Testing the TK Tar Baby

[picowatt]

I like your idea of a control system that is pulsed at the same duty cycle as the NERD apparatus.

But I take it then that the pure oscillation mode, that we have been discussing here for so long, is to be abandoned in favor of a continued BIPOLAR gate drive situation as described in Figure 7 and the other high heat trials. This will of course require the use of a function generator or some other arrangement for pulsation and duty cycle control and negative bias current and positive gate voltages on the part of the NERDs. It is unclear how they propose to deal with this issue.

I'm not sure why you would want to do your calorimetry in air. Is it because the original NERD report just had the element hanging there in the breeze, like some unclothed body part, and their data reported "temperature _over_ the resistor" ? This is unreliable, as you know. Your proposed scheme of airflow and so on will be an interesting introduction to the pitfalls of calorimetry. I'd suggest doing it in oil from the beginning, using a fairly heat "leaky" but still stable container for your temperature-Wattage data. It might take a bit longer but it will be easier to get reliable data from a reasonably easy to construct system.

You might like to take a look at the MOAC description here:
http://www.earthtech.org/experiments/ICCF14_MOAC.pdf

But seriously... why bother? Surely you've noticed.... in those periods when you can pry yourself up out of your armchair, using your never-yet-received mosfets, as Ainslie would have it.... that your batteries are discharging when your build of the circuit is running. Therefore, you have not replicated the Ainslie circuit at all, so all your other efforts are moot.

TK,

I agree that if the goal was to quantify the actual amount of heat produced that alternate methods would be more suitable (such as heating a known quantity of fluid to given temp).

However, as the goal, I thought, was to determine if the NERD circuit can generate more heat for a given amount of battery capacity than other methods of applying current to the load, and as these tests could run for some time, I considered the air cooled chimney.  Keep in mind that I would use a fairly large thermal mass of aluminum which effectively replaces the oils' thermal mass.  Basically I am just increasing or maintaing a faster "leak" rate.  I have a location where the apparatus can be placed where the temp and humidity are very constant, changing only slightly and over month long periods.  Also, as I want to automate the setup so it will require no intervention on my part, I would like to maintain fairly low temps.

Even if I use a slightly "leakey" cylinder filled with oil, I would likely wrap it in a chimney to ensure consistent convective air flow about its entire circumference.

As for pulsing and the comparison test, I assumed that the NERD circuit as maintained in the papers is the test that is to be performed.  I would definitely consider a cleanly pulsed low RDSon MOSFET at an appropriate duty cycle using the same Rload to do rundown comparisons if that circuit is used.

As for a comparison load if only the osc portion of the circuit is tested, it becomes a bit more difficult to envision what would be the best comparison load for that.  Possibly the switched load would be suitable as well, but at a higher frequency and increased duty cycle.  In other words, attempt to duplicate the switching period and duty cycle of the osc but with cleanly and well dampened switching that does not present a neg mean.  Still thinking about that one.

A lot will depend on what the claimed increase in battery capacity is. 

Also, if you have noticed, the claims have become a bit fuzzy.  COP=infinity, not claiming to exceed the battery's capacity, no observed drop in Vbatt throughout all tests, all in the same post.  So, honestly, I do not really know what she means by COP=infinity, or, frankly, what is even claimed.

PW 

[picowatt]

TK,

Took a bit longer for the graph to appear.  This sounds reasonable, but would the argument not be that the method used to measure the circuit's power draw "accurately" is somehow flawed?

What method did you use to determine the circuit's power draw to generate that graph?

PW

[TinselKoala]

Yes, you are right. At present there doesn't appear to be an actual testable hypothesis that is clearly formulated. The actual claims are disappearing, like the 5.9 megaJoule claim.

I can anticipate a claim that would go something like this: Well, TK, your comparison graph like those above don't go long enough. The NERD circuit may heat at lower efficiency in realtime, but it heats much _longer_. In other words, say the control circuit would heat to a stable 50 degrees, maintain that for 100 minutes, then die out and cool off, while the NERD circuit might only heat to 35 degrees using the same power, but could do it for 200 minutes or more.But I think this kind of performance would require some different kind of behavior than is presently claimed, I think, because you would have to postulate some mechanism that _prevented_ the full power drawn from the battery from reaching the load initially but also somehow prolonged that same state of affairs longer than normal.

You asked earlier how hot my oil load has gotten. I have seen it get to nearly 200 degrees C. But under oscillations alone it rarely rises to over 50. Here's a photo of it running fairly warm, using a bipolar triangle pulse drive. Recall that there are 250 mL of mineral oil in there, density 0.83 gm/mL and specific heat 1.67:

[TinselKoala]

TK,

Took a bit longer for the graph to appear.  This sounds reasonable, but would the argument not be that the method used to measure the circuit's power draw "accurately" is somehow flawed?

What method did you use to determine the circuit's power draw to generate that graph?

PW

I did it by using a digital oscilloscope to generate an instantaneous power trace and average it across multiple waveform cycles. The power input to the circuit was directly related to the "on" time of the duty cycle, as you can imagine. I used a fast risetime pulse generator to pulse the circuit (this is the "Quantum magazine" circuit, presented as "paper 3" on the new forum alongside the present two papers) so the pulses are far more stable than those provided by the original 555 timer (which is another story altogether. I invite you to build the timer circuit as shown in the schematic with the article and see what happens.) I used a regulated power supply, no pulsations, for the DC trials, and measured its power with voltage and current meters, and I used the exact same load for all of those runs. It was a long time ago and I don't have my notes here, but it's all documented in old threads that are still archived somewhere on this forum and on energeticforum as well.

ETA: this circuit also radiated power in RF and heated the single mosfet quite effectively; that's why the load heating efficiency was less than the DC level. This circuit's claims of OU were linked to the inductive spike caused by the collapse of the field in the inductive load and other circuit elements; this spike contained real power and could be siphoned off, as I showed, to charge capacitors to high voltages and even charge batteries... but not the running battery. Once again it was a matter of being able relatively easily to reproduce the basic data, once the major errors had been identified, like the duty cycle issue, but not being able to reproduce the claimed major effect of the perpetual battery charge. I think FuzzyTomCat went a lot further with this than I did and got results that were more in line with Ainslie's claims, but still not supporting her completely. That is another interesting story best told by him.

[picowatt]

TK,

I am assuming she means to claim that a given battery lasts longer when producing heat at the load using her circuit than it does when the load is powered "conventionally".  But again, at this time, a claim of exceeding a battery's capacity is not being made, but the upcoming tests, we are assured, will demonstrate that the capacity is exceeded.

In light of this, I remain unsure as to what COP=infinity means.  I'd ask, but I know I would not get an answer.  If the claim were merely that the battery does not discharge, all manner of fairly simple tests can be utilized to prove/disprove this.  Proving to exceed the battery's capacity will be a bit more difficult to prove.  But again, it will depend on if we are talking about a 20% increase or orders of magnitude.  Even in one of her posts, she states it could take a couple days for the battery to discharge (or something similar) and I assume she meant while testing the NERD circuit (although she could have been referring to comparison tests, I do not know).

PW