest that you try and see if you can get a Math trace and then get your automatic Measurements to find the mean and other values for the Math trace.
Performing the instantaneous multiplication of V and I and then getting the mean of the resulting trace is the proper way to compute "average power" with the scope. This is basically integrating over time and then dividing by the time, to obtain an average Joules per second value, ie Watts.

However.... the Current Sense transformer that you cited is probably introducing a Phase Shift of its own in the measurement it displays. You could test for this by doing a comparison of the signal from the CST, with the Vdrop across an inline current-viewing-resistor, on a pulsed or sinusoidal signal, to see if there is a phase shift between the readings. If you do have a phase shift this means that you can't use the Math to do a simple IP multiplication without first correcting for the phase shift.

This is why it is often better to use a CVR rather than a transformer when doing power measurements. Even expensive actual Current Probes will have propagation delays and phase shifts too, so even when using these kinds of probes you still have to be aware of and correct for these factors when using them for Power measurements.


The scope can probably compute the RMS value of a complicated signal properly, so you should be able to use and compare the RMS values of both current signals in your scopeshot. How useful this information is, is questionable though. I'd suggest that you do a proper _average power_ comparison of the input and output powers, since as TinMan's experiment shows, simply comparing the current values can lead you down a garden path chasing after those slippery Red Herrings.

Getting usable power out of the circulating Reactive power in a tank circuit is something I've addressed in some videos. Basically, the tank is being pumped up by the input source to a value where losses equal the input. Extracting power from the tank can be considered a further "loss". So you can extract real power from the tank, and sometimes this substitutes the extraction loss for some other loss like radiation or heat dissipation in the tank, so the tank can stay "pumped up". But if your extraction exceeds the input source, then the tank will deflate, and perhaps the input source will have to supply more power to keep it pumped up, or perhaps the tank will simply collapse.

[TinselKoala]

My bad Tinsel, It was the Function Generator then and not a scope, I remembered falsely. Concerning the Math trace and no calculations my RIGOL DS1052E scope does produce a math trace but there is no way to get any Math measurement from it directly, as a readout. I remember someone urged me to produce a CSV file to get a calculation from it but I had no idea what to do with it. My bad again. I should learn. The limited Math function of my scope is a sore point with me as I thought it would do it considering it offered the Math trace/function. But no dice. I sure cannot afford to buy another better one.

Thanks for clearing that up Tinsel, appreciated muchly. Still no Cheeseburgers ?

Grr. Yes, I've just reviewed the Manual for the DS1052E, and you are right, it does not allow automatic measurements of the Math channel. But it can at least use the Cursors on the math, so you can get Peak and P-P values anyway. I'm flabbergasted that these scopes don't do Measurements on the Math!

Using the saved CSV values requires some experience with spreadsheets like Excel or OpenOffice/LibreOffice.  The CSV stands for "comma separated variables" and what you should get is a file that has two columns representing voltage and time-slice, and the files can get very large if you have the entire buffer rather than just the screen data in the file. This "CSV" file is then imported into the spreadsheet program, then you can use the spreadsheet to graph the data, and/or to do various math functions on it.

I couldn't tell from the Manual whether the scope will save a CSV file of the Math trace, or just the CH1 and CH2 traces. If you can generate a CSV file from your scope (just the screen data not the whole buffer!) and attach it to a comment, I can try to put it into a spreadsheet for you and do some analysis and send it back to you so you can see what the process is.
But I've noticed that some people are complaining that some Rigol scopes don't even save the CSV file properly...  I haven't tried it on my DS1054Z so I can't speak to that issue yet.

[Farmhand]

I think if I use the Ultrascope software on a windows computer it will do it, not sure about save to a file on a usb stick but I have a feeling it will, I can't recall. I haven't set anything up at the new residence as yet but I might just get out the scope and an existing circuit I have here just to check the CSV file to usb stick thing. I need to swap some computers around before I can connect a more powerful PC to the scope than the old laptop, the program will run better then, it was laggy at times on the laptop and seemed out of sync with the scope, kinda sus. i'll try it with a better pc and see, maybe the cable could have been faulty too, I think I remember a faulty connection cable.

[Farmhand]

Yep it does save CSV file to usb stick and it can do displayed or maximum data also there is a PARA SAVE option, which I haven't read about yet. Incidentally I have the scope on an aluminium benchtop and I hung the scope leads over another thing with no power to it but the blue probe tip was touching the bench top without the ground clips touching the bench top which resulted in the scope displaying a 20 v peak to peak sine wave, and when I touched the bench top it increased a fair bit, so obviously if the scope is not used correctly some really inaccurate measurements could be made, or assumed. 

I just saved to CSV file what the scope displayed. 

P.S. There is a huge difference in file size when comparing the Displayed data CSV file (20 kb) to the maximum data CSV file (268 kb).

Oh I misread your post TK, I'll get a math trace on screen and save the CSV.

[Farmhand]

Sorry I misread your post, Here is a CSV file with a math trace on the screen AxB. I had to change the file extension/type to bmp as the site won't allow the uploading of CSV files, you should be able to just change it back.

[Farmhand]

Below is an example where power measurements become non-trivial.

I have a 92uF cap connected to an air-core coil, approximately 17 meters of wire.

Scope channel A (yellow) is monitoring the amperage circulating in the tank circuit.

Scope channel B (blue) is monitoring the current driving the tank circuit from my PA amp.

I have tuned the frequency for the lowest possible amperage required from the PA amp.  Notice the signal shape.  The actual amperage here is something less than the peak of 2.88 amps and more than 720mA RMS, but by how much...?   Very difficult to say unless you integrate across one complete cycle.  I seem to recall some high-end scopes can actually do this calculation, but for the rest of us...

However, the amperage in the tank circuit shows a nice sine wave and we can be pretty sure the 10 amps RMS is quite accurate.


Of course the trick now is, can I extract this cyclic 10 amps in the tank circuit without adding any more amps to the driver...?   Supposedly Dally did it and Ruslan, but can we do it?

Maybe if the circuit is floating or grounded in a certain way and somehow synced to the house power phase, it might be possible to light some LED's by utilizing some stray mains power. 

Matt here is my set up that I first put together for another reason but ended up getting serious about measuring accurately some parts of it in use, in the clip  I am using inductive resisters for CSR's but I did order and fit the circuit with non inductive resisters and have it all setup but it is still up the mountain, illness stopped me suddenly from continuing. But I am much better now. I'll go get it tomorrow and see if I can get some accurate measurements from it, one tank each side and each can get up to 1600 volts peak to peak from a 12.5 volt input and less than one amp input (with no load), but being as it operates at about 420 kHz a lot of energy is radiated and measuring is very difficult due to HF stray stuff and all that bizzo. But I'll have another go and see if things look logical with non inductive resisters and a better layout scheme.

HF circuit, you can skip through to see the more interesting parts by using the preview function on you tube-run the mouse along the time line.
https://www.youtube.com/watch?v=vFOHk_0IDZg

[TinselKoala]

Sorry I misread your post, Here is a CSV file with a math trace on the screen AxB. I had to change the file extension/type to bmp as the site won't allow the uploading of CSV files, you should be able to just change it back.

OK, so I've played around with your CSV file in the LibreOffice Calc spreadsheet. There is something weird going on for sure. Did you change some scales or something, or maybe the CSV file isn't reflecting the actual channel voltage scalings properly? Do you have a screenshot of the scope's display of this same data? The screenshot up above appears to have more data on it than what I got in the CSV file, did you change the timebase?

Anyway, here's the process. First I just doubleclicked on the CSV file (after changing the extension back to CSV). It automagically opened the LibreOffice Calc spreadsheet import box. Then I imported it, with the "detect special numbers" box checked and the "delimiter" selected as '  .  (Otherwise the CSV's expo notation is imported as text rather than numbers.) Then I widened some columns, selected 5 places after decimal for the number format.  The file only contained columns for the time and the two channels, no math. So I added a Math column and had it compute the multiplication of the two channels values. Then I made a plot of the data range, using the "x-y" style plot format, and played around with the axes and etc. to get it to display properly. The spreadsheet math functions did not include RMS, so I had to make up a function to compute the RMS values of the two channels data, and this required having the Squared values, so I added two more columns and had the spreadsheet compute the squared values, then I computed =SQRT(AVERAGE(columnvalues)) to get the root-mean-square values for the CH1 and CH2 data. Getting the average Math was easy using =AVERAGE(columnvalues).

But the values for the CH1 and CH2 voltage values that went in in the first place don't seem right when compared to the "Stray Voltage" screenshot you posted up above. So I don't know really what's going on. I'll have to try to set up something on my scope and look at its CSV values to see if I can figure out if there is any discrepancy in scaling, or what.

[Farmhand]

Umm yes and no to those questions respectively. I may have changed the scales as I produced the waveform in a haphazard way, and no the scope shot above  won't match the CSV data because I used a slightly different method, I touched one probe tip to the bench top (aluminium) and the other probe tip to my right thumb. Which produced a more sinusoidal yellow trace, then I think I had to adjust some vertical scaling to fit the wave forms on the screen but keep them as big as possible, and I did not capture a scope shot of that. My head was not completely in the game. I'll look at the file you produced now, I have Apache open office.

P.S. WOW, TK that is frickin awesome, the math trace looks almost exactly like it did on the screen. It is quite possible the scope does produce an accurate CSV file.

I just woke up so I'll go get a circuit from the mountain in a few hours and try to get it set up early tonight, I'll have to get some kind of table down here to work on. I just gotta hope my cars tow ball is high enough to tow a big trailer meant for a 4x4. Then hope they will lend it to me. Or I could just make up a tank and use the function generator to excite it so I can get a sensible pair of wave forms at a lower frequency from a small setup to test the validity of the CSV data, and to learn to do it myself. I don't have a non inductive resister here but I do have one or three spare ones loose up the mountain.

As those were two voltage measurements out of phase the Watts figure is only demonstrative. But it looks very good. I am suitably impressed.   
First time I've used the scope in almost a year, time fly's.

Thanks, you've kicked started me again, I feel the need to learn how to do that myself quite acutely.