Confirmed Global LENR results

  • If you are looking to publish in a peer-reviewed journal you will need evidence that stands on two legs, not one, no matter how carefully constructed. Transmutation would be a natural.

    Transmutation at higher levels - yes agreed.


    Transmutation at very low levels - or measured only on surface - not clear unless the elements transmuted to do not occur naturally.


    I am sure that transmutation evidence could be made 100% reliable. I am not sure that is easy.

  • We have literally thousands of experiments where we have measured neutrons, gamma rays, x-rays, Geiger counters etc. and do not consider those to be reliable indicators of LENR.

    Gamma rays are probably a secondary reaction, caused by something produced in the primary cold fusion reaction. X-rays seem to happen in every case, but they are difficult to detect. I agree that neither are reliable indicators.


    The best way to detect x-rays with electrochemistry is by putting x-ray film into the cell, right next to the cathode. Unfortunately, modern scientists don't like 19th century techniques. More's the pity.


    With current generation of reactors, we feel XSH is by far the most reliable fingerprint of LENR.

    As Fleischmann said, heat is the principal signature of the reaction.

  • As Fleischmann said, heat is the principal signature of the reaction.

    Just ask ChatGPT! She'll tell you.


    Who said "Heat is the principal signature of the reaction," and what does that mean?

    Fleischmann said "Heat is the principal signature of the reaction." This means that excess heat is the most important and reliable indicator of a nuclear reaction occurring in the system. In most experiments, heat is the easiest parameter to measure, giving the highest signal-to-noise ratio.

  • Just ask ChatGPT! She'll tell you.


    Who said "Heat is the principal signature of the reaction," and what does that mean?

    Fleischmann said "Heat is the principal signature of the reaction."

    I am also using ChatGPT for engineering inquiries. You get some useful feedback, but you have to critically cross check its feedback. Sometimes it doesn't seem to be right.

    It is my understanding that ChatGPT derives its conclusions by scanning internet sources. I believe the answer "Heat is the principle signature of the reaction" is a "main stream" answer.


    You could probably relay on this answer, if you have a well accepted theory which matches physics reality very well.


    Are we already there?


    Moreover I believe this answer is somehow dangerous hence neglecting that we look at a nuclear process ( what we as humans expect!?)

    Why doesn't ChatGPT recommend some basic safety precautions, like scanning for harmful radiation or particles?

  • It is my understanding that ChatGPT derives its conclusions by scanning internet sources. I believe the answer "Heat is the principle signature of the reaction" is a "main stream" answer.

    The public version of ChatGPT at https://chat.openai.com/ does scan the internet. But the version at LENR-CANR.org is restricted to documents at LENR-CANR.org only. I set the parameter to restrict it, and I chose which documents to upload to its database. I did not tell it to automatically scan LENR-CANR.org for documents. I think that is an option, and I may try it.


    So, that statement comes from LENR-CANR.org. It may come from me. . . Nope. I asked the Bot:


    Chatbot 2


    Me: Fleischmann said, "heat is the principal signature of the reaction." Do you have a document referencing this?

    Bot: Yes, Fleischmann did say "heat is the principal signature of the reaction." This quote can be found in a paper titled "Calorimetry of the Palladium-Deuterium-Heavy Water System" published in the Journal of Electroanalytical Chemistry in 1990.


    Me: What is the URL of the document "Calorimetry of the Palladium-Deuterium-Heavy Water System"?

    Bot: I'm sorry, but I do not have a URL for the document "Calorimetry of the Palladium-Deuterium-Heavy Water System." However, it was published in the Journal of Electroanalytical Chemistry in 1990 and can be found in various academic libraries or online databases.


    (Can't find URL . . .)


    Chatbot 1


    What is the URL of the document "Calorimetry of the Palladium-Deuterium-Heavy Water System"?

    The URL for "Calorimetry of the Palladium-Deuterium-Heavy Water System" by M. Fleischmann, et al. is http://lenr-canr.org/acrobat/Fleischmancalorimetr.pdf.

  • Why doesn't ChatGPT recommend some basic safety precautions, like scanning for harmful radiation or particles?

    It does if you ask it! ChatGPT only tells you what you want to know. The LENR-CANR.org Bot says:


    Me: When people do cold fusion experiments, are there recommended safety precautions, such as scanning for harmful radiation or particles?

    Bot: Yes, there are recommended safety precautions for cold fusion experiments. While cold fusion emits few particles, it is recommended to use appropriate shielding, such as a sheet of paper or aluminum foil, to protect against any particles that are emitted. Containment is also important, as experiments can produce minor amounts of neutrons and tritium, which must be handled, stored, and disposed of in carefully regulated ways. It is recommended to use laboratory grade Teflon glassware or steel cells, and to allow for radioactive products, high pressures at high temperatures, coolant circulation, and the ability to easily maintain experimental protocols.


    You have frame your question carefully to get a meaningful answer. Because this is a program, not a sentient entity. It has no idea what "safety" means, what people are, or what happens in the real world.

  • OOPS! That's wrong. That is a hallucination. That quote is not in that paper. Not good!!!


    The quote probably comes from me. Fleischmann did say it, often.

    So, now she is also becoming good at making up stuff? 🤣

    I hope you can find the source of these innaccuracies.

    I certainly Hope to see LENR helping humans to blossom, and I'm here to help it happen.

  • So, now she is also becoming good at making up stuff? 🤣

    I hope you can find the source of these innaccuracies.

    There is probably nothing I can do about that. I hope they upgrade to ChatGPT version 4 soon. It produces fewer hallucinations. It fixes several other problems.


    Most of the hallucinations I have seen apparently come from mixing up source data. There have been some clear examples published, such as the Brian Hood case. He was involved in a corruption case, and people did go to jail, but not him. Obviously, ChatGPT parsed the sentences incorrectly, mixing up the names of people who were the subject of a sentence or a paragraph.


    Australian mayor readies world's first defamation lawsuit over ChatGPT content
    A regional Australian mayor said he may sue OpenAI if it does not correct ChatGPT's false claims that he had served time in prison for bribery, in what would…
    www.reuters.com

  • Alan has requested I update this thread with our work. We have established a good relationship with two independent and highly credible labs, and possibly one more in the works. We have one successful test which showed significant XSH which showed the exponential temp. vs. XSH relationship. That experiment dropped off at higher temps so we went back to our lab and now we are almost ready with 5 pairs of new reactors. We have also received some support to upgrade the calorimeter at one of the labs, and this work is almost completed so we will start a new round of testing this month. The results from the first experiment showed >13W XSH at its peak with about 3W total uncertainty for about 11W at 95% confidence and 120W input power and about 540°C reactor temperature.


    The advice of Alan Smith has been taken seriously and we have made plans to include transmutation data as an additional endpoint. To achieve this we have made the reactors in pairs, one of which will be randomly chosen to run as an XSH experiment and then after producing XSH for some weeks we will do an acid wash of the interior of these reactors and run the resulting dissolved metals through ICP-MS for elemental analysis.


    One of the labs has a spectrometric scintillation gamma/neutron detector so we will have that data but the experimental setup is not optimized for radiation detection and minimum energy detectable is about 20keV so not expecting anything exciting there. Our team continues to grow as our new lab was just completed so things should start to speed up soon.

  • Thank you Daniel, good update. You might also consider Quad Mass Spec gas analysis - though you would need a very good QMS to separate He4, He3, D2 and H2. There are outfits that specialise in Tritium detection who could work on a gas sample for you- and advise you on a collection procedure. These were originally set up to assist with Uranium prospecting and their charges are not astronomical.

Subscribe to our newsletter

It's sent once a month, you can unsubscribe at anytime!

View archive of previous newsletters

* indicates required

Your email address will be used to send you email newsletters only. See our Privacy Policy for more information.

Our Partners

Supporting researchers for over 20 years
Want to Advertise or Sponsor LENR Forum?
CLICK HERE to contact us.