Shop OBEX P1 Docs P2 Docs Learn Events
Capacitor Charging Question — Parallax Forums

Capacitor Charging Question

ajwardajward Posts: 1,130
edited 2015-12-12 22:31 in General Discussion
Hey All...

Per the title... I want to charge a bank of capacitors. The bank is 10-1000uF/35 Volt in parallel. I don't think I want to just switch them across the supply source without an inline resistor.
Am I correct? What value would be sufficient? I have a couple of 100 Ohm/ 10 W critters.

My application is to demonstrate the rudimentary principles of a railgun. :innocent:

Any wisdom is appreciated!!!

Amanda

(The latest "Shiny Thing")

Comments

  • jmgjmg Posts: 15,173
    ajward wrote: »
    My application is to demonstrate the rudimentary principles of a railgun. :innocent:
    Hehe, charging is the easy part, Cap discharge is where the fun is :)

    Usually in a high power design, you use a switchmode charger, as that can draw a regulated power from some source, and deliver that to the capacitor with lowest loss.

    However, sure you can use resistors for very early testing - check the peak power as V^2/R (12.25 W here)
    time constant R*C gives a rough guide of the order of ramp.

    https://en.wikipedia.org/wiki/Rise_time
    " In a pure RC circuit, the output risetime (10% to 90%) is approximately equal to 2.2*R*C ]"
  • kwinnkwinn Posts: 8,697
    edited 2015-12-12 22:44
    100 ohms at 10W is more than enough to limit the peak current at any voltage a 35V capacitor can handle. Peak current would be 0.35A with a 35V supply.
  • ajwardajward Posts: 1,130
    Many thanks Y'all!!!

    @
  • koehlerkoehler Posts: 598

    IIRC, there are a couple of posts on the net regarding 'Gotchas' when constructing such a device.
    Can't find them now, however some unexpected things can occur so might be worth trying to find.
    Hospital visits during the holidays are a bummer.
  • kwinnkwinn Posts: 8,697
    If the capacitor were charged to it's rated 35V (not recommended) the total energy would be 6.125 joules per capacitor, so not a great hazard individually. Just don't aim the projectile at your eye.
  • GenetixGenetix Posts: 1,754
    Amanda,

    Rather than build a charging circuit why not use something like a car battery as your power source.
  • ercoerco Posts: 20,256
    ajward wrote: »
    My application is to demonstrate the rudimentary principles of a railgun.

    LIKE! I'll be following this closely to check your progress Amanda. I already have a Christmas laser and Christmas flamethrower, so a Christmas rail gun will be a great addition.

  • Beau SchwabeBeau Schwabe Posts: 6,566
    edited 2015-12-13 23:02
    Take that voltage up a notch by a factor of 1000 and you have a Marx Generator ... :-)

    This one produces about 450kV with each stage being about 35kV. The idea is that each stage is charged in Parallel via a high Ohm resistor (5 Megs in this case)... When the voltage reaches enough to jump the spark gap, then an over voltage is produced instantaneously on ALL of the remaining spark gaps. You can control the initial spark gap by having one spark gap slightly closer than any of the others. When a spark is initiated, the spark itself acts like a dead short (a few Ohms) and effectively places ALL of the capacitors in series ...

    HighVoltage.jpg

    There is Video also:

  • ajwardajward Posts: 1,130
    erco wrote: »
    ajward wrote: »
    My application is to demonstrate the rudimentary principles of a railgun.

    LIKE! I'll be following this closely to check your progress Amanda. I already have a Christmas laser and Christmas flamethrower, so a Christmas rail gun will be a great addition.

    <sigh> Yeah... and I worry about you! :-)

    @

  • ajwardajward Posts: 1,130
    edited 2015-12-19 13:38
    A bit of experimentation this evening proved a little disappointing. The idea was to impart a little motion to an aluminum bar across two aluminum conductors.
    The capacitor bank was charged to 12 volts and I gave the bar a nudge onto the conductors... nada. -_- A couple more tries with the same "exciting" result.
    Of course at 12 volts (my present limit) I'm only getting 702 milliJoules. Wow!

    Wanted to try one more thing before calling it a night. Since I couldn't do more with the voltage, I replaced the aluminum bar with a neodymium magnet. Aha! As the magnet got close to the conductors, it wanted to twist slightly. I suspended the magnet above the conductors and when the current was applied the bar swung parallel to the conductors. Progress... of sorts. :innocent:

    Anyhow, I'll post a couple of pictures of the setup tomorrow in a few days.

    'Tis getting late and the downstairs neighbors are getting cranky! :smile:

    Amanda
  • kwinnkwinn Posts: 8,697
    Amanda, did you read the wikipedia article on rail guns? Lots of helpful info and math formulas.
  • ajwardajward Posts: 1,130
    kwinn wrote: »
    Amanda, did you read the wikipedia article on rail guns? Lots of helpful info and math formulas.

    I did actually. I have a fair understanding of the mechanics, but the math is still kind of fuzzy. Keeping the design pretty low key till I have a better grasp of things.

    @

  • kwinnkwinn Posts: 8,697
    edited 2015-12-19 14:01
    It was the kind of currents required that stood out for me. The article lists current requirements in the millions of amps. Even though you have a much smaller projectile you will need more current than 100,000 uf capacitors can deliver. Another consideration is the ESR of the capacitors. Standard electrolytic capacitors have fairly high ESR. You will probably need low ESR capacitors like those used for xenon flash lamps to get the high peak currents required.
  • ajwardajward Posts: 1,130
    kwinn wrote: »
    It was the kind of currents required that stood out for me. The article lists current requirements in the millions of amps. Even though you have a much smaller projectile you will need more current than 100,000 uf capacitors can deliver. Another consideration is the ESR of the capacitors. Standard electrolytic capacitors have fairly high ESR. You will probably need low ESR capacitors like those used for xenon flash lamps to get the high peak currents required.

    Yeah... started out modestly. The design was partly based on a demonstration using a 9 volt battery. I think that demo "had some er...serious help" moving the little projectile down the conductors.

    I read that the Navy device developed about 33 megaJoules. Yikes! The projectile hit something like Mach 7. Heck, I'd be happy if my little Aluminum bar would just twitch a little.
    Located my 24 volt power supply. I'll play again today!!!

    @


  • MicksterMickster Posts: 2,693
    edited 2015-12-21 13:06
    ajward wrote: »

    My application is to demonstrate the rudimentary principles of a railgun. :innocent:

    Amanda

    (The latest "Shiny Thing")

    LOL..kinda reminds me of when I was "experimenting" with 2,000PSI of helium in my PCP (pre-charged-pnuematic) airgun. That huge .44 calibre chunk of lead went supersonic and the whole darned neighbourhood knew about it.
  • ajwardajward Posts: 1,130
    Well... I found out my little 24 volt wall-wart puts out close to 33 unregulated volts and I can get the capacitor bank charged to a little over 32 volts... a little over 5 Joules. Cool!

    When I try to get my rods (Aluminum and Neodymium magnet) to move down the conductor rails, I get a noticeable arc and the "projectiles" stop immediately. If I understand correctly, the objects have to be inserted between the conductors with some considerable initial velocity. Not sure how to do this with my setup
  • ercoerco Posts: 20,256
    edited 2015-12-22 04:51
    ajward wrote: »
    Not sure how to do this with my setup

    When in doubt, add fire.

  • kwinnkwinn Posts: 8,697
    Perhaps add a spring loaded finger behind the projectile that also holds a micro switch open. When the finger is released it gives the projectile a push and closes the switch.
  • ajwardajward Posts: 1,130
    A quick image of my test setup. It's sort of like something I saw used to demonstrate the principles of a railgun. But I think that video exaggerated the performance... a lot.

    Anyhow... the picture shows the aluminum conductors, my capacitor bank, knife switch and two armatures.

    20151227_162955%20copy.jpg

    I used the knife switch because... well, what mad scientist project doesn't have one. :cool:
    In one position the switch powers up the cap. bank and the in other connects the bank to the "rails".

    Right now, when an armature contacts the conductors, there is a nice arc and the foil burns through to the adhesive.
    I'm going to replace the foil conductors with two - 1/16" x 3/4" aluminum bars, although my expectations aren't very high.

    I have another 13 - 1000uF x 35 volt caps on the way, but the total capacitance isn't the issue. The Navy had some successful tests with just four 850uF units... rated at 11kV!!! (Plus a whole lot of other (*expensive*) equipment.)

    I guess this qualifies as my Christmas project. :innocent:

    Amanda
    20151227_162955 copy.jpg
    2521 x 969 - 522K
  • ajwardajward Posts: 1,130
    erco wrote: »
    ajward wrote: »
    Not sure how to do this with my setup

    When in doubt, add fire.

    I'm almost there... just need to add a little accelerant.

    @

  • jmgjmg Posts: 15,173
    The Physics of this setup has me puzzled.
    I thought a Rail Gun was essentially a commutation problem, where you have something very like a series of solenoid coils, and the projectile accelerates through the centre.
    Movement Physics are identical to a solenoid, and (very) careful activate/release of each current impulse gives the acceleration.
    ajward wrote: »
    Right now, when an armature contacts the conductors, there is a nice arc and the foil burns through to the adhesive.
    Given that setup, what else was expected ?
    Google Spot Welding :)

  • ajwardajward Posts: 1,130
    edited 2015-12-28 22:40
    I believe you're thinking of a coil gun. The railgun is based on the Lorentz Force. From on what I saw on the internet (everything there is true, isn;t it? :->). the armature should have moved a bit along the conductors. Further research indicates this should not happen... but the arc is kinda cool!

    Anyway, I've located a source for some copper bar. Looking for some decent 300-400V capacitors and a design for a ~400VDC power supply.

    Using these resources from the Naval Postgraduate School:
    Design, Fabrication, and Testing of a Scalable Series Augmented Railgun Research Platform
    An Experimental Study of Electromagnetic Lorentz Force and Rail Recoil
    Modeling of the Electric Ship

    Amanda
  • kwinnkwinn Posts: 8,697
    Maybe put the new rails closer together and use a smaller projectile?
  • RITESH KAKKARRITESH KAKKAR Posts: 254
    ajward wrote: »
    A quick image of my test setup. It's sort of like something I saw used to demonstrate the principles of a railgun. But I think that video exaggerated the performance... a lot.

    Anyhow... the picture shows the aluminum conductors, my capacitor bank, knife switch and two armatures.

    20151227_162955%20copy.jpg

    I used the knife switch because... well, what mad scientist project doesn't have one. :cool:
    In one position the switch powers up the cap. bank and the in other connects the bank to the "rails".

    Right now, when an armature contacts the conductors, there is a nice arc and the foil burns through to the adhesive.
    I'm going to replace the foil conductors with two - 1/16" x 3/4" aluminum bars, although my expectations aren't very high.

    I have another 13 - 1000uF x 35 volt caps on the way, but the total capacitance isn't the issue. The Navy had some successful tests with just four 850uF units... rated at 11kV!!! (Plus a whole lot of other (*expensive*) equipment.)

    I guess this qualifies as my Christmas project. :innocent:

    Amanda

    Hello,
    What are you building of these Capacitor??

  • ajwardajward Posts: 1,130
    RITESH KAKKAR wrote: »
    Hello,
    What are you building of these Capacitor??

    Hopefully, something to demonstrate the Lorentz Force.

    Amanda

  • ercoerco Posts: 20,256
    Amanda: Your Youtube channel is long overdue (2 years?) for some new videos. How about uploading some rail gun fun when you can? :)
  • jmgjmg Posts: 15,173
    ajward wrote: »
    Hopefully, something to demonstrate the Lorentz Force.

    Since the losses will be very large here, I'd start in a controlled manner :

    * make some 'staples' - bent wire with slightly flattened feet, and check those always fall the expected way.
    (reverse the current to prove to yourself it really is Lorentz Force)
    * add a paper ruler to your tramlines
    * Add a parallel-mosfet current controller, so you can vary the current impulse, and also eliminate physical bumping effects.
    * scope or sampling ADC grab the current peak
    * tune for largest fall-over nett movement.

    With losses being I*I*R and available energy being (1/2)*C*V*V, and RailGun wiki says Force = L*I*I/2
    - but also impulse = F*dT, and dT is related to Tau, where R is V/I, which gives Impulse = k*L*I*V*C

Sign In or Register to comment.