Implications of Signal, Seeing into the Cat with X-Rays - Video from MFMP

Quote from Paradigmnoia

I have fiddled around with the spectra files, even last night a bunch. The background doesn't remove very neatly at all. I tried time periods that were roughly equal to 7 and ones twice as long as background.
I selected and compared ROIs between the…

The different traces have different times. Are you normalising each to counts/hour and taking average of background then subtracting value multiplies by length of trace 7 (about 4 hours I think)?

When determining how "extreme" the signal is it is worth dividing total number of counts/hour by background number of counts/hour.

@Thomas Clarke
I thought that I was fairly clear, where you quoted me about time periods.
I tried a couple different periods ( 6, 16 and 22 ) that were close or slightly longer than 7 to try as background. I also tried some periods that were multiples of the time period of 7. The multiple-long periods give slightly higher overall readings than the similar-time periods in most cases. However, neither type works well at scrubbing the K40 bump when used as as background for subtraction using the software suggested with the spectra files, when subtracted only once. Two consecutive subtractions of the same period used as a background works reasonably well, but the K40 bump is still fairly noticeable. Three consecutive background subtractions leaves an obvious Signal pattern (which looks like the one I posted a few pages back, blue background) every time.

I hope that this week I will measure spectrum from my reactors. 2"x2" NaI should give a nice data.
I have made many improvements there. Sensitive measurement for Beta will work as well.
A dust inside reactor will be excluded since a decent vacuum pump with a filters will be installed there.

@Thomas Clarke
(This might not be so clear...) It might be that since these are mostly random background events (although with similar causes), that the chances are as good that adjacent channels will be excited in any one selected file as the channel that records a similar event in any one other file, so that a single background subtraction won't work as well as might be expected. I'll try two consecutive background subtractions from two different similar-period files and see if that gives a better spread of events that makes a better overall background subtraction.

Edit: Only marginally better. Here is spectra 7 minus sequentially 6, 16 and 22:
(The little cluster on the high keV end is 45 counts combined)

@Thomas
Once again I must correct serious misinformation you posted, apparently without checking facts.

First, regarding the gamma spectrum #7 recorded during the GS5-2 experiment, the peak signal observed was 59.9 CPM at 23.6 keV. Background level at the experiment was repeatedly measured over a period of two weeks prior to the experiment, and was found to be 2.32 CPM at 23.6 keV. Thus the signal seen was 25 times background, not 6 as you claimed. Furthermore, the shape of the spectrum is visibly different than the typical background spectra, and bears none of the signature peaks of Radon or its progeny, or of any other proposed contaminant.

Second, regarding the alleged ground loop, great care has been paid to this possibility in refining the experimental apparatus over the course of six prior experiments in 2015. For details, please refer to my paper at https://goo.gl/VsFidF
and to the system wiring diagram on page 3 of the experiment document at https://goo.gl/tidwWK. The increased noise seen in the temperature data is believed to be due to induced AC from the magnetic field of the heater coil, and is an unavoidable consequence of the cell design. However, this is irrelevant in the case of GS5-2, for which no excess heat is claimed. Unlike the thermocouples, the NaI gamma detector is physically distant from the heater and is therefore not subject to this problem.

Alan Goldwater

@magicsound
What is the low keV cut off or attenuation value?
I was thinking it was 50 keV for some reason.
Sorry to be a bother. I'm sure this was discussed earlier, somewhere, but there are now a lot of places to go looking for that value again.

@StephenC
Regarding high energy beta shielding, I think the idea is to use a material composed of atoms that will recoil easily if hit, thereby avoiding a strong X-ray emission. Lots of grazing would be good to slow the betas down, so even a dense gas of light elements would be good, or water.
Alumina is probably OK, but not great as a shield.
Once (if) the beta energy gets converted to an X-ray photon, blocking it is a lot harder.

@Para.

The low energy cutoff depends to some degree on the calibration settings of the instrument. When auto-calibrated with CS137, the cutoff is 11 keV, appearing on channel 11 of the MCA. The highest channel (1024 pending upgrade this week to 4k) is 1016 keV.

For the GS5-2 spectra, a 2-point calibration was done to extend the range at the high end. The low end filter cutoff is 22 keV in channel 14. The highest bin is nominal 1428 keV. By setting up the instrument this way, the K40 peak at 1460 keV can be used for measuring the calibration drift as the NaI detector heats up.

@magicsound
Thanks.
I was thinking that channel 15 was 50 keV, but it seemed a bit too close to the bismuth peak at 78 keV.
I was doing a two point, post calibration by using the K40 peak at 1461 keV, and the 78 keV bismuth (Pb?) peak, but thought that maybe doing a three point re-calibration using the filter cut off would help stabilize things in the Signal area.
Just for messing around, and trying things out.

@Para.
Keep in mind that the TC of a typical NaI detector is -0.3%/°C. In the GS5-2 data, this resulted in the K40 peak appearing at ~1350 keV with the detector at 42 °C.

@magicsound
That's why I was attempting to re-calibrate the spectra files. I could see that the K40 bump was wandering around a bit from file to file.

Unfortunately the software doesn't seem keep the calibrations locked in by keV if you import them as a background, so I had to look for ones that lined up the Bi 78 and K1461 peaks better.
I was fiddling with that last night for quite some time to see if I wasn't doing something wrong, or if I had try to find a decent work-around.

I might be able to mess with the files... but at some point it is best to move on to better things and not mess with the data beyond its usefulness.

@Paradigmnoia In the plot you provided above there is what looks like a small step and slight change in profile in the peak. Do you know what causes this? Is it just an artifact of removing the background? Or is it a real effect? Do we know the energy rang in keV where it occurs?

Regarding the small cluster around 800 keV I'm wondering if they can be gamma emissions at this energy not normally in the background?

I think there are some gamma emissions for some Cu/Ni isotopes around 1300 keV or so, so I'm wondering if we see a cluster there if that can also be indicating excited states of those element. I suppose it could easily be confused with the Compton edge for K40, if it does occur.

@Paradigmnoia Regarding the shielding of beta, I suppose we should be careful with materials since I suppose they would introduce bremsstrahlung emissions that might confuse the signal we are looking for.

Would not the mullite tube already be sufficient to block any beta generated inside the device though?

If a shield is used wouldn't boric acid be a good option... Also good for neutrons.

@StephenC
The small dip I am certain is caused by the 78 keV peak getting removed three times, and so is an artifact. I suppose it demonstrates extreme background subtraction.

I would be pretty surprised if MeV range betas were made, but even low keV betas can be a problem if they hit a high Z object. That is exactly the principle of an X-ray device. The conversion rate to X-rays is pretty low, though. The vast majority of betas or electrons in a beam scrub their energy as heat, which is why rhodium or tungsten (sometimes combined with gold) are used as targets for the electron beam in X-ray devices, rather than lead.
The ceramic probably deals with almost all betas, at least to some fairly high energy level, but I cannot say that with absolute certainty.
Air deals with most low energy betas fairly quickly.

@StephenC
Sorry, I misread your post regarding 800 keV, and was thinking 80 keV for some reason.
Where is the 800 keV cluster? If they are the bump, left of the valley which is left of the K40 hump, then that is a typical background shape. (It might be Bi214 from radon decay, at 1120 keV). If there were more hits hiding in there from something else, I suppose it would be hard to tell unless a large signal was pushing through.

Below is another handy gamma ray energy list, but it doesn't include the 78 keV peak which is obvious in the MFMP spectra, which is actually from cosmic rays hitting lead ( probably their Pb cave). This is a handy indicator that a cosmic burst probably did not cause the Signal, since this Pb x-ray peak was not greatly enhanced.
https://www.cpp.edu/~pbsiegel/bio431/genergies.html

@Paradigmnoia The 800 keV cluster I was mentioning is actually closer to 900keV on re checking and was in fact the small cluster of points or so near the end of your graph you mentioned to Thomas. I think the graph ended near 1024 keV.

i agree there are few MeV betas normally as even beta emission with high Q values and hence endpoint in the MeV range such as Ni65 have only a few emissions at this high level and most emmissions in the 100s keV and below.

Yup I guessed it might be due to the peak removal... I was curious if it could show an absorption feature around 66.9keV but it makes sense what you said since this part of the back profile was removed 3 times. I think in Eccos and Bobs charts it was quite smooth here apart.from the background bump with not much noise too.

regarding this peaks at around 78 keV in these original charts I wonder if it is it the same size as would be expected from background over the time frame? The one in spectrum 7 is at much higher over all counts but still visible on the log scale at this high level

Do you know the tool on this link by the way:

https://www-nds.iaea.org/relnsd/vcharthtml/VChartHTML.html

Its really informative and clever when when you get into it. The search function is very powerful and provides a lot more information than is in the chart itself. There is a very nice app too

Quote from Alan Goldwater

Once again I must correct serious misinformation you posted, apparently without checking facts. First, regarding the gamma spectrum #7 recorded during the GS5-2 experiment, the peak signal observed was 59.9 CPM at 23.6 keV. Background level at the experiment was repeatedly measured over a period of two weeks prior to the experiment, and was found to be 2.32 CPM at 23.6 keV. Thus the signal seen was 25 times background, not 6 as you claimed. Furthermore, the shape of the spectrum is visibly different than the typical background spectra, and bears none of the signature peaks of Radon or its progeny, or of any other proposed contaminant.Second, regarding the alleged ground loop, great care has been paid to this possibility in refining the experimental apparatus over the course of six prior experiments in 2015. For details, please refer to my paper at goo.gl/VsFidFand to the system wiring diagram on page 3 of the experiment document at goo.gl/tidwWK. The increased noise seen in the temperature data is believed to be due to induced AC from the magnetic field of the heater coil, and is an unavoidable consequence of the cell design. However, this is irrelevant in the case of GS5-2, for which no excess heat is claimed. Unlike the thermocouples, the NaI gamma detector is physically distant from the heater and is therefore not subject to this problem.Alan Goldwater

I admit (me culpa) to the 25 vs 6. I was remembering the 6 from before times were factored in - although since some error mechanisms are time independent 6 is also relevant.

For the rest of your comments, I don't see how they bear on what i was saying:

(1) I have never ever claimed the MFMP results could come from radon. It is silly.

(2) Re ground loop. "Great care has been spent" merely means that it is indeed an issue, not that the issue has been resolved. We are talking here about a potential one-off glitch that will not normally evidence itself. Also, it is interesting to look at the ambient temperature TC raw data. You find it is noisy, as you might expect, due to electrical noise - probably ground current noise. What is interesting is that it is much MORE noisy in some periods than others, and trace 7 corresponds to a more noisy period. Not a smoking gun, but something to be resolved and understood before seeking way-out explanations for this one reading. I realise that there should be no noise. That is not the point: strange things happen regularly and what should happen and what does happen often do not correspond, for reasons only in retrospect understandable.

(3) Re the spectral shape. indeed, it is vastly different from background spectrum, but not from "no input" spectra. These have exactly the same bias towards low energy buckets. The very smooth distribution looks plausible therefore some sort of software or electrical noise induced glitch.

(4) I don't have a good explanation for this signal. But I've lived my life finding strange observations that i could not explain - and later - with a bit more evidence - or a bit more determination - explaining them.

People who research LENR often do not have that determination, because for them any unexplained observation related to excess heat or radiation or isotopic anomaly can be fitted into an "LENR catch-all" hypothesis. I'm not generalising here, and no doubt you do not fall into this camp. Actions speak louder than words and when MFMP have tightened up instrumentation, not replicated their result (or replicated it in a form that shows it is an artifact) they will no doubt be back on the straight an narrow path from which they have currently strayed. Of course, if there were a genuine radioactivity anomaly as claimed it would be a smoking gun for something extraordinary possibly though not necessarily involving nuclear reactions. I'd be as excited as anyone in that case. Perhaps more so since I would have no baggage to get rid of.

My judgement based on existing evidence is that that looks highly unlikely.

@StephenC
The plot I posted has the X axis in channels, rather than keV. So that was the remains of the K40 hump at 1461 keV, and related scatter, and partly my misalignment.
The plot reverts to channels after subtracting the backgrounds.

There are probably far better ways to clean off the background than the way I did it.
I was mostly interested in what the Signal looked at when as much normal background was removed as possible.

Thanks for that link.

Quote from Thomas Clarke

Of course, if there were a genuine radioactivity anomaly as claimed it would be a smoking gun for something extraordinary possibly though not necessarily involving nuclear reactions.

Please indicate how radioactivity might be induced without involving nuclear reactions?

Quote from Longview

Quote from Thomas Clarke: “Of course, if there were a genuine radioactivity anomaly as claimed it would be a smoking gun for something extraordinary possibly though not necessarily involving nuclear reactions.”

Please indicate how radioactivity…

Well, it would require something extraordinary and probably impossible, rather like LENR. If you allow one you must allow the other. Thus, some sort of collective phenomena that resulted in high speed electrons...

The point being that the premise here is something extraordinary, and therefore it does not do to make any assumptions about its cause.