FacebookTwitterDiggStumbleuponGoogle BookmarksRedditTechnoratiLinkedin

Stainless Steel and Glass Dual Walled Cells

Keep up to date with developments relating to the Celani inspired Stainless Steel and Glass Dual walled cells developed with support from donations and team resources.

LIVE DATA | Experiment Log | Latest Video | RSS feed of blog posts | RSS feed of all comments

Steel & Glass Cells : Preliminary Test Findings [UPDATE #1 - Wire capability?]

on .

First, can we say a HUGE thankyou to you guys for making this cell construction possible. Your donations have made a big contribution to the costs of developing this flexible cell that has great potential. Nicolas Chauvin has worked extremely hard to put this together with input from other members of the team particularly Mathieu and Ryan. Also the instrumentation and control package was put together by the US team and will also be funded by your donations. THANKS - FOR MAKING THIS POSSIBLE!

The purpose of the stainless steel and glass dual envelope cell design is to improve the output heat measurement of Celani’s cell design by capturing as close to all the heat coming from the cell.

The internals of the cell is similar to Celani design with two wires winded around mica support. The glass tube is also made of borosilicate and can support up to 8 bars at 350°C.

The main difference is the use of a single ended tube to reduce the leakage of hydrogen or other choices of inner cell atmospheric composition.

The tube is sealed onto a stainless steel flange with special Teflon gasket with a silicon core.

Between the glass tube and the stainless steel envelope, it is also possible to choose the gas composition and pressure. This allows fine tuning the heat flux conductivity between the cell and the outside environment. With 0.15 bar of helium in this middle chamber and 80 Watts input, the cell temperature can go up to 350°C when the cell is immersed into ambient temperature water. So far, in preliminary tests, the pressure inside the glass tube and in the middle chamber have been very stable with very limited leakage even when using hydrogen and helium.

With this configuration most of the heat flux going outside the cell is from radiation through the glass onto the stainless steel outer envelope. Since the cell is almost completely immersed in water, the heat is accumulated into the water.

To get a good feedback of radiated heat, 3 thermocouples have been placed on copper rings around the glass tubes. These rings act as heat integrator to avoid measuring temperature at a very specific point on the glass surface.

Finally, we measure the evolution of water temperature in large volume baths (big water buts) to get a good idea of the output heat.

To get a precise measurement of the excess power, we have built two cells that have the exact same construction. Our preliminary test has been configured with one passive cell which is used as a control cell. It is loaded with 2 Nichrome (NiCr 80:20) wires. The second cell is active and is loaded with 2 Celani Constantan wires (low number of layers). In order to avoid the need for cell calibration and modelling of temperature to output power, we have chosen to manually fine tune of the input power of the passive cell in order to match the water temperature in the 2 setups (active & passive).

In our preliminary experiment, after successfully loading the Celani wires with Hydrogen, we have been applying 84.8 Watts input power in the passive cell and 80.0 Watts in the active cell to maintain very similar water temperatures (± 0.1 °C) in the same large volume of water.

This early test suggests that it takes 6% more power to maintain the same volume of water at the same temperature in the control cell when compared to the active cell. This is also the first experiment where we have two lengths of Celani wire ( one 14L one 16L and around 550mg of wire) in the same active cell.

When applying the same 80 Watts in both cells, we could measure about 1.5°C temperature difference in favor of the active cell. That is to say, 80W raised the water temperature from 20 to 30 degrees, but in the active cell, it was 1.5°C higher so the gain is of the same order as the preliminary findings where the same temperature is maintained by a lower power in the active cell's bath.

The evaporation has shown to be practically identical for both cells from. As we move forward we want to ensure high accuracy on the amount of water in each of the baths and to ensure they are carefully mixed. A first check will be to swap the cells into the opposite water bath.

We hope to get the live Feed up and look out for up and coming video!

As ever - add your comments and suggestions below.

UPDATE #1 - Wire capability

Celani called today to say how positive he was about these preliminary results, it was suggested that we have some things to tidy up with the design of the experiment but that it was encouraging. He then went onto say that the reason was that we are using the least capable wires, so we asked for an explanation.

Basically, the 2L wires are better than the 14/16L wires because of the positive effect of the nano-diamandoids. The process used to add the extra layers destroys these diamondoids and the extra layers do not benefit sufficiently until the number of layers are far higher. In terms of Mechanical stability, the 2L is the best and the 700L is extremely fragile. However, the order of claimed excess power potential superiority of wires is like this, from the top down:-


So.... in theory, the only way is up!

Add comment

Security code


0 #16 Ecco 2013-05-12 12:37
This page uses imperial units but gives a rough idea of what to expect with heat losses from an open water container:

www.engineeringtoolbox.com/... /
0 #15 Ecco 2013-05-12 12:29
If the aim is calculating energy through the rise in water temperature over time, it's important that the container is well insulated. An ideally insulated water container (= no heat loss) would make it possible to calculate output energy just by knowing the amount of water and its temperature over time.

As far as I understand heat loss from a real world open water tank occurs mainly through evaporation and radiation from the water surface, and the higher the temperature is, the worse this problem gets. So it's important that the container is closed and of course also well insulated.

So, what you want is an insulated tank/barrel with a closed lid, containing a relatively large amount of water (50-100 liters) and a relatively large goal temperature difference from ambient (for example 25-30 C), but not too large as to avoid having to deal with excessive evaporation losses.

This would be a slow, but pretty "safe" experiment to perform.
0 #14 bob 2013-05-03 01:45
I wouldn't get diverted to non steady state experiments with this apparatus. In a steady state where mixed water bath is at the same and constant temperature, your heat balance equations simplify significantly to something like:

electrical energy in (passive) = electrical energy in (active) + heat generated (active)

The large thermal inertia of water bath nulls out the effect of transient air drafts. Would be better to increase the water contact with the lid surface, but one can make a convincing argument that heat losses here are similar in active and passive containers at steady state (as determined by a constant and equal water bath temperature in active and passive cells) and thus cancel each other out.
0 #13 Ryan Hunt 2013-05-02 16:48
Lots of good suggestions in these comments for improving this experiment. Little details in this apparatus such as water interaction with the steel top where most of the heat flows, or in the stratification of the water temperature, or in drafts in the room cooling the cells differently will make big differences.
0 #12 bob 2013-05-02 12:24
If you insulate the water buckets you will cause the water bath to reach higher temperatures before heat losses balance again. I don't see a gain in doing that. However what about adding a thermoelectric cooler to an insulated water bath system. That way you could cool to a fixed temperature and use the electrical power to the cooler to calculate heat flux out of your system.
0 #11 Robert Greenyer 2013-04-29 20:37
@Edwin Pell

Thanks - and yes... submerge them! We are cautious in operation in case we break anything!

We to look forward to taking several experimental threads forward.

Please do not hesitate to offer up any ideas as to how we might improve this type of cell.
0 #10 Robert Greenyer 2013-04-29 20:35

We are right on it, though, purely because of the physical location of wires - this type of experiment may first occur in the Concentric Calorimeter ... and VERY soon!
+1 #9 Edwin Pell 2013-04-29 19:42
Beautiful experimental hardware. Why not submerge the top of the top flange? If it is vacuum tight it is water tight.

I find your result compelling with the setup as is. Adding the stir is nice. Looking forward to many more results with many variables studied across their range. Thank you.

+3 #8 Ecco 2013-04-29 19:38
@Robert Greenyer: about update #1, in my opinion the answer to this issue can be as simple as using more 2L active wire inside the cell. A good start would be quadrupling the amount ie using 2 grams of it. Establish that there is an effect first, optimize it later.
+1 #7 Robert Greenyer 2013-04-29 19:17


Thermally insulating the water buckets is trivial and a good addition to the rigour of the experiment design.

In terms of excess energy generated, Nicolas has said that the preliminary runs that have been going for a number of weeks have effectively generated an excess power that is equivalent to the storage capacity of a couple of standard car batteries in Watt hours.

It will be much better to evaluate this when we do a complete run, learning from all the comments posted and suggestions from collaborating scientists in a second run with full data access. When this data is available we will likely take the formulae you detail and hard code it into a live google document so people can track the progress of the cell.
0 #6 Mika 2013-04-29 18:31
Hi, my first post here!

The comparison between active and passive cells are a good start, but does not provide enough quantitative data about the energy production.

If you could provide verified data about the mass of water and the heating time while heating with 80W, we could calculate the true amount of energy that went into the water.
E=m * deltaT * HC
where deltaT is the temperature difference
and HC is the heat capacity (~ 4.181 J/(g*K) @ 25°C)

If you get the resulting E above the energy input (80W*t), we have proof of energy production together with clear quantitative data!

The container has to be as insulated as possible and we need a water body where side effects from the instrumentation are negligible. I think 1 m³ of water would do the job best. Target deltaT of ~10° is also good, since the HC changes a bit.

Greetings, Mika
0 #5 Robert Greenyer 2013-04-29 17:56

Nicolas is going to introduce a stirrer (with a motor outside the bath to remove any measurable energy input). A prop shaft will drive a propeller in the bath.

The low pressure in the outer cavity between the glass and the steel is to insulate the main length of the cell so that lower power in achieves a higher wire temperature.

I agree that it is likely for heat to be lost from the top of the cell and given this is easy to address by insulating the top - we could see even higher favourable disparity between the active and control cells after insulation has been applied (well, assuming that it isn't a heat loss artefact anyhow!)
0 #4 Ecco 2013-04-29 14:31
Just an observation - if you want heat to be transferred to the external steel tube wouldn't it be better to increase the gas pressure in them to improve heat convection, or better yet, using a liquid (although this might bring problems with high temperatures).

I imagine that much of the heat if being transferred by conduction to the exposed flanges (and thus unaccounted for, lost), at the moment.

Of course, submerging the cells entirely and insulating the water containers would solve most problems/discus sion points about this.
0 #3 Robert Greenyer 2013-04-29 10:17

Hi Ecco, thanks for the link to the RSS - I have added it above.

I propose that the top of the cells are insulated fully to force all thermal output into the water and avoid questions of variance due to this point. A thermal camera used at this point will provide a good indication as to the need for this.

The cells are in their steel tubes which also has a controlled environment composition of 0.15 bar of helium as indicated in the blog post.

The water buckets have been placed to give them as near exact environment as possible. Switching the cells over will be a good metric to add weight to the preliminary findings.

This cell provides the opportunity to wrap a heat exchanger on the outer steel jacket and insulate the whole lot, thus making a good fluid based mass flow calorimeter.
+1 #2 Ecco 2013-04-29 10:03
I think it might be useful to use a thermal camera to check for heat loss differences between both cells, if there are hot spots, etc.
0 #1 Ecco 2013-04-29 09:18
Swapping both cells and making sure that the water is well mixed is a good idea. I suggest checking with a dummy load if the water buckets' position in the testing room affects in any significant way their heat loss. Also make sure that the water temperature sensors don't "see" the glass tube (no need if the cells are being submerged in water with the external steel tube. This isn't clear from the photos).

This kind of set up inspires way more confidence that the results are real, but since the preliminarily measured excess heat is still relatively small, extra care has to be taken in order to reduce the chance of errors. The good news is that this shouldn't be as complex as with previous cells.

By the way, I got here through the most recent posting on E-CatWorld. The Steel&Glass cells section on quantumheat.org has a different RSS feed, so it didn't appear on mine:


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