Motto
Only smart questions can generate good answers
Infinite Energy Magazine created by Gene Mallove is not a strictly poor reviewed scientific
journal. There are not many experts in the unknown and those who think the known has to be applied for judging and everything make many destructive errors therefore peer review has a limited validity for the new ideas in LENR. Some of the best, most important, impactful, high level Cf/LENR paper were published in Infinite Energy and a new bright example is in the recently launched issue 113- it is
" Questions About Lattice Enabled Nuclear Reactions: Experiments, Theories and Computations" by David J. Nagel
If you want to build a ship, don't drum up people to collect wood and don't assign them tasks and work, but rather teach them to long for the endless immensity of the sea. (Antoine de Saint-Exupery)
I want now to teach you to long for a high quality LENR paper.
Only smart questions can generate good answers
Infinite Energy Magazine created by Gene Mallove is not a strictly poor reviewed scientific
journal. There are not many experts in the unknown and those who think the known has to be applied for judging and everything make many destructive errors therefore peer review has a limited validity for the new ideas in LENR. Some of the best, most important, impactful, high level Cf/LENR paper were published in Infinite Energy and a new bright example is in the recently launched issue 113- it is
" Questions About Lattice Enabled Nuclear Reactions: Experiments, Theories and Computations" by David J. Nagel
It is the second of a troika of papers; the first is here:
http://www.infinite-energy.com/images/pdfs/Nagel118.pdf
In No 2 the following questions are asked:
Q13. What are the roles of various stimuli applied to LENR
experiments?
Q14. Do resonances of any kind play a role in occurrence
of LENR?
Q15. What are the reasons for the lack of reproducibility
in many LENR experiments? (
SEE THE ANSWER OF DAVE NAGEL BELOW!
Q16. What are the control parameters for production and
variation of excess power?
Q17. What are the prospects for long-term reliable energy
generators based on LENR?
Q18. What parametric variation and other experiments
might be done to advance LENR empirically?
Q19. What experimental tools should be employed for
LENR experiments?
Q20. What might be done theoretically to speed the
understanding of LENR?
Q21. What are theoretical LENR energies and rates?
Q22. Which computational tools should be employed for
LENR theories?
Q23. What is the value of computing material characteristics
and properties?
Q24.What is the payoff from data analysis and data mining?
Now about question 15- some LENR-ists, immune to optimism and with serious flaws in ceative thinking (I am one of these) think that reproducibility is the Mother of All Questions in LENR
and this has absolute priority for solving, therefore I will show here the complete answeer of Prof Nagel:
"Reproducibility is commonly and sensibly taken to mean that, if a particular experiment is redone properly, the results will be repeatable, that is, close to what was originally measured.
Reproducing an experiment means that the equipment (in and around an experiment), protocols (procedures and practices) and materials (fuels and others) are either the same, or sufficiently similar, in the key features that determine the outcome of the experiment. Hence, achievement of better reproducibility requires careful attention to everything that goes into, and is done before or during an experiment.
Reproducing an experiment is easier if it is done in the same laboratory as the initial experiment. However, the reproducibility that is most meaningful is that between laboratories
with different scientists and organizations. Related questions are what most controls the reproducibility of LENR experiments, and what is needed experimentally to achieve full reproducibility?
There are two major reasons why experiments are difficult to reproduce, in general, and especially in a new field where the key variables are not known adequately. The first is the
difficulty in matching the setup and input factors. The second is the natural tendency of scientists to vary these factors, either because they think they have a better idea, or
because they do not have similar equipment and materials.
Some of the factors already discussed above must be responsible for the early and remaining problems with lack of reproducibility in LENR experiments. Variations in seemingly identical experiments, as well as changes during an experimental run, have proven to be major problems in the field.
They must be due to lack of knowledge of the conditions necessary for the production of LENR and ignorance of the details of material requirements, which are almost certainly
key to LENR.
Reproducibility is so important, and so challenging for LENR experiments, that it has been continually discussed in the literature and commentaries on the subject. Reports on specific experiments have cited the percentage success rate, that is, the fraction of a group of experiments that give excess heat or some other evidence of LENR. They are widely scattered
in the large literature on LENR experiments. The topic of the reproducibility of LENR experiments deserves a new study and report. Here, we will cite only one paper from a decade ago.
Letts and Cravens wrote: “The thermal response of the cathode is typically 500 mW with maximum output observed of approximately 1 W. The effect is repeatable when protocols are followed and has been demonstrated in several laboratories.” This is one of several published assertions about the good reproducibility of some experiments.
Variations in the chemical makeup of the experimental equipment can introduce inconsistencies. However, good control over experimental equipment is usual, and it is
unlikely that variations in apparatus are responsible for the irreproducibility of many LENR experiments. If the same equipment is used by the same scientist in the same laboratory,
there remain two candidates to explain the lack of reproducibility, protocols and materials.
The protocols in LENR experiments include what voltages are applied at what times and which measurements are made when. It is possible to automate entire protocols, including
additions or removals of electrolytes during the course of an experiment. A computer program, with no variation from run to run, could drive the entire experiment. Complete computer control is not commonly done, which leaves open the possibility that variable human participation in the protocols is responsible for variations in experimental outcomes. So, it is
Work at ENEA showed quantitatively that the production of excess power depends sensitively on fine details of cathode structures. Violante and his colleagues discovered that Pd cathodes, which they manufactured, were more likely to give excess heat if they had surface roughness in the range of 106 to 107 m-1. They also found that having dominant (100) surface crystal orientations favored the production of excess heat. These findings encourage a series of experiments in which metallurgical techniques are used to insure the correct crystal orientation, and subsequent ion bombardment or other surface treatment methods are employed to produce roughness with favorable spatial frequencies.
Even if materials with well-known composition and structure, and also uniformity, were available, reproducibility of LENR experiments is not guaranteed. Fortunately, there are
systematic approaches to the problem of reproducibility.
One of them is to obtain good statistics on experimental behavior by running large numbers of nearly identical experiments and using as many time-dependent diagnostics for each as is reasonably possible (affordable). Having many measurements for numerous runs would permit various types of statistical analyses. If any of the experiments gave excess power for at least some of the time during their operation, the analyses could point to what is different and useful
about them. Remember the many materials tried and numerous tests made by Edison before a reliable filament for light bulbs was discovered.
Multiple parametric experiments can be run serially or in parallel. So-called matrix experiments, in which many experiments are run simultaneously, with one or two parameters
being varied between the different experimental setups, can be very valuable. However, they are impractical (too costly) for complex experiments that require expensive equipment. Parametric variation experiments with such equipment must be done sequentially, preferentially in the same apparatus, if it is stable over time and use. Sequential experiments can also be conducted in different apparatus. At this time, operation of many small and relatively inexpensive setups with simple but adequate diagnostics should be very useful for empirically improving reproducibility of LENR experiments, even before full understanding is
achieved. The prospect of running multiple LENR tests simultaneously in the same or very similar conditions is attractive.
Understanding, that is, having a theory that is clear and well tested, should also solve the reproducibility problem. But, that approach contains a Catch-22 in that the experiments
needed for adequate testing of a theory might be problematic, if all of the relevant parameters and their values are not specified by the theory being tested. The situation can be imagined as similar to a system of roads. If a person starts at a point that has no connection to the desired
goal, there is no hope of getting to it. But, even if there is a connection between the starting point and the desired destination, there are many forks that will lead off to places other
than the goal. Some of the “topography” is known for LENR, such as high loading for electrochemical experiments. But, key guidelines for achievement of reproducible excess power
in various experiments are still missing.
Before leaving the topic of reproducibility of LENR experiments, it should be noted that there has been little attention to quantification of whether or not excess heat or some other evidence for LENR has been observed. The question, yes or no, is part of the important subject of “binary classification.”
That topic is critical to medical testing for questions about whether or not a person has cancer or some other malady. If heat, radiation or other effects occur at low levels, there is naturally
a question of whether or not an experiment produced LENR. That decision is commonly the opinion of the experimenter.
Usually, there are not enough runs to achieve statistics adequate to apply the common 3σ rule. Then, an effect is considered present if its magnitude is more than three standard deviations (σ) of the noise in the measurement system above the average noise level. No LENR experiments have come close to being run enough to apply the detection methods that go by the name Receiver Operating Curves. The classification challenge is toughest for heat data. Not only are
there the normal statistical considerations, but sometimes small levels of LENR power production are not distinguishable from power due to chemical reactions. In general, there
should be more attention to measures of reproducibility, beyond subjective determinations of success or not, and to distributions of those measures.
In summary, not all of the key factors for the production of LENR are known, let alone adequately controlled. Hence, it is difficult to replicate experiments, even within one laboratory
and, especially, between laboratories, even if they seem to be similar in their construction and operation.
Despite this fundamental challenge, the problem of imperfect reproducibility of LENR can be systematically confronted.
This requires two things, a serious and adequately funded attempt to have similar equipment and procedures, and the use of very well characterized materials. Both the composition and structure of materials have to be known in detail before and after experiments.
Lack of assured reproducibility is both a scientific and practical problem for LENR. And, it still figures in the unwillingness of the broader scientific community to accept LENR as a legitimate field of science. It can be noted that, even after more than a half century of dramatic industrial success in a multi-billion dollar industry, the production of integrated circuits does not produce yields of 100%. Achievement of even good reproducibility will speed acceptance of the reality of LENR, even in the absence of full understanding. So,reproducibility remains a critically important goal. But, it alone does not guarantee the development and success of
commercial LENR power generators. Control of such systemsis also mandatory.
Please compare this objective, systematic, consistent presentation of the facts and rational search for a solution with my incessant whining due to the extreme wickedness of the R-problem in LENR and rhetoric about "actionable" and "non-actionable parameters".
Eventually both say the same thing: we will need a cut-the -Gordian knot type solution for this problem.
2) News, info for today.
Playing LENR Catch-up ...already
National Instruments – Serious about Cold Fusion:
In No 2 the following questions are asked:
Q13. What are the roles of various stimuli applied to LENR
experiments?
Q14. Do resonances of any kind play a role in occurrence
of LENR?
Q15. What are the reasons for the lack of reproducibility
in many LENR experiments? (
SEE THE ANSWER OF DAVE NAGEL BELOW!
Q16. What are the control parameters for production and
variation of excess power?
Q17. What are the prospects for long-term reliable energy
generators based on LENR?
Q18. What parametric variation and other experiments
might be done to advance LENR empirically?
Q19. What experimental tools should be employed for
LENR experiments?
Q20. What might be done theoretically to speed the
understanding of LENR?
Q21. What are theoretical LENR energies and rates?
Q22. Which computational tools should be employed for
LENR theories?
Q23. What is the value of computing material characteristics
and properties?
Q24.What is the payoff from data analysis and data mining?
Now about question 15- some LENR-ists, immune to optimism and with serious flaws in ceative thinking (I am one of these) think that reproducibility is the Mother of All Questions in LENR
and this has absolute priority for solving, therefore I will show here the complete answeer of Prof Nagel:
"Reproducibility is commonly and sensibly taken to mean that, if a particular experiment is redone properly, the results will be repeatable, that is, close to what was originally measured.
Reproducing an experiment means that the equipment (in and around an experiment), protocols (procedures and practices) and materials (fuels and others) are either the same, or sufficiently similar, in the key features that determine the outcome of the experiment. Hence, achievement of better reproducibility requires careful attention to everything that goes into, and is done before or during an experiment.
Reproducing an experiment is easier if it is done in the same laboratory as the initial experiment. However, the reproducibility that is most meaningful is that between laboratories
with different scientists and organizations. Related questions are what most controls the reproducibility of LENR experiments, and what is needed experimentally to achieve full reproducibility?
There are two major reasons why experiments are difficult to reproduce, in general, and especially in a new field where the key variables are not known adequately. The first is the
difficulty in matching the setup and input factors. The second is the natural tendency of scientists to vary these factors, either because they think they have a better idea, or
because they do not have similar equipment and materials.
Some of the factors already discussed above must be responsible for the early and remaining problems with lack of reproducibility in LENR experiments. Variations in seemingly identical experiments, as well as changes during an experimental run, have proven to be major problems in the field.
They must be due to lack of knowledge of the conditions necessary for the production of LENR and ignorance of the details of material requirements, which are almost certainly
key to LENR.
Reproducibility is so important, and so challenging for LENR experiments, that it has been continually discussed in the literature and commentaries on the subject. Reports on specific experiments have cited the percentage success rate, that is, the fraction of a group of experiments that give excess heat or some other evidence of LENR. They are widely scattered
in the large literature on LENR experiments. The topic of the reproducibility of LENR experiments deserves a new study and report. Here, we will cite only one paper from a decade ago.
Letts and Cravens wrote: “The thermal response of the cathode is typically 500 mW with maximum output observed of approximately 1 W. The effect is repeatable when protocols are followed and has been demonstrated in several laboratories.” This is one of several published assertions about the good reproducibility of some experiments.
Variations in the chemical makeup of the experimental equipment can introduce inconsistencies. However, good control over experimental equipment is usual, and it is
unlikely that variations in apparatus are responsible for the irreproducibility of many LENR experiments. If the same equipment is used by the same scientist in the same laboratory,
there remain two candidates to explain the lack of reproducibility, protocols and materials.
The protocols in LENR experiments include what voltages are applied at what times and which measurements are made when. It is possible to automate entire protocols, including
additions or removals of electrolytes during the course of an experiment. A computer program, with no variation from run to run, could drive the entire experiment. Complete computer control is not commonly done, which leaves open the possibility that variable human participation in the protocols is responsible for variations in experimental outcomes. So, it is
desirable to remove this possibility by a series of experiments without human intervention, once the experiment is set up and the computer programmed. However, it is not expected that differences in hands-on participation by scientists can account for all the experimental irreproducibility.
If it does turn out that protocols are not the main suspect for irreproducibility, materials are left as the prime candidate.
The experience at SRI International is directly relevant. Pd from a wide variety of sources, treated in diverse ways, was subjected to tests at various temperatures and many different additions to electrolytes for loading of either protons or deuterons. Success in loading of hydrogen isotopes and the production of excess heat varied widely. A more recent article, also from SRI International, deals with the topic of replication of LENR experiments.
It is possible that low levels of impurities within the solid materials in an experiment are basic to the outcome of LENR experiments. This could be the case because the impurities create conditions needed for LENR, or because they catalyze the heat-producing reactions, or even because they participate in the reactions. It is very expensive to measure accurately the quantities of low-level impurities in the materials that go into and come out of an experiment. So, few experiments in the field have done a defendable job of assaying impurities that might influence or even determine the outcome of experiments.
The Naval Research Laboratory was unable to produce excess heat with pure Pd cathodes in over 200 experiments. However, they had some indications of heat production with old Pd materials that contained Pt and Rh as impurities. That observation led to experiments with Pd alloys, which contained 10% Rh. Those cathodes gave significant excess heat in 6% of 61 experimental runs. The composition of cathodes is clearly important to the production of power by LENR.Work at ENEA showed quantitatively that the production of excess power depends sensitively on fine details of cathode structures. Violante and his colleagues discovered that Pd cathodes, which they manufactured, were more likely to give excess heat if they had surface roughness in the range of 106 to 107 m-1. They also found that having dominant (100) surface crystal orientations favored the production of excess heat. These findings encourage a series of experiments in which metallurgical techniques are used to insure the correct crystal orientation, and subsequent ion bombardment or other surface treatment methods are employed to produce roughness with favorable spatial frequencies.
Even if materials with well-known composition and structure, and also uniformity, were available, reproducibility of LENR experiments is not guaranteed. Fortunately, there are
systematic approaches to the problem of reproducibility.
One of them is to obtain good statistics on experimental behavior by running large numbers of nearly identical experiments and using as many time-dependent diagnostics for each as is reasonably possible (affordable). Having many measurements for numerous runs would permit various types of statistical analyses. If any of the experiments gave excess power for at least some of the time during their operation, the analyses could point to what is different and useful
about them. Remember the many materials tried and numerous tests made by Edison before a reliable filament for light bulbs was discovered.
Multiple parametric experiments can be run serially or in parallel. So-called matrix experiments, in which many experiments are run simultaneously, with one or two parameters
being varied between the different experimental setups, can be very valuable. However, they are impractical (too costly) for complex experiments that require expensive equipment. Parametric variation experiments with such equipment must be done sequentially, preferentially in the same apparatus, if it is stable over time and use. Sequential experiments can also be conducted in different apparatus. At this time, operation of many small and relatively inexpensive setups with simple but adequate diagnostics should be very useful for empirically improving reproducibility of LENR experiments, even before full understanding is
achieved. The prospect of running multiple LENR tests simultaneously in the same or very similar conditions is attractive.
Understanding, that is, having a theory that is clear and well tested, should also solve the reproducibility problem. But, that approach contains a Catch-22 in that the experiments
needed for adequate testing of a theory might be problematic, if all of the relevant parameters and their values are not specified by the theory being tested. The situation can be imagined as similar to a system of roads. If a person starts at a point that has no connection to the desired
goal, there is no hope of getting to it. But, even if there is a connection between the starting point and the desired destination, there are many forks that will lead off to places other
than the goal. Some of the “topography” is known for LENR, such as high loading for electrochemical experiments. But, key guidelines for achievement of reproducible excess power
in various experiments are still missing.
Before leaving the topic of reproducibility of LENR experiments, it should be noted that there has been little attention to quantification of whether or not excess heat or some other evidence for LENR has been observed. The question, yes or no, is part of the important subject of “binary classification.”
That topic is critical to medical testing for questions about whether or not a person has cancer or some other malady. If heat, radiation or other effects occur at low levels, there is naturally
a question of whether or not an experiment produced LENR. That decision is commonly the opinion of the experimenter.
Usually, there are not enough runs to achieve statistics adequate to apply the common 3σ rule. Then, an effect is considered present if its magnitude is more than three standard deviations (σ) of the noise in the measurement system above the average noise level. No LENR experiments have come close to being run enough to apply the detection methods that go by the name Receiver Operating Curves. The classification challenge is toughest for heat data. Not only are
there the normal statistical considerations, but sometimes small levels of LENR power production are not distinguishable from power due to chemical reactions. In general, there
should be more attention to measures of reproducibility, beyond subjective determinations of success or not, and to distributions of those measures.
In summary, not all of the key factors for the production of LENR are known, let alone adequately controlled. Hence, it is difficult to replicate experiments, even within one laboratory
and, especially, between laboratories, even if they seem to be similar in their construction and operation.
Despite this fundamental challenge, the problem of imperfect reproducibility of LENR can be systematically confronted.
This requires two things, a serious and adequately funded attempt to have similar equipment and procedures, and the use of very well characterized materials. Both the composition and structure of materials have to be known in detail before and after experiments.
Lack of assured reproducibility is both a scientific and practical problem for LENR. And, it still figures in the unwillingness of the broader scientific community to accept LENR as a legitimate field of science. It can be noted that, even after more than a half century of dramatic industrial success in a multi-billion dollar industry, the production of integrated circuits does not produce yields of 100%. Achievement of even good reproducibility will speed acceptance of the reality of LENR, even in the absence of full understanding. So,reproducibility remains a critically important goal. But, it alone does not guarantee the development and success of
commercial LENR power generators. Control of such systemsis also mandatory.
Please compare this objective, systematic, consistent presentation of the facts and rational search for a solution with my incessant whining due to the extreme wickedness of the R-problem in LENR and rhetoric about "actionable" and "non-actionable parameters".
Eventually both say the same thing: we will need a cut-the -Gordian knot type solution for this problem.
2) News, info for today.
Playing LENR Catch-up ...already
jd_sweeney trying to convince Canadians to join the LENR race
http://www.drboblog.com/national-instruments-serious-cold-fusion/
Open Power Association Newsletter #17: Roy Virgilio honored; collaboration with Francesco Celani moving forward Hydrobetatron.org speaks well about two friends of mine
Open Power Association Newsletter #17: Roy Virgilio honored; collaboration with Francesco Celani moving forward Hydrobetatron.org speaks well about two friends of mine
Recently, Piantelli spent a lot of time informing MFMP on the all important matter of credibility in LENR experimental claims.
ReplyDeletePrincipally about the HUGE amount of energy that can be stored in various forms of Hydrogen and that must absolutely be excluded before a threshold for anomalous heat is met. Below that threshold no meaningful conclusion could be had about anomalous heat.
He talked about ionization, absorption, re-combination, para and ortho and various charge states etc. Just ionization energy of 1.008 g (1 mole of Hydrogen) is 1,312 kilojoules, the re-combination is 423 kilojoules and so on. Without a full account of the amount of potential hydrogen in a reaction, results are a fantasy and will not be taken seriously.
A weak LENR system that operates at or near that observation threshold cannot use heat as a indicator of a positive LENR reaction.
Other criteria must be used to mark the start of the LENR reaction.
The nano-magnetic theory of LENR supports two other more sensitive indicators for the start of the LENR reaction.
1 - onset of superconductivity which reduces the resistance of the heater wire circuit.
2 - production of RF radiation as another energy emanation product of the LENR reaction.
These two more sensitive indicators can be used to supply experimental feedback affording the capability to improve the LENR reaction in an experiment until anomalous heat greatly surpasses the heat detection threshold.
All the recent TPR2 replication efforts have shown #1 above, and #2 has yet to be checked for.
Axil,
ReplyDeleteAs usual thanks for your well worded and informative insights.
DSM
Whether an experiment is reproducible or not can depend on the definition of the experiment. There is a reproducible cold fusion study with PdD cold fusion. That is, set up conditions known to generate the effect, at least some percentage of the time, and use the state of the art to see the Fleischmann Pons Heat Effect. Capture and measure any generated helium.
ReplyDeleteThis has been done many times. The anomalous heat effect correlates with the helium. That's reproducible. It generates a measure, a ratio. Nearly all experiments reported find the ratio to be the same within experimental error.
If one is looking for predictable results as one measure, typically heat, with PdD cold fusion, even the best protocols don't do that. And, you know, if I throw a die, how many times does it come up a six? There is no reproducible experiment, as to single outcome. However, the outcome can be statistically found with many experiments.
We shoot ourselves in the foot when we acknowledge and excuse lack of reproducibility unless we clarify what *is* reproducible. Yes, there are explanations for the chaotic nature of the heat. But that does not mean that there is no reproducibility.
Reproducibility is not a rubber concept, it is something practical as starting your auto, TV, grass cutter machine> you push a button and the gizmo has to start, any time all times. If it don't start it must be a reason you have no gas, some piece is out. The same must
Deletebe in LENR/CF -excess heat any time, all times- because if we will
You say about correlation with helium- no practical, value, helium is the effect not the cause of CF.
Not much to define on reproducible, it starts at will and gives very similar or easily interpretable results each time.
Not we create definitions- they are ready at the "customers" No repro- not buy.
Peter