FacebookTwitterDiggStumbleuponGoogle BookmarksRedditTechnoratiLinkedin

Welcome

The Martin Fleischmann Memorial Project is a group dedicated to researching Low Energy Nuclear Reactions (often referred to as LENR) while sharing all procedures, data, and results openly online. We rely on comments from online contributors to aid us in developing our experiments and contemplating the results. We invite everyone to participate in our discussions, which take place in the comments of our experiment posts. These links can be seen along the right-hand side of this page. Please browse around and give us your feedback. We look forward to seeing you around Quantum Heat.

Join us and become part of the project. Become one of the active commenters, who question our work and suggest next steps.

There is a great opportunity to learn everything about the project and actively help it with every day progress. The great feedback from some of our volunteers will tell you more. Now we search for someone to come to HUG, Minnesota to spark the Sparky cell and see with own eyes whatever exciting will happen.

We are in a phase where we are developing and testing our methods before we test the next active Celani Wire.  Per some excellent suggestions from our followers we are trying several new little experiments while we work towards some calorimetry also.

Stepping up and down

First, we are testing a new way of doing the calibration where we step the power up and then back down again.  This is simple and easy to do and will help determine if the cell was at adequate equilibrium for the calibration test.  It was suggested by Ed Storms.  The results look like this:

When we plotted the upgoing curve vs the downgoing curve and found them to overlap pretty closely.  To see the difference, we subtracted one set of points from the other, and got the following graph.

Here is the resulting data set in an xls format stored in google drive:  https://docs.google.com/open?id=0ByUU8sU37DYsbVRLRnFyQ2JMODg 

Better thermal connection to the glass:

Second, we are trying to enhance the validity of the temperature measurements.  Towards this end, we took a small piece of aluminum (roughly 12mm X 36mm X 2.0mm), rolled it to match the glass, and carved a groove for the thermocouple tip to fit into. 

Then we installed it on the glass with a little thermal grease under it and some some kapton tape over it.  That should make some good thermal contact.  Too bad we can't do that inside the glass, too.

Third, more temperatures sensors to explore more about the ambient

Then we also added more temperature sensors.  Due to the limitations of the number of columns in the instrumentation software and the fact that we don't know if we'll keep these sensors, we put them in a different test.  That test should be available in the data viewer today.  

In Summary:

  • T_A1 - Left side of air stream in vent hood
  • T_A2 Middle of air stream in vent hood
  • T_A3 - Right side of air stream in vent hood
  • T_G1 - Exterior of glass on the left side (plumbing side)
  • T_G2 - Exterior of glass in the middle
  • T_G3 - Exterior of glass on the right side
  • T_F1 - Middle of the flange on the plumbing side
  • T_F2 - Middle of the flange on right side

Look for the range of variation in each part of the air stream.  The sensors may have slight offsets.  I guess we'll see.

Look for the variation of the temperatures on the glass and the range of noise on each one.  Compare that to T_glassout which is now under a piece of aluminum with thermal grease between it and the glass.  

Look for the slope of the temperatures on the flanges when the glass looks constant - expect different time constants


 

Results of Pressure Sweeps at a (mostly) Constant Power

The violet line was a sweep done with the new borosillicate cell with Macor ceramic wire supports.  The orange line and light blue line below are with the insulation at the ends of the cell.  The orange line is at 48 W and the light blue line is 47W.   It showed a tiny amount of improvement in the slope, but definitely shows a trend for the temperature to rise as the pressure decreases.  There are still some differences in the geometry between our cell and Celani's but I was expecting the insulation we put in to make more of a difference than it did.  I don't really know why the T_glassout sensor seemed to read lower after the insulation was in place, though.  

When we divide by the input power, we get a slightly smoother shape.  It shows the two runs after the insulation was in place to behave very similarly, but shows the other run in violet to be of a significantly different baseline.

Next steps:

We have been advised to try to mount the cell in a vertical orientation in order to facilitate more consistent convection currents on the inside of the glass.  We are going to try to figure out how to do that next. 

Update #1:

 

From the Euro Team:  As the team in Switzerland wait for the Celani wire in the EU cell to de-load, we see Nicolas working on a Mizuno type cell and Mathieu talking to Dr. Edmond Storms who is offering his assistance to the project.

 

Update #2:

With our Euro Cell, we are trying to unload hydrogen from Celani's wire as much as possible.
Unfortunately, we have broken the Nichrome wire 2 days ago.
So we cannot heat the wire indirectly, but rather by applying current into it while pumping the cell to vacuum.
As we do not want to put too much current in the wire, we wrapped a copper sheet around the cell to maximize the internal temperature.

Cell covered with copper

Obviously, in these conditions we are far from the calibration baseline.
Therefor all excess heat data that you might have seen during the last 2 days are not correct.

 

Comments   

 
+1 #29 Chuck 2012-12-21 18:53
Good idea with the copper--it will certain even out the spot temperature differences. Do you also plan on placing the rig in an insulating box? While it's true that IR won't pass through solid copper, it will heat the copper which will then re-radiate in the IR spectrum.

I think what's really needed is a way to completely thermally isolate the cell from external temperature influences.
Quote
 
 
+1 #28 Ged 2012-12-21 16:23
@123star,

I concur. I found some silver wire for K type thermocouplers that was only $13 on Amazon, so seems price isn't too bad. Also found this information on silver thermocouplers, http://www.isotech.co.uk/files/document_library_file-51.pdf where it could measure upwards of 970 C with an accuracy around 1 to 4 mK. I think that's rather acceptable for our purposes if they decide to go that route.

Though, if they stick with the copper wrap, then we should be good with none of those worries with our current ones.
Quote
 
 
+1 #27 123star 2012-12-21 12:51
I found a picture that shows the reflectance of Nickel. I haven't found a good reference for the reflectance of NiCr alloys yet (the meaterial of which a thermocouple type K is made), but this graph strongly supports the idea that a naked (i.e. without the aluminum patch) type k thermocouple may be sensitive to the temperature of the wire.


www.freepatentsonline.com/6554854.html

I think polished gold or silver are the best choices for our glass cell.
Quote
 
 
+2 #26 Ron B 2012-12-21 04:47
On 12/14/2012 11:50:00 The USA Cell 1 seemed to have a jump in pressure from 0 to 12.5 Bars. I didn't find anything in the USA log for that time. According to the sensors at that time, the mica temp jumped to over 500C along with the ambient yet the board temp fell to zero.This anomaly lasted several minutes on a 30 Sec sample resolution.

At 11:49 Power Blue was set from 0 to 125MW
At 11:50 Resistance dropped from 18Ohms to -3200 ??
At 11:56 Pressure raised from 0 to 150 and t-mica jumps to over 500C.



I'm having a hard time trying to figure out what this is telling me. Did the mica really see over 500C for several minutes?
Quote
 
 
+2 #25 Ged 2012-12-21 01:52
@123star,

Great find! I had been digging everywhere, but only found a graph for copper metal that went as long as the 800 nm range. That graph did seem a little different, as it showed copper as having an absorption peak around 800 nm, but who knows.

Either way, you're absolutely right that there will be no other way for heat to escape, so it won't matter (IR, like most of the electromagnetic spectrum, cannot pass through metal). This is also why the copper wrap will help the cell to be warmer per input power, which is also something we very much want.

We definitely want that copper wrap around the cell, in my opinion; much for much cleaner, more consistent, and far simpler data.
Quote
 
 
0 #24 123star 2012-12-21 01:09
@Ged
I think that in the case of complete metal wrapping any metal is good (because heat "has no other way to escape", I can't find a good way to explain that now), that's not the case if we make a small probe which is touching a piece of metal.
I found a reflectance a curve for copper by the way (even if it is about "metal films" I think it's ok anyway)
www.tvu.com/PNextGenTFWeb.html.
I'm still looking for a reflectance chart for NiCr 90% 10% (the meaterial of which a thermocouple type K is made).

@Alan G
Low reflectance for the probe is perhaps desirable, but what we need is constant reflectance all along the IR range. That way we are sure that the power absorbed is independent from the wire temperature.
Silver seems a good choice.
Quote
 
 
0 #23 AlanG 2012-12-20 22:50
Hard black anodize would make the aluminum piece a pretty good IR absorber, less than 10% reflectance up to 10K nm
See figure 3 at http://www.noao.edu/ets/gnirs/SDN0003-26.htm

Because it conducts heat far better than the glass, a larger piece of aluminum carefully fitted over T_GlassOut would reduce the possible error from turbulent convection inside the cell, and if black anodized it would also eliminate the unknown energy leak of IR radiation at the point of measurement.

123star wrote:
I noticed that aluminum absorbs shorter wavelengths better, I don't know if this effect is negligible or not. What is your opinion? Remember that the glass is a poor heat conductor after all. (to read the graph, remember that IR spectrum starts from 750 nm)
Quote
 
 
+1 #22 Dieter Seeliger 2012-12-20 18:54
@Ryan,
thanks for the clarification about this. At the first look, I did not found a deeper description in your files and the connection cables could be made of copper too. I pointed you to this, because I do not want that anyone could critizese your results
picking up this point. And I want this wonderful project to be a 100% success.
23 years of critizism and suppression is enough !
BR Dieter
Quote
 
 
0 #21 Ged 2012-12-20 18:27
@123star,

Oh, do you have a graph like that for copper? I want to see if we can convince the MFMP team to keep that copper wrap around the cell!
Quote
 
 
0 #20 Robert Greenyer 2012-12-20 18:26
@Ged

That is an option - we are going to have to re-build this cell as the NiChrome wire is broken. That means we will have to re-calibrate and so this all becomes a possibility.
Quote
 

Add comment


Security code
Refresh

Here is your generous contributions so far towards our $500,000 target, thanks everyone! : $40,780   Please Donate
See the current state of our booked costs here

MFMP Facebook Feed

Martin Fleischmann Memorial Project's Facebook Notifications

Martin Fleischmann Memorial Project's Facebook Notifications