FacebookTwitterDiggStumbleuponGoogle BookmarksRedditTechnoratiLinkedin

US/EU Cell Calibration Results

Written by Ryan Hunt on .

Both the US and the EU cell have completed their calibration cycles. The protocol is summarized in this spreadsheet here: V2.0 Protocol Table

The resulting calibrations are summarized in this spreadsheet, NOTE: the wire characteristics including R0 (initial resistance value at room temp is on the second tab sheet: V2.0 Protocol Calibration

 

In summary, the US Cells should be able to detect an excess energy signature of 0.5 watts or more. The EU Cells should be within 0.25 watts.

The US cell was in step 5.c, where we just hold the power steady and watch for deviation over time. The HUG team has decided to abbreviate this step in order to expedite the real test.  In the last 48 hours, we have seen that the US Cell is sensitive to ambient temperature changes. While the measured ambient around the cells is supposed to be isolated by the protective shell and the heated air flow, it obviously is not working well enough. We will also do our best to control the room temp better, but it still means we will need to account for real lab temperature changes (as logged by the T_ambient on the CTC test). In my opinion. if we are seeing possible results that are just above the noise like that, anyway, it will be a null result.

The EU cells are also sensitive to ambient changes, the same way Celani's were.

We also see that the pressure sensor on the EU cell has something periodically adding an offset to it, but we do not have a clear understanding of why it is happening. It certainly appears to be a ground loop issue or some other wiring problem. That is unfortunate, but not critical to the experiment, so we will not be troubleshooting while the experiment continues.

 

Lessons from Mathieu’s loading will lead to deviations from the protocol for step 6 - the loading phases

Go to higher temp - just barely made it this time - maybe 35 watts to get to 200+C instead of 170. Matt’s wire loaded, but only at the very peak temperature the cell was able to achieve with the hydrogen in it. To give the protocol more robustness, we will up the power to 35 watts for loading in case the next wire requires higher temp to start absorbing.

Change the cycling, reloading time to(3 bars H2 and 35 watts) 6 hours hot, 1 hour cool, during that one hour we will measure the cold resistance and that would be when we vacuum out the hydrogen and any contaminants and replace it with fresh hydrogen.
Do the entire loading with totally passive heating, no active heating. Note, however, that we are running about 0.25W in the active wire to keep the resistance reading clean.

Increase the power and heat of loading because the black oxides are still visible near the mica supports where the wire is slightly cooler.

Add comment


Security code
Refresh

Comments   

 
0 #24 Ecco 2013-06-23 16:29
The resistance of the active wire on US cell A appears to be decreasing, however most of this might be accounted by the internal pressure decrease due to the leaking problem this cell is having.
Quote
 
 
0 #23 Mathieu Valat 2013-06-23 14:01
Ed.
I made the necessary changes on the protocol to fit between both versions (table and text).
I also inverted the step power after loading to have indirect heating then direct heating.
This way we put current into the Celani wire as late as possible.
Quote
 
 
0 #22 Robert Greenyer 2013-06-23 13:57
@Edwin Pell

It was correct on the document rather than the table. We think we might however start with passive heating and then move to active heating as this is the most cautious approach.

The thing about passive heating in the V2.0 Protocol, it will really just be IR as due to the dynamic vacuum there will be no convective heating. It will be interesting to see if we see anything in this mode.

Fortunately, we anticipated this during internal discussions and hence the calibrations for both passive and active wire cell performance.
Quote
 
 
0 #21 Edwin Pell 2013-06-23 02:45
Step 8 of the protocol says apply power to the NiCr wire. Should is say the NiCu wire?
Quote
 
 
0 #20 Ecco 2013-06-22 17:20
@Robert Greenyer: how was R0 measured?

With a 250mW measuring current under pressurized hydrogen atmosphere before loading started, R0 on the active wire in EU Cell A was ~18.45 Ohm. I think this value is more comparable to current conditions.
Quote
 
 
0 #19 Robert Greenyer 2013-06-22 17:04
NOTE: the wire characteristics including R0 (initial resistance value at room temp is on the second tab sheet)

for those that missed it
Quote
 
 
0 #18 Ecco 2013-06-22 08:48
If I remember correctly the bump in pressure after the first power cycle in the graph was due to input power being manually shortly disabled before being bumped up to 35W.
Quote
 
 
0 #17 Ron B 2013-06-22 03:12
Ecco,
I see what you mean, yea the second cycle didn't seem to get the same change in rate of absorption as the first one did.
The VERY first heat/pressure "on" really gets a drastic change.
Quote
 
 
0 #16 Ecco 2013-06-22 02:48
I meant total absorbed hydrogen and I was referring to the second power cycle in the graph without H2 flushing.
Quote
 
 
0 #15 Ron B 2013-06-22 02:43
Actually, it looks the same as before. There has been a more significant change in resistance fall rate when hydrogen got flushed (however power was slightly increased, and this might have been a significant factor. H2 diffusion in metals increases with pressure, heat and time after all. Increasing input power increases heat and pressure):

Do you mean the diffusion rate or total absorbed hydrogen?


The data would tend to indicate that the diffusion rate of hydrogen decreases with time if the resistance value is a function of total diffusion. ???

I
WRT the slope of resistance change before and after the first power/pressure cycle, they don't look the same to me. I think we must be looking at different things??
Quote
 
 
0 #14 Ryan Hunt 2013-06-21 21:59
One in in temperature and one is translated to equivalent uncertainty in watts.
Quote
 
 
0 #13 Edwin Pell 2013-06-21 21:56
You have two columns under CI 95% which is the +side and which is the -side?
Quote
 
 
0 #12 Robert Greenyer 2013-06-21 07:39
@Edwin Pell

We intend to stay true to the course. We may add or reduce extra loading cycles with potential Hydrogen re-charges to optimise as much as possible loading. The heating and cooling allows for cycle stress at the material structure level. We need to protect the wires and proceed with caution.

In the current cool down the resistance has reached 14.222 Ohms which is good, but their still seams to be more good loading to be had.

The takeaway from passive heating based loading of the wire is that it would be minimal input energy difference to load 1 wire as 100s in the same Hydrogen chamber. This is an important consideration for those that would later claim that only a better battery had been created.
Quote
 
 
0 #11 Edwin Pell 2013-06-21 02:48
I hope you stick with the protocol and do not apply power to the active wire for loading. Sure it is interesting but finishing a complete experiment first is also interesting.
Quote
 
 
0 #10 Ecco 2013-06-21 01:30
Actually, it looks the same as before. There has been a more significant change in resistance fall rate when hydrogen got flushed (however power was slightly increased, and this might have been a significant factor. H2 diffusion in metals increases with pressure, heat and time after all. Increasing input power increases heat and pressure):


i.imgur.com/mFCxTKx.png

Again, if the experimental protocol hasn't been set in stone, for the loading phase (since we don't care about calibrations at this stage) I would try increasing power on the active wire a bit - even just 5 watts should help - and 10+ watts on the heater wire, at the same time (if the increased power to the heater wire isn't somehow going to damage the active wire). Then perform again a similar power on-power off cycle to sample cold active wire resistance @ 250 mW at regular intervals just to check out how faster the wire loads this way.
Quote
 
 
0 #9 Ron B 2013-06-21 01:23
Ecco says: "Active wire resistance on the EU Cell A, after a short pause due to the temporary input power removal, resumed from almost the exact value it reached before power was removed.

Ecco, that's very interesting and also, look how fast the resistance is dropping after the power/pressure cycle, the rate of change difference from before the cycle is significant. : )

Mathieu had good luck on the older version of this cell. I hope he can repeat that with this better calorimetry.
Quote
 
 
0 #8 Ecco 2013-06-21 00:54
Active wire resistance on the EU Cell A, after a short pause due to the temporary input power removal, resumed from almost the exact value it reached before power was removed. This is the "memory effect" Celani also talked about in 2012.

I don't think that cycling power on and off alone is going to help much active wire loading, though.
Quote
 
 
0 #7 Ecco 2013-06-21 00:03
It appears that off-load active wire resistance on EU Cell A is significantly lower than pre-hydrogen flushing conditions. About 14.5 Ohm vs 15.1 Ohm.
Quote
 
 
0 #6 Ryan Hunt 2013-06-20 21:25
We will let it cycle a few times and then flush the hydrogen, I think. That way Matt can get some sleep.
Quote
 
 
+1 #5 Ecco 2013-06-20 21:11
Then it might an idea to add like 10W to the passive wire and 10W to the active wire to check if resistance decreases further/quicker with the higher temperature and some power to the active wire, but I guess you don't want to potentially screw up this experiment as you want safe results for ICCF18.

By the way, if I understand correctly, you are going to flush and refill hydrogen again next time the cell will be temporarily powered off, right? (that would be in about 2 hours, I think).
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