FacebookTwitterDiggStumbleuponGoogle BookmarksRedditTechnoratiLinkedin

Celani Replication

Keep up to date with our replication of experiments based on Celani's cell right here.
Please feel free to add your comments to the blog entries.

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

EU Differential cells running [UPDATE #8 - Video postulation on recent data]

am .

Mathieu Valat has re-started the differential Celani cells in France and they are currently undergoing Helium calibration. The log book for them under the second tab of the EURO LOGBOOK

The live data is under "Cell EU1.3: Protocol V1.3" on HUGNetLab


  1. The leak was fixed from the previous run
  2. The cells have far more accurate thermocouples with standardised bonding pads that will conduct heat and thermalise IR from the same area on both the active and control cells
  3. As noted before, the glass extrusions were specifically cut and aligned from the same tube which was demonstrated to cause a non linear offset between the cells results previously

After a Celani style "free air" replication to confirm/test previous results, the intention is to do a foil wrapped experiment despite the capability of the new thermocouples.

This is the first of our internal re-run experiments, we will keep you updated in this blog thread.

UPDATE #1 - Calibrations completed

There is satisfaction when you see such a predictable system response. Now is the time to have a real absolute measurement of the effect that Celani's wires are able to produce. The maximum temperature gap between the two cells is 5°C maximum and 4.2°C in average (please, correct me if you would). But at least it is measuring lower temperature for the active cell, Cell A.

Because there is a risk off trapped He in the wire nanostructures, I will briefly pull vacuum inside the cell while power is still on. Then hydrogen will be introduced, a final check for remaining leaks with the detector, then let it be and show us all what these wires are able to produce. Now that it is characterized, understood and predictible, it will be easy to swap from one wire to another to test performance of various samples.

Be prepared for some Live Open Science!

I hope you like the endeavor, at least I really liked it! I just hope we will have a realistic answer now.

UPDATE #2 - Picture of thermocouples

Here is an image [Fig 1] of the external thermocouple configuration on the cells. First the glass was cut from a single extrusion with alignment marks and then these new thermocouples were carefully placed in the same locations. They have a pre-made foot with adhesive, which is machine die cut to ensure equivalent exposed surface. It effectively presents a opaque square to IR and the acquired temperature readings from them have been cross checked with a laser thermometer.

Fig 1: Image showing the new machine made thermocouples with standardised beige adhesive patch


UPDATE #3 - Are we starting to see life?

In the screen captured chart attached, whilst the cell temperatures seem not to be materially affected by ambient changes and the passive cells temperatures seem solid - shortly after a change in resistance from drop (loading) to rise in the active wire, there is an increase in the active cells external temperature measurement and the differential closes from mean. There are what appears to be several events in Fig 2 below.

Fig 2: The active cells resistance is periodically departing from its downward 'loading' trend. Small temperature variations are present in active wire that are not seen in passive signal.

UPDATE #4 - Cross talk?

Mathieu has been on a much needed holiday for a few days now, and before he left he turned off both the air conditioning and the heater in the room and locked the lab door. The reason for turning of the HVAC (heating, ventilation and air conditioning) was that they were both independently controlled and both attempting to target a temperature.

For a few days now, there has only been atmospheric drivers to temperature variation, as a result, ambient data is smoother and more predictable. The remaining variations in temperature are driven by sun/cloud/wind/rain/day/night/atmospheric pressure/humidity etc. many of which are integrated by the thermal inertia of the building and room in which the experiment is located. In hindsight the HVAC was adding a lot of complexity to the temperature variation, and also positional variation which made it difficult to see pure signal data from the cells. 

With this in mind, Bob was looking at a 24 hour plot of the cells yesterday on a 15 minute average, when he noticed some interesting apparent features in the data which you can see in Fig 3.

Fig 3: Active and passive cell temperatures compared to ambient

Firstly, just before the beginning of the 23/05/14, the active cell started to get hotter. This clearly affected the ambient as the room was cooling down and shows as a bump in the data - but the size of this bump was just 0.05 ºC - over 1 hour after which the room temp continued down on its small day/night variation.

Since there is no direct thermal coupling (well, other than bounced IR and convection) between the active cell and the polystyrene ball insulated ambient sensor, Bob wondered if the active cell, with its large approx. 3ºC variation was affecting the passive cell, considering there is a direct open H2 gas channel linking them. In the vacuum protocol, this did not matter as vacuum is a very good insulator, but in this experiment, the active cell getting hotter would pump hotter air at higher pressure into the passive cell and also H2 is a very good conductor of heat.

It appears that in Fig 3, the increase in the active cell temperature is increasing the passive cell temp by a lower amount. The problem with this is it is flattening the differential. In Fig 4, a longer period of time is shown.

Fig 4: Active and passive cell temperatures compared to ambient over 52 hours.

Again, in Bob's opinion, the active cell seems to be influencing and leading the passive cell temperatures on the up tick, other opinions may vary. Incorporating the 1.5ºC ambient influence on both cells may shed further light on what is going on - but we probably need to do some tests to understand what, if any, cross talk is going on.

  1. One suggestion is to shut the gas bridge and turn off the power in the active cell and let the passive cell run through a day cycle to establish its long term average temperatures in relation to ambient variations.
  2. Another might be to shut off the power in the passive cell and leave the bridge open to see how much above ambient the active cell is able to influence passive cell
  3. A third approach would be to power both cells but with the bridge closed (there would be no pressure equalisation but the leak is small now)

Pressure pulsing/compensation

One interesting capability of this experiment configuration in light of this finding is that the passive cell could be used as a pressure pump by powering on and off its nichrome wires enabling easy changes from equilibrium state (with potential to trigger events) or gas pressure compensation for leaking gas.

Excess heat

Our best guess without detailed analysis, and accounting for this potential cross-talk, is that if this wire is producing excess heat, it is less than 3% so far at peak.

After running cross-talk tests, we can raise the input power to 50W - which would be at the limit of their capability as previously identified.

UPDATE #5 - Run up

There has been a little bit of a run up in cell temperatures over the past 12 hours with the active cell clearly above the passive as the ambient fell to its minimum.

Fig 5: Run up?

UPDATE #6 - Bridge closed

In Fig 6, on a multi-day window whilst Mathieu was on a break, one can see the active cell spending increasingly more time above the passive and at a more elevated differential.

Fig 5: Near 5 day plot where it appears the active cell was pushing through the passive

Despite this, Mathieu decided to start to test the possibility that there may be cross-talk muddying the differential as discussed in blog update 4. He chose option 3 ... "power both cells but with the bridge closed" to start with and at 9.30 UTC he closed the bridge and you can see the result in Figure 6 below.

Fig 6: Effect of closing the bridge

You can see the big oscillations in temperature at 9:30 UTC when Mathieu went into the lab, the passive cell rapidly stabilised but the active took a little longer, then, one hour after the action, the active cell broke upwards and, this time, there is no discernible effect on the passive cell.

As ever, we'd love to hear your thoughts on this.

UPDATE #7 - Above the line

So, since an hour ofter the bridge was closed, on a 15 minute average, the active cell sat clear above the passive cell for over a full day which you can see in Fig 7.

Fig 7: Temperature profile of the cells after closing of the bridge

When looking in more detail, there is a short period of time around 07:10 - 07:20 on 27th when the lines cross, we should note however that in calibration, the temperature of the cell that is now active, was well below the passive cell.

Do you think we should let this run for a little longer like this, or test the cross-talk hypothesis in other ways before upping the power?

UPDATE #8 - Video postulation on recent data

Embedded below are two videos where Bob Greenyer is discussing recent data, other opinions may vary, please make your case in the comments.


0 #114 Ron B 2014-06-18 15:09
Has there been any discussion about making S&G cell parts available for purchase? I'm interested in setting up some experiments of my own.
0 #113 Robert Greenyer 2014-06-16 09:35
@Ed Pell

Spot on.


We are still waiting for the fresh wires, so I think Mathieu is going to put in some other old ones, to fill time and generate some more data.
0 #112 Edwin Pell 2014-06-14 22:41
All, I think it is just like Mike McKubre describes PdD. There must be a minimum level of H loading AND there must be a flux of H.

A high loading will not work if it means the flux drops to zero. A low loading will not work even if the flux is high.

As to whether one can move the loading up and down within some critical range and keep the effect alive indefinitely is an open question for NiH, the answer is yes for PdD.
+1 #111 Robert Greenyer 2014-06-06 19:01

Great discussion.

For LENR a certain level of loading is required for LENR to take place - this is the same for PdD systems and the reason for the failed initial replications of P&F's Work, below a certain level of D saturation NO exess heat will ever be observed.

From Celani we understand that resistance drop is a proxy for LENR potential, below an un-quantified level there will be NO apparent excess heat no matter what environment or stimulation is applied. Additionally, hydrogen is disorbed over time depending on various variables like temperature of the wire, environment pressure etc. That will cause an active wire to become inactive over time anyhow.

Resistance can be monitored during active powering or a small monitoring current. Both have been done in the past. By running scores of experiments, we can build up a data set that allows us to understand what level of loading and unloading are critical and how also temp and direct applied current affects a wire.

There are several variables - one of which is no two wire is the same - this exacerbates the challenge.

Mathieu was concerned about the age of the wire, there is simply no understanding of how the wires change over time, but it is known that nano structures can merge together over time even at room temperature either by chemical or temperature means. Additionally - they were simply stored in a snap seal plastic bag - hardly cleanroom, with unknown impurities in the bag.

Mathieu's other big concern was that the calibration gas could have been adsorbed.

We are expecting fresh wires, additionally, we understand Celani is sending us some of his powdered accelerant that he cited use of in the 2014 colloquium.
0 #110 Kapytanhook 2014-06-06 17:31
@Ecco Cool, very interesting stuff :) Yeah it would be hard to test insulated if the effect is not very stable, it could snuff itself out.

Can't wait to see more testing.
+1 #109 Ecco 2014-06-06 16:50
@Kapytanhook: in Celani's presentations and papers on these experiments, the provided data show that excess heat gets produced even just by supplying indirect heat to the active wire. In the past months Ubaldo Mastromatteo from STMicroelectron ics has claimed to have replicated a version of his cell in a thermally very insulated variant which produced excess heat with only indirect heating, meaning no current passed through the active wire.

What I get from this information, whether true or not, is that these active Celani wires are supposed to produce excess heat just with indirect thermal triggering (upon hydrogen loading), which most likely has to be strong enough in order to work (ie room temperatures do not provide enough energy for this excess to happen at a detectable level).

As you write, a thermally well insulated setup made for indirect heating only would nicely and simply demonstrate the effect. The only caveat is that it might be more complex than it sounds. The powder cell experiment being performed in Minnesota shows that it isn't, if measurement and testing conditions consistency is required - or in other words: good calorimetry.
+1 #108 Kapytanhook 2014-06-06 16:30
@Ecco I never really got that dead wire thing, figured if was pretty unambiguous it was the electrons or / EM / EMF interacting with the hydrogen in the metal / on the metal surface.

If they worked without any current wouldn't they show signs of excess heat already at room temperature? if not wouldn't a great test be to heat it up, seal it airtight, put it in something really insulating like styrofoam, if any heat is generated and the insulation is good enough it should stay hot permanently.

Seems to me like a way better test then mucking around with all this electric guff.

Did Celani have any theories on this, does he assume it is heat or the electricity that is triggering this excess?
+1 #107 Ecco 2014-06-06 16:16
@Kapytanhook: actually, I was suggesting to make the active wires electrically "dead", not connected at all to the outside world.

Since they're supposed to generate excess energy even just with heat supplied by a different, inert wire, that should still work especially if their amount is increased.

Having many "dead" wires in a twisted bundle arrangement would allow to increase their total surface area inside the cell without performing significant modifications to it, further reducing testing time and cell building costs.
+2 #106 Kapytanhook 2014-06-06 16:04
@Ecco Ahh yes, with multiple wires it would be a bit pointless or very complex. I always assumed it was all about the surface of the wire, so instead of multiple wires id suggest just trying to get a longer bit to coil up. But multiple wires does make a lot more sense, to optimize surface area you'd want wires with the lowest radius possible. And in order no to evaporate those instantly you are going to need a lot in parallel so you can keep the amperage low on each wire. And yeah it would really average out wire inconsistencies . Thanks for making me see that.

If the effect is now 3% excess (big if) trying to get 10 times the wire surface in there might be the answer to getting 30% excess heat if it actually is a surface area problem.

It's nice to see your insightful comments on most posts, they are appreciated.
+2 #105 Ecco 2014-06-06 15:45
@Kapytanhook: what I'm saying is that measuring the electrical resistance of the active Celani wires can be useful, but in the end if the point of these experiments is measuring demonstrably significant amounts of excess heat, that should be given less priority than other, possibly more helpful things, like for example increasing their amount (and therefore increasing the chance that one or more *will* generate excess heat).

They could be arranged in a twisted bundle like this, in order to save space and keep things similar to the single active wire version:


No meaningful resistance reading would likely be possible with the active wires bundled together like this, so such measurements could be completely dropped, avoiding possible issues with the passthrough(s) needed to bring the them outside the cell.

Having more wire passthroughs means that potentially more issues with gas leaks especially when temperatures increase could arise, added costs, additional cell assembly time, etc. Therefore, the opposite would also be true: having only the heating wire to come out of the cell(s) would likely make things simpler and cheaper.

Of course, if the active Celani wires are not electrically connected to the power supply, no direct current heating or resistance measurement will be possible with them, I'm aware of that (duh!).

This being said, I guess MFMP want to first check out if the wires are able to produce excess heat with the cell in its original configuration. My opinion is that given that these wires are reported to perform very differently between each other, testing them just one a time might take long.

Add comment

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