FacebookTwitterDiggStumbleuponGoogle BookmarksRedditTechnoratiLinkedin

Calibration Data for Ox wire shows interesting behaviour (UPDATED)

Geschrieben von Mathieu Valat am .

I just finished collecting and stripping the data from the unfinished calibrations of Cell#2 showing the results with different gas and gas-mixtures with this oxide coated constantan wire.

Looking at the variation of the irradiated heat generated by the cell and pressure shows pretty convincing results:

We can clearly see that between He an H2 the difference exists, but lines are not really crossing, they look more parallel to me ; which is a very good sign that the heat irradiated by the cell is less influenced by the gas nature/compositions. I still have to make up my mind about the difference between H2-Ar gas mixture vs He at 3.5 bar. Also, it might be related to the variation of resistivity...

That is the fun part. We are almost convinced that resistivity is acting strangely for this wire but here we have the proof that the very first run changes the resistivity of the wire in a very different way that the rest of the runs.

This very first run influences the wire resistivity and changes it forever. This change occurs at a temperature above 230°C. I mention "above", it is because the temperature of the wire must be far greater than that.
I will record the changes in the active wire during the first run to enable a comparison.

The good thing is that the H2+Ar looks pretty stable -after the first run-, which is encouraging.

But that is not the most important point to make: The pressure and the nature of the gas changes the variation of resistivity inside the wire! Look at how every similar gas seems to follow the same variations and ends almost at the same point, above 250°C. Terrific!!! ;)

I still struggle to understand completely the reason why the resistivity with He @ 3.5bar is far above from the others. It might explain glimpses of what Ryan is seeing at higher pressures.

Feel free to share your thoughts!

EDIT

I baked the wire @ 1.5A up to 1.9A. I did this after the hydrogen calibrations where finished to release as much hydrogen from the wire if any before doing the helium ones. This may explain why the first He resistivity curve is different from the other He curves on the above graph. The micro-structure has been set to something else during this baking.

As a matter of fact the maximum resistivity drop are smaller in an order of magnitude from what we say with the active wire. Let me know if you think that hydrogen got in and if this calibrations are good enough to go on with the active wire. Thanks!

I have added what should be the future calibration plot that present how the thermal output energy is as a function of the power input. Still the very first calibration is way off the rest of the measurements, I actually want to redo it again. It sounds to me that the gap we have between the different lines is dependent to the variation of T_ambient throughout the measurements. We went through some deep cold in Europe and some of these calibration days where very much. I am still trying to make my lab room better regarding these variations.

I will plot the Stefan Boltzmann calculation soon and make public the master file after I have the final results of the current calibration.

Thank you for reading. ;)

Add comment


Security code
Refresh

Comments   

 
0 #40 Eric Walker 2012-12-10 05:27
@Ecco, several of the commenters (myself included) are from vortex. This experiment is being actively followed.
Quote
 
 
0 #39 David Roberson 2012-12-09 05:37
Quoting Paul Hunt:


One nice thing about a calorimeter is that it allows us to wrap the cell in heat reflectors or insulation and still measure the heat accurately.


It is good that you can capture most of the heat and direct it to the moving air. My concern is that it might be difficult to establish the correct amount of air flow. Too much and the outside glass will cool down which will then reduce the temperature of the hydrogen.

You might actually get valuable information by adjusting the outside glass temperature and determining how that effects the process. It seems that most times when new instrumentation is available, new discoveries are made.
Quote
 
 
0 #38 Paul Hunt 2012-12-09 04:23
[quote name="David Roberson"]Have you decided upon an ideal T_GlassOut to use with the air flow calorimeter? I wanted to caution you that the air flow will most likely have a major effect upon the glass temperature as heat is extracted by convection. That temperature reduction will be reflected within the device. It has not been determined how important the hydrogen gas temperature is toward the LENR process.

If you plan to keep the air flow very low so that all the temperatures remain nearly identical to the recent tests then your equipment downstream of the device must withstand that high temperature. I am thinking about the baffle material and the heat sensor wires, etc.


One nice thing about a calorimeter is that it allows us to wrap the cell in heat reflectors or insulation and still measure the heat accurately.
Quote
 
 
0 #37 Ecco 2012-12-08 17:39
@Ged: for the sake of completeness have a look at this too re 1.16w excess with 5w input. Credits to the original poster on vortex-l:


i.imgur.com/L9CV7.png
Quote
 
 
0 #36 Ged 2012-12-08 17:33
@Ecco,

Oh wow, thanks for the graphs! The one in that PDF wasn't as clear, but those definitely clarify the behavior as power and temperature go up.

Interesting to see it's response is much more sensitive to temperature than it is to power.

At any rate, this suggests that even this small piece of a two layer wire still has plenty of headroom in these conditions. I really wonder where the max output point is.
Quote
 
 
0 #35 Ecco 2012-12-08 10:38
@Ged: I think somebody from the Vortex-l mailing list is watching these discussions. I was about to write a comment (which I quickly redacted since I wasn't very confident I was correct) on how excess heat appears to increase with heat, according to data in the pdf you linked, when this new chart, supposedly clarifying this, got posted there:


i.imgur.com/A0OBf.png

The other one from the same discussion, based on temperature data that can be seen on the left portion of slide 2:


i.imgur.com/pDJoY.png
Quote
 
 
+1 #34 Robert Greenyer 2012-12-08 09:41
@Rats

Yes we have known this for many weeks but we remain impartial and focussed.

@Ged

The point of multiple replication, beyond incontrovertibl e proof from exact same experiments is to then to be able to establish the parameters that make this baby tick!
Quote
 
 
0 #33 Ged 2012-12-08 08:57
Also interesting to notice from that link (unfortunately I don't know how to edit my own posts) and the data, is that the output excess seems to taper off as the input power increases, and is not strictly linear. This may imply there is a maximal rate the wire can work at (which is realistic), and any input after that point suffers from diminishing returns. However, it could be bimodal or more, too, where after a certain threshold it kicks back up again.

So many questions, so much to explore. It's very exciting. What all this implies for our test cells here I'm not sure yet (at what point do these full wires hit diminishing returns vs input power/temperatu re?) -- we'll have to discover that ourselves I think.
Quote
 
 
+1 #32 Ged 2012-12-08 08:53
@Rats,

Here's the link 22passi.it/.../...

It's from Celani's slides on the ST data, and indeed, the 1.16 W excess is when the input was at 5 W (second page). That is hugely important to know.
Quote
 
 
+1 #31 Rats 2012-12-08 08:26
Just read on e-cat world that the 1.16W achieved by ST Microelectronic s was with 5W input. This changes everything. That's 23% excess power!

Edited: I'm sure I read this somewhere. I can't find the reference. :-(
Quote
 
 
0 #30 Robert Greenyer 2012-12-08 08:20
@David

Thanks again for your hard work. Can you transfer the model to a Google Doc spreadsheet and make it open as well as provide a xls for download on Google Drive.

This will help others to see your work and look for possible improvements.
Quote
 
 
0 #29 123star 2012-12-08 07:51
What happened to Cell #1? Why the power regulator isn't working? Just curious.
Quote
 
 
0 #28 David Roberson 2012-12-08 06:12
I have been putting together a model of the system and believe that the latest one does a reasonable job of simulating the operation. I can input the Power drive to the device and obtain the curves of T_GlassOut and PowerOut versus time.

The non linear relationship between the two variables causes the rising edges of these waveforms to have different time constants than the falling edges. At first this seemed strange, but now my model reproduces the result.

The time constants are dependent upon the pressure to a degree, but can be determined quickly from the data you publish.

You are doing an excellent job of measuring the performance of the device which is demonstrated by my model.
Quote
 
 
0 #27 Robert Greenyer 2012-12-08 04:32
@Sanjeev and others

I understand that Mathieu is currently re-running the Calib-75H2-25Ar -P3.5 however the blog entry is not correctly showing this.

@Ron B

Thanks that looks interesting - is it applicable in H2?
Quote
 
 
0 #26 Ron B 2012-12-08 02:02 Quote
 
 
0 #25 Chuck 2012-12-07 23:28
Let's assume that the heating and treatment of the wire changes it, as evidenced by the change in resistance.

Is there any assurance that doing so changes it uniformly along its entire length? If not, how does one ascertain exactly where to take temperature measurements?
Quote
 
 
0 #24 Pierre 2012-12-07 21:01
Could this fist time effect be caused by passing the Curie point
of this wire?
Quote
 
 
0 #23 David Roberson 2012-12-07 18:29
Have you decided upon an ideal T_GlassOut to use with the air flow calorimeter? I wanted to caution you that the air flow will most likely have a major effect upon the glass temperature as heat is extracted by convection. That temperature reduction will be reflected within the device. It has not been determined how important the hydrogen gas temperature is toward the LENR process.

If you plan to keep the air flow very low so that all the temperatures remain nearly identical to the recent tests then your equipment downstream of the device must withstand that high temperature. I am thinking about the baffle material and the heat sensor wires, etc.

Perhaps you are delivering the ability to vary an additional parameter which is internal gass temperature. This might yield important new data.
Quote
 
 
0 #22 Ron B 2012-12-07 15:34
Are you guys recording data for the temperature of the wire when exposed to the hydrogen and the temperature of the hydrogen?

I ask this because of the research done at Univ of Missouri involving production of neutrons with hydrogen and metals.
Quote
 
 
0 #21 Gipsel 2012-12-07 13:23
Quoting Sanjeev:
I guess a wire not made of Ni would be a better control, such as steel or Al. I was wondering why Celani chose a wire more or less similar to active wire as a control.

One may think about replacing the supposedly inert NiCr heating wire with some sheathed heating element (like this ones: thermocoax.com/.../..., they can be tailor made with non-heating cold ends and such). It may serve two purposes: (i) increased protection of the heating element from reaction (formation metal hydrides, Ni is quite susceptible to this) and (ii) increase of the surface area and therefore lowering of the temperature of the heating element (at the same output power and the same gas and Constantan/Cela ni wire temperatures) for a reduction of the near infrared radiation emitted from the heating wire (for which the glass tube is transparent).
Quote
 

Here is your generous contributions so far towards our $500,000 target, thanks everyone! : $45,020   Please Donate
See the current state of our booked costs here