10KW LENR demonstrator (new thread)

[tomclarke]

nuclear power: miracle sold: real
quantum physics: miracle sold: real
...not let's go back a bit.

microscope: miracle sold: real
cathode ray tube: miracle sold: real
evolution: miracle sold to all industrialized countries except for the u.s.: real
heliocentrism...
electricity...
flight...

the list goes on.

As pointed out above none of these things are miracles.

More importantly no-one sold them as miracles. They made money only after they were accepted by science as real phenomena.

true.

until a source for the anomolies can be thoroughly explained scientifically, it is still a legitimate and worthy and respectable field of scientific investigation.

Of course. It is what science does. But remember 99% of anomalies come down to misunderstanding of subtle experimental issues, and are profoundly unexciting when finally tracked down. So don't expect rational people to spend money investigating weird experimental results unless they are coherent or beyond likely experimental error.

also true. i was reminded of "polywater" when writing the above. it turned out to be just a contaminant. but until it was discovered and demonstrated that that was the source of the anomaly... well, that would never have been discovered and demonstrated if people had never investigated the phenomenon empirically. so you see, the empirical investigation was good science, and led to practical results. perhaps not the most exciting results, but results that provided solid answers, nonetheless.

on that point, i _don't_ expect rational people to spend money investigating weird experimental results unless they are coherent or beyond likely experimental error. many LENR experimental results are coherent and beyond likely experimental error. and, true to your statement, rational people are spending money investigating them.

I've got nothing aginst this. I do dislike the typical LENR methodology which is NOT investigatory.

To investigate you take an anomaly and try to break it by adding controls etc, until you either have cast iron anomaly, or you have discovered cause.

LENR take an anomaly and try to enhance it by altering experimental conditions.

[happyjack27]

I've got nothing aginst this. I do dislike the typical LENR methodology which is NOT investigatory.

To investigate you take an anomaly and try to break it by adding controls etc, until you either have cast iron anomaly, or you have discovered cause.

LENR take an anomaly and try to enhance it by altering experimental conditions.

From my experience that's what confuses a lot of people. You're right in both what they are doing and what they need to do. So then why aren't they doing what they need to do? Well because before they can break down the anamoly, they have to first be able to create it, consistently, and honestly. You can't study stars without the stars, and flames aren't acceptable substitutes. So that's why they're working on "enhancing" it: so that they can then break it down. They've got it pretty good now. They need more people to reproduce to confirm the context, then they can start breaking down. Science is a slow process. You can't skip steps.

[happyjack27]

You can't claim your answers to be solid if you can't show that they are truly answers to the anomaly rather than some superficially similiar context. Well, you can claim, but no rational person is going to accept that as logical proof.

[tomclarke]

I've got nothing aginst this. I do dislike the typical LENR methodology which is NOT investigatory.

To investigate you take an anomaly and try to break it by adding controls etc, until you either have cast iron anomaly, or you have discovered cause.

LENR take an anomaly and try to enhance it by altering experimental conditions.

From my experience that's what confuses a lot of people. You're right in both what they are doing and what they need to do. So then why aren't they doing what they need to do? Well because before they can break down the anamoly, they have to first be able to create it, consistently, and honestly. You can't study stars without the stars, and flames aren't acceptable substitutes. So that's why they're working on "enhancing" it: so that they can then break it down. They've got it pretty good now. They need more people to reproduce to confirm the context, then they can start breaking down. Science is a slow process. You can't skip steps.

Time will tell. They claim to have had reliable large results for many years now. But these all from methodology which is suspet - so that when tested likely to vanish. I ask when they will take these reliable large results and validate? When this happens I will watch with great interest. Till then it is rather like somone claiming to win at a game of dice but refusing to check whether he is using loaded dice till his winnings are big enough.

[ladajo]

Looking at the photo that Axil posted regarding the latest information "leak" from Rossiworld, did anyone else wonder about the claimed power input (147VAC @ 24.5Amps), and the photographed electrical connections? The left hand connection looks like standard lamp cord, going into a screw-down connector, which then switches to twisted pair instrument wire. The right side looks like all high-temp instrument wire.

Personally, I would would a little nervous to push 25 Amps continuous through lamp cord as it is normally 18 or maybe 16AWG. For his application he should be using at least 12, but more likely 10AWG or larger. It also looks like he is using T or E type, which is probably not the smartest in terms of high temp application. Not to mention his connector block is suspect at high loads for a cook off. Once more, Rossi "posts" ("allowed the leak") evidence he really doesn't know what he is doing.

Let us not yet get into the "pressure not an issue" as we heat an sealed steel enclosed volume to 1200C and greater statement.

[happyjack27]

Time will tell. They claim to have had reliable large results for many years now. But these all from methodology which is suspet - so that when tested likely to vanish. I ask when they will take these reliable large results and validate? When this happens I will watch with great interest. Till then it is rather like somone claiming to win at a game of dice but refusing to check whether he is using loaded dice till his winnings are big enough.

Please substantiate your accusation that the " methodology which is suspet - so that when tested likely to vanish.", esp. in light of your apparently contradictory statement, "They claim to have had reliable large results for many years now." that "claim" that "they" make is pretty easy to substantiate, and pretty easy to prove false. what evidence do you have that it's false?

also, you ask "when they will take these reliable large results and validate?" they have already done so. "Till then it is rather like somone claiming to win at a game of dice but refusing to check whether he is using loaded dice till his winnings are big enough" eh, sort of. i'd say it's more like building giant particle accellerators to try to find particles and refusing to claim that you found anything until your results reach a certain level of statistical significance, or conversely have been proven false at a certain level of significance. For instance, if the same anamolous results can be reliably created using light water as heavy water. however, in all experiments where light water was used as a control, light water never showed anomalous results. It's clear that the dice are somehow loaded, in comparison to light water dice. the question is, what is loading them? To answer them, one must roll them a lot. so a good first step is to improve the rate and reliability at which you can roll them, and then have independent groups test your methodology in order to confirm at. there has been a little bit of that, and the confirmations do hold up. but it's not enough to satisfy LENR researchers. check out SPAWAR's co-deposition experiments, for instance. these are experiments that are relatively easy to replicate and have very reliable results.

[tomclarke]

Time will tell. They claim to have had reliable large results for many years now. But these all from methodology which is suspet - so that when tested likely to vanish. I ask when they will take these reliable large results and validate? When this happens I will watch with great interest. Till then it is rather like somone claiming to win at a game of dice but refusing to check whether he is using loaded dice till his winnings are big enough.

Please substantiate your accusation that the " methodology which is suspet - so that when tested likely to vanish.", esp. in light of your apparently contradictory statement, "They claim to have had reliable large results for many years now." that "claim" that "they" make is pretty easy to substantiate, and pretty easy to prove false. what evidence do you have that it's false?

I don't have any evidence that it is false. However the claim is extraordinary and therefore needs to be supported by extraordinary evidence. The evidence quoted is all tendentious. It would be a matter of going through each bit and showing why - which would take a long time since there are 100s of papers claiming experimental evidence. To take a typical example, most of the claims of excess heat (beyond chemical) fall into the following categories:
(1) grossly suspect experimental method (e.g. Rossi)
(2) calorimetry with inadequate controls: Hagelstein, the Italian guy whose name I forget.

These deficiencies could in all cases be remedied quite easily, but have not to my knowledge been. Till this happens the evidence is unclear, which combined with the extraordinary nature of the claims makes then very unlikely.

If you like you post details of one of these experiments and I will say why it is deficient.

[Skipjack]

In all of the cases the observed excess output heat has been a small fraction of the input heat which could mean experimental error or simply a measurement issue. In the cases of higher claimes excess heat such as Rossi and Defcalion, the experimental setup has to be doubted. Also these have never been independently verified.

[Carl White]

In all of the cases the observed excess output heat has been a small fraction of the input heat which could mean experimental error or simply a measurement issue. In the cases of higher claimes excess heat such as Rossi and Defcalion, the experimental setup has to be doubted. Also these have never been independently verified.

What about the NANOR created by Dr. Mitchell Swartz and being demonstrated since January by Dr. Peter Hagelstein of MIT? Which displays 14 times the input energy (according to Hagelstein)?

Why not go independently verify it?

[Skipjack]

What about the NANOR created by Dr. Mitchell Swartz and being demonstrated since January by Dr. Peter Hagelstein of MIT? Which displays 14 times the input energy (according to Hagelstein)?

IIRC, it was very low output and the methotology was not very solid.

[tomclarke]

In all of the cases the observed excess output heat has been a small fraction of the input heat which could mean experimental error or simply a measurement issue. In the cases of higher claimes excess heat such as Rossi and Defcalion, the experimental setup has to be doubted. Also these have never been independently verified.

What about the NANOR created by Dr. Mitchell Swartz and being demonstrated since January by Dr. Peter Hagelstein of MIT? Which displays 14 times the input energy (according to Hagelstein)?

Why not go independently verify it?

Indeed. It has been doing this for some time. Although the power generated is very small it could be validated as follows:

(1) take current experiment.

(2) enclose entirely in double-walled metal box with known thermal resistance insulation (maybe polystyrene) separating inner from outer, thermocouples on outside of inner wall and inside of outer wall. Easy to calculate total thermal resistance, though we calibrate as well.

(3) Use fine wire for thermocouple outputs and heater inputs, lead it at angle through insulation if necessary so as to make heat conduction through wire insignificant. easy to calculate this.

(4) Use two heating elements in experiment, one active with magic powder, one control. Calibrate using the control. Check both response time to heat flow change, and relation between total power and temperature difference. Do the calibration throughout the anticipated range of power levels. As sanity check compare measurements with expectation from calculated thermal resistance.

(5) Measure electrical power in to heating elements carefully (V & I - don't assume anything about element resistance staying constant). Use same equipment to measure both active and control heater.

(6) Now use the active element, do the experiment, measuring power in and out.

(7) Calibrate again immediately after measurements, to be safe, though nothing should have changed in box insulation.

This is not much different from what he has now - the extra stuff would cost $100 or so.

Note that errors in power in & temp out measurement are both controlled, and that the double-walled design with metal walls means that heat out from either element must have the same effect on the box, and the thermal gap used to measure the output power is clearly independent of which heater is used.

Note that careful electrical measurement of input power makes this parameter bomb-proof. If oscillating power is required at low frequency that is still OK, but you need dual channel sampled V & I going to a PC which does numerical integration. If dV/dt is high you need to be careful to avoid spurious measurement. In fact if at all possible avoid high dV/dt since this might corrupt thermocouple mesurements. Make sure the control input power waveform is the same as the active one, which controls for measurements errors on input or output due to electrical interference.

Students in Hagelstein's class could do this extra instrumentation and take measurements as a Term project. It requires some care, but is neither particularly difficult nor expensive.

If this were done and written up carefully as a paper it would be very convincing. It could then be replicated by others (using internal bulb with heaters provided by Hagelstein and sealed if needed to prevent IP leak).

Best wishes, Tom

[tomclarke]

Polystyrene foam 0.033 W/mK

say box is 20X20X20cm, with 2cm between inner and outer wall

thermal conductivity is: 0.033 *0.2*0.2/0.02 = 66mW/K

We can get a thermocouple accuracy of 0.1K => can measure with 6mW resolution from type J thermocouples.

If lower powers need to be measured just increase the thickness of the insulation to 10cm for a X5 increase in resolution.

Check that thermal bridging due to wires is not significant:
suppose wires are 0.1mm dia. ratio of cross-section areas is approx (200/0.1)^2 = 4E6. Thermal conductivity ratio between copper and polystyrene is:

400/ 0.033 = 1.2E4 << 4E6.

Check that power dissipated in heater wires is not significant:
rho = 1.7E-8 ohm*m

R = 1.7E-8 *0.02 / ( 0.0001 * 0.0001 ) = 0.034 ohm.

P = R *I^2 so for these wires we can manage I=300mA with only 3mW extra dissipation between inner and outer boxes.

We still have a factor of 300 on the thermal conductivity ratio so we can use thicker wires, say 0.3mm dia and I=3A heater current OK.

That all looks very comfortably within spec required for Hagelstein's experiment where input power is in range 0.5 - 5 mW and output power in range 1mW - 100mW. You would need the thicker box for output resolution down to 1mW.

One more check, at these low powers what is the expected response time? Assume total inner weight is 500g. (We have the bulb etc from glass as well, but this will do ball park)

specific heat steel is 0.5 kJ/kgK => 250J/K. So time constant is:

250 J/K / 10mW/K = 2500s. So you have power averaged with 1 hour time constant. That is Ok for a 10 hour experiment, as Hagelstein runs. After all, it is total input energy versus total output energy that matters. It will make calibration more tedious, with a calibration run lasting maybe 24 hours. The long time constant from the box and bulb can be accurately determined and compensated from the calibration data, so with a bit of care calibration can be accurate with only say 30 minutes between power steps. We then get calibration with 20 power steps in 10 hours real time.

Looks like I've done most of the work H's students should be doing to design this experiment.

[cg66]

tomclarke - i would be curious to hear your thoughts on measurement approach in the Godes report from ICCF-17.
Link can be found here:
http://newenergytimes.com/v2/conference ... F-17.shtml

[tomclarke]

tomclarke - i would be curious to hear your thoughts on measurement approach in the Godes report from ICCF-17.
Link can be found here:
http://newenergytimes.com/v2/conference ... F-17.shtml

It is interesting. This experiment achieves positive results not through bad calorimetry - the calorimetry looks pretty good - but through measurement errors due to the HF pulses. They have noticed that they get positive results by pumping the system with high power 100ns pulses! The high frequency components from this are very substantial, and it is almost imposible to stop interference from this to other electrical measurements. This could affect either the input power measurement or the RTD output measurements - due to rectification of high frequency components inside voltage amplifiers.

Anyone who has played with high power RF pulses knows how difficult it is to screen coupling to other electrical circuits, and whereas the exprimenters seem to have been very careful with calorimetry issues, they appear not to have not considered electrical interference! Since this is undoubtedly a real issue, if they had considered it they would detail screening, control measurements, etc used to make sure it was OK, just as they do with the calorimatric issues.

You do not get this class of problem in experiments that have DC power input. Want to bet that these will instead have calorimetric problems?

Best wishes, Tom

[tomclarke]

Considering the experiment above, which is from Broullion work, what is the bet that ALL their apparent positive results are due to measurement errors caused by the use of high power HF pulses? Such errors are very difficult to track down and in the writeup they are not checking whether they exist or not with the same care as they apply to controlling calorimetry errors.

Note that to get apparent positive results from experiments you don't have to be fraudulent, or even stupid. You just have to ignore one possible source of experimental error.

Since the phenomena they expect is supposed to be difficult to control almost any results can be explained as due to variability in the phenomena.

This is one example of how the "unfalsifiability" of LENR theories makes validation difficult. Any experimental error giving positive results (about 1/2 of all such errors) can be interpreted as LENR heat.

Want more evidence in this case? The observed output depends on the pulses and peaks at a specific pulse rate. That is exactly what you would expect from errors caused by HF pickup via parasitics.