584,805 active members*
4,831 visitors online*
Register for free
Login
IndustryArena Forum > Mechanical Engineering > Mechanical Calculations/Engineering Design > Lets build Computer Controlled Graphical Fountains
Page 3 of 17 1234513
Results 41 to 60 of 325
  1. #41
    Join Date
    Feb 2009
    Posts
    2143
    Here's a how-to of doing it "in reverse"...

    Bubble Display - JBProjects.net

  2. #42
    Join Date
    Apr 2006
    Posts
    3498
    Quote Originally Posted by RomanLini View Post
    Hi Khalid.
    No that's not the way to connect it. The simplest way is to use a resistor between the power supply +32v and the + terminal of the solenoid. Then the capacitor from that point to ground.
    First of all, thanks for taking time and wrote such a descriptive to the point post. I was able to sketch the resistor and capacitor. Please correct me if the sketch attached i drew after reading your post is right/wrong?.

    Quote Originally Posted by RomanLini View Post
    When the FET and solenoid is OFF the cap will charge to 32v. Then when the FET turns on the cap will supply the pull-in current to operate the solenoid, and after a short time (some milliseconds) the cap will be semi discharged and the resistor will supply enough holding current to keep the solenoid pulled in, and solenoid voltage will be about 20 to 24v.
    That is really logical.

    Quote Originally Posted by RomanLini View Post
    I also think as the IRF540 FET is rated for just 100v it's not a great choice for switching the solenoid.
    Each Solenoid (24VDC 260mAmp) is connected to its own IRF540, so 40 solenoid having 40 IRF540. Why do you think the 100Volt IRF540 is not a good choice.

    Quote Originally Posted by RomanLini View Post
    As for the resistor and capacitor values, you need to switch the solenoid on/off at the fastest frequency it will get in normal use. The look on the 'scope waveform to make sure the resistor is small enough so it is fully charging the cap during the shortest OFF time. The cap needs to be big enough to give a good pull-in effect but small enough so it is discharged during the shortest ON time.

    Sorry I know it sounds complex but it is actually quite simple in practice.
    I have no scope i am not electronics engineer... can you calculate the capacitor/resistor value?
    Attached Thumbnails Attached Thumbnails capacitor location.jpg  
    http://free3dscans.blogspot.com/ http://my-woodcarving.blogspot.com/
    http://my-diysolarwind.blogspot.com/

  3. #43
    Join Date
    Apr 2006
    Posts
    3498
    Quote Originally Posted by PEU View Post
    Khalid! I always wanted to do a project like this, at one point I began searching for the acceleration of an object in free fall due to gravity and estimated that I need a fountain of about 5 meters high (if I recall correctly) for a drop of water to reach the floor in one second (this is a 1st year of physics problem, but I'm not a physicist )

    So, you need to know if your solenoids can open/close fast enough to keep the pace of the drawing at certain dpi. For example: if the solenoids are able to open/close at a pace of 50ms that means you can have a resolution of 1000/50=20DPS (dots per second) as you may realize that means you can only have 20 dots per 5 meters and thats not that good IMHO so you need faster solenoids, at least with a 10ms response time.

    Food for thought, hope you can test your rig soon.

    Oh, I almost forgot, there are at least three patents for this kind of fountain, you can google them and check how they did it, I recall doing this too


    Pablo
    Hi pablo, thanks for posting. I will make a rig for a single solenoid valve and test with different frequency of time. I will make the video and post here. I will also check and compare the simple circuitary without capacitor/resistor with capacitor/resistor, and show the effects.

    please keep watching and sharing your ideas with us. this is a community project and i will try to make the software available to all of CNCZone members.
    http://free3dscans.blogspot.com/ http://my-woodcarving.blogspot.com/
    http://my-diysolarwind.blogspot.com/

  4. #44
    Join Date
    Apr 2006
    Posts
    3498
    Quote Originally Posted by mcphill View Post
    Here's a how-to of doing it "in reverse"...

    Bubble Display - JBProjects.net
    I have seen this before. the system is very easy to implement as compared to waterfall. you can vary fluid density and other parameters... and can easily get that results..but in our case a bit hard as the gravitational force and the water droplets plays an important role.
    http://free3dscans.blogspot.com/ http://my-woodcarving.blogspot.com/
    http://my-diysolarwind.blogspot.com/

  5. #45
    Join Date
    Apr 2011
    Posts
    0
    Hi, Jason here, from Australia.
    I've been following your project since a week ago when I came across it. Love the software you wrote!

    Ever since I first saw these waterfall printers on TV, i have wanted to attempt to make one, and so my search began last week to find a point to start from.

    It's a shame I wasn't onto this forum earlier because I would have liked to throw in my help on providing code for a micro-controller as I am ok with Pic coding. But it looks like you have it sussed now anyway.

    I would love to have a copy of the software you wrote for the imaging, if you would be so good as to share your work. i could learn alot from it. I'm semi proficient with VB in MSAccess, but haven't attempted any standalone executables yet, so I'd love to use your piece as a learning tool.

    I joined up like you said.... Can't wait to see your first run with water and everything hooked up. The fun bit will be when the time comes to play around with the amount of water to have in the holding tank attached to the solenoids and calibrate the flow rate of the falling water. Have you thought of a way to Set a float level and water pump and all that stage yet? The size of the orifice being open/closed, and the way the water acts as it comes out of the opening will be rather interesting too. I'm sure you will make this work, what ever it takes!!

    To be totally honest, it's so good to be able to sit back and watch you do all the work while we sit here and learn how to make one of these homebrew waterfall printers!!!! Cool tips thrown in of course from the audience, I guess its a team effort, but your doing most of the work. I'll be building one of these as soon as its perfected.

    I can't wait for that post with the first test vid Khalid.

    Thanks, See Ya!

  6. #46
    Join Date
    Apr 2006
    Posts
    3498
    hi Jason,
    I will be away for a week so the progress will be haulted. It is good to know you are also interested in PIC programming and waterfall. i will comeback to your post with detailed reply. now you only help us in sizing of capacitor and resistor with some neat sketch for installation of only one solenoid.
    Regards
    http://free3dscans.blogspot.com/ http://my-woodcarving.blogspot.com/
    http://my-diysolarwind.blogspot.com/

  7. #47
    Join Date
    Oct 2005
    Posts
    2392
    Hi Khalid, I have attached the schematic for one solenoid below. I hope it makes sense.

    To calculate the resistor;
    32v input;
    24v on solenoid 240mA
    12v on resistor 240mA

    resistor resistance;
    R = E/I = 12v/0.240 = 50 ohms (use 47 ohms)

    resistor power;
    P = EI = 12*0.240 = 2.88W (so use 5W resistor)

    Re the FET, I think using a 100v 40amp FET to drive a 32v 0.24A load is not the right choice. The solenoids will produce some high voltage spikes when they turn off, so a diode 1N4007 is needed (see schematic) and also you are better off with a high voltage FET for reliability.

    Because your application is high voltage and low current, something like a 600v 5A 2SKxxx FET would be best, depending what you can get locally that will be cheap.

    Capacitor value is still unknown and will depend a lot on the amount of energy your solenoids need to turn on fast with whatever water pressure etc you are using. If you don't have a scope you can do some testing with different cap values... I would probably start with about 220uF 50v cap, and maybe go larger uF if needed.

    Also you will probably need to maintain a constant water pressure inside your tubular manifold. If you maintain a constant pressure at the inlet of the solenoid valve you should get a constant flow of water over time through the valve orifice.
    Attached Thumbnails Attached Thumbnails capacitor location.jpg  

  8. #48
    Join Date
    Jun 2007
    Posts
    3757
    Circuit is close, but put a 36v or 48v zener in series with the diode, cathode to cathode, and the solenoid will turn off much quicker.
    As Roman say's use high voltage transistors, that can handle the inductive spike.
    Then no diode is required at all.
    With the diode there, the current will continue to flow for quite a period of time, while magnetic field is collapsing.

    Using a gate resistor, maybe 470 ohms, you can slow the transistor just enough so that it snubs the spike. To test this just start with a small supply voltage increasing until the spike reaches, say 80% of the transistor rating, and if you haven't got to your chosen supply voltage, increase the gate resistor and try again.
    The slower the switch off, the smaller the spike. Turn on speed doesn't matter matter, so no diode needed on the gate resistor.
    The gate should have an 12-15v Zener to ground, to protect the MOSFET. The fast changing voltage on the drain can capacitive couple destructive voltages to the gate.
    Super X3. 3600rpm. Sheridan 6"x24" Lathe + more. Three ways to fix things: The right way, the other way, and maybe your way, which is possibly a faster wrong way.

  9. #49
    Join Date
    Oct 2005
    Posts
    2392
    I *was* trying to keep it simple Neil, he needs to make like 40 of these things.

    I agree with adding the 36v or 48v zener in reverse series with the diode to give faster turnoff. And still use the HV FET.

    As for the FET gate protection it probably won't be needed due to the low currents being switched and the HV FET generally being pretty good spike wise. The FET gate is directly driven from a PIC output pin which will have quite low source impedance to either 5v or 0v and switching speeds in the nS range so FET gate voltage should be pretty much under control.

  10. #50
    Join Date
    Jun 2007
    Posts
    3757
    Yes Roman,
    I am accustomed to doing much more brutal >20A circuits.:drowning:
    I must think small again!!:banana:
    I suppose the gate resistor testing, would allow the use of existing 100v devices, sans zener and diodes.

    It will be interesting to see how fast this can be made to operate.
    A one channel birds nest test is really the starting point, don't you think.
    Super X3. 3600rpm. Sheridan 6"x24" Lathe + more. Three ways to fix things: The right way, the other way, and maybe your way, which is possibly a faster wrong way.

  11. #51
    Join Date
    Apr 2006
    Posts
    3498
    Hi Neil and Roman,
    Thanks for your brainstorming on the subject and sharing your experience... You peoples are really knowledgable and great... As i am away from my bench for couple of days so i can read your posts but unfortunately can't experiment...As my next version of this fountain will contain 200+ solenoid valves so i need this "40 valves version" optimized and complete in all respect... So, whatever you suggest for the fast switching of the solenoid i will get it experimented so keep throwing your ideas here please.

    What about IN4148 and IRF840 combination? I have both available in my locality.

    Again thanks for sharing your experiences with all of cnczone beneficiaries..

    PS:
    In the video i already installed IN4007 across the Solenoid to supress high voltage spikes and Happy to see majority of posts by Austrailians in this thread...
    http://free3dscans.blogspot.com/ http://my-woodcarving.blogspot.com/
    http://my-diysolarwind.blogspot.com/

  12. #52
    Join Date
    Jun 2007
    Posts
    3757
    Hi Khalid,

    Post your proposed circuit here, and we may be able to help.
    It is a bit tricky guessing how the parts will connect. The IRF840 is a good move. Roman ?

    Neil
    Super X3. 3600rpm. Sheridan 6"x24" Lathe + more. Three ways to fix things: The right way, the other way, and maybe your way, which is possibly a faster wrong way.

  13. #53
    Join Date
    Oct 2005
    Posts
    2392
    If you are using the diodes, 1N4007 will only be a tiny bit dearer than 1N4148 and is many times better. I buy 1000 quantity 1N4007 for about $5.

    IRF840; MOSFET N-CH 500V 8A TO-220AB
    Rds On (Max) @ Id, Vgs 850 mOhm @ 4.8A, 10V

    That looks good. Driving it with only 5v gate drive will make the Rds on about 1.5 ohms, which sounds high in FET terms but with only 240mA Id that's E=I*R = 0.240*1.5 = 0.36v Vds. Still looks good.

    For reliability I would use the IRF840, and as Neil says then you can probably eliminate the diodes on the solenoid coil. You could tack a 3300 or 4700 ohm 1/4 resistor across the solenoid coil, they will only draw 5mA in use and will give a big reduction in the spike voltage when the coil turns off. That will be cheap and avoids the cost of 2 diodes.

  14. #54
    Join Date
    Jul 2006
    Posts
    1256
    Sorry you guys have me lost on all the electronic stuff.I worked for a company that built a water wall for projection.Open top and atmospheric pressure was not enough.A tube with holes was used and a pump to 100PSI or so.Drilled holes were not good enough.I beleive they had to be drilled with a lazer.Hope this helps.
    Larry
    L GALILEO THE EPOXY SURFACE PLATE IS FLAT

  15. #55
    Join Date
    Apr 2006
    Posts
    3498
    Hi Neil, and Roman:
    Attached is the schematic i have drew in paint program, not pretty but understanadable ... The Shiftout Registers (SIPO) getting seial data from microcontroller and spitting out the data in 8-parallel bits. I have 5- 74HC595 installed and the microcontroller uses only three pins to control 40 data lines of 74HC595.
    The attached schematic includes the latest sketch posted by Roman for IRF540.

    Roman
    As you have suggested IRF840 for this applicationis and also suggested to remove IN4007 and install 4.7K 1/4 watt resistors. The Capacitor values are still unknown .. and as i do not have Oscilloscope so starting with 50V 220mF and increasing a little higher without seeing the effect of timing will not help me anymore .. Without Oscilloscope i can't judge the openinhg closing timing of my Solenoids..BUT,... I have prepared a test rig for single solenoid valve and i can check the drop formation and timing visually ..This will help me great...

    So as soon i reach to my working area i will experiment and will show you different videos with different combinations of the Capacitor..

    Neil, I am sorry i can now attach that bad hand drawn sketch and could not post unless i reach at the work place.. As this is OpenHardware project so in the last i will be able to share the Software, the microcontroller source code and answer the questuions...
    Keep posting and help me improving the PCB...
    Attached Thumbnails Attached Thumbnails shiftout schematic.GIF  
    http://free3dscans.blogspot.com/ http://my-woodcarving.blogspot.com/
    http://my-diysolarwind.blogspot.com/

  16. #56
    Join Date
    Jun 2007
    Posts
    3757
    Let's see what it does to the water.
    Super X3. 3600rpm. Sheridan 6"x24" Lathe + more. Three ways to fix things: The right way, the other way, and maybe your way, which is possibly a faster wrong way.

  17. #57
    Join Date
    May 2005
    Posts
    925
    Khalid, download ExpressPCB - Free PCB layout software - Low cost circuit boards - Top quality PCB manufacturing their program is free and very user friendly for schematics, then you can do the pcb too, only limitation is that you cannot print it because they want to use the service they provide. But you can screencapture and vector trace
    ● Distribuidor Syil en Argentina ● "www.syil.com.ar" ●

  18. #58
    Join Date
    Apr 2007
    Posts
    1955
    So, you are drawing me into another projects - hey Khalid and Pablo ?

    It looks very fun.

    I watched the original video a few times. Approximately time 1:54 you can begin to estimate the pressure used on the water. I didn't do a real calcuation, but it must be at least 5 bar, perhaps 10 bar to work. I also think they used real nozzles, not just holes.

    I had not before seen such a fountain, which is quite pretty to look at.

    Once you get the fountain itself working, then adding more LED colors will be easy.

    I am not sure, but I suspect that the solenoids must be getting driven hard both open and shut to operate at that speed.

    :wave:

  19. #59
    Join Date
    May 2005
    Posts
    925
    Harry :cheers:
    ● Distribuidor Syil en Argentina ● "www.syil.com.ar" ●

  20. #60
    Join Date
    May 2006
    Posts
    11
    I think that the water needs to be under considerable pressure so that it leave the nozzle with a good initial velocity that wont increase much as the water falls. If you rely on gravity, the speed of the water will increase considerably as it falls and will distort the image that you are trying to produce. The valves you have should be good for at least 4 bar, dont know about your plumbing though?

Page 3 of 17 1234513

Similar Threads

  1. A computer controlled etch a sketch for fun
    By jfong in forum Hobby Discussion
    Replies: 1
    Last Post: 07-05-2014, 07:04 AM
  2. Lets Build CNC Time Fountain
    By Khalid in forum Mechanical Calculations/Engineering Design
    Replies: 2
    Last Post: 03-08-2014, 12:49 PM
  3. Lets build a cnc robotic arm!!
    By patriciooholegu in forum DIY CNC Router Table Machines
    Replies: 30
    Last Post: 02-25-2012, 12:37 AM
  4. One axis (z-axis?) computer controlled testing device
    By TsThorsell in forum Stepper Motors / Drives
    Replies: 3
    Last Post: 02-07-2008, 07:38 PM
  5. Replies: 9
    Last Post: 06-17-2006, 06:23 PM

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •