Effer, a spare parts story.

Discussion in 'CNC machining projects' started by BattyZ, Jun 21, 2021.

  1. BattyZ

    BattyZ Silver Banner Member

    I have been thinking about this. Belt slap may be an issue if they are undersized. I am aiming for a 3/4 to 1 in wide timing belt so should handle the 850 oz-in torque from the 10:1 stepdown I am currently planning. I could also do 20:1 reduction and that would still give me 150 rpm...did I mention the C axis should be continuous? Might have the option of being a mini vertical lathe if I can figure out the cam.
     
  2. Monty

    Monty Silver

    I'm definitely in the right place....nice. :D
     
  3. BattyZ

    BattyZ Silver Banner Member

    Well yes, this is thee Home Foundry Forum. lol And thanks!

    So after way too many hours scouring fleabay for harmonic drives and right-angle gearboxes under 10 arc-min backlash that I could possibly afford... I decided to try my luck with timing belts this second go around. On the head-head combo the belts did really well once tensioned, I feel like all of my play was coming from the single-bearing setup for the 4th and 5th. I guess we are going to find out!

    Screenshot 2021-12-12 233338.png Screenshot 2021-12-12 233424.png Screenshot 2021-12-12 233457.png Screenshot 2021-12-12 233534.png
    Some thoughts:
    -Current pulley ratios are 10:1 which yields 850 oz-in on the driven side. Have room to go bigger underneath
    -Belts will be T5 timing belts 16mm in width.
    -Yeah I know the C axis servo is just plopped right in the middle of the way. I struggled with placement on this for a while. Really wanted to do a right angle gearbox or a twisted belt setup to get parallel with the bed but simple is best for starting.
    -After modeling, I don't think it will be too much of an issue, there will be a vice or work holding of some sort lifting the workpiece up anyway, might be able to clear with long pieces.
    -That circle platten in the bed rotates if that wasn't clear, it is 15 inches in diameter.
    -Both rotary axis have paired and preloaded angled-roller bearings for concentricity and rigidity.
    -I have additional skateboard bearings on the driveshaft so the servo bearings won't see much load at all.
    -Limit switches will be fixed to the Z-axis and moving bed so no moving wires there
    -Only moving wires are the power leads for the spindle and maybe spindle control, thought about utilizing a clockspring here (from an automotive steering wheel)
    -Since I will be running a cable tray to the moving bed I might move to a rotating or driven ballscrew underneath the table
    -I have yet to design the casting connecting the linear rail slides to the bed
    -In real life, the gantry is placed more towards the back of the rails so I have full access to the bed in the Y direction, 35 inches of travel at the very maximum IIRC.
    -The bed is slated to be a for part casting
    -Trying to make most of these pieces using the LF casting process. Both servo mounts will be machined from stock. Cableways 3d printed.


    Thoughts are welcome, better ideas are even more welcome.

    Oh! Here is a photo from 4 years ago today on where my CNC journey was...
    Snapchat-1516382853.jpg
     
  4. BattyZ

    BattyZ Silver Banner Member

    P.S. I have even started provisions for way covers. I know, not really my style. Thought they might be nice for a change.
     
  5. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    It's quite the piece of motion control and sizable work envelop. Forgive me for saying, but looks like it might not be the top of the chart as far as rigidity. What are you building it to cut? Foam? ...and what/why do you need all the axis and freedom of motion?

    Best,
    Kelly
     
  6. BattyZ

    BattyZ Silver Banner Member

    Yeah as it looks the gantry setup is a bit weak to the eyes. The cad is more of a representation. Here it is welded up.
    20211213_111252.jpg

    It surprised me with how rigid it turned out. I can always redo it with bigger thicker, maybe epoxy granite filled, metal later. The final goal is an epoxy granite-style fixed gantry. This will be a not-on-casters setup. lol

    Mold-making materials will be the primary focus. Foam, MDF, composites, wood. Aluminum is still on the table but mostly as flat stock and will build a big riser table. The Z-axis will be as close to the linear bearings as it can be.

    I agree with your doubts. The overarching question here is will it work? Will the final tool-to-workpiece tolerance be tolerable....toleranceable??

    That humble pile of metal and acme screws I posted earlier did do nice work. A guy would not think so looking at it. Each one of these pockets took about an hour, if one of them was off the board was ruined. I remember clutching a hand oiler can and the air gun...

    Snapchat-1471971470.jpg

    As always, time will tell. Thanks for the question.

    -Batty
     
  7. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    It's asking a lot out of that z-axis, but like you say, if you have a convenient way to raise the table when the z envelope isn't needed, may reap more rigidity and still have the work envelop available when needed. Not a criticism, just a matter of being fit for purpose. Heck I'm machining foam and that's a joke........thank goodness or I would have broken a bunch of tool bits by now LoL!

    Best,
    Kelly
     
  8. BattyZ

    BattyZ Silver Banner Member

    I hear this! When I make mistakes vutting metal on my big cnc all I hear is an audible *tink* and I know immediately I have made a mistake. With the cnc router in foam, can just hear the router bog down more. Lol
     
  9. BattyZ

    BattyZ Silver Banner Member

    Monday night this week I met up with a programmer friend of mine. We worked through converting a similiar kine.atica file to the one I would need. Tom from Dynomotion actually gave us the changes. He has been very helpful! Got the whole solution to build.

    So now have the correct kinematics file for the machine!

    Software wise only the post processor is left to sort out. Hardware wise I have but just begun...the final 2 servo motors are on their way and the new board has arrived.
     
  10. DavidF

    DavidF Administrator Staff Member Banner Member

    I haven't had a " crunch" in over a year!!
    Then again, I haven't powered up my mills in a year either :oops:...
    Hoping to remedy this soon....

    Edit, not the crunch part!
     
  11. BattyZ

    BattyZ Silver Banner Member

    "DO IT. DO IT NOW. GET TO THE CHOPPA." haha
     
    DavidF likes this.
  12. BattyZ

    BattyZ Silver Banner Member

    Was able to sneak in a little time to get one of the pieces ready for coating and casting.
    Resized952021121795154505(1).jpg
    Should help stiffen the Z axis significantly. Below is a picture of what made me make the jump to closed-loop servos...
    Resized952021121795120852.jpg
    I think I restarted this side 3-4 times before giving up. Had I slowed the speed way down I could have possibly made it but then it would have take over 2-4 hours. This is when I started to calculate the actual torque (or lack thereof) of the steppers at the speed I was trying to move the axis to machine foam. Even my big stepper could barely whip itself out of a wet paper bag.

    So far the servos have been a dream. They are barely warm after cutting for hours. Mind you that there are little to no cutting forces at play. There are 2 more servos in the mail for the 4th and 5th! I need to get around wiring all the alarm outputs together so if something would go wrong they all stop together. This will most likely wait for the dynomotion setup.

    Should have some free time coming up to make progress this week!
     
    Al2O3 likes this.
  13. DavidF

    DavidF Administrator Staff Member Banner Member

    I'll be working on it soon. Swapping in a smooth stepper either net drive and an mpc.
    With any luck I'll be finished by 2023 :confused:
    20211220_105805.jpg
     
  14. BattyZ

    BattyZ Silver Banner Member

    David, that MPG will be nice! let me know how easy it is to get going.
     
  15. DavidF

    DavidF Administrator Staff Member Banner Member

    I'm hoping it goes easy. It's supposed to wire right up to the eithernet board I have.
    I still have a few more weeks of overtime to deal with before I get a chance to play with it.
    Heading in now for another 16 hour shift :confused:
    SmartSelect_20211220-154109_Chrome.jpg
     
  16. DavidF

    DavidF Administrator Staff Member Banner Member

    Last edited: Dec 25, 2021
  17. BattyZ

    BattyZ Silver Banner Member

    Had some unexpected somewhat-success casting some parts yesterday.
    20211223_124246.jpg 20211223_163702.jpg 20211223_175442.jpg 20211223_175459.jpg

    Unexpected, because the big pour I ran short of metal. Singed some leg hairs too! (Aluminized pants have been ordered.) The first little pour went great, it is also the cast with the best surface finish. The last one I did not have the pouring basin secured enough and it floated during the pour. It rotated on me so I just kept pouring to the side and the metal went around. Good thing that stuff is non-wetting!

    I still have work to do on my coating and overall process but nice to see some results. Going to change the way I handle crucibles, the first pour I ran short because I currently have to support the crucible and charge with one arm, which is typically above my shoulders for the duration. Moving towards a two handed, two arm approach. I will start a thread with the kiln kart upgrades & tooling that will allow this. Should facilitate way easier pouring for bigger castings. I am a big guy (6ft+), should be able to comfortably pour more than I am right now and that is due to the current setup.

    Quite excited about the piece that turned out really well. On the other piece, the tooth profile had some sections that are ready-to-use from the casting but will need to be post machined. I have a new strategy to try for that! Plunge milling.

    Merry Christmas fellas!
     
  18. BattyZ

    BattyZ Silver Banner Member

    That looks intriguing, would make for a nice all-in-one look. I wonder what post processor they can use. Most likely a generic linuxcnc or mach post?
     
  19. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    When you have a part with large surface area, it's best to place it such that is has the least projected area to the sand surface. This along with depth in sand will minimize the tendency to float. In round terms, dry silica sand is about .06lbs/in3 (~100lbs/ft3). Aluminum .1lb/in3. So if your part happened to be the same cross section as your flask, and there was no binding friction in the sand, it would float the mold media every time. It quickly gets worse with higher density metals.

    Of course, positioning in this manner this isn't always very practical depending upon what you have for a flask, but it's one of the reasons I used barrels for my flasks. The other reason is the cylindrical shape is very rigid compared to a flat wall box which will bulge and flex.

    Regardless of the shape, the molds quickly get very heavy and for lost foam casting, that means it takes a lot of vibratory energy to pack them well. My molding flask holds 400+lbs of sand. That's why my rig is on wheels and tilts to dump. I vibe as I fill it so the mass is lower initially so the vibe energy is more effective. The top of the mold is just sprue height so packing is not as critical as it is on part/casting features. A tall flask is harder to pour by hand.

    KC's Lost Foam Molding Rig

    The other advantage to vertical positioning, as the metal progresses deeper in the mold, the metalastatic (hydrostatic) pressure increases. So the deepest part of the mold sees the coolest metal but also the most head pressure pushing on it. Lastly, it promotes good directional solidification from the bottom up.

    It's great when it all comes together. How many pounds is your biggest pour?

    Best,
    Kelly
     
  20. BattyZ

    BattyZ Silver Banner Member

    55 gal drums cut to certain heights, stackable and on wheels is on the list! Has been for a while. How my plastic buckets will I have to melt through before though? lol

    Ya know, I never thought of it this way. Makes sense, clear ice cubes come from one-directional freezing. I would know, I machined some ice presses! (Need lots of freezer space since the gigantic riser never actually solidifies.)

    I am headed to the shop this afternoon, could weigh one of the swords with the way too big of a feeder to get a good ballpark.
     

Share This Page