Electronic home switches made easy! - Page 2


Page 2 of 62 FirstFirst 123451252 ... LastLast
Results 21 to 40 of 1237

Thread: Electronic home switches made easy!

  1. #21
    Member jalessi's Avatar
    Join Date
    Feb 2007
    Location
    United States
    Posts
    4553
    Downloads
    1
    Uploads
    0

    Thumbs up

    RomanLini,

    I should have read the entire thread, sorry I missed several posts.

    Excellent repeatability, cant ask for better than that.

    Thank You,

    Jeff...

    Patience and perseverance have a magical effect before which difficulties disappear and obstacles vanish.


  2. #22
    Member
    Join Date
    Apr 2006
    Location
    na
    Posts
    1016
    Downloads
    0
    Uploads
    0

    Default

    question are the very smalll ear earth magnet too strong for this?
    would they still work? any feedback on using them?
    the resistor is for led or switch bounce pull up?
    or the r values changes the gap setting?


    do u have a video with it working with the plastic mounts....the led goes on when switch triped?
    the new mounts do they use the led?



  3. #23
    Registered
    Join Date
    Oct 2005
    Location
    Australia
    Posts
    2392
    Downloads
    0
    Uploads
    0

    Default

    Quote Originally Posted by eloid View Post
    1 question are the very smalll ear earth magnet too strong for this?
    would they still work? any feedback on using them?
    2 the resistor is for led or switch bounce pull up?
    3 or the r values changes the gap setting?
    4 do u have a video with it working with the plastic mounts....the led goes on when switch triped?
    5 the new mounts do they use the led?
    1. Possibly, the stronger magnets i tested triggered the sensor from further away. There is no benefit to that. Because of the higher strength magnetic field everything is increased; sense distance, hysteresis distance, error.

    My thoughts for "perfect" magnets would be small weak magnets. They pick up less crud and will give better accuracy and repeatability.

    Even the weak bar magnets I used trigger the hall switch from 5mm to 7mm away, if I had used weaker magnets still that triggered it from say 3mm away the accuracy would be doubled.

    2. The resistor is for "pull-up" use, it makes the sensor output a +5v when no magnet is near, and output 0v when magnet it near. It turns the sensor into "logic-level" output that can be tied directly into any logic input and means you never have to fuss with resistors later.

    3. No, the resistor does not change the gap setting.

    4. Sorry I don't have a good video camera right now, I will take a video at some point in the future.

    5. Thank you, that is a REALLY COOL idea! If the LED is soldered onto the sensor and clear plastic is used for the mount then it will light up when the magnet is near!

    Apart from the cool fun factor that would be really valuable for setup and fault finding and would really improve the design!



  4. #24
    Member
    Join Date
    Apr 2006
    Location
    na
    Posts
    1016
    Downloads
    0
    Uploads
    0

    Default

    Quote Originally Posted by RomanLini View Post
    1. Possibly, the stronger magnets i tested triggered the sensor from further away. There is no benefit to that. Because of the higher strength magnetic field everything is increased; sense distance, hysteresis distance, error.

    My thoughts for "perfect" magnets would be small weak magnets. They pick up less crud and will give better accuracy and repeatability.

    Even the weak bar magnets I used trigger the hall switch from 5mm to 7mm away, if I had used weaker magnets still that triggered it from say 3mm away the accuracy would be doubled.

    2. The resistor is for "pull-up" use, it makes the sensor output a +5v when no magnet is near, and output 0v when magnet it near. It turns the sensor into "logic-level" output that can be tied directly into any logic input and means you never have to fuss with resistors later.

    3. No, the resistor does not change the gap setting.

    4. Sorry I don't have a good video camera right now, I will take a video at some point in the future.

    5. Thank you, that is a REALLY COOL idea! If the LED is soldered onto the sensor and clear plastic is used for the mount then it will light up when the magnet is near!

    Apart from the cool fun factor that would be really valuable for setup and fault finding and would really improve the design!
    try a blink led.... ie red or blue?

    if some where in the sensor there was a cicrcuit to change the gap range it would be cool
    ie a pot.... for gap triming or adifferent hall sensor that has that feature

    Last edited by eloid; 03-25-2010 at 07:52 PM.


  5. #25
    Registered
    Join Date
    Oct 2005
    Location
    Australia
    Posts
    2392
    Downloads
    0
    Uploads
    0

    Default

    I'm curious why you would need adjusability? Surely all control software has the ability to adjust for the home switches, there's no need to adjust at the switch itself. Please explain if there's a need to adjust the sensor setpoint, they are very sensitive so any mechanical or electrical adjustment would be difficult to get accurate to 0.001"! It just seems better to do in software...

    One of the big advantages is that these switches don't have to be placed at the end of travel because the magnet doesn't contact them. I'm thinking of mounting my home switches in tne MIDDLE of the travels, to set zero there (in software), and then the maximum error from the ballscrew thermal expansion will be halved ie +0.5/-0.5 instead of +0/-1.0.



  6. #26
    Member Bubba's Avatar
    Join Date
    Mar 2004
    Location
    lagrange
    Posts
    1804
    Downloads
    27
    Uploads
    0

    Default

    The problem with putting the home switch in the middle of the travel is then the software doesn't really know which direction to go "home"! This is one of the main reasons to put it at the end of an axis. Then you always know to go in either the plus or minus direction.

    Art
    AKA Country Bubba (Older Than Dirt)


  7. #27
    Member
    Join Date
    Apr 2006
    Location
    na
    Posts
    1016
    Downloads
    0
    Uploads
    0

    Default

    I'm curious why you would need adjusability?

    if magnet is to strong or weak, you may get dirt grime get on limit point you may want a samller gap.... thus .. ajust sensity to trigger the limit
    switch etc after you bolt it on .........

    Surely all control software has the ability to adjust for the home switches, there's no need to adjust at the switch itself. Please explain if there's a need to adjust the sensor setpoint, they are very sensitive so any mechanical or electrical adjustment would be difficult to get accurate to 0.001"! It just seems better to do in software...

    One of the big advantages is that these switches don't have to be placed at the end of travel because the magnet doesn't contact them. I'm thinking of mounting my home switches in tne MIDDLE of the travels, to set zero there (in software), and then the maximum error from the ballscrew thermal expansion will be halved ie +0.5/-0.5 instead of +0/-1.0.

    Last edited by eloid; 03-26-2010 at 01:34 PM.


  8. #28
    Registered Jesse B's Avatar
    Join Date
    Aug 2009
    Location
    Canada
    Posts
    392
    Downloads
    0
    Uploads
    0

    Default

    These are very cool! I'm considering going with some of these rather than their mechanical counterparts...

    I also think an LED would look pretty cool



  9. #29
    Registered
    Join Date
    Oct 2005
    Location
    Australia
    Posts
    2392
    Downloads
    0
    Uploads
    0

    Default

    Thanks for the feedback people!

    LED
    I've been thinking about the LED, if the 3.5mm hole for the wire is drilled further, into the inside pocket wall, then a 3mm LED could be inserted there. The LED would still be enclosed in the pocket, but under the hall switch and it's light would shine out directly opposite the wire, on the opposite side of the mount.

    I think the LED shining out in that direction is best, as the mount will be under the magnet when the LED lights up (and probably under part of the machine) but the LED will be visible shining out the end of the mount.

    Update
    The home switches on my machine are going well, I've cut a few complex jobs on it including toolchanges, stopping for dinner and turning the machine off, then when turning the machine back on the home sensors showing machine has not skipped any stepper steps. I turned the machine off again to work for an hour on CAD, then back on and re-homed it and kept cutting the same job, new CAD pattern, exactly the same place on the job.

    It's going to be nice to be able to make jigs for regular jobs, and know I can come back to a job and re-run it at any time and the machine can be relied on to cut in exactly the same spot on the jig etc.

    Perfect repeatability (give or take some thermal expansion) is pretty cool!



  10. #30
    Member dertsap's Avatar
    Join Date
    Oct 2005
    Location
    canada
    Posts
    4230
    Downloads
    1
    Uploads
    0

    Default

    brilliant idea , I've been trying to come up with a homing solution for my 4th axis ,I was going to go reluctantly with a proximity switch but this looks like a much better solution

    A poet knows no boundary yet he is bound to the boundaries of ones own mind !! ........


  11. #31
    Member
    Join Date
    Apr 2006
    Location
    na
    Posts
    1016
    Downloads
    0
    Uploads
    0

    Default

    now that you have a cnc working i woukd get a small v bit and make a small pcb using surface mout resistor and 2 surface mouted high intenesty "flashing" blue or red smt led and make mounting pads for ss441 hall and connections for cable..


    then encapsel in clear plastic resin. the pcb would no bigger than .625x.625



  12. #32
    Registered
    Join Date
    Oct 2005
    Location
    Australia
    Posts
    2392
    Downloads
    0
    Uploads
    0

    Default

    Quote Originally Posted by eloid View Post
    now that you have a cnc working i woukd get a small v bit and make a small pcb using surface mout resistor and 2 surface mouted high intenesty "flashing" blue or red smt led and make mounting pads for ss441 hall and connections for cable..

    then encapsel in clear plastic resin. the pcb would no bigger than .625x.625
    I appreciate you suggestion Eloid, but I don't think that is the best way to do it. I work as an electronics product designer, and use SMD all the time.

    But in this case SMD would be a step backward. That requires a PCB so that adds cost and/or hassle, and makes it harder for some people, remember this is an "easy open-source" design for anyone to build. There's no gain in time from SMD on this simple device, it's just as easy to solder a resistor and LED wires as it is to solder a resistor and LED SMD, actually it's probably easier to solder the wired versions for most people.

    As for the flashing LED, I see that would add a fun factor. But the main real use of the LED is to set the sensor up, and a non-flashing LED comes on immediately when the sensor is triggered. But there is nothing to stop someone using a flashing LED if they prefer it, that's what open-source is about.

    I'm not sure the size thing is too relevant either. I made 2 sizes thinking the smaller one would be better. After they were finished I prefer the larger one; larger stronger screw mounts, less likely to break or move when cleaning with a brush etc etc. It seems that most people have large MDF routers and a sensor 1" x 1.25" with 2 strong screw holes is a pretty reasonable size. If people REALLY want a tiny sensor they can just solder 3 fine wires to the hall sensor and glue it to their machine with a tiny spot of 5-minute epoxy.



  13. #33
    Member
    Join Date
    May 2006
    Location
    Australia
    Posts
    2420
    Downloads
    0
    Uploads
    0

    Default

    Nice work Roman, thanks for sharing, this is definitely on my "to do" list for my next machine.

    I was thinking of using a proxomity switch along with the Z index pulse on the servo encoder for homing but it looks like your setup works as well and is nice and simple.

    Russell.



  14. #34
    Registered
    Join Date
    Jan 2010
    Location
    USA
    Posts
    110
    Downloads
    0
    Uploads
    0

    Default

    Im thinking that I may use this in the design of my router.



  15. #35
    Member
    Join Date
    Apr 2006
    Location
    na
    Posts
    1016
    Downloads
    0
    Uploads
    0

    Default smt soldering is easy when you use glue dots

    I appreciate you suggestion Eloid, but I don't think that is the best way to do it. I work as an electronics product designer, and use SMD all the time.

    I work in electronic industry also ...with a cnc a vbit and some 1oz copper clad pcb it would fun to easy and very professional look.... ( if didnt have a cnc your sol) when I get some free time i design a pcb .... but for those that are interested here a little electronics 101 . You just need smt devices.. pre tin the pads of your cnc made pcb brd where compont are to be mounted... use some crazy glue ... place a small pin drop under the device use needle nose or tweezers to place and hold few sec ... let it dry ....then use heat gun or hair dryer to reflow the pre tined pads and glued device.... and your done. they will "not move" as you reflow just dont over heat them...
    and the solder will reflow so nice... lol when you see solder starting melt remove heat...


    But in this case SMD would be a step backward. lol how do you figure that????????? most people have a cnc... already.

    That requires a PCB so that adds cost and/or hassle (you mean fun) ..., cost ...? cost $2-$5 buck tops . Thats few buck for 6x6 sheet of blank copper double side brd 1 oz. which you could easy make 20-30 pcs


    that works out to less $0.20 cent per pcb ( for a diy pcb) so cost is not a issue... lol and to reflow with heat gun or hair dry way easyer than solder iron... trust me been doing for years... no fancy equipment required.

    "easy open-source" design for anyone to build........ I know most people can reflow 30 brd faster than solder 30 brd.. trick is the small glue dots.

    There's no gain in time from SMD on this simple device, ( i dissagree lol... fun factor,, and 2 smt device is not hard?)

    it's just as easy to solder a resistor and LED wires as it is to solder a resistor and LED SMD, actually it's probably easier to solder the wired versions for most people. (read above again. lol)


    As for the flashing LED, I see that would add a fun factor. But the main real use of the LED is to set the sensor up, and a non-flashing LED comes on immediately when the sensor is triggered. BUt its eye catch attention and you want that!!!!!!!!!!.... and cost is the same.. most case cost is the same at a surplus electronic stores

    But there is nothing to stop someone using a flashing LED if they prefer it, that's what open-source is about.

    I'm not sure the size thing is too relevant either. I made 2 sizes thinking the smaller one would be better. After they were finished I prefer the larger one; larger stronger screw mounts,

    I would make it the same foot print as a mechanical switch with same mounting holes. easy upgrade feature.

    Last edited by eloid; 03-28-2010 at 07:46 PM.


  16. #36
    Registered
    Join Date
    Oct 2005
    Location
    Australia
    Posts
    2392
    Downloads
    0
    Uploads
    0

    Default

    I'm interested to see the SMD design you come up with Eloid, please post pictures etc in the thread.

    You might be right that some people would prefer an SMD version, to my mind a simple wired version is better but we all have different ways we like to make things.



  17. #37
    Registered
    Join Date
    Oct 2005
    Location
    Australia
    Posts
    2392
    Downloads
    0
    Uploads
    0

    Default

    Mounting block with LED

    I got a couple hours free today so I drew up my improved mount block design with the added LED indicator.

    This diagram should have enough info, dimensioning etc for anyone to make these mounts. Sorry I don't use any standard CAD program so I can't offer CAD files or DXF, my machine works direct from Corel so I just work in 2D or write C code to do complex 3D cutting. Maybe someone who uses the popular CAD program (I don't even know which CAD program is most popular here!) can whip up a CAD design.



    Cutting info.
    The mount is cut from 10mm (3/8") clear acrylic, using a 1/8" (or 3mm) diameter end mill capable of plunge and cutting the full depth of the sheet.

    1. Cut the shallow pocket 24x11 mm (2mm depth)
    2. Cut the deep pocket 20x11 mm (to a total 7mm depth)
    3. Cut the screw relief step, 30x10 mm (4mm depth)
    4. Cut the 2 screw holes or elongated slots as desired
    5. Cut around the mount 30x25 mm (to remove it from the sheet)
    6. (in drill press) drill a 3mm hole for the wire (and deep enough for the LED if LED is to be used).


    Mount design inprovements

    This mount has a few small improvements from my design further up the page;
    1. larger and thicker screw area for strength
    2. shallow area for Hall switch is full 11mm width now; for better epoxy flow and less epoxy shrinkage there around the sensor
    3. LED hole, is just a continuation of the wiring hole
    4. deep pocket is not quite as deep now; saves epoxy, stronger, less shrinkage

    At this point I can't see how to improve it much. Apart from redesigning it for special needs like miniature size, or changing the screw mounts to suit the equipment it looks pretty much done to me.


    How the LED works

    The design uses a 3mm LED, this is the smaller size (other popular size is 5mm). The 3mm size has become the most popular size these days and should be available anywhere cheap.

    It won't matter too much which LED colour you choose, you can still use the same resistor values. You can even use different colour LEDs on different axes.

    It would be best to use a "high intensity" LED, these cost maybe 50 cents more each but are MUCH brighter so will be much more visible when buried down between moving parts of machinery.

    The resistor values shown below only run about 4mA through the LED, so the old style LEDs will not be very bright at that current but the new high intensity LEDs will be very bright at 4mA as they are really "high efficiency" as well as high intensity.


    Other mount materials

    If you want to use a different plastic than clear acrylic, like you may choose a high strength plastic like Delrin or just use plastic you have handy then it will be more difficult with the LED.

    In the clear acrylic, the LED will light up the whole mount and be very visible.

    But if using a black or coloured plastic, you have 2 options;
    1. drill deep enough so the end of the LED actually comes out of the mount (this requires sealing before pouring epoxy)
    2. drill ALMOST through, so there is only a thin skin of plastic in front of the LED

    I would try method 2 first! Most plastics (even black) are quite translucent when they are very thin like 0.5mm (0.020") and the high intensity LED will shine out easily. If not, then just continue the hole all the way through.



  18. #38
    Registered
    Join Date
    Jan 2010
    Location
    USA
    Posts
    110
    Downloads
    0
    Uploads
    0

    Default

    Guess what I am doing tomorow! Awesome job!!!!!!



  19. #39
    Registered
    Join Date
    Oct 2005
    Location
    Australia
    Posts
    2392
    Downloads
    0
    Uploads
    0

    Default

    Wiring up the LED mount

    Below is the wiring diagram for the new mount, the left version is just the hall switch and a pull-up resistor. The version on the right has an added LED and one more resistor. Both use the same mount block, so the LED is optional.





    Construction method;
    1. trim and bend the hall switch legs
    2. glue the hall switch with a spot of superglue
    3. (LED option) trim the LED legs
    4. bend the K leg of the LED as shown
    5. glue the LED in the hole with a spot of superglue
    6. trim the wires and pre-solder their ends
    7. insert the wires and solder to the hall switch
    8. trim and bend the resistor legs as needed
    9. solder the resistor(s) to the other parts

    At this point the sensor can be tested, just connect +5v and ground to the RED and BLACK wires and when a magnet comes near the LED with light up.


    Filling with epoxy

    If sensor tests ok, put hot melt glue around the wire to stop epoxy leaking out. You can use modeling clay for this but use a LOT and make it physically supported to the mount AND a base plate so nothing moves until the epoxy sets, this may take 24 hours. I much prefer the hot melt glue. You can also use silicone (silastic) to seal it but this takes longer to set and is harder to remove than hot melt glue.

    Fill pocket with epoxy or 5 minute araldite. The 5 minute araldite is much thicker and won't tend to leak as bad, but it will be yellow and less attractive and is typically a low quality material with less sealing ability and worse temperature properties. Other 2-part epoxies will take longer to set, and can leak if not sealed well, but when fully set are superior.

    Don't use polyester resin (the stuff you put a few drops of hardener in). This has poor shrinkage properties and will often pull away from the sides of the pocket allowing moisture in or allowing the hall switch to vibrate or move. It can also be corrosive to wiring and components.



  20. #40
    Registered
    Join Date
    Oct 2005
    Location
    Australia
    Posts
    2392
    Downloads
    0
    Uploads
    0

    Default

    Quote Originally Posted by Ernie34 View Post
    Guess what I am doing tomorow! Awesome job!!!!!!
    Thanks Ernie!

    Please post pics of anything you make, especially changes and improvements.



Page 2 of 62 FirstFirst 123451252 ... LastLast

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
  •  


About CNCzone.com

    We are the largest and most active discussion forum for manufacturing industry. The site is 100% free to join and use, so join today!

Follow us on


Our Brands

Electronic home switches made easy!

Electronic home switches made easy!