MFMP Provides Update About Me356

Quote

No plausible mechanism for a measurement error or artifact has been found, though it's still possible.

Given the circumstances, a measurement error or artefact is absolutely plausible. I posted a hires plot of the ambient temperature measurements during the occurrence of "the signal" at the time on ecatnews. Here it is again: https://postimg.org/image/nza0n6nub/full/

I wouldn't trust any measurement done during those hours with any instrument in that room. Looks like static discharge against a ground leak to me. The other TCs seemed to be similarly confused.

Bremsstrahlung gives the wrong spectrum. You need the less common "inner Bremsstrahlung" to match the given spectrum even approximately.

And my hypothesis of electrical noise has not yet been ruled out. Mild confirmatory evidence comes from the coincident power glitch. Only mild because you maybe need noise over a longer period of time than the glitch to generate this waveform - but I have to say that is just a guess. Coincidences are worth something, so it may well be implicated.

The idea that because an exact mechanism is not obvious, therefore noise-related artifact is unlikely is frankly just wrong. The point is that noise-related artifact is possible, and matches the spectrum.

Re TCs being confused. The voltage sense lines going into ADCs share a ground with the USB bus, as far as I can see. How bad is that? Answer - we don't know. it can be made perfectly Ok but just one connection with ground intermeittent and you get large ground noise.

How is some unprecedented and never before observed weird x-ray spectrum a better hypothesis than noise in this case?

Quote from THHuxley

Bremsstrahlung gives the wrong spectrum. You need the less common "inner Bremsstrahlung" to match the given spectrum even approximately.

Not necessarily. In the case of 137Cs, as pointed out to me by gameover several days ago, you would have two bremsstrahlung spectra overlaid on top of one another: one beta with a 0.51 MeV endpoint (94 percent) and another beta with a 1.2 MeV endpoint (5.4 percent).

https://en.wikipedia.org/wiki/File:Cs-137-decay.svg

But that just begs the question, where's the 0.6 MeV gamma peak? (Or is there one hiding there?)

At any rate, it seems worth the time to use the 137Cs calibration source to investigate what the spectrum would look like in this setup, even if the possibility of a match is remote.

Coming back to the big picture here.

No normal scientist would spend much time trying to analyse this artifact. They would repeat the experiment and either measure it to death (if it exists) or reckon it was one-off artifact if it cannot be repeated.

I can't understand why MFMP do otherwise in this case.

Quote from Eric

Not necessarily. In the case of 137Cs, as pointed out to me by gameover several days ago, you would have two bremsstrahlung spectra overlaid on top of one another: one beta with a 0.51 MeV endpoint (94 percent) and another beta with a 1.2 MeV endpoint (5.4 percent).

Perhaps I'm wrong, I'm not putting much effort into this for the reasons I state above.

But surely that would make for a clear "wiggle" in the spectrum, which we do not see?

Quote from THHuxley

But surely that would make for a clear "wiggle" in the spectrum, which we do not see?

There's several things going on. There are the two bremsstrahlung spectra with different endpoints; there is the increasing cross section with increasing energy; and there's the underlying beta spectrum tailing off at the endpoint. I have no idea what it would look like. This is something that should be measured.

You don't have to worry about it if you don't want. I find it interesting.

Quote from Jami

Given the circumstances, a measurement error or artefact is absolutely plausible. I posted a hires plot of the ambient temperature measurements during the occurrence of "the signal" at the time on ecatnews. Here it is again: postimg.org/image/nza0n6nub/full/

This is an unlikely source of error applied to the gamma spectrometer. The scintillator is surrounded by hundreds of pounds of lead that prevent anything other than a slow daily drift in temperature for the scintillator it surrounds closely. The spectrometer itself is a standalone instrument that is largely digital.

The temperature measurement used a 16 bit A/D with its own processor, and what you are seeing is probably largely a slow drift with quantization noise in the thermocouple voltage acquisition. The slower drift may partly be due to imperfect cold junction compensation. This DAQ system is a completely disconnected system from the spectrometer. The spectrometer is a stand-alone system that is periodically [digitally] read into a file using the same computer. The spectrometer is only read out after the spectrum is completely integrated and is read by reliable data methods.

I am not saying that there is no possibility that the signal arose from unknown noise input, but what you describe is not it.

The ambient temperature variation Jami plotted is typical of the environment in my "lab", which is not climate controlled. It includes both the daily variation of temperature and the measurement noise bands of type K thermocouples in close proximity to the AC-powered heater coil. Jami also suggests "ground leakage", which is prevented by use of a medical-grade isolation transformer on the entire heater system, and "static discharge" which is equally impossible given the mounting of the scintillator inside the conductive metal of the 240 kg lead cave.

Regarding the other issues raised:
* I tested the possibility of a spectrometer signal resulting from a power supply glitch. I switched the heater system on and off repeatedly at up to full power, while recording a spectrum, and saw no resulting effect.
* I ran several week-long background spectra both before and after the experiment, and saw no evidence of anything other than typical background.
* The suggestion of the signal being a result of stray air-borne radioactive debris from Fukushima is absurd, more than 4 years after the event. Any such occurrence in environmentally-conscious California would surely have been noted by others, and well publicized. Santa Cruz is officially a nuclear-free zone and is alert for such a possibility.

Quote from magicsound

The suggestion of the signal being a result of stray air-borne radioactive debris from Fukushima is absurd, more than 4 years after the event.

This will not be the first absurd suggestion I've made. How about this: are you using the cesium calibration source that Bob H. referred to?

Quote from BobHiggins

The spectrometer is a stand-alone system that is periodically [digitally] read into a file using the same computer. The spectrometer is only read out after the spectrum is completely integrated and is ready by reliable data methods.

Quote from magicsound

* I tested the possibility of a spectrometer signal resulting from a power supply glitch. I switched the heater system on and off repeatedly at up to full power, while recording a spectrum, and saw no resulting effect.

Interesting. I was thinking for some reason that the spectrometer might be fed by the same power supply as the heater coil.
Since it is stand-alone, I doubt it would be influenced by the heater power.

Edit: I just looked it up. Seems to be AC powered. So what if its AC supply was interrupted? This is the expensive experiment I was thinking of earlier.

http://www.spectrumtechniques.com/ucs30.htm

Quote from THHuxley

No normal scientist would spend much time trying to analyse this artifact. They would repeat the experiment and either measure it to death (if it exists) or reckon it was one-off artifact if it cannot be repeated.

I can't understand why MFMP do otherwise in this case.

MFMP has already run one approximate replication and did not see the signal reappear. I say approximate because the experiment was not automated, so there is no script to drive the experiment in exactly the same way. MFMP will continue to try to replicate. We have been pushing for other experimenters to instrument themselves with sensors for radiations and it is happening. Jiang and Zhang Hang did not test for radiations. Unless you test for them, you will not see them.

MFMP is also adding neutron monitoring to their experiments.

Quote from BobHiggins

In fact, the 137Cs is a standard calibration source for the gamma spectrometer - the 662 keV gamma is all that shows - no Bremsstrahlung is seen at all.

The idea is not that you would see bremsstrahlung with a simple cesium calibration source. The idea is that you might see it when the electrons are sent into the wall of the lead cave. If there does turn out to be a bremsstrahlung spectrum, what would the 662 keV gamma peak look like within it? Perhaps it's an absurd suggestion. Seems like something that could be looked at.

Quote from Eric Walker

The idea is not that you would see bremsstrahlung with a simple cesium calibration source. The idea is that you might see it when the electrons are sent into the wall of the lead cave. If there does turn out to be a bremsstrahlung spectrum, what would the 662 keV gamma peak look like within it? Perhaps it's an absurd suggestion. Seems like something that could be looked at.

I don't think you could thread out a small Bremsstrahlung effect from the huge Compton scattering seen when measuring the 137Cs check source. The gamma signal detected from this check source is huge.

On the recheck of the signal, was excess heat produced? The production of excess heat may be a prerequisite of signal production.

Quote

The ambient temperature variation Jami plotted is typical of the environment in my "lab"

It isn't. In fact the plotted period is totally atypical of the environment in your "lab". The rest of the data looks very similar to the first three hours of the plot.

Quote from axil

On the recheck of the signal, was excess heat produced? The production of excess heat may be a prerequisite of signal production.

It could well be that radiation and excess heat are not seen at the same time. We have heard before that the radiations seem to occur in some start-up transition - when the operating parameters are out of whack, but on the verge of producing excess heat. Like Celani's report of a large radiation burst he detected outside the lab while Rossi was starting up his eCat.

Quote from Jami

The rest of the data looks very similar to the first three hours of the plot.

I'll have a look at this Jami. Did you get the data for your plot from the HUGnetLab archive? What data interval did you use?

Quote from Jami

the plotted period is totally atypical of the environment in your "lab".

Here's a 48-hour plot of the ambient TC vs. RTD cold junction reference, 30-second average data from the GS5.2 experiment. They track pretty well. The noise level on the ambient TC is ±1°C peak, which is the typical rated accuracy for a Type K TC. That represents about 80 microvolts of p-p noise, mostly caused by stray AC fields on the unshielded TC wires.

http://goo.gl/1tWX5X

Quote

Regarding the other issues raised:* I tested the possibility of a spectrometer signal resulting from a power supply glitch. I switched the heater system on and off repeatedly at up to full power, while recording a spectrum, and saw no resulting effect. * I ran several week-long background spectra both before and after the experiment, and saw no evidence of anything other than typical background.* The suggestion of the signal being a result of stray air-borne radioactive debris from Fukushima is absurd, more than 4 years after the event. Any such occurrence in environmentally-conscious California would surely have been noted by others, and well publicized. Santa Cruz is officially a nuclear-free zone and is alert for such a possibility.

I'm not sure if this is arguing against my point above that in this cases a noise issue is the most plausible candidate for explanation of the data?

If it is then I'd point about that no such testing can ever detect an intermittent noise issue. suppose that a faulty ground connection, for example, was the culprit. Disconnecting and reconnecting a USB connector would restore it. In fact with this type of typical intermittent fault almost anything (including ambient temperature cycling) can switch an intermittent error on or off.

From my own experience, such weird errors (in electrical circuits) are real, always surprising, and relatively common. That does not prove this issue was such an error, but it is a plausible candidate.

Trying to reverse engineer from this limited data a specific detailed noise mechanism is a mugs game, and not needed for the general point that noise is a decent candidate.