Testing the TK Tar Baby

[TinselKoala]

Now, look at this. Groundloop asked Ainslie what was the level of the GATE SIGNAL.
He simply and succinctly asked, "And I also would like to know what voltage your function generator was at when you got the 320mA current?"

And this is her entire answer, with my emphases and comments interspersed.

Groundloop - I STILL do not understand why you even did that experiment.

That's right... because she is naive to the scientific method, she cannot understand why people ACTUALLY do experiments: to see what happens, NOT TO PROVE SOME THEORY.

We have discussed the 'voltage' evident across that Q1 as illustrated in that scope shot - AT LENGTH - in the LOCKED thread.  picowatt went to extraordinary lengths to explain that the zero reference indicated in the 'boxes' at the base of the screen ONLY related to their position with respect to the central 'horizontal' line.  I on the other hand said that the values in those boxes related to the chosen 'coupling' being AC or DC.  In effect the voltage evident over that shunt is shown as a DC coupled value when it should - in truth - be an AC coupled value.  TK is well aware of the argument.  He actively engaged.  picowatt's reference to the zero crossing was CORRECT.  My reference to the coupling was CORRECT.  But the point was made - by me - that the zero crossing, as indicated, is NOT appropriate to an AC coupled value that was actually REQUIRED.  The coupling at that channel is DC. 

But GL asked you about the GATE SIGNAL, not the shunt. Fortunately he is able to see through your hallucinations and take home the "4 volts" value as representative of the voltage APPLIED TO THE GATE.

It was subsequently explained to me that I COULD, in fact, have coupled that channel to AC.  In which case you most assuredly would be looking at something in the order of 4 Volts - as you, yourself - determined.   

What TK is trying to do here is to REVIVE that argument with the 'inference' that we had NOT dealt with this subject - AT LENGTH.

That is an out and out lie. LOOK AT THE ANNOTATED BLOW UP I POSTED OF THAT SCOPESHOT. I AM THE ONE WHO POINTED OUT THAT THE GATE SIGNAL WAS NOT TURNING THE MOSFET FULLY ON BECAUSE IT WAS ONLY 4 or 5 VOLTS. I CALCULATED THE CORRECT 320 mA CURRENT, TO WHICH AINSLIE AGREED. I EVEN POINTED THIS OUT TO GL WHEN HE USED 12 VOLTS IN HIS FIRST EXPERIMENT.

Then he would have used YOUR finding to suggest that you, a known and skilled member and participant - would PROVE that 12 volts enables that current flow.  Which would then 'spin' the story that our SCOPE VOLTAGE VALUE has been misrepresented. It WAS misrepresented.  It IS only 4 volts or thereby but it is 4 volts AC.  Again.  Our coupling on all our channels was DC.  And I will GLADLY demonstrate this when we get out demonstrations up and running.

You don't even know what you are talking about. The only trace where AC coupling could be valid is the CVR trace.... but the 4 VOLTS IS ON THE GATE TRACE. And that trace should most definitely NEVER be AC coupled. You have no idea about the purpose or the behaviour of DC vs. AC coupling; you are parroting words without understanding them at all.
And of course what you say about me is total crap, responding again to what YOU HALLUCINATE that someone says, rather than what they actually say.

What is sad is that you were engaged at all without first finding out - from me - our own stance on this.  The more so as you then also went to such extraordinary lengths to prove his point. 

You are sad. He is not proving "my point"... he is illustrating the truth, by experiment, something that YOU have never done.

But I'm personally grateful for your efforts.  And as ever, you are essentially correct.  But you are also, correctly representing the zero crossing line.  Our scope shots representation of this is NOT correct.

WHAAT? Is this coming from the same person who said, "There is no way the scope can be used improperly.  A setting is a setting. " 

But our REFERENCE to that channel 2 was to show evidence of NOT the voltage - but the frequency.  And that was NOT misrepresented.  Had the emphasis been on the VOLTAGE then I would indeed, have MISREPRESENTED that value. 

Kindest regards,
Rosie

Can anybody decode that last part?  What difference does the presenter's "emphasis" have on the DISPLAYED DATA? Does the emphasis change the numbers? Well, in Ainslie's mad mind apparently so.

Now... what _I_ believe GL has shown with his experiments is simply this:

In the first screen below, the 320mA current is entirely reasonable for the level of gate drive shown.
And in the second shot below... the level of current shown (ZERO) is not reasonably expected from a functioning mosfet wired properly in the circuit receiving a gate drive level shown. And that's all.

Some explanation therefore must be sought for the mosfet's strange behaviour. There are multiple explanations possible. My current favorite is that the mosfet is simply permanently open, fried from its high-heat experiences, and thus will pass no current regardless of the gate drive level. However I can also think of other reasons as well. Without an explanation from the experimenter, I will continue to "assume" that the mosfet is simply blown.

[MileHigh]

We are all trapped in circles.

I must have commented at least five times that Rosemary's "thermal story" is purely anecdotal and not scientific at all.  I am sure the irony in that statement is self-evident to most readers.  She will persist to tell the story anyways.  It's just like the battle over the scope traces and the power averaging.  It just goes in circles.

The only way out of this trap is to move forward and talk about something new.

I was thinking about how to do the dim bulb testing in as simple a manner as possible.  The only thing that has to be estimated is how long you keep the batteries connected to the NERD circuit before you do the dim bulb test.

So, let's assume that we want to "hypothetically" burn off 2/3 of the energy stored in her batteries before you do the dim bulb test.  Here is a proposed simple procedure.

1.  Rosemary sets up her negative oscillation mode and measures the current flowing through the main loop with a digital multimeter or an ammeter.
2.  Using the measured rate of current flow and the ampere-hour rating of the (presumably) new and smaller batteries, you determine how long the NERD circuit has to run to "hypothetically" burn off 2/3 of the energy stored in the batteries.
3.  Run the NERD circuit for the calculated amount of time.
4.  Do the dim bulb testing.

Note that if she uses a function generator then we can factor in the duty cycle for the negative oscillation mode - without drama.

So if anyone has any comments or suggestions please feel free.

You can imagine the setup running for say, 20 hours, with the DSO recording negative power, a.k.a.; "COP infinity."  However, we also know that the ammeter shows a net current flow from positive to negative, and we used that information to calculate the length of the test run.

Right there you see two seemingly contradictory pieces of data.  You can infer from the net current flow that the batteries are discharging, while the real-time power averaging done by the DSO is indicating that there is a net flow of power back into the batteries.

So which is right, the inference made from the basic current flow measurement, or the high-tech DSO average power measurement?  That is the crux of the issue.

All of this drama could be resolved by the simple 4-step test procedure outlined above.

MileHigh

[Rosemary Ainslie]

Guys, it seems that our TK is now getting as far out of control as FTC.  IF that's possible.  Certainly they've both been working on this for some time now.  And all that 'passion'.  If I didn't know better I'd think they're both in love.  LOL.

I keep revisiting this "calculation" because it is so very hilarious. Ainslie refuses to correct it, instead preferring simply to admit that it is "a tad out". Come, let us dissect it and see what a "tad" is in South African English. It is much larger than a Texan "tad", that is for sure. This is the definition of the calorie, no problem here, since it's just parroted from a text.

I certainly endorse that 'reason' that he's proposed for his continual reference to this.  That he finds it 'funny'.  It's always a good thing to laugh.  And  I think the essential difference between us is that I freely acknowledged a mere computation error.  TK the poor soul needs to concede a profound misunderstanding related to power analysis.  Understandably he DARE not.

Have we? But when the experiment was reported Ainslie said " This was an exciting test.  We took water to boil 0.7 litres. " (Blog post #117). So let's use 700 mL in our calculation.

Actually it started off at nearer 1 liter.  But with that rather dramatic insertion of that element resister there was a significant amount of it splattered and spilled out.  But I think that 0.7 liters is certainly a conservative and fair assessment.

Did we? Normally when reporting time-temperature data one records the time it took to arrive at that temperature, and since Ainslie proceeds to attempt to calculate that way, we are justified in taking 90 minutes as the time it took to raise the 700 mL of water to 104 degrees C using a submerged heating element. HOWEVER... that is not how the experiment was actually performed, is it. " It rose from 66.9 degrees C to it's (sic) final temperature of 240 degrees C.  That's when I put it in water.  And then it took the water up to 80 degrees centigrade where it pretty well stabilised. " (Blog post #117). And this temperature is not actually the water temperature but the temperature of the heating element, isn't it, because the thermocouple is still mounted to the element. No matter... let's continue as if it took 90 minutes to get all the water to 104 degrees as Ainslie implies with the rest of the calculation. But does " a little over 80" mean 82? "The water temperature then stabilsed (sic) after an hour or so - at a little over 80 degrees centigrade." (Blog post #117).

This is somewhat confusing.  YES it's the temperature of the element resistor.  YES it appears to have reduced after its immersion in water.  YES I think it's a fair measure of the temperature of that water as it was evident OVER TIME.  I'm satisfied that this much he's got right.  Which is REMARKABLE.

Did we? But.... any child with a calculator could tell you that 82 plus 20 is not 104. (An error of 2 percent is a Texan "tad".) And that water at sea level atmospheric pressure boils at 100 C and you CANNOT get it hotter than that without pressurizing it. And.... "Then Guys - and in conclusion - in the space of a few short minutes - with an increased frequency - it then took the temperature to boiling point - I think.  It wasn't actually boiling but it had small bubbles. And the temperature recorded at 104 degrees C - or thereby, from memory." is how it was reported at the time the experiment was done. (Blog post #118). And of course the 104 degrees is the temperature of the element not the water. And it has small bubbles but no sound of boiling. Ho. Hum. Yawn.

And guys.  RED ALERT - MORE SPIN.  The rise in 20 degrees took less than 10 minutes - AFTER we changed the the applied switching frequency.  And it certainly DID NOT represent a stable temperature of the water - as it didn't have enough time to stabilise.  We HAD to stop that test.  It was just WAY more energy than I was comfortable with.  ANd 104 degrees centigrade was STILL the measure of the heat on the load resistor.  That's PRECISELY where that probe was attached.

No, "Joules" is NOT equal to one watt per second. Most definitely not, and the rest of the "calculation" that Ainslie commits illustrates just why the terms Watt and Joule are NOT interchangeable and mean different things. A QUANTITY is NOT A RATE.-

And GUYS.  We ALL know this.  A Joule is the quantity of POWER computed over 1 second.  And a WATT is a unit of power that is representative of the energy and voltage applied over a representative sample of voltages.  I think TK is simply trying to remind himself of this.  It's possibly the kind of repetition he needs to get something 'drilled' into that mind of his.  Hopefully he won't AGAIN try and claim that 0.32 mA x vbatt is the wattage when it's calculated over a mere 12.4% of a single duty cycle. LOL.

This is so overweeningly arrogant and ignorant that I am flabbergasted every time she uses the phrase, because nearly every time she does so, she errs in one way or another. MOST ESPECIALLY in this post of hers right here, but many times elsewhere as well.

LOL I LOVE the word 'flabbegasted'.  And I'm neither arrogant nor ignorant.  I think the word he's looking for is 'indifferent'.  Notwithstanding all the CRUEL treatment.    And INDEED I'm prone to the occasional error.  Aren't we all?  I don't think I'VE ever misrepresented myself as an EXPERT.  Let alone a SCIENTIST with MULTIPLE degrees. 

Kindest regards guys.  And I've spared you that RIDICULOUS math reference.  It's so LOADED with spin and misrepresentation that it'll likely confuse God himself.

Rosemary

[TinselKoala]

All you need is an alarm clock, an automatic battery charger, and her oscilloscope.

Charge up all batteries to the same level using an automatic automotive battery charger that shuts off when it senses a full charge has been reached (by whatever method it uses, usually decrease in charge current at some specific voltage applied). Take two batteries at random and set them aside, in a bullet proof safe under lock and key, so TK can't fly quickly from Texas to SA, sneak into her mansion at night, drug her little rat dog, and discharge those batteries.

Then use the remaining 4 batteries to run the Ainslie circuit, tuned by the scope and ACTIVELY BOILING 700 mL of WATER. That's no problem, right? RIGHT??
Set the alarm clock for 24 hours from start (not including the tuning time, of course), and walk away. Hire some kids to keep the water container topped off overnight with instructions to call the fire department if a battery catches fire. When the alarm clock rings, stop the run and perform the Dim Bulb test on all the batteries, using time-lapse webcam and clear labels of which battery is which.

Repeat the test 5 times, taking about 2 weeks total to perform all 5 tests, recharging and randomly selecting the batteries each time.

And 3 out of 5 sets wins the match. That is, if on 3 out of 5 tests the Ainslie run batteries last just as long -- or longer -- as the set-asides, Ainslie's claim is PRELIMINARILY vindicated and further testing should be performed, and a LOT of people will be very excited. OTOH... should Ainslie's batteries run down first, on 3 out of 5 trials.... then somebody has some thinking to do, nobody will be very excited, and we can all go on to other more important work. Well... most of us will be able to, anyway.

Note that this isn't even a stringent test. Since Ainslie has repeatedly claimed that her batteries are all still fully charged, a STRINGENT test of her claims would not permit her to charge the batteries conventionally AT ANY TIME. Even before her "current" set of proposed tests, which are unlikely ever to be performed anyway.

[Rosemary Ainslie]

And here's another doozy.

My dear TK, 

All you need is an alarm clock, an automatic battery charger, and her oscilloscope.

Charge up all batteries to the same level using an automatic automotive battery charger that shuts off when it senses a full charge has been reached (by whatever method it uses, usually decrease in charge current at some specific voltage applied). Take two batteries at random and set them aside, in a bullet proof safe under lock and key, so TK can't fly quickly from Texas to SA, sneak into her mansion at night, drug her little rat dog, and discharge those batteries.

Then use the remaining 4 batteries to run the Ainslie circuit, tuned by the scope and ACTIVELY BOILING 700 mL of WATER. That's no problem, right? RIGHT??
Set the alarm clock for 24 hours from start (not including the tuning time, of course), and walk away. Hire some kids to keep the water container topped off overnight with instructions to call the fire department if a battery catches fire. When the alarm clock rings, stop the run and perform the Dim Bulb test on all the batteries, using time-lapse webcam and clear labels of which battery is which.

Repeat the test 5 times, taking about 2 weeks total to perform all 5 tests, recharging and randomly selecting the batteries each time.

And 3 out of 5 sets wins the match. That is, if on 3 out of 5 tests the Ainslie run batteries last just as long -- or longer -- as the set-asides, Ainslie's claim is PRELIMINARILY vindicated and further testing should be performed, and a LOT of people will be very excited. OTOH... should Ainslie's batteries run down first, on 3 out of 5 trials.... then somebody has some thinking to do, nobody will be very excited, and we can all go on to other more important work. Well... most of us will be able to, anyway.

Note that this isn't even a stringent test. Since Ainslie has repeatedly claimed that her batteries are all still fully charged, a STRINGENT test of her claims would not permit her to charge the batteries conventionally AT ANY TIME. Even before her "current" set of proposed tests, which are unlikely ever to be performed anyway.

This would ENTIRELY satisfy your standards of proof.  We all know this.  It is NOT, however, scientific.

Rosie Pose