User talk:Kirk shanahan/Archives/2009/February

Weighing validity of opposition - lost comments

I noted in some diffs that some comments seem to have been lost. I have copied them here to be able to respond:

"The work that you have published in the peer-reviewed literature does not match Wikipedia's definition of forbidden original research, because it was published in a secondary source. Improvements to the article are the purpose of discussion here, and helping everyone to understand your sources should lead to improvements in the article.

I am trying to understand your "calibration constant shift" technique, which you say is potentially applicable to non-electrolytic cold fusion. We have established that you start by assuming a hypothesis of no excess heat, and then you design a general theoretical argument in support of that hypothesis. You then evaluate your argument's credibility based on conformance to available data and your expectations about quality control. If you were to later learn that there was data which was not consistent with your argument, would that invalidate it? 69.228.201.246 (talk) 21:57, 10 December 2008 (UTC)"

My argument so to speak is nothing but a mathematical demonstration of the analytical chemistry 'Golden Rule' (of which there are many) that "You can't calibrate an unstable system". As such, what I have done is show that a CCS (which occurs because the system has changed) can explain Storms' (2002 paper) and Szpak, Fleischmann, et al's (2005 paper) results. That is a done deal, and won't change. However, it is always possible that new data will shed more light on the problem and point elsewhere than a CCS, even to the point of proving CF. But, at the same time, since it has been shown twice in cases where the calibration data was available that a CCS has the potential to explain the results, _every_ excess heat claim must show their calibration results in order to eliminate a possible CCS. To put it another way, I have delineated a systematic error that seems to explain a lot, and any claim to excess heat must prove that a CCS is not active, otherwise the CCS remains a preferrred explanation (since it is a conservative one). When you go to different experimental protocols like the D2 flow through membranes, the error bar on their results is unknown at this time. But as long as they are calibating, a CCS could be present, so again, they have to supply the data to be able to refute its presence if they want their results accepted. Kirk shanahan (talk) 13:25, 11 December 2008 (UTC)

When applied to electrolysis, does your calibration constant shift theory imply a reduction in the volume of evolved output gases? If so, how much? 69.228.201.246 (talk) 17:02, 11 December 2008 (UTC)

Again, potentially. To explain, my 'theory' has 3 levels. The first is the simple CCS, i.e. by using slightly different calibration constants, the apparent excess heat signals can be eliminated. The 2nd level is a proposal of one of possibly many ways to get a CCS. That specifically is to move heat from a zone of the calorimeter where it is detected less efficiently to a zone of higher heat detection efficiency. The zone of lower efficiency is most likely where heat losses are greater, which is most likely where heat loss paths exist, such as penetrations of power leadss and thermocouples, etc. That normally means the gas head space. In open cells, the gas just leaves this area and exits the cell. In closed cells, the recombination catalyst is found there. The first and second levels are nothing but algebra, nothing complicated there, and no one has challenged any of that. But I also proposed a third level, which was a physical/chemical mechanims that I felt explained how a CCS could be obtained AND also explained many other observations on the functioning of P&F type cells. That mechanism is what was attacked by Szpak, Fleischmann, Storms, etc., in the literature. I feel I answered all criticisms and showed how the critics were making many erors, but you read the papers and decide for yourself.

So, IF my proposed physical/chemical mechanism is correct, one would see a reduction in the volume of evolved gases from an open cell. There would be no noticeable change in the closed cell, with the possible exception of gas space pressure.

The only time gas space pressure in a closed F&P cell has been measured to my knowledge was by McKubre as reported in his 1998 EPRI report. Unfortunately, the signal is not explained, i.e. calibrated, and the numbers didn't make sense to me. I asked McKubre for help on his cell calibrations twice in 1999, and he declined to help (which may be reasonable given the fact that the data presented in the report was from 1993-4). In my 2005 paper I point out that the Szpak et al paper I am commenting on attempts to measure this by recombining the gases external to the cell/calorimeter, but they end up with 7% MORE water than they should based on their power consumption. The cause of this is proposed to be entraiment of water droplets in the gas stream, which would be another systematic error in these type of measurements.

So the key point is here that if it was measured correctly, evolved gas volume would potentially show that excess recombination was occurring. If you look here http://home.netcom.com/~storms2/fig1.html you can see a plot that Ed Storms has posted showing that the parasitic electrochemical recombination reaction drops off as current density increases. But there are two zones of anomalous data in this plot - the fourth point reading left to right and the several points in the .03-.07A region. These points lie above the line that goes through the rest of the data as indicated by the Will model (note that the Will model is not linear and would curve up to fit the Jones data if plotted to that region). Both regions lie about 20% or so above where the Will model would say they should lie, and I contend this is the excess non-electrochemocal recombination needed for my 3rd level mechanism. But there are no excess heat data available for these runs and both sets of authors told me they saw no excess heat. So, at this point my physical/chemical mechanism remains untested, but it certainly could be. However, if it can't be proven, that does NOT negate the CCS, just forces us to come up with a new mecahnism to get it. Kirk shanahan (talk) 19:05, 11 December 2008 (UTC)

I must repeat my question about the quantity of the evolved gases. Based on your understanding of quality control in calorimetry, what is the chance that those who have said your explanation is in error because of the quantity of evolved gases observed are themselves in error? 69.228.201.246 (talk) 05:47, 12 December 2008 (UTC)

The 2004 publication by Szpak, Mossier-Boss, Miles and Fleischmann contained an attempt to measure the quantity of evolved gases. Given that all these folks are 'experts' in the field, I believe we should assume that their results are the best of the typical. They were using an open cell configuration but they subsequently converted the outflowing gases to water and measured the amount formed. They observed that "the total consumption of D₂O was 7.7 cm3 instead of 7.2 cm3, assuming 100% Faradaic efficiency, which is within experimental error.". In my 2005 comment on their paper, I noted that this is a 7% error (while the CCS is a 1-2% error), and that it is in a positive direction. In other words they 'created' 7% excess water. How could that happen? The simplest explanation is a process called entrainment (Wiki's page on this is a little lame), which is where one phase of material is carried along in another phase. In this case small water droplets in the gas flow. I believe it is reasonable to assume this process is active in all other cells. So, this means that for an accurate measure of the actual gas flow rate, it must be 'demisted' before being measured. I know of no experimental setup that did that. So I think it is safe to assume an error bar on such measurements of 1 sigma=7%, 3 sigma=21%. That's not very good. In the few other cases where people try to measure this, as I recall, the errors were consistent with this estimate or worse.

In other words, I am saying that a) very few actual measurements of this type exist, and b) they are very error laden, to the extent that they really don't offer any reliable evidence of any kind. That might be a bit strong, but that's where I am at today. Perhaps you can find a study that disproves me. So in relation to your question, based on the above, I would say the chances are near 100% that those who claim they have disproved my CCS mechanism by measuring evolved gases are in error. 192.33.240.30 (talk) 13:56, 12 December 2008 (UTC)

The 2004 publication by Szpak, Mossier-Boss, Miles and Fleischmann claims agreement with calculated volumes within "1.0%", not 7% (page 102.) 69.228.201.246 (talk) 21:16, 12 December 2008 (UTC)

The sentence quoted above comes from page 105, section 4.1, 2nd paragraph of that report. 0.5 cc (7.7-7.2) is 6.94% of 7.2. Seems we have found a logical inconsistency in their paper. Kirk shanahan (talk) 14:09, 13 December 2008 (UTC)

The 1% figure refers to a reference. I looked it up and could not find the data to support the 1% figure. ~Paul V. Keller 23:39, 18 December 2008 (UTC)

arbitrary section break

Let's step back a moment. What is the rough difference in the expected volume of evolved gases with and without recombination in a Navy SPAWAR co-deposition cell, holding the observed excess heat constant? Isn't it on the order of several dozen if not multiple orders of magnitude? 69.228.213.202 (talk) 05:45, 20 December 2008 (UTC)

Years ago I may have calulated this, but I don't recall right now. The answer to your question lies above. They expected 7.2 cc of water (that's 0% recomb.) and got 7.7, that's a 7% excess in the wrong direction to prove excess recombination is occurring (100% recomb. would be 0 cc water). Based on the excess heat signal, I recall I estimated about 20% or so recombination had to be occurring, which made the total 'error' in their water measurement more like 25%. If you want gas volumes, convert the liquid water formed to gas at STP. Kirk shanahan (talk) 19:45, 22 December 2008 (UTC)

Melvin Miles, who made the measurements, told me that "the initial volume was 91 +/- 1 cc, the final volume was 83.3 cc, thus 7.7 +/- 1 cc was consumed versus 7.2 cc calculated by Faraday's Law. A quick approximate calculation indicates that the volume of D₂O consumed would have to be smaller by about 5 cc if the excess heat in this experiment were due to recombination. This was NOT observed. The amount of D₂O consumed by electrolysis is within experimental error (+/- 1 cc) of Faraday's Law. A slightly higher D₂O consumption is expected experimentally because the D₂ and O₂ gases bubbling out of D₂O are always saturated with D₂O vapor. This is an evaporation effect.... Fleischmann's calorimetric equations have always included this evaporative effect." GetLinkPrimitiveParams (talk) 16:57, 19 January 2009 (UTC)

It always amuses me how single-minded people can be. IF WE ASSUME NO COLD FUSION, then the excess water in the outflow would be ~5+(7.7-7.2)= ~5.5cc. Thus the measured +0.5cc grossly understates the amount of water loss by entrainment, it should be more like 5.5cc. So, we now have two explanations for the observed 7.7cc water loss; cold fusion OR under-the-surface recombination produced excessive entrainment (which is now disconnected from a direct caclulation by the complexity of the process of microdroplet formation). Also we note that IF THERE IS NO COLD FUSION, the experimental error on water loss is ~5-6 cc out of 7-8cc, or minimally about 62.5%. Clearly not adequate to base scientific conclusions upon.

So now, you choose. Is there an unexplained revolutionary new phenomenon occurring that will necessitate rewriting physics and chemistry textbooks, or is there an interesting but completely explainable phenomenon happening? It has to be a choice, not a determination, because the two proposed explanations both fit the tidbits of data available. The sensationalist will choice the first choice, the boring, conservative scientist the second. Kirk shanahan (talk) 17:28, 20 January 2009 (UTC)

What do you mean by "entrainment"? GetLinkPrimitiveParams (talk) 20:56, 20 January 2009 (UTC)

Two-phase flow, i.e. here, liquid microdroplets carried along in a flowing gas stream. Check any chemical engineering text. Kirk shanahan (talk) 21:41, 20 January 2009 (UTC)

What is the evidence for that, and how is it any different from typical evaporation? GetLinkPrimitiveParams (talk) 06:27, 21 January 2009 (UTC)

See above paragraph beginning "It always amuses me...". Also, read any chemical engineering text. Or, think about it the next time you flush a toilet. You might also want to look at the thread on Vortex found here

http://www.mail-archive.com/vortex-l@eskimo.com/msg29656.html

especially the one here:

http://www.mail-archive.com/vortex-l@eskimo.com/msg29675.html

{BTW, Jed's comments about evaporation being computable are amusing, as when I used that argument to cast doubt on a psuedoresult produced by Mizuno, Jed wouldn't accept that. It devolved into a discussion of a rat pool party on spf.} Kirk shanahan (talk) 14:20, 21 January 2009 (UTC)

If I am understanding you, you are suggesting that 20% of the evolved gases are recombining. That means that there would have to be 5 cc more water left over. But there isn't; there is less water than predicted. So you explain that by droplet production "entrainment" -- is that right? So your theory is that recombination, which doesn't occur during other forms electrolysis, occurs with D-Pd electrolysis, produces 5 cc more water, and 4.5 cc more water droplet entrainment in addition to the expected amount of evaporation. And the empiricists haven't embraced this explanation? Do you really think they are incapable of measuring both vapor and droplets? GetLinkPrimitiveParams (talk) 12:22, 22 January 2009 (UTC)

First off, there is less water than predicted LEFT IN THE CELL. This means the LOSS was GREATER THAN PREDICTED. 100% Faradaic Efficiency, i.e. no recombination, predicts 7.2cc loss. So 20% of that would be 1.44cc, not 5 as you suggested.

What I am saying is that, based on input power measurements and known physics/chemistry, a loss of 7.2 cc fluid was expected and 7.7 cc was found. A 0.5cc difference is non-trivial, so saying 'that is within experimental error' simply points out that the person saying it doesn't want to refine his error bars down any further at that time. I do, and as 0.5cc is non-trivial, it should be easily done, especially since there are well-known physical chemical processes (entrainment) that could account for an excess water loss. I also pointed out a report of observed evidence that directly supports the entrainment proposal. Since Miles does not acknowledge entrainment as a viable process, it is clear he has evaluated its impact as 0. This conflicts with the observation that such a process occurrs in CF cells, and with the base knowledge of the field of entrainment. If the effect is not quantified, no comment can be made by Miles on its impact. I also say that observed fluid loss is made up of at least 3 components: electrolysis, evaporation, and entrainment. So, I am saying, 'YES, Miles made a mistake". The actual water loss, being composed of 3 components, is a 'confounded' measurement. The components must be measured separately, not in a lump sum fashion as Miles did. Depending on the degree of entrainment, the major fraction of water loss could be due to it. We don't actually know because no reliable measurement was made. So, I don't know if they are incapable of measuring both vapor and droplets, because they haven't tried!

AND, in my publication on this I point out that the maximum available recombination heat, uncorrected for CCS problems, is about 0.5W, while SMMF measure 0.27W maximally. Thus the 7.2cc water retention in the cell would be ~1/2 that. So my memory based 20% figure was low. It was probably more like 50%, or 3.6cc excess retained water in the cell. Is that an outlandish amount to be taken out by entrainment?? I have no idea, it has to be measured, and it isn't. The point is: WE KNOW ENTRAINMENT HAPPENS. If we know it happens and it can impact the numbers we are discussing, IT MUST BE MEASURED before any intelligent discussion can occur, and certainly before any conclusions are drawn.

I would also like to point out a mental error I have made in this discussion, which does not impact the conclusions, but may explain why Miles thinks I don't know what I am talking about as reported in the Vortex message. I had confounded two Szpak et al publications. The 2004 one with Miles and Flesichmann (and Mossier-Boss) is where the numbers come from. In fact they do not indicate how they determined them therin. However, in a lenr-canr version of a 1998 Fusion Technology publication, and open cell with an attached (but ouside the calorimeter boundary) recombiner is shown. I assumed they were collecting water thusly instead of measuring water loss in the cell, but the 2004 paper I commented on did not make that clear. Lack of experimental details in publications is an endemic problem in the CF area. Kirk shanahan (talk) 17:41, 22 January 2009 (UTC)

Melvin Miles in email reports that he has measured evaporative and droplet loss and it is simply not as you predict. If recombination were occurring, then there would be more water not less. Instead there is less water, so you have to assume that the entrainment is huge in comparison. I would hope you will contact him and Pam Boss to discuss this, and perhaps tour SPAWAR's codeposition research facilities to observe the experiments in progress as any U.S. citizen may. GetLinkPrimitiveParams (talk) 18:13, 22 January 2009 (UTC)

Sorry for the slow response. I was 'involved' with V.

OK, a) email don't count b) especially when I don't have a copy of it, but c) I think you are referring to a Miles message that was at least partially posted to Vortex, to whit:

"vortex-l [Vo]:Melvin Miles comments on recombination Jed Rothwell Sat, 03 Jan 2009 14:29:06 -0800

Here is a message from Melvin Miles about recombination.

Subject: Shanahan's misunderstandings of recombination . . . If Shanahan's statements about my NHE co-deposition measurements are typical of his lack of understanding of cold fusion experiments, then I would not give credence to anything he says. For example, I did not convert the outflowing gases to water as Shanahan states.

For my three NHE co-deposition experiments, I simply measured the initial and final volumes of electrolyte to obtain the volume of D2O consumed. For the cell in question (Cell A-2), the initial volume was 91 +/- 1 cc, the final volume was 83.3 cc, thus 7.7 +/- 1 cc was consumed versus 7.2 cc calculated by Faraday's Law. A quick approximate calculation indicates that the volume of D2O consumed would have to be smaller by about 5 cc if the excess heat in this experiment were due to recombination. This was NOT observed. The amount of D2O consumed by electrolysis is within experimental error (+/- 1 cc) of Faraday's Law.

A slightly higher D2O consumption is expected experimentally because the D2and O 2 gases bubbling out of D2O are always saturated with D2O vapor. This is an evaporation effect and not Shanahan's entrainment. Fleischmann's calorimetric equations have always included this evaporative effect.

Full details of all three co-deposition studies are in my NHE Report, pp. 22-25 (see LENR-CANR.org). There was no correlation of the excess heat to recombination in any of these three cells.

[This refers to http://www.'the Wiki banned site'.org/acrobat/MilesMnedofinalr.pdf to which I should add page numbers. - JR]

I hope this helps to discredit Shanahan's false statements about my experiments as well as his statements about cold fusion in general.

Mel"

where Miles does not mention measuring entrainment, and in fact seems to think the idea silly. Which is why I posted a follow-up message to that post, with special reference to the sentence:

"Horace rightfully mentioned electrolyte droplet entrainment, I have witnessed the reality of this phenomenon myself (visible traces of salt remained after letting dry a spoon exposed to the outgoing vapor). "

where Michel Jullian remarks on his direct observation of entrainment. So, since entrainment is a real process, and since it seems to have been observed, it should be considered, but it never is, therefore the water loss measurements of Miles are non-conclusive.

With regards to my visiting them, why would I spend the money and the time to go visit a lab of people who denigrate my work in print and in email without ever discussing _anything_ about it with me first? They have already set the pattern, I'll just follow it. Kirk shanahan (talk) 12:57, 27 January 2009 (UTC)

I am certain that the tit-for-tat strategy is incompatible with effective scientific communication. GetLinkPrimitiveParams (talk) 04:05, 1 February 2009 (UTC)