MFMP Provides Update About Me356

Quote from THHuxley

Not if the voltage drop is less than 0.5V. What voltage do you need to generate spurious spikes from the photodiodes? I'd guess a few 10s of mV.

Your meaning isn't clear here. Are you referring to photodiodes in the USB data connection from the spectrometer to the computer? Higgins' comment and your own understanding of digital systems should rule that out. The data is accumulated in the UCS30 MCA and is sent to the computer in real-time merely for display. On completion of the integration, the data is downloaded to the computer as a file in.csv format, or a similar default file format that includes setup parameters. In either case, it's a digital file and defects in transmission of that data would cause one or more packet CRC errors and eventual abort of the file transfer. As pointed out by others, it's impossible to conceive of such a peculiar digital error as to cause incorrect but coherent data like what we saw, without also causing a system error. If USB was that fault ignorant, no one would use it.

If on the other hand, you were referring to a photodiode in the scintillator head, that is not how they work. The photomultiplier tube typically operates at around 800 volts and would not be affected by ground leakage resulting from the mechanisms you described.
Regarding possible ground loops, the data system is star grounded and also has a 5 ma GFCI on its mains power source. Subsequent to my study of GS alumina conductivity leakage, I completely isolated the heater power system with a medical-grade transformer. I then tested the ground leakage current and found it to be 50 microamps or less at up to full heater power.

Further comments on the spectrometer: the scintillator head in its lead cave is on a separate cart that doesn't touch the cart holding the experiment. Its only ground is through the connections to the MCA chassis. The scintillator crystal face is protected from IR by a heat shield of ~.002" aluminum foil at the aperture of the lead bricks. This keeps the heat rise at the sensor to about 20 °C above ambient at full heat.

Quote from Eric Walker

MFMP are doing themselves a disservice by going on a fishing expedition like this and not holding themselves to a more difficult standard.

Eric, how else could we have dealt with this event (Greenyer's comments aside). As I said previously, I just run the experiments and publish the data. Are you suggesting that we should not have released this data until having replicated it? In that case we would have been denied the useful and productive suggestions made here that incrementally help improve our experimental technique. And those of you that care whether we ultimately succeed would not have had an opportunity to participate in our shared quest to understand LENR, or disprove its reality, whichever the data shows us.

Quote from Paradigmnoia

This is the sort of thing I am thinking of, like perhaps if the power to the detector body gets interrupted for a brief moment, right at the start of a new file.

This sort of statement is of no help, as it clearly shows no understanding of how a computer works...

I recommend to the LENR sceptics to point to events with a far greater likelihood:

- A black bird flying over the laboratory just before an new file opens...
- RMF from a pacemaker
- Refrigerator always starts at the wrong moment...
- magicsound did cut and paste an angry bird...

Quote from THHuxley

inner Bremsstrahlung spectrum

Aren't k,l, etc "inner" transitions inherently "characteristic" and not broad spectrum as is Bremsstrahlung? Please clarify.

Quote

If on the other hand, you were referring to a photodiode in the scintillator head, that is not how they work. The photomultiplier tube typically operates at around 800 volts and would not be affected by ground leakage resulting from the mechanisms you described.Regarding possible ground loops, the data system is star grounded and also has a 5 ma GFCI on its mains power source. Subsequent to my study of GS alumina conductivity leakage, I completely isolated the heater power system with a medical-grade transformer. I then tested the ground leakage current and found it to be 50 microamps or less at up to full heater power.

OK - fair enough. If the scintillator signal (which is what I was considering) is a current source then ground noise injected directly there is less likely. However ground noise flowing through ground connections everywhere (between PC ground and whatever grounded bit of metal happens to be touched by the cave) remains an issue: loop currents can cause errors in instruments - though depending on internal design if this is good will not. But, overall, that reduces likelihood of analog PM noise from ground loops being the issue by quite a lot.

Quote

Aren't k,l, etc "inner" transitions inherently "characteristic" and not broad spectrum as is Bremsstrahlung? Please clarify.

My understanding was that classic B has a cutoff energy, and inner B is complex but might model the spectrum. But as I indicated above I'll not argue if somone who knows more differs. I'd need to see the precise equations and assumptions made to get them before thinking this spectrum was a close fit to Bremsstrahlung, since there seem to be a lot of variables in how it is produced.

Quote from magicsound

Eric, how else could we have dealt with this event (Greenyer's comments aside). As I said previously, I just run the experiments and publish the data. Are you suggesting that we should not have released this data until having replicated it? In that case we would have been denied the useful and productive suggestions made here that incrementally help improve our experimental technique. And those of you that care whether we ultimately succeed would not have had an opportunity to participate in our shared quest to understand LENR, or disprove its reality, whichever the data shows us.

My difficulty with this experiment has not been with MFMP's decision to report the data, which I think has been helpful for the reasons you mention. It's been with the conclusions that have been drawn and that continue to be drawn. An anomaly that is suggestive and might be artifact is being transformed into something that is seen as likely to be evidence of LENR. The distinction might seem like a small one, but it's a big one. As the anomaly was the result of a fishing expedition (trawling through the data and finding something unplanned and interesting after the fact), there are too many variables that have gone uncontrolled for and leave open the door to more mundane explanations. I'll wager a guess that many scientists coming to LENR fresh would have issues with this mode of reasoning and would be harsh in their judgment of MFMP.

I can scour hours and hours of security cam video footage and find evidence of a UFO, and then argue on the basis of various things I know about what was going on at the time that it was not a bird (it moved in a straight line) and not an airplane (there were no commercial routes running at that time, and the logbooks for the local airport don't show anything) and not a malfunction in the recording hardware or software (it's working fine now, and there were various measures that were taken to protect the circuitry) and still not get much closer to a conclusion that other people will agree with that there was a UFO. The footage remains suggestive at best, and probably artifact.

So my difficulty is not with the reporting of the data, but the process of reasoning that is being used to show that it is probable evidence of LENR.

I'm still curious: You must have used a radioactive standard to calibrate that scintillator. What did you use for that, and at what points in time was this done?

@Wyttenbach,
What happens in the PMT when the power is interrupted? Will the computer calibrate for the PMT output at the start of each file?

Quote from Eric Walker

So my difficulty is not with the reporting of the data, but the process of reasoning that is being used to show that it is probable evidence of LENR.

That conclusion has never been part of my reporting or comments, and has rather been offered mainly by third-party observers here and elsewhere in the community. Bob Greenyer has suggested it in his promotional videos, but that is his personal assertion and does not represent a consensus of MFMP. Our recent internal discussion of the issue has led us to say only that like the possibility of excess heat, it MAY represent such evidence, and that further confirming work is needed.

Quote

I'm still curious: You must have used a radioactive standard to calibrate that scintillator. What did you use for that, and at what points in time was this done?

A two-point calibration was done prior to the experiment using a ¹³⁷Cs check source. The calibration only affects the x-axis scale of the spectrum by assigning an energy level to each bin of the MCA. As previously discussed, that parameter is subject to some drift with temperature, and correction was applied to the data post-experiment to account for it, using the characteristic 1461 keV ⁴⁰K line that is always present in the environment. Other tests were performed post-experiment to explore the lower energy bound of the scintillator, using a variety of check sources. The entire kit of check sources is stored in a 1-cm lead lined box, outside the lab, about 15 meters from the experiment.

Quote from THHuxley

However ground noise flowing through ground connections everywhere (between PC ground and whatever grounded bit of metal happens to be touched by the cave) remains an issue

As I pointed out, other that the coaxial cables to the MCA there is no ground connection or path to the scintillator head or the lead cave in which it is placed . The cart holding the lead cave is a separate structure not touching the experiment cart. It has rubber wheels, and with the cables disconnected, shows greater than 40 Megohms isolation from system ground. So I am confident there is no ground loop in the scintillator.

Earlier, in response to something I said, you wrote:

Quote from magicsound

Eric Walker wrote:
that particular run was suggestive at best, and possibly artifact. Do you disagree with this assessment? Note that the announcement that was made at the time sounded very different than this summary.

BobG is sometimes overly enthusiastic, as befits his role of "chief facilitator". After the post-investigations I described, I now think the data is more than suggestive. No plausible mechanism for a measurement error or artifact has been found, though it's still possible. That is why I encourage ongoing informed questions, and will add suggested tests as time and facilities allow.

"More than suggestive" is different than "suggestive". You also talk about "no plausible mechanism for measurement error" being found. The obvious implication is that we should conclude that the run was evidence for LENR. Bob confirmed this impression on my part by not backing down on the the run being only suggestive. I think I've made my point at this point and don't want to quibble over small nuances.

Quote

A two-point calibration was done prior to the experiment using a 137Cs check source.

Is the source still around? Can you aim it at the lead cave, away from the spectrometer, and see what the spectrometer picks up? This should be easy to do.

Quote from Eric Walker

Is the source still around? Can you aim it at the lead cave, away from the spectrometer, and see what the spectrometer picks up? This should be easy to do.

Yes, that can be done. In the test you propose, there will be ~10 cm of lead between the check source and the scintillator. How long a sample integration do you think necessary for useful results? How far do you want the sample source to be from the aperture, keeping in mind that Compton scattering may be an issue.

Quote from Eric Walker

Is the source still around? Can you aim it at the lead cave, away from the spectrometer, and see what the spectrometer picks up? This should be easy to do.

The 137Cs check source is a plastic disk about 1" in diameter with a spot in the middle with 1uCi of 137Cs. You do not "aim" it - it is an omnidirectional gamma source at about 662keV. Are you looking for the solid angle shielding attenuation function at 662keV?

Keep in mind that the signal was, by energy, much stronger at lower energy. At 100keV there will be near zero penetration of the lead cave and it will be a pure shadow shield. There will be no diffraction. At 662keV, the sensitive cone will be slightly larger, but not much different. The cave will not be a brick wall filter because of the finite size of the crystal and the crystal becomes gradually shadowed by the cave as the angle from the axis increases.

From the outside is interesting, but not the specific test I was thinking of. I was thinking of something like this (the "=" is lead, and the "o" is the scintillator):

============
============
==
== ooo
==
============
============

From the inside of the cave, aim the source towards the wall, way up close to it, so that the opening of the source is not in view of the scintillator, but anything coming from the lead wall will be in view. If possible, integrate long enough to get any kind of clear signal. If the shape is way off, or if Compton scattering produces an obvious peak that was not seen in the original run, then this test is a failure. I expect the test to fail, for the reasons Bob H. mentions. But since it's so easy to do, seems like it should be done.

Quote from Eric Walker

I expect the test to fail, for the reasons Bob H. mentions.

If you're proposing to look for secondary emissions from the lead (including Compton scattering), I can open a crack through the lead cave, pointing to a spot inside the aperture but not at the scintillator. Then if I place the check source at the outside of the crack, a beam of gammas will strike the lead near the front of the NaI crystal, but little or none will reach it directly. It will probably be necessary to post-process the spectrum to remove background if it is to show any signal other than BG. Since this involves added work and planning, I would like agreement on a formal experiment plan before starting.

Bob, I now see the omnidirectional disk. The sources I recall in a textbook I read were shielded metal tubes, with a cap that can be removed.

Alan, I like your idea of creating an opening in the lead cave so that only reflected radiations will make it through. But if this is something that is going to require a lot of effort in the way of post-processing of the data, and it cannot be tackled in an exploratory fashion, then perhaps best not to worry about it. I do not know, even to a first-order approximation, what one might expect. I just noticed that there are the Pb/Bi lines, that lead has a high bremsstrahlung cross section, that the curve doesn't go far beyond 1.2 MeV and is heavily weighted towards the low energy side, and that the 137Cs was a radiation source that was around. So there are some tantalizing coincidences there. But the human mind sees patterns in everything.