me356: Reactor parameters [part 1]

  • I plan to build own reactor based on work of Dr. Parkhomov.


    Parts are relatively very cheap and available. First from all I want to build heater that will work for longer periods.
    Heater must be driven with electronics.


    I plan to build control system based on PI(D) regulation with microcontroller so it can work without any additional stuff.
    Maybe I will try DC with pulse width modulation to control temperature so it can work from a different power sources.


    Heater Wire: I will start with Kanthal A1 wire or hopefully I can get Super Kanthal that is able to stand even 1900°C continuously.
    Heater Ceramic tube: 1x ceramic with min. 99,7% Al2O3 - ∅D = 16mm, ∅d = 12mm, L = 80mm, Tmax = 1800°C
    Reactor Ceramic tube: 1x ceramic with min. 99,7% Al2O3 - ∅D = 10mm, ∅d = 5mm, L = 290mm, Tmax = 1800°C
    Reactor Ceramic filings: 2x ceramic with min. 99,7% Al2O3 - ∅D = 5mm, L = 100mm, Tmax = 1800°C
    Thermocouple: unknown yet, maybe R, S or B?


    Do you think that diameters of 10/5mm are fine? What do you suggest?


    Thank you Dr. Parkhomov for perfect work, I hope you will continue!
    Sharing your findings can make our world much better.

  • interesting.
    PID controller (or any good control) is, if well done, a good way to prove energy gain. Read the paper on McKubre isothermal calorimetry...
    It seems multiple TC are good to check that there is no hotspot, or strangeness.
    Controlling convection, and room temperature may avoid artifacts.


    Do you plan a better calorimetry than just few TC? mass calorimetry à la parkhomov?



    What are the safety measure you planned... It is dangerous chemical and potentially explosive (à la "Bang") ? maybe it would be nice you propose your measure for "peer review" by experienced people here ? (not me :huh: ).

  • All tests should be done in a special room made by conrate so even fire or explosion should do nothing.
    I want to control and observe the process remotely, at least 50m away from the reactor.
    In case that gamma or neutron radiation will occur it will be not dangerous for us.


    Stage 1. building the reactor that is able to stand at least 1200°'C for long periods (at least weeks) with atmospheric pressure (without fuel). When this criterium will be fulfilled, we can go to Stage 2. All data will be logged, especially temperatures for future comparison. In this stage TC calibration and PID regulation tuning will be performed.


    Stage 2. testing reactor with fuel, with 12h temperature rampup to not exceed pressure - everything will be controlled by program, so it should be perfectly reproducible.
    We will use opensource tools and hardware so everybody can possibly reproduce it later with little effort.


    Stage 3. long term tests with fuel, trying to achieve the highest possible COP with SSM. Improvements in the control algorithm.


    Stage 4. trying to harness heat for usefull usage to achieve the highest efficiency.

  • Note for those who can interpret it :
    Airbus patent report (in the description) that gamma are produced when the reactor is too cold and one trigger the reaction.


    One naive question to chemist/lab experts : can a CO2-ice extinguisher be useful to freeze the reactor in case of unexpectable event? (it is a question, can be stupid)

  • Regarding Fuel, what Nickel powder may be the best? I believe that it should be clean. Do you think that 10um particle size will be OK?
    Vale Type 255 is approx. 3um.


    I have measured, that fuel for the reactor costs only approximately $0.24! Thats awesome. I have Ni + LiAlH4 from local chemist laboratory.

  • OK, I've checked various solutions and it seems to me, that the best solution for controlling output power may be with Solid State Relay and PWM.
    Alternatively we can use triac that will cut AC phases and thus can change output power.


    But I think that SSR is the easiest solution that will just work with the best latency.


    I will use thermocouple K type with 1300°C Tmax because of good availability and fine parameters.


    Radiation measurement tool will be most probably Gamma Scout, because interfacing is quite easy and well documented.

  • I don't understand why people fear to use PWM (triac, switched power supply). with a good wattmeter , and some adapted filters it is very precise...


    of course don't use kit chen voltmeters or even RMS am/voltmeter. never "assume" the load is pure resistive... (pure resistor are rare, and hard to build). see possible RF with oscilloscope, and filter it eventually.


    if you add a PID controller with TC/thermistors as sensor, even if no component is precise or calibrated (have to be stable), you can regulate a heater quite precisely, especially in comparison mode.
    designing/tuning the PID controller is not simple (there is some theory, have to calibrate the transfer function first), but there is known method.


    independently add few good TC to calibrate the regulator.. I cannot help ther, it is not my education.


    Note to barty: kit chen is forbidden word :D

  • I have found Silicate based cements that are able to stand 1300°C. It is used for ovens and stoves, so I think it can work.


    I am struggling with Fuel Container little bit at the moment.

    Quote

    The AGP fuel container was a stainless steel tube slightly smaller in diameter than the bore of the reactor tube. The length was about 50 mm ( note this was eyeballed from photo) The tube was sealed by heat-resistant cement.
    The tube or the sealant had three small openings. These were necessary to release the pressure from the hydrogen which is generated as the LiAlH4 decomposes when the reactor is heated.


    I dont know, if the whole reactor should be sealed so hydrogen "can't escape" at all. If AGP made 3 holes in the fuel container, then there is just little bit more room for the hydrogen gas. But still it can't escape. It will be blocked by ceramic fillers from both sides.


    If it was 5cm long tube then there is lot of room for the air, so ceramic fillers can't push the air out. Was the fuel container sealed somehow first from both sides (with loaded fuel) and then inserted into the reactor?


    Do you think that pressure is the key factor in the process? At least before melting point of the lithium.


    Why do we need fuel container if the fuel can be placed directly to the reactor ceramic tube? Is it because of easier refueling?


  • ...I am struggling with Fuel Container little bit at the moment. ...


    We lack information to answer your questions about the fuel container described by [lexicon]Alexander Parkhomov[/lexicon] in his latest short presentation.


    You could cut down the stainless tube which comes with this thermocouple.



    I have purchased this thermocouple and used it with an Arduino. The size is appropriate for a reactor. One end is sealed as manufactured - it is like a test tube. After dismantling the parts, you would still have a working thermocouple to use in the experiment. Since you are likely to destroy thermocouples during long runs, buying several is advised. This way you also are buying stainless steel tubes, so your cost is low.

  • Wow, many thanks for sharing the info!


    To not destroy thermocouples I was planning (for the future) to make calibration inside the reactor tube and outside where temperature is not so high.
    With some calculations about heat transfer in the material I can get hopefully precise temperature measurement from outside of the reactor.
    So very naive calculation may be temp = temp_tc2 * 2; in case that temperature will be 2x smaller than in the core.
    With this thermocouple can last for very long time.


    Disadvantages:
    - possible increased delay in getting temp of the core.
    - decreased precision with same decoder.

  • Good, ceramic tubes arrived, everything is OK. Quality is perfect 99,7% Al2O3 and 1800°C guaranteed. Each tube including fillers is made of this material. Price for these tubes was $23, not bad. With higher quantity it can be much cheaper.


    My heater will be made of:
    ∅D = 15mm, ∅d = 10mm tube, I have also ∅D = 16mm, ∅d = 12mm, but first one fits fine on the reactor tube.
    20AWG (0.812mm) Kanthal A1 wire, 3569.8mm length (R = 10Ω), 66-67 wraps with 50-100mm total leg length (<3% power loss)
    Wire sealed with a stove cement - based on water glass, 1300°C max. temperature.


    Input Voltage 230AC @ 50Hz -> Transformer 2:1
    -> Solid State Relay controlled by microcontroller with fast AC PWM or Phase Angle Triac circuit.


    Max. output power cca 1210W @ 110V, 11A.

  • ... My heater will be made of:
    ∅D = 15mm, ∅d = 10mm tube, I have also ∅D = 16mm, ∅d = 12mm, but first one fits fine on the reactor tube.
    20AWG (0.812mm) Kanthal A1 wire, 3569.8mm length (R = 10Ω), 66-67 wraps with 50-100mm total leg length (<3% power loss)
    Wire sealed with a stove cement - based on water glass, 1300°C max. temperature. ...


    Since you only mentioned diameters, it is unclear which generation of Parkhomov reactor you are emulating - the newest epoxy sealed version, or the much shorter modular version.


    I also wonder how many tubes you have. Do you have, for instance, 5 of each size? I realize we all hope for immediate success and wishful thinking leads to owning no spares.


    No worries. Each experiment is an adventure that can lead to great things. Especially when results are widely and accurately shared.

  • I am replicating the latest version.


    If everything will be fine, calibration tests will be performed this week.
    I am prepared for unsuccessfull tests, it is very normal.


    But I strongly believe, that in the end, it will work.

  • Good news, I have found that company that is producing ceramic tubes is also producing special cements from very same material. So it is designed for this stuff.
    Max. temperature: 1500°C


    I have also programmed reactor controller with PID.


    It supports few stages:
    1. Slow pre-heating procedure to configured temperature.
    2. Maintaining required temperature.
    3. Self Sustain mode - TODO
    4. Turn off



    Everything can be controlled and configured remotely in real-time. All the data are remotely transmitted and also logged.

  • "Gamma Scout geiger counter"
    Is this a windowless counter? If not a better choice would be a scintillation counter. Alpha beta and gamma radiation would be interesting. A standard Geiger counter is only sensitive to gammas.

  • Calibration test is coming soon!


    I have all needed parts, yet I have to build the reactor now.


    I have succesfully tested my electronics now which include:
    - Transformer
    - Triac Board (will be better than thyristor) - SSR solution does not worked as expected.
    - Reactor Controller with sensors
    - Data logger and Streaming machine


    Calibration tests will show if there could be any problem, but I hope it will work very well.
    Streaming machine will stream online including realtime reactor data. So everybody can see exact measurements without any delay.


    Reactor Controller has own firmware that supports many modes, can be updated remotely, stopped, restarted, modified without touching it. It has PID regulation - already tuned.
    Same with Streaming machine. It will work separately, but will be connected online.


    Firmware is prepared even for SSMs.

  • Yes, except that I have bought cheaper model. Logging feature is still there but you can't read the data in realtime. So I will be reading the data by camera and periodically from the log via USB.


    Interestingly this counter is turned on all the time, it can't be turned off. And it should last for 10+ years with this one battery charge.
    Basically it measure radiation "endlessly".

  • Can somebody share tips for wrapping kanthal wire on the ceramic tube?


    It should be done very precisely to avoid local overheating.
    I have tried some methods, but it seems that wounding on the screw-thread is working fine.
    For 16mm tube it seems that M16 screw with fine pitch (1.5mm) could be good.
    So to get 10 ohms 20AWG Kanthal A1 wire 64 wraps is needed. This will take 96mm of the 100mm tube - perfect fit.

  • Can somebody share tips for wrapping kanthal wire on the ceramic tube?


    It should be done very precisely to avoid local overheating.



    I would suggest to make the coil on the ceramic with 2 wires all together in close contact. The kantal one and a dummy cooper the dia of which will precisely set the pitch of the coil.
    Then remove the cooper one.

  • Well, I think it could work.


    Firstly I have used lathe machine that wrapped the wire on 14mm metal tube.
    Then after placing the coil on 16mm ceramic tube I have used copper wire with proper diameter and wrapped it by hand to place kanthal wire very precisely.
    Lastly I removed copper wire carefully and the result is quite good.


    Now it is time for sealing kanthal wire with alumina cement.