Glass / Steel hybrid cell and other progress [UPDATE#10 - Pressure testing going well]
We are progressing on many fronts.
- Julian gave notice that Community Interest Company status for Quantum Heat has now been confirmed by Companies house in the UK. This means we are now in the right legal structure to launch our Kickstarter. This is a not for profit entity, the directors can not receive a dividend and if there is a problem the assets are legally obligated to be donated to the international charity "Save the Children".
- Celani confirmed that he is finishing the wires 25/02/13 we need for the US and EU Dual V2 protocol experiments in the V1.3 cells.
- Mathieu was baking all the Mica 25/02/13 and was testing the Macor pressure cut end pieces for the V1.3 cells.
- Ryan is working on the steel cell and a new calorimeter - more on that soon.
- Recent donors can expect to start receiving their draw sequence numbers soon.
Nicolas has been assembling the components from Ryan and the engineers in switzerland for the steel and glass hybrid cells. He says they will be put together 26/02/13 after which the cells will be run through a number of weeks of pressure and other testing and tweaks. Here are a few photos for your to feast your eyes on.
 |
 |
 |
 |
 |
 |
 |
 |
UPDATE#1 - Celani has shipped the 400+ layer wires!
We have a tracking number and they are on their way to France. Some will need to be sent to US along with the Mica and Macor parts Mathieu has made and there are a few components to ship from US to EU, but it is all coming together.
UPDATE#2 - Strange things happening
Whilst we wait for the delivery of the 400+ layer wires and other elements to enable the construction of the cross atlantic dual duelling differential tests, we are going to periodically be maintaining the pressure of the existing two US cells in their calibration range. It is now the second time we have done this and we are seeing some interesting results.
In the graphs below, you can see the Pin and the Pout based on Central European Time. In these graphs - the pressure was reset just after 20:00 hours on the 26/02/13, we would encourage you to look at the Pxs and Pressure graphs over the same kind of time period and think about what might be happening. In both cases the cells produced a transient dip before getting on with their usual cyclical variance. It is too early to call and a few more re-pressurisations and patience will tell us more - but the trend seams to be positive.
Of course, when we do have two identical cells running side by side, with no gas issues and only one loaded, things will be easier to rationalise.
UPDATE#3 - Surprise wire
The wires have arrived - and there are 4 X 400+ layer wires in the package - enough to perform our dual duelling differential tests... however... Celani gave us all a present...
One wire is 720 layers
Now we REALLY want to test this wire... However, we were intending to have 400+ layer wires in each of the cells with the difference in layers between a cell pairs being no more than 10%.
Since we are running one cell loaded and one cell unloaded as a control - for that part of the v2.0 protocol - it actually matters little what wire the control cell is running, just that it is running in every other way the same. The purpose of having the control cell containing a Celani wire is that it can be loaded later down the line and hopefully show Pxs so adding weight to the evidence.
So the question is, should we say run the initially active cell in the EU with a 720L and the other with a 400+L wire.
You opinion is valued
UPDATE#4 - Heat flux?
We want the best determination of energy output in calibration and active runs right? We don't want to have to think about moving TCs, TC contact issues, temperature gradient effects, stability over temperature range. We also want to know that the TC is accurately representing the energy averaged over a decent sample area without any worries about Infra Red (IR) thermalisation clouding the debate.
In our coming V2.0 Celani protocol experiments, we don't have to worry about convective and gas conduction effects. The active and calibration runs are both reliant on IR output from the wire only. This makes thermalisation of IR even more important for the experiments accuracy. So what to do?
Well as mentioned previously a few weeks back, a site follower called "bobicanprogram" did a bit of web searching and came up with the following device family that seams to solve many of our outstanding issues and is particularly suited to the new protocol. You can see information about it here:
RdF Heat Flux Sensors, for the precise measurementof heat loss or gain through materials.
But there is a catch - given that they have the patent on this tech - it is 2 orders of magnitude more expensive that our normal themocouples for the ones we need that operate up to 260 degrees centigrade.
Here is their cost sheet.
So should we go for them? We need the 20457-TC variants so it will cost $2960 for just 4! Should we just get 2 and run them on one dual cell set-up? Then we can have a comparison with normal TCs to assess their added value and cross reference measurements.
We could have them in 1 week and they would then be part of these key experiments.
What do you think? Should we go for them?
UPDATE#5 - PXs vs Pressure
The cells are continuing to produce some really great data that will help us to understand things in the fullness of time. We are now going to periodically re-set the pressures in the two cells and in the graphs below, you can see the cells response to these events. As we do more and more cycles like this, we might even be able to draw some conclusions. Let's see what happens.
UPDATE#6 - You win some...
So Mathieu goes to the lab to retrofit the first EU cell for the V2.0 Celani protocol. When he arrives, he finds out he has not got his wallet with him, so he will not be able to buy the syringes from the Pharmacy to glue in the pass-throughs! Doh!
Anyhow, he calls a friend to borrow the money which goes well and he finds himself at the local pharmacy. There he is confronted with the Spanish inquisition about why he wants them, because, like, he has a beard and so is clearly a user! Not offended by the line of questioning, Mathieu explains all about the collaborative project, its purpose, hope and aims. The pharmacist was so inspired by the endeavour, she said it fully made her day and gave the syringes for free! Big win!
Then... following clear instructions. Mathieu found out it was not always easy to handle rapidly polymerising resin. In his own words...
"Gluing the passthrough did not go well.
The epoxy didn't get into the tubing on both flanges, it stayed on top of the ceramics. My heat gun worked at 350C, so I kept it at good distance. I don't know what I did wrong!
One the second time, my syringe spilled glue on the flange, that's my bad. I attached the picture.
Connections are good. It should be able to maintain pressure/vaccum normally
Do you think I should redo this?"
Life is not always plain sailing when you are a doer!
UPDATE#7 - Did we ignite?...
In the past six and a half days we have seen a few bursts of PXs and re-pressured the cells twice, as of 2 - 3 days ago, both of the cells appear to have switched to a different mode and PXs is now trending upwards with a more focussed direction - we are about due for another re-pressurisation - let's hope it does not break this move higher.
UPDATE#8 - Coming together
Nicolas has been working hard on assembling the Glass/Steel hybrid cells ready for the critical pressure testing. Ryan and Nicolas will be working together to upgrade the software on the data acquisition boards and laptop.
 |
 |
 |
 |
 |
 |
UPDATE#9 - Getting connected
As you can see below - Nicolas has been busy hooking everything up! One down, one to go!
 |
 |
UPDATE#10 - Pressure testing going well
Nicolas reports that the pressure testing of the steel/glass cells is going well, this was a major area of concern that is melting away right now - way to go nico!
Comments
don`t know if this has something to do with the calculation.
There could be a defective contact in the measuring circuit, a defective soldering place on the PCB or something else.
These kind of changes of one value I recognized sometimes during long time setups with my own experiments.
Br Dieter
measuring artefact as this has been found several times during the last runs
Do you mean the artifact that comes with the resistance calculation or another one? It's just curious that it dropped for 12 hours and then popped back up after that.
There seems to be some relationship between pressure and Pxs although I've seen examples when the pressure changes are not consistent with changes in Pxs. I am starting to wonder if there's not a secondary effect with pressure changes that affect resistance/Pxs (such as shock to the micro-structure s)
take a look at the pressure and the Pxs. Since the ambient temperature is more or less constant for the last few days, the dependence of the pxs to the pressure could be seen pretty clear.
The sudden resistance drop at late 0703 could be simple explained by an measuring artefact as this has been found several times during the last runs.
I`m sorry, to say this, but in my opinion, we are riding a dead horse with these vertical cells...
But I hope to have much more stable results with the new reactor line and the upcoming new Celani protocol.
Cells have been performing well for a few days now.
Looks like we might have "switched off" the effect in Cell 1.0 with it bouncing around 0W PXs, but Cell 1.1 is holding steady near 3W and threatening to creep up again.
Should the experiments that capture all the energy out prove positive then other experimental findings will be easier to accept.
I understand that the Heat Flux sensors sold by Omega are Rdf ones, they sell a small selection of their product through third parties but not the high temperature ones.
@P.D.
Thanks for your comment.
Omega makes a heat flux transducer which costs $263 each and has a substantially higher sensitivity (I think) than the RDF variety. That's found here:
omega.com/.../HFS-3_HFS-4.html
Note the upper temperature limit is 150 degrees C. If this is not enough, you can contact them to see if they know about other heat flux transducers for higher temperatures.
If you run your experiment inside a Seebeck calorimeter "envelope", you may not need discrete heat flux sensors. The Seebeck is merely a collection of such sensors wrapping around the entire experiment.
Good luck with this. You're on the right track. You MUST account for all of the heat flow in the system, one way or another.
I was not clear. Not saying that the P_Xs is of chemical origin. I hope it is not. Not saying that the reaction NiO to NiH is exothermic.
My thought was that you certainly have a chemical process every time you start the cell which as it goes yields more NiH from NiO which coincides with the increase of P_Xs.
The same process could be responsible for the change of the resistance.
Good luck.
Can`t wait to see them on the bench
Please take a look at the following blog post that discusses the potential for chemical.
quantumheat.org/.../...
We are seeking a residual gas analyser to see the composition of the gas in the cell after an experimental run.
In any case, the Celani V2 protocol removes gas from the equation, so these questions disappear.
The Ni wire has been initially exposed to air and part of the NiH has been oxidized to NiO. After the replacement of the air with H2 the reversed process takes place. Reduction of the oxide to hydride. If the reaction takes place inside the NiH lattice we should have an increasing volume of NiH to facilitate the reaction. Or could be an oxide layer on the outside separating the bulk from the H2.
The same effect may be responsible for the change in the resistance if the NiO and NiH have different resistivity.
A product of the reduction is H2O. For the sake of the experiment it could be interesting to heat up the cell as much as possible for a period of let's say 24 hours constantly flowing some H2 in order to remove the water vapor taking that variable out of the equation.
I don't believe that hypothesis is right, or something to be worried about (though doubling pressure to see what happens would be a wonderful experiment).
This is because it is -pressure-, not number of atoms, that determines heat transfer rates in convective (non-mass transfer, like we have here) situations, as pressure is the number of atomic collisions occurring between the gas and its container. It is those collisions alone that can transfer convective energy. So if you increase pressure by increasing the moles, or by increasing the temperature, you should still have the same convective energy transfer rate whichever the situation for that same given end pressure. Unless there's an obscure equation on the matter, I think that's the right interpretation.
Thermal capacity on the other hand would be related to the moles of gas. However, the heat capacity of the hydrogen will only affect the rate of the temperature response, not its magnitude.
Not sure that you would have to double the pressure, but this idea would help address the gas density arguments without shutting down any reaction that might be occurring (which seams to take longer to get going than the time it takes for the gas to go below the calibration range).
@Rats
The active wire resistance going down is either a cooling or a 'loading' effect, the wire resistance increasing is either a heating or potentially the New Fire 'burning' the H2. In either case these variables are related and all are related obviously to PXs - they have to be.
What I make of the data is that when the trend of both wire resistances are increasing, we're going to have excess power.
Pressure changes seem to have an impact on the immediate value of the resistance but the trend over a week doesn't appear to change directions.
Ecco: To test this hypothesis, both cells would have be fully cooled before every hydrogen refilling operation.
Would doubling the cell pressure at temperature provide information about the density vs excess heat correlation?
As per Ecco's post (#62) P_xs appears to have a strong correlation with resistance. If there is excess power then P_xs should not follow any resistance trends, should it?
1) Hydrogen leaks out of the cell, wire temperature naturally increases because of decreased convection towards the glass tube.
2) MFMP restores pressure to 1.5 bar without turning the cells off, but since the wire (and the mica which has some thermal inertia) has slowly increased temperature due to the small leak, it takes a slightly less amount of gas to reach 1.5 bar.
3) Over time, after repeating step 1-2 a few times, P_XS, based on internal temperatures, appears to be increasing because there's increasingly less hydrogen inside the cell. However if we were to base calculations on T_GlassOut temperatures, we would see that there would still be a net heat balance.
To test this hypothesis, both cells would have be fully cooled before every hydrogen refilling operation.
Another alternate hypothesis is that the linear scaling is due to temperature build up from the current reaction speed, and it is the temperature that is speeding up the reactions, and they stay at whatever speed point they got to within a certain temperature range (system memory, or strong autocorrelation ).
P_xs is definitely raising on my screens despite the re-pressurizati on. So, now we know the raise over time we've previously seen was -not- due to the loss in pressure, magically.
I guess we should leave these two running a day or two more, or until we are ready for the new wires and better protocol? This is getting very interesting to watch... keep up those repressurizatio ns, as they have removed the whole pressure change variable, and indeed it seems we may have ignition again!
Also, this suggests that these cells are slow to ignite (i.e. days). We might get a sense of the time scale now. Interesting to suppose why it takes awhile, and even more so why ignition times may be different between different experiments (e.g. does it ignite faster/slow/the same if the pressure is slowly dropping versus if it is repressurized periodically? So far ignition time seems to be independent of that variable, though).
RSS feed for comments to this post