Step 3: X Stage Frame and Drawer Glides

We built the frame from 1x4 and 1x6 pine - we used the fancy grade to limit the amount of warping etc. The length of the 1x6s was 40" to allow for the length of the belt plus a little extra room. The 1x4s were 38" long, but it looks like they could be 40" too. Maybe someday... Screw the 1x4s on top of the 1x6s with a couple drywall screws in each corner. If the screws get in the way when mounting the pulleys later, you can move them. The 1x6s are on top to allow the most room for the mechanics. Make sure to make the frame nice and square - measuring the diagonals works pretty well. Once the frame was complete, we mounted the timing belt pulleys and timing belts. We used a timing belt on the opposite side to help prevent racking as it moves, and it turned out that we only needed one shaft to go across, though we had originally used a long shaft on both ends.

Once we had that working and tested (OK, we cheated and did step 6 first :-)), we added the two drawer glide combos. As described in Step 1, each one was two drawer glides attached top to bottom to give a longer extension with less expensive glides. The ones we had allowed you to remove the top glide (of three), making mounting much easier. We used 6-32 screws and cut off any excess with a Dremel tool.
probley the worst CNC ever built. No z Axis? Servos? I bet it only has enough power to use a pencil on paper and thats it. There is no way this thing is doing anything useful! This is terrible!
<p>Hey cliffy,</p><p>Post a Picture of your design!! Carl, really nice work!</p>
<p>How do you coordinate the movements of the the two motors so one of them doesn't arrive at the destination earlier or later than the other..?</p>
You divide up the full move into move increments or time increments like 100 steps or 2 seconds etc. - those can be any number based on how fast the motors can move and how far you need to go. <br><br>Then you step through those and calculate the X &amp; Y position for the current step in the full path.<br><br>The sample code in this Indestructible has a simple version, based on time increments: http://www.instructables.com/id/Internet-Arduino-Controlled-T-Slot-XY-Table/<br><br>This one has similar code: http://www.instructables.com/id/Table-Sized-Arduino-Joystick-Controlled-T-Slot-XY-/<br><br>For stepper motors, you often use the longest dimension in steps.<br>
<p>Looks a little complicated, but thanks!</p>
<p>I saw this jig for routers, https://m.youtube.com/watch?v=ulblySFElo4 then i realized it could be controllable by servos, </p>
Could be - not sure how much torque is required for that setup. DC motors would work. It's a cool design, and relatively low cost, though! Lot's of fabricating special parts, though. Check out the T-Slot version of this project - very easy to make.<br>
<br> <br> Thank <br> you for this great tutorial! Thought I might share this website I came <br> accross which I found very useful:http://www.directindustry.com/industrial-manufacturer/xy-stage-76555.html
This is an awesome project, thanks for sharing.
Hey Carl amazing, i have an idea, and this will be very useful, <br>is there a way you can attach a powerful led light on the pointer end, and have it fade in and out? <br>the idea i have is to capture a lens flare. <br>as in do your camera move (in the computer) do a 2d track on the image of the light source.(track the cgi light source) <br>export the 2d tack coordinates to the XY table, <br>Have a camera pointing the other end. once the light is in position take an picture on the camera. <br>As to simulate depth/ occlusion the light source need to fade in intensity. <br>Problem i have, i have no idea how to program this for the XY table to read. any ideas?
Cool idea! Take a look at the T-Slot version of this project - there is a link to it in the intro. Since that one uses an Arduino, you can drive an LED with a transistor - just like the side lighting that project uses, but an LED instead of the Z-Axis servo. Maybe try to get an LED working and dimming with an Arduino first to see how it goes since you are learning? I have another instructable for permanent holiday LED lighting that shows how to dim an LED strip - same thing for an individual LED, and maybe even easier. There are a number of examples of a single LED dimming with an Arduino on the web. Good Luck!
Thanks - looking forward to seeing how it goes since I am curious about using stepper motors too.
I really enjoyed trying your sand CNC app--thanks! I am <a href="http://www.instructables.com/id/DIY-CNC-Router-Plans-How-to-Build/" rel="nofollow"><strong>building my CNC</strong></a> to use drawer slides and NEMA17 steppers.
Just had a thought; you could mark lines on the belt and use a light sensor to detect the light reflected. This method gives no wear at all and the parts are cheap. to get the markings accurate im sure you could use a contact transfer to mark the lines on.
could also use a rotary encoder at the shaft, gray code or something, it'd be like, the reverse of a stepper motor, unless the shaft jumped a whole rotation, you'd have an absolute position... same technique, but without worrying the belt will get the ultra tiny dots rubbed off. there's an ible for a jog shuttle made from a vcr head, the same printout dots and encoder would work perfectly.
Nice idea, but marking the belt would be a little tedious, to say nothing of what Skaar said, the possibility of them rubbing off, also makes it less ideal.. the contact transfer, would also need to be 100% accurate, even if using just 2 or 3 already-marked areas. Most inkjet printers, use a clear plastic disc, and yes, there is that danger of the shaft jumping a tooth, or two.. but those usually match to the splines on the gear, and have an 'End Of Travel' optical sensor at both ends. I imagine the multi-turn POT is also susceptible to the 'Jump a tooth' problem..
Indeed - I have added hose clamps on the tubing connecting the potentiometers since they can slip. There is a trade-off between locking everything down and allowing some parts of the system to slip in case something goes wrong. For a CNC machine, locking things down more may make sense since they are generally attended. For TeleToyland, they need to run for weeks or months without intervention. Having a stepper and a home switch may work - it can home itself periodically, thus adapting to small variations.
Please help me......, <br>Am a Engineering student from INDIA <br>i want to construct a Desktop CNC as my Mini Project <br>I have buy 3-axis Nema23 287oz-in stepper motor CNC kit and constructed my own frame with exess of 1000,700 &amp; 100mm in X,Y &amp;Z Directions. <br>i have some doubts in WIRING &amp; Mach3 operation........if u have any Manuals and tutorial videos please sent me. <br>My mail id: krissh0009@gmail.com.........Thank you.
hey arent potentiometers supposed to go only 180 degrees? in this case it goes round and round
Good catch. You are right - normally only one turn, but we are using 10 turn pots. It was mentioned in the intro, but we updated the parts list and this page to avoid confusion. Thanks!
Can u show us the circuit of this thing. how it been connecting to ueach other.
i used the same drawer glides too. they're perfect for the job. although they do dip if you go past their contracted length. they keep things neat and low cost when compared to aluminum rods and linear bearings.
Thanks. The drawer glides are still working on the web site, but I have an updated design that uses nylon screen door rollers on t-slot extrusions. The advantage is that they don't extend past the table when in use.<br><br>http://www.instructables.com/id/Internet-Arduino-Controlled-T-Slot-XY-Table/<br>
Love your innovation using &quot;off the shelf&quot; items! <br> <br>Grrreat job! <br> <br>
Great Instructable!!! Ive been working on a machine of my own but I have some questions about my driver board. Has anyone used the HobbyCNC EZ Driver Board Kit or the PRO series?
Have you considered Kerk? I was aware of them having very low cost linear guide systems about 5 yrs ago...
Hi, I am a new-registered member in your site... <br>I have a question here regarding to the PID X-Y table... <br> <br>How do you know the maximum distance that the plate on the X-axis and Y-axis can reach when the potentiometer (the potentiometer in your diagram) is in saturation (potentiometer reach maximum)?? <br> <br>If I am using another two potentiometers as the controller to control the X-axis and Y-axis movement respectively, when I tune the controller to its maximum, it is expected that the potentiometer that linked to the servo motor will reach maximum as well, vice versa. <br> <br>Hence, if my hardware is built without priorly knowing these information, it will happen that when I tune my controller to its maximum, the plate that move along the X-axis or Y-axis will reach to the edge before the potentiometer (the potentiometer that linked to the servo motor) reach its maximum if the area of the hardware is too small. <br> <br>So, how are we going to know the exact length and width of the X-Y table that enable the potentiometer to rotate freely from 0% to 100%?? <br> <br>Please be tolerated while reading due to my poor english, haha. <br> <br>Thanks a lot, hope to hear from you soon.
CNC news! DIY CNC hits mainstream in an O'Reilly Radar tech blog report today. See<br><br>http://radar.oreilly.com/2010/12/diy-fabrication-hits-a-new-pri.html
The linear bearings could be a smooth flat metal track, with the carriage mounted to a standard ball bearing.
i run the bs2 stamp chip i got over 60 servos around&nbsp; i been trying to find a way to run servo's instade of stepper motor&nbsp;&nbsp; im working on a nothere project cnc tipe&nbsp; but i don't won't to run a 300 board i like to find a nothere way to run it or convert the servo wire's to run with a home made board
Very creative way of doing CNC ;-)<br /> <br /> ...but I still believe that using steppers is the way to go. <br />
I have an idea here...will it work&nbsp;fine if i replace the drawer slider with a robot castor&nbsp; attach under the top X-axis bar?
<p>if i were going to use roller-chain-sprockets instead of timing belt, is that possible to clamp the chain like what you've done here?</p>
Maybe you could even just put a screw through one of the links?&nbsp; We could have done that with the timing belts too, but we were reluctant to put a hole in the timing belt.<br />
another question here. in this step 5 (picture number 2), does the small wood block, which attach on the Y-axis top glide, glue/screw together with the X-axis ?
Yes, I think its described in Step 4 - we attached small blocks to the X Axis glides, then the Y axis board to the tops of those blocks.&nbsp; That was so the Y Axis was above the X Axis timing belt.<br />
I would like to use your table design in an upright position, i.e. vertical.&nbsp; I cannot tell whether the design is mechanically consistent with this configuration.&nbsp; Thank you!!<br /> &nbsp; <br />
Thank you!&nbsp; I will let you know how it goes.<br />
It should be since the drawer glides work in most orientations, but the weight of the Y axis would be a fairly big strain on the X axis.&nbsp; Maybe you can add a counterweight to it? <br />
very cool! i would love to see how you did your web interface with the SSC-32 board.
We use a SitePlayer Telnet device to go from the Ethernet to the serial port.&nbsp; That let's us use simple PHP code to send commands via telnet.&nbsp; Then, we connect a Lynxmotion SSC-32 right into that serial port to drive the servos.<br />
<p>Hi,carls, i have a few Questions here:</p> <p>Does the Y-axis Drawer Glide system which you did here slide from the bottom to the top of the whole mechnism? I dont understand how did you make the drawer glide slide in the position 'bottom-top'?what does it mean by 'bottom-up'?<br /> <br /> As my project is about a pick and place&nbsp;mobile robot,so i would like to make the Y-axis&nbsp;slider to slide from the bottom up to where&nbsp;the top end. Could you teach me?&nbsp;</p>
Both axes use the same approach:&nbsp; we took two drawer glides and bolted the thin sides (the part that gets connected to the drawer) together, so that leaves the two wider cabinet sides out.&nbsp; We did this just to save the cost of buying a really long drawer glide.&nbsp; Another way to go is get linear bearings from a place like http://www.vxb.com.&nbsp; We may try that on a newer version of the TeleToyland Sandbox.<br /> <br /> Just a note that this system is very low precision, so it may not be well suited to CNC apps.<br />
I do not want to hijack your comments here but I have a few projects in mind, and you and your commentors seem to be the right people to ask.<br/><br/> I want to build an X, Y, Z, stage for a webcam-based microscope.<br/> It does not need to move more than a few inches in each of the axii (sp.?) so I was going to use the servos themselves to move them (as in the same way they would be used in a model plane). This should give me the minute control I need as well as a &quot;center&quot; position and joystick control.<br/> I will use the throttle channel for the z (zoom in this case) axis so that the zoom can be set and left, and the other stick for x and y respective to the stick so it will be logical to control the thing from the RC transmitter. I won't be able to control the focus, even though I have one more channel, because I have no way of connected a servo to the focus knob at this time.<br/> I am using an XBox360 webcam modified for higher focus control. I have no way offhand of measuring the total zoom, but with the focus adjusted all the way out (close) you can focus on an object about a half inch away from the camera. if it works well I am sure I can add more lenses later.<br/><br/>Another project I am working on is a very large Radio Control Truck, based around a Honda 1000 watt generator. I know it will be difficult and odd to use the generator as the power source, but I have my reasons. I need two POWERFUL motors and controllers because I want it to be a tow truck and powering the rear wheels individually allows me both the ability to switch between series and parallel wiring, as well as making it unnecessary to use a differential.<br/> I also need to build and run a winch capable of ... I don't know, 500 pounds?<br/><br/> The truck will be based on a 1920's chain-drive, Mack &quot;C-Cab&quot; and will have multiple beds that bolt or pin in place and will include a tow truck, a 5th wheel (semi truck or tractor trailer), a flat bed, and possibly a dump bed or &quot;rollback&quot;.<br/><br/> I am taking sponsors for the truck if anyone is interested.<br/><br/> <a rel="nofollow" href="http://diecastoms.blogspot.com">http://diecastoms.blogspot.com</a> if interested, but I haven't been updating it as often as I should.<br/><br/> Thanks for reading my comment and not flaming me ;P<br/> Mike from &quot;DC&quot;.<br/>
Why not just use DC motors if you're killing the pot?
You are right. As noted in the Instructable, we are using a separate DC motor on the X axis currently since the system is in such heavy use on TeleToyland. We still need the PID control, and prefer the servo signal control, so we are using a board from Pololu that works like the hobby servo electronics. But, it's worth mentioning that the hobby servo approach does work fine. We are using it for the Y axis, and we used high speed + high power servos just to make the system work faster. Using regular servos (maybe a high power one for the X axis) does work fine, so the timing belt servo control is a nice low cost way to get lower precision linear motion. Using DC motor drivers would be great too - just gets into more cost like traditional CNC machines.
Wow simple enough for me lol if i get some cash i may maje one of these will be great but i suppose you could wire a dremel to it And run a router bit Through polystyrene or something Great rated 5/5
Great 'Ible! How much do you think you spent total on this project?
Off the top of my head: timing belts: $12 x 2 timing belt pulleys: $7 x 2 shaft collars $3 x 8 bearings: $9 x 8 - the patio door glides may be a less expensive solution 1/4" shaft: $10? drawer glides: $15? x 6 wood: $20? misc hardware etc.: $50? servos: $80 x 2 for the fast digital ones. $15-$30 ones worked too, just slower. potentiometers: $15 x 2 for the wirewound ones - worked fine but we have ordered $40 hybrid ones, hoping they work better. Hope that helps. As usual, we spent more with parts we eventually didn't use, and we could also probably find even less expensive ways to do some of the above.
i'm building a cnc mill and its gonna cost me 30 for the x y an z and thats counting the stepper motors and the router i can get for free and the stepper driver for 50 i will stick to my cnc mill but yours works great if i was rich i would make it your way hahaha

About This Instructable


569 favorites


Bio: A Maker since childhood with all the classic symptoms, a robot builder, and an Internet software CTO by day.
More by CarlS: Smoking No-Chemical Pumpkin Quick Digital LED Jack-o-lantern Light Free-form Soldering LED Tree
Add instructable to: