another small breakthrough on our NERD technology.

[picowatt]

Rosemary,

After reading my posts regarding how to visually make measurements from the screen of an oscilloscope, do you now agree that the offset values displayed on the screen next to "ofs", have absolutely nothing to do with making measurements off the screen?

Or do you stand by your original assertion below?  Which is the very first post mentioning those offset numbers by anyone.  If you recall, you were in disagreement with my 'scope reads because I was not factoring in whatever is stated by you in the following quote:

Picowatt

Look again at those screen shots.  You'll see channels 1 through 4 - on the base line display.  It indicates that the display is DC.  Then look at the offset values that follow.  It indicates the PEAK TO PEAK VOLTAGES with their 'bias'.  That's the number to apply to the Channel 3 display as the Channel 3 display is actually and obviously AC.

Regards,
Rosemary

PW

[Rosemary Ainslie]

And MileHigh I LOVE your posts.  Always a pleasure to answer you.  Somehow I get it that your posts are not as MALICIOUS as your 'friends'. Perhaps it's because you use better English.

TK:I am also curious to know what the 2.34324 MHz means.  In looking at the DSO capture you can see that it is capturing 500 seconds and it can store 500K points (I assume per channel).  So that makes sense to see that the "1kS" on the display corresponds to a sampling rate of 1 kHz.Note that means that the oscillation waveforms are grossly undersampled because we know the oscillations are in the 2-4 MHz range.  The DSO in this case shows the negative oscillation mode "gate" signal is at 6.17 MHz.  Personally I would want to double check this because of the nature of the setup but let's assume it's right for now.

You're absolutely right MileHigh.  But the number of samples per the LeCroy is actually 500 000.  Over an extended period which applies to that first test - we certainly have an 'undersampling' - or better put - a reliance on an 'average'.  But that average is also correct - as shown in the average DC value over the batteries.  It's well within the margins of errors required to evaluate that waveform.  But well done for being the only one to point out anything valid at all.

Even through the waveform is way undersampled, in theory "Monte Carlo" methods could come into play here and the "VV" calculation is actually legit.  But I personally would much prefer to do a "VV" calculation on a tiny tiny sliver of the waveform with at least 8X Nyquist sampling, and try to line up a perfect number of cycles, to see what the "VV" calculation would say in that case.  I am not sure if the scope can sample in the 50-100 MHz range though.

We have another sample of precisely this - which is also included in that paper.  Take a look.

Going back to the subsampling in the DSO capture, I am just not comfortable with a "Double Monte Carlo" "VV" calculation.  I have never played with a DSO and explored what happens when you are subsampled so I am just guessing.In a way this is all academic because the battery "voltage" is a fake-out.

So much good sense and then this?  How is the battery voltage a 'fake out'?  It most certainly is NOT.

You have to be aware of a pattern:  Instead of trying to engage and understand the function generator current flow issue, you are belligerent and put up a fight.  Instead of trying to understand how you read voltages off of the DSO and engaging and trying to learn, you say that you are going to take this up with your "experts."

MileHigh - give me a break here.  My reference to experts only relates to the question put to me by picowatt - with a repetition that boarded on 'combative' was how it could be that the OFFSET did not correspond to the Channel 3 display of the gate voltage?  I had NO CLUE.  But now I most certainly DO understand.  The waveform across the gate should be AC coupled.  Our machine defaults to DC coupling - required for the analysis of the battery supply and for the shunt that determines the current to and from that supply.  That then explains all.  But somehow it seems that out of the 3 of you - or is it 4 - I'm the only one who understands my explanation.

It's all just you being combative for no reason.  I read PW's discussion of the offset voltages for the DSO and he is absolutely correct - and that's coming from somebody that has barely even used a DSO!  Almost no companies had DSOs in my time.  They were too expensive and probably couldn't store more than 1/1024th the number of samples that today's DSO can store.   

I'm not an expert.  But I assure you that I am well able to find my way around the LeCroy.  The Tektronix not so much.  But then I didn't have that machine for more than one week - I think it was. 

You are just wasting time and energy.  You are talking to experts but you don't want to listen to them and instead you want to fight tooth and nail all the way.  It's the old cliche, you are your own worst enemy.

WHO are the experts?  Picowatt?  Or TK.  And what exactly is their expertise?

Kindest regards MileHigh
Rosie

[Rosemary Ainslie]

Rosemary,

I'd be interested to hear what your experts say of my explanation. You could print the annotated scope shot out or email it to them.

What is your objection to the annotated scope shot Rosemary?

No Poynty.  You actually wouldn't want me to circulate that.  Or if you do - then you must also know that you're simply circulating a great number of utterly erroneous annotations.  The OFFSET value has NOTHING to do with a 'graticule' reference.  It has EVERYTHING to do with that little line at each channel reference.  It indicates the zero crossing line.  THAT'S IT.  The numbers displayed in the box - relates to the peak to peak voltage which only has relevant to an AC coupled voltage.  Therefore it ONLY applies to the AC input from the signal generator.  The center graticule 'vertical' is to help the user to select the appropriate waveforms over whatever period is intended.  Nothing else.  And I can scroll - as mentioned to the left or the right to include or exclude more or less - as required.  The trick is to scroll to give as balanced a display as possible.  And then one simply takes the screen shot from that point.  This is better enabled if one freezes the shot and then adjusts as required to get that sample best represented.  If one takes a shot in real time then one may not have a representative shot of those waveforms.  But frozen or in real time - it still gives an accurate value of the mean averages where they're DC coupled - or the bias where they're AC coupled.  We never 'touched' the AC coupling as our interest is in determining the power from the battery.  And that needs a precise reference to ZERO.

Kindest regards,
Rosie

[MileHigh]

I was not factoring in whatever is stated by you in the following quote:

I had to LOL about that one!  We need secret decoder rings!

C'mon Rosie Posie you need to shift gears.  Did you see that nice annotated diagram explaining the meaning of the DSO offsets to you that Poynt made?

You have to grok Rosie....  Grok grok grok....  And then thank Poynt for making the effort to explain the concept to you.

Try getting over some small learning humps.  Take some baby steps.

The goal should be to draw up a simple test plan that everyone agrees on and do the actual testing.  You are going to be in shock when the light bulbs dim out sooner than you expect.  We dare you to follow through and prove us wrong!

You know what I have said... We are not mindless guppies swimming against the glass in a fishbowl.  Logic and reason have to prevail.  When you say stuff like, "I assure you that if the current flow through the resistor was that 'out of phase' then there would be absolutely NO evidence of any heat WHATSOEVER."  it's mindless guppy talk.

Do you want to draw up a test plan and then do the dim bulb testing?  If you do you desperately need our help.  You need an epiphany and you have to open up and engage and try to learn.

Right now you are inside your own fishbowl and don't realize how you are being perceived.  You have to engage and be civil and express a desire to learn.  Without that you are toast!

That's my last try.

MileHigh

[Rosemary Ainslie]

And...ONCE AGAIN.... I offer Tar Baby to any interested third party with the equipment and knowledge to test it, as long as Ainslie's device is tested alongside, by the same protocols and analyses. I don't even care what they are, right or wrong.... because my ONLY CLAIM is that Tar Baby performs just like Ainslie's device in all significant respects.

SIDE BY SIDE TESTING. I'll gladly send this box of stuff off to anyone for testing, as long as Rosemary does the same, the devices are tested and analyzed the same, and the results are published openly.

I will NOT associate with you TinselKoala - on any formal basis.  You are utterly disqualified as a representative of anything other than an absurd attempt to DEBUNK.  And I would NEVER presume to associate our NERD circuit with that absurd apparatus that you IMPLY is a REPLICATION.  It fails in ALL aspects - starting with the voltage reference and the TYPE of LOAD that you're using.  And it ENDS with the entire inability of your equipment to manage the required power analysis - OF ANY KIND let alone the required and careful tuning to establish the test parameters. 

Rosie Pose