Calculating COP

  • Hey there!


    I am running one experiment and I can find quite interesting results. Fuel is mixture of many elements.


    Take a look at the attached plot and data. Could you calculate COP from this?

    Measurement was made by IR gun that logged the temperature. You have to just ignore step around 200 deg C. This is measurement error (wrong E) and apply for both plots.


    In both runs everything was same except the red plot - reactor has no Hydrogen inside - vacuum. Next comparison must be done with Argon.


    What you can see is basically 600W of applied power at the heater -- identical for both cases -- and the power was cut completely. Before cut the power was stabilized and temperature equilibrated.

    You can see Heat after Death for 30 minutes. There was no insulation whatsoever but with some wool around it could last few times longer.

  • An event that lasted a few minutes transiently slowed down the exponential temperature decline normally expected for a standard resistor. After that, temperatures declined as expected; possibly slightly faster compared to null due to the gas contained inside the tube increasing thermal conductivity.


  • What result would you get if you integrate both curves? Isn't the area below the curves a measurement for energy?

    If yes then excess energy would be:


    excess energy = area of blue curve - area of red curve

  • COP, no idea from that, but the derivative i.e. dt/dT would show better what happened; here is a quick graph of that with a 15 samples (seconds) rolling average.


    I think the derivative i.e. dT/dt only shows how steep the temperature curve rises or falls (pitch of temperature curve), but this no real indication for the energy.

  • I wonder if JohnyFive is obtaining a figure of 1.2 by doing something like:


    Σ(T_Active4) / Σ(T_Null4)


    In this way (using Kelvin temperatures) I'm obtaining 1.22


    That's basically pretending that the temperature recorded is an average value for the entire reactor in order to integrate the pseudo-power/energy obtained for every second in the data with its fourth power.

  • Without knowing how the temperature is recorded and the positions of the (presumed) thermocouples it is pretty meaningless anyway. A picture of the reactor would be helpful in judging the data, but that may not be possible.

  • As I written at the first post the temperature was measured by IR gun. The measurement spot was center of the heater. And inside the heater there is tube with my fuel mixture. Before any Excess Heat could be apparent the fuel must heat through 3mm ceramic tube and 1mm SS container.


    In a few days I will make more precise measurement and can take some photos. There is just a coil, ceramic and SS tube.

    Probably I will cover it with some insulation.

  • Alan Smith

    I don't have any real reason to wait for photos, just a few words would be fine.


    I'm mostly curious because all affordable IR probes I've come across recently don't seem to have data logging capabilities or USB output and I was interested in getting one for quick tests (with all possible caveats considered due to emissivity issues, etc). If I have to spend in the few hundred euro range just to have this feature which could be very cheaply added by the manufacturer, then the money will be most probably be better spent elsewhere, at least on my part.

  • can I have never found a cheap one with data-logging, but you could probably break into the system and steal the feed to the display, There are some google posts that deal with this issue though it does require some modest electronics and programming skills. The photographs suggestion btw was just my little joke.


    http://www.exisle.net/mb/index…cen-tech-ir-thermometers/


    And here's the kind of budget thermometer these articles describe.


    https://www.harborfreight.com/…er-thermometer-63985.html

  • Alan Smith

    Ideally one would need to know in advance which ones offer such relatively easy way to get temperature values as in the link above. I suspect many of the cheap models I have considered getting might not have such guarantees. Thanks for the possible suggestion, though; worth considering when I'll decide to also get an Arduino or similar experimental board to play with.