Robert Horst Verified User
  • Male
  • Member since Mar 15th 2016
  • Last Activity:

Posts by Robert Horst

  • Unconventional electrolysis

    Quote from can

    I remember I tried that a good while ago but I recall I found it complex and not as "interactive" as the circuit simulator I've been using above, with the latter quality being a godsend when one only wants to explore the basic behavior of relatively simple circuits for educational purposes. I recognize that the former is more powerful, however. I have been using Linux for quite some time now, but it should be possible to install that here. Alternative should exist too in case that won't work, although in the worst case I could use a Windows virtual machine to run that.


    Before purchasing other equipment I could try a different power supply which I already planned using in more tests later today. It's of the same type, but a bit nicer than the one I've been using so far. It also supports up to 40A (combined) on its 12V rail. However in early testing several months ago I found it to be less tolerant to abuse/short circuits and would shut down often.


    Shorting is definitely a condition that will occur often in these experiments, both of transient (recoverable) and more serious (not easily recoverable) nature. Sometimes I've had that even interrupting power and restoring it immediately after caused large enough changes at the electrode interface as to cause a more serious short-circuit that requires clearing the electrodes manually. I'm aware that LiPo batteries can provide very fast rates of discharge but also that they can easily explode or catch fire if abused. I'd also need a dedicated charger.

    LTspice is a little harder to learn, but gives much better simulation of real components. That gets more important as you use MOSFETs and other active components,or need to simulate the core saturation of inductors. It also allows much more complexity in stimulating the circuit with different waveforms. I run it in a Windows10 VM on a Mac.


    Regarding your power source, for either a power supply or battery you may want to add a resettable fuse (also known as a PTC or positive thermal coefficient device). Here is one that might be a good choice:

    https://www.digikey.com/produc…500/RHEF1500HF-ND/5029799

    With this one, you can sustain 15 but it it can take 20 sec to trip at > 28A. Max resistance is just 9 mohm. It will not bother your circuit normally, but will prevent fires or damage during a dead short. You could pick one that trips at a lower current depending on the peak current your supply can deliver.

  • Unconventional electrolysis

    You may want to try LTspice which is a free, full function Spice simulator.

    https://www.analog.com/en/desi…UvlzR6RuKBdBoC0oEQAvD_BwE

    I have use both Mac and PC, versions, but found the PC version to work better.


    Regarding interference from your 12V power supply, I would suggest running the experiment with a battery to see if you get the same results. You could get a 3S (3-cell , 11.1V) Lithium Ion or Lithium Polymer of the type used for RC cars and planes. They will deliver 20A or more, and any AC would be due to your experiment and circuit alone. It would be a good idea to put a fuse in the circuit. You do not want to be near if the output leads are shorted.

  • FP's experiments discussion

    Thanks for spotting this. The method is OK. Near 100V the graph has a very fat line because temperature and voltage are right next to each other. One side of the fat line is about 1655 and the other side 1665. I probably guessed wrong on which half of the line was voltage, but that does not affect the main results. The main point of the graph was to figure out where the voltage drop had to be from something other than bubbles on the cathode, and that is 26.7 hours before this point.


    If I get time, I will try to do a similar analysis for another cell.

  • Unconventional electrolysis

    This circuit operates much differently. This is a boost circuit where the short causes energy to be stored in the inductor, then when the short clears, very high voltage appears at its output. If the high voltage then causes ionization of air across a gap, there is a low impedance path for current to flow. The current is only limited by the total resistance in the path. If you generate 1 KV this way and the total resistance is 1 ohm, you get a spike of 1 KA. But the current it goes right through the battery or supply and may burn it out. (I should also point out that you need to be very careful with circuits that can generate high voltage and high current. They can be very dangerous.)


    For the inductor, an air core avoids the saturation problems but takes many more turns for equivalent inductance. You might not care about saturation if you are just trying to dump the stored energy in one shot. You would only need a core that does not saturate for the charging current.


    The circuit design depends on what you want to accomplish which is may be hard to figure out at this point. If you have voltage and current targets and know what you will use for a power source, then you can design a circuit to provide the desired output.

  • Unconventional electrolysis

    This circuit will certainly produce a big current spike.

    You would need to use a cap with a very low ESR and connect it with heavy wires to keep the wiring resistance low. Maybe a cap like this would work:

    https://www.digikey.com/produc…8M000/495-6089-ND/3545213


    The coil inductance will be much lower than your calculation without a good core, and it is hard to find a low cost core for such a high value of inductance and current. It looks like the only function of the inductor is to limit the current to or from the power supply when the cap discharges. Maybe you could just replace the inductor with a rectifier to protect the supply. If the supply then draws too much current after cap discharge, either use a current-limited supply or a small series resistance to limit it.


    Or maybe I am misunderstanding your reason for the coil. If you are trying to use the the coil to stimulate the system, the current spike will not appear there. The simulation of a circuit with high currents in an inductor will require a more sophisticated simulation because the core is likely to saturate at high currents.

  • FP's experiments discussion

    Quote from Ascoli65

    Praiseworthy attempt to remedy to the lack of more detailed data, but you should pay attention to the time values. Probably, to get these values, you have expanded the time axis. In such a case the timelines are no longer vertical due to a very slight inclination in the original figures. So you have to estimate the time on the basis of inclined timelines like those shown for Figure 6A in the jpeg posted at the beginning of this page (1). But beware, each one of the 4 Figures 6 has its own inclination. For instance Figure 6B (2) has an opposite inclination with respect to Figure 6A. What you need to do is to expand the ordinate axis on the left by the same amount and use its final inclination to read on the abscissa the time of the points selected on the Vcell curve.

    I took this into account with a very tedious but accurate way of measuring the times.

    I took a screen shot of Fig 6B and pasted it into Visio, then rotated it to get the Y axis vertical (-0.35 degrees).

    Then drew a line from (0,0) to (250,100) mm and moved/stretched the graph to make the (0,0) and (2500,100) points of the voltage graph exactly line up with the (0,0) and (250, 100) endpoints of the line.

    Then locked the pasted graph's position and size and locked the (0,0) point of the line.

    Then moved the other end of the line by typing a voltage (Y value) into the measure box, and iterated the X values until the end of the arrow was centered on the graph line. Visio allows you to zoom in very tight to get it centered.

    Then record X and Y of that point and add a zero to X to make it Kseconds. Repeat for next point.


    Next time, I would probably use a capture tool and save lots of work.


    You also said "However, these early vapor bubbles condense within the water mass, so they can't be considered to calculate the vapor which leaves the cell. IMO, there is no water loss due to vaporization until Tcell reaches 90-95 °C."


    This does not seem right. If the vapor condenses, then the water level does not change and there is no mechanism for the steady increase in resistance.


    Water does not need to boil in order to evaporate. There is an average of 10W keeping the water very hot for 24 hours. My graph shows that the evaporation should start to be significant around 80C. That is where there is no more room on the cathode for more bubble coverage and another mechanism needs to take over to increase the resistance. If the temperature outside was 80 C (176F), some pretty deep puddles would evaporate in 24 hours.

  • FP's experiments discussion

    Here is an analysis of the Fleischmann 92 paper based only on the voltage plot in Fig 6B.


    First, the cell can be modeled as a constant current source feeding a 1.54V zener diode in series with a variable resistance. The resistance is a function of the conductivity of the electrolyte, volume of electrolyte and air (bubble) coverage on the cathode. Even if bubbles are forming over the entire surface of the cathode, the time-average of air (H or vapor) coverage should never be more than about 50%, because on average half only could be covered by bubbles while the other half is covered by liquid that is yet to be decomposed or vaporized. If you imagine a continuous stream of bubbles from a fixed area, even if the bubbles in a vertical column are nearly touching each other, the time average of that cathode area would still have a max of about 50% H or vapor. See the diagram of this model below. I take the minimum resistance from the left-most point in Fig 6B, which is about 11 ohms. At that point, I assume the entire surface is liquid. Then, assuming the resistance is linear and determined by the liquid surface area, the resistance rises as the liquid coverage drops until resistance approaches infinity with no liquid at all on the cathode. The key point of this graph is to see that the voltage starts at 7.1V and rises to 12.6V at the 50% coverage point.




    Now looking at Fig 6B, I carefully digitized the points for the last 24 hours of the experiment (accuracy is about +- 2000 sec but is sufficient for this analysis). This is the part of the graph with voltage above 12.6V indicating greater than 50% bubble coverage on the cathode. At any higher voltages, the drops in resistance need to come from somewhere else, namely the liquid level starting to drop or the liquid starting to be replaced with foam. See the plot of the digitized points below




    Then I integrated the enthalpy input by summing the contributions from each plotted point = (V-1.54)*.5A*deltaTsec = watt-sec = Joules. The sum over the entire graph comes to 988 KJ.




    The paper calculated that boil off of half the cell requires 102.5 Kj. So the energy input during the time above 12.6V is more than 4X the amount required to vaporize the entire volume of the cell. Someone could analyze the expected losses to ambient, but that seems unlikely to change the conclusion here that the vaporization was caused by the positive feedback loop that ramped up of the input power as the resistance increased.


    Regardless of the losses to ambient, the cell could not have stayed liquid until the last 600 sec given the reported cell voltages.

  • LION-AG Experiment

    Quote from Bruce__H

    I don't think I am looking at an artefact of plotting. You and I seem to be dealing with datasets that are actually different. For instance, in the 53 seconds between 7:52:73 and 7:53:36 my datasheet only shows 24 valid ThermoA data values (the rest show as "Error"). In your plot, on the other hand, there are 53 separate values. This leads me to think that your plot is showing data where interpolated values have been imputed for about half of the data points.

    My data came from this post:

    LION-AG Experiment


    I downloaded the zip file called

    LION4_noholes_interpolated.zip


    and then plotted the data from 003.csv


    Where did your data come from?

  • LION-AG Experiment

    Quote from Bruce__H

    I am also puzzled by the double-line nature of the thermocouple readings at this time. This is not due to straightforward discretation errors because there are occasional intermediate values.

    The double line structure looks like an artifact of the way you plotted it (points without connecting lines). The temperature just has small variations around 33C. See my plot of the same region below.




    The way I read the current line is to think of it as set points of the PID.

    Negative = too hot, cut all current

    Positive = too cool, add current to heat up

    Zero = keep at this temperature by cycling current on and off

    So it makes more sense if you think of the current plot as the commanded current, not measured current. I don't know if the spreadsheet really has commanded current, or if the zero point of the current sensor is wrong. That would be critical to know in interpreting the data.

  • LION-AG Experiment

    Here is a plot of the current and ThermoA from a short portion of the run 3 raw data spreadsheet:




    A couple of odd things. Most of the time the current oscillates around zero from +.19 to -.30 (Amps presumably). I assume there should never be negative current, so the zero point of the hall current sensor is not set correctly. When I have used that type of current sensor, I always had to have a calibration step to zero out the sensor. The zero can also drift with temperature and voltage.


    The big spikes in current (over 1 A) make it look like the PID is not tuned correctly. It is acting like a bang-bang control system, or maybe the current level cannot be set to intermediate values.


    The period when the current is off is interesting. It stays well above room temperature (32-35C) for over a minute with zero power input. There is much lower variance in temperature when the current is off. When the current is cycling, the variations do not exactly follow the changes in current, so something interesting may also be happening then .


    Here is another plot from later. It is ramping to a higher temperature set point and the current stays on at 1 A.



  • FP's experiments discussion

    The 1992 paper has more data supporting the foam hypothesis.


    In the Enthalpy calculation, they compute an average input power of 37.5W for the 10 (or 11) minute boiloff period. The graphs show that they are using a .5 A constant current supply, which means that the cell voltage must have averaged 75V during those 10 minutes of decreasing levels (of either foam or boiling liquid). Note that the Fig 6 graphs show the voltage at less than 5V, then quickly rising towards the 100V limit when boiloff is finished.


    It is easy to see how the cell voltage could rise sharply when it is mostly foam. It could even be oscillating between 100 V and about 5V as the foam bubbles evaporate and re-form connections.


    The question is whether the cell voltage could average 75V while boiling. We do not need to do an experiment to figure this out because I found an excellent paper that examined this in excruciating detail:

    Evaluating the Behavior of Electrolytic Gas Bubbles and Their Effect on the Cell Voltage in Alkaline Water Electrolysis, Dongke Zhang and Kai Zeng, 2012. This paper is behind an ACS paywall, preventing me from uploading the whole thing, but here is one of the key figures:


    They did both theoretical calculations and measurements with different current and electrolyte concentrations and the highest voltage they ever saw was around 5V. The hightest delta-V due to bubbles was never more than a few volts.

    Zhang also writes:
    "Similar to gas bubble evolution caused by boiling or desorption, electrolytic gas bubble formation involves nucleation, growth, and departure. All these steps determine both the residence time and the diameter of the gas bubbles, which are important parameters for determining the resistance effect. It is thus important to gain a detailed understanding of the electrolytic gas bubble behavior which will help to alleviate the resistance of the electrolytic bubbles"


    It seems possible that vigorous boiling could increase the voltage above 5V, but very unlikely that it could reach 75V. If it got that high, there would have to be so many bubbles that observations of the level could no longer be used to determine the volume of liquid. I think this is solid additional evidence that the last 10 or 11 minutes must have been mostly foam.

  • Unconventional electrolysis

    How about a high current PWM LED controller?


    The MAX16818 is designed for up to 30A, 7-28V, and a switching frequency set by an external resistor from 125 KHz to 1.5 MHz. You could use poor filtering to get the PWM cycles to show up at the load.


    They have development kits, but only for low current. You would need some higher current MOSFETs to drive 20A. See:

    https://datasheets.maximintegrated.com/en/ds/MAX16818.pdf and

    https://www.digikey.com/produc…rt=MAX16818EVKIT%2B&v=175

  • FP's experiments discussion

    I think Kirk was pointing to his reply to your question (from Sept, 2017). It seems like he proposed an experiment to show how much the calibration constant changes when the location of the recombination changes. The question is whether the LXH reported in the 1992 paper is less than the maximum error from uncertainty in the calibration constant.


    Alan Smith wrote:

    | Or describe a decent experiment to prove it in sufficient detail for it to be replicable.


    Kirk's reply:


    "To reiterate, replace the Pd and Pt cathode and anode in a standard F&P-type cell with a Joule heater, i.e., a resistor. Make the leads long enough so that you can bend it up out of the electrolyte and into the gas space of the cell. Assume 20W total input power derived from 2A at 10V. Use maximally 2A at 1.54 V in the gas space heater, and 2A at (10-1.54) V in the liquid heater. Calibrate by varying current.


    Now change the voltage distribution to less heat in the gas space. Be bold, take it to 0. So 2A max at 10V in the liquid. Calibrate by varying current.


    Now, wait 3 days and repeat.


    Wait 3 more days and repeat.


    Report results.


    Note that Ed Storms proved Jed wrong in his 'it don't matter where or how' comment. In the experiments that I reanalyzed he reported 3 different calibration results. (The only researcher I have seen do this BTW. Kudos to him for honesty and thoroughness.) He reported that a Joule heater gave a calibration equation of Pout = 0.072107 * DeltaT -0.23893. Electrolytic calibration (what I describe above) however gave Pout = .071221 * DeltaT -.177146 *initially* and Pout = 0.070892 * DeltaT - 0.14405 *finally*. (So it makes a difference how, where, and when.)


    Compare to my extracted separate calibration equations for runs 3 and 6, which both displayed zero or nearly so 'excess heat' (which means they used 'inactive' electrodes and thus are equivalent to calibration conditions). I obtained Pout = 0.070672 * DeltaT - 0.177146 for run 3 and Pout = 0.071320 * DeltaT - 0.0.131471 for run 6. Pretty solid evidence that my zero excess heat assumption gives calibrations well within the normal variation of the experimental setup, isn't it?


    Correction: The electrode used in Runs 3 and 6 was an 'active' electrode that had become inactive and was immediately revitalized by an anodic strip for Runs 4 and 7, which showed maximum excess heat signals."

  • Unconventional electrolysis

    For AC tests, you might try a development board with a stepper driver like the ST L6208. It can drive up to 52V, 2.8A RMS and 100 KHz.

    see: https://www.digikey.com/produc…6208PD/497-4136-ND/724253

    The ST motor development boards also have a nice motor control GUI that runs on a serial-connected PC.

    This may be a better choice than an RC motor ESC. Sometimes they have protection circuits to prevent them from driving currents to other than balanced 3-phase loads.


    But any high-current motor driver, will generate lots of EMI. Before investing in one of these, you might do some quick tests to make sure your Geiger counter is not too susceptible.


    Get a cordless drill, preferably one with a brushed motor, and then start and stop it close to your Geiger counter to make sure it does not react.


    I recently did an experiment with a scope connected via coax to a circuit in a sealed thick aluminum box. After getting funny readings, I disconnected the battery in the box and still got the funny readings. Then I noticed that flipping a light switch would cause the scope to trigger. Then when I tried the cordless drill within a few feet of the box, it went crazy with 10s of mV noise. Aluminum Faraday cages are not perfect shielding for RF, and they provide no magnetic shielding.

  • FP's experiments discussion

    The foam issue is not the only problem with the HXH (boil off) phase in the Fleischmann 1992 ICCF paper.


    The Enthalpy Input calculation says:

    "By electrolysis = (Ecell - 1.54) × Cell Current ~ 22,500J"


    It should not be hard to measure power to an accuracy of 1 mW and to take samples every 1 mS. That means it should have been easy to measure input energy to an accuracy of 1 uJ. Why did they throw away 8 orders of magnitude and estimate input power to the nearest 100J? Did they round up or down? It is not even clear which of the 4 experiments they are talking about. Why do they not show calculations for all 4?


    Similarly, they show boil off time as exactly 600 sec, not 599 or 601. But earlier in the paper, they say it was 11 minutes (660 sec). Why did they choose to round that down and make the excess power look higher instead of more conservative?


    They do not show a graph of the critical 10 or 11 minutes of boil off that shows input current, voltage and temperature. They also do not show those details for the claimed 3 hours of HAD (heat after death).


    I would expect a high school science project to be more careful than this.


    The carelessness of the analysis for these phases does not give me much confidence about the way they analyzed LXH phase, but I did not attempt to study that part of the paper.


    Is this really the best paper written to support excess heat with electrolysis? If not what is?


    By the way, I do not consider myself either a skeptic or believer in CF, and do not even think it is appropriate to label people that way. We should all analyze the papers in the field and look for possible errors. It will not help the LENR field to promote the conclusions of papers with obvious flaws. I still hope that some avenues of LENR will succeed, but mostly I am still waiting for definitive papers and experiments.

  • Media/News/Video Library-No discussions please

    The Lindau-Nobel Meeting, held annually, brings Nobel Laureates together with top graduate students from around the world for a week of lectures and small-group discussions. Each meeting has a theme, and the 2019 theme is Physics. The meeting will take place from June 30 to July 5, 2019.


    It would be great to get some LENR students there. The application process usually requires nomination by an academic partner (students selected to represent their university), but there is also a way for exceptional students to bypass that selection process.


    See:

    https://www.lindau-nobel.org/

  • FP's experiments discussion

    Quote from Ascoli65

    Are you able to locate the time of that frame on the graph in figure 8? It is not an expanded graph, it is in the original size as it appears on the F&P paper (3). Its time axis ranges from 1590000 s (=18 days + 9.40.00) to 1660000 (= 19 days + 5.06.40). The 3.26.14 time of the above video frame + 19 days gives a total of 1654718 second. So this video frame is far on the right (at least a couple of hours) with respect to the vertical arrow which indicate the "Cell dry" time.


    How do you explain this discrepancy?

    It looks to me like the video timestamps are time-of-day because they all are for hours 0-23. But the graphs in the paper appear to be time from the start of the experiment (in seconds or Ksec). So to line them up, you need to know what time of day corresponds to 0 Ksec.


    However, I looked at the video a couple dozen times and am inclined to agree that the arrows are foam levels, not liquid levels. The cells seem to transition through three clear phases. In the first phase, you can see that it is mostly liquid with gradually increasing bubbles as the liquid boils. In the second phase is is mostly foam and in the third phase, the foam level rapidly decreases to zero. You can tell the foam phase because sometimes the level decreases and then increases again, which could not happen with liquid. For instance, look at Cell 1 at 21:23 when it is full of foam, 21:40 when the top of the foam is a little lower, then 21:55 when it is full of foam again. Several times the video cuts away for hours between phases 1 and 2. For Cell 1, there is a cut between about 11:30 and 18:36.


    The Enthalpy Balance in the paper is based on only the last 10 minutes and assumes the liquid is boiling then. Even though I have great respect for Fleischmann's work in general, I would have to agree with Ascoli that this paper is likely flawed.


    For ease of finding them again, here are links to the video and the paper. (It is hard to get much out of stills. You need to run the video to see how the levels are changing.)


    External Content www.youtube.com
    Content embedded from external sources will not be displayed without your consent.
    Through the activation of external content, you agree that personal data may be transferred to third party platforms. We have provided more information on this in our privacy policy.

    http://www.lenr-canr.org/acrobat/Fleischmancalorimetra.pdf