Welcome to Our Community

Some features disabled for guests. Register Today.

Hello OB forums and many ?'s from a noob.

Discussion in 'General Talk' started by Joe Santarsiero, Oct 30, 2014.

  1. Joe Santarsiero

    Joe Santarsiero OB addict
    Staff Member Moderator Builder

    Joined:
    Oct 30, 2014
    Messages:
    902
    Likes Received:
    196
    Hey guys,

    Just signed up, but I've been eying the site after checking out the OX build.

    So far, I love the V-slot design, site layout, and open source lovin I've seen around here, so I've decided to park it with you guys.

    First off, I'm fairly technical minded. I can read and write G-code, have some C++ experience, and am a competent machinist. Digital electronics I can hook up and most of the time understand whats going on. I'm glad to have found the site and a community of folks with varying levels of experience.

    So far, I'm looking to build an OX based machine which can handle aluminum and 3d wood at a decent rate while being able to plow through 2d wood.

    So many questions have come up that I have a hard time finding answers for.


    I have ordered:
    3x 425 oz-in nema 23's from automationtechnologiesinc:
    http://www.automationtechnologiesin...stepper-motor-425-oz-in-14”-shaft-with-a-flat

    Their nema 17 76 oz-in:
    http://www.automationtechnologiesin...-stepper-motor-kl17h247-150-4a-for-3d-printer

    A TinyG v8 from synthetos

    and a cheap-o 24VDC 15A 350W power supply off ebay:
    http://www.ebay.com/itm/12V-24V-15A...41?pt=US_Light_Bulbs&var=&hash=item462b03e7bd


    Randomized thoughts and questions: (I'm all over the map here, so bare with me)

    1. What should I expect as far as characteristic from these higher torque motors compared to say a 170 - 280 oz-in motor?
    They had a 495 oz-in 3A monster, but I felt I was pushing my luck already as the tinyG is rated for 3A max while sufficiently cooled.

    2. I feel this power supply will be sufficient for juice and the bigger concern is whether or not It'll crap out because of quality. Agree or disagree?

    2. I got the Nema 17 76 oz-in stepper for the Z for $6 and see that it is being used successfully. I do have concerns as to whether or not it'll be able to handle heavier spindles..say if I go with a heavy router. Anyone?

    3. As far as spindles go I am still undecided. I really would love to have rigid tapping into Al (fingers crossed for form taps) off the bat, but would consider it as a second spindle upgrade option or for a future CNC. From what I know so far, I'd be looking at a lower RPM ranged spindle and some extra hardware all of which might not whip up saw dust as fast if I choose to work wood in 3D. Any guidence here would be great. Cost is not a huge limiting factor, but I'm trying to be conservative as this is a first project (and I have a wife). I would love to have a spindle start, spindle speed, spindle stop functionality, but I guess I could deal without it to begin with. A relay is always another option.

    4. I would like at "least' a 2" - 3" Z range. I plan on making some sectioned molds with alignment pinholes out of foam. I'll layer them and vacuum mold over some and fiberglass others. Considering the current OX gantry plate heights and the loss of rigidity while at deeper Z depths, What length Z screw should I be considering. I know things are based on spoiler board thickness and all, but I'd love some guidance and input as to what I should be looking at as far as Z screw lengths. I have been considering a ball screw set-up for the Z because I am not particularly fond of lead screws from experience.

    5. As far as any plates needed for an OX build that are not sold here. What's the best way to get a hold of these? Would I be better off obtaining the stock and having my local shop whip them up or are there builders offering them around here? I'd love some aluminum to start, but could deal with another rigid material to start. Anyone that has experience with prices and quality plz let me know.

    6. should I expect any issues with conversion if I'm using a metric gt3 pulley and belt and machining in imperial? I'm guessing no, but might as well through it out there.

    Thanks for any feedback on any of these issues.

    I'm happy to be here and look forward to conversing with the group.

    Joe.
     
    Mark Carew likes this.
  2. steelspinner

    Builder

    Joined:
    May 2, 2014
    Messages:
    24
    Likes Received:
    12
    Welcome Joe!!

    I was reading over your list of items, there are a few things you should look at. First, the steppers you have chosen are pretty strong at VERY low speeds. Doing the motor calculations (assuming you are use the 24v supply) the max rpm you will get from those steppers (no load) will be 189 rpm bipolar parallel (2.8 amps/phase) and 94 rpm bipolar series (1.4 amps/phase). Honestly I think you would get a wider useful range from the 175 oz/in motors from the parts store.

    The nema 17's you have will work very well for the Z axis if you are using a lead screw. If you are using the 8mm 4 start (8mm / rotation) screws from the parts store you have a theoretical max load of about 56 lbs on the z axis (more than is practical with an extrusion based machine).

    Unless you want to spend a TOOOOON of money on a spindle, forget rigid tapping. You would need a very high torque servo motor that can supply that torque at very low speeds to 0 rpm (when reversing). You need about 40 in/lbs continuous torque to tap a 1/2-13 unc with a 75% tap drill in aluminum, as an example. Pricing servos and controller, which will be a real ***** to make work with a TinyG, your looking at a min. of $1000 usd. Also, they are very heavy and have so much instant torque that it would probably twist your gantry assembly. You can get smaller lower priced servos (still tough to make work with TinyG) that will give you great results as part of a spindle assembly, but not tapping.

    If you are really wanting to go in the direction I think you are heading, you may want to look at more robust motion controllers than the Tiny G. You said you can do some C programing. I would suggest taking a look at Kflop as a motion controller. It is a C based FPGA with great specs. There are a bunch of others out there you just have to look around.

    You may want to look at using a higher voltage power supply for your steppers. The best way to drive steppers for best performance is to use a power supply that is 32*SQRT(Motor Inductance). This is limited by your driver of course. The steppers you listed from automation technologies are 6.8mH parallel and 27.2mH series. In order to drive them to their potential you would need ideally 83 vdc for parallel and 167 vdc for series. Usually bigger isn't better when it comes to stepper motors if you want any kind of speed (rpms) and torque above VERY slow speeds. http://www.geckodrive.com/support.html is a very good primer for stepper motor theory and usage.

    I think for a first machine, the parts you have will get you started. Be aware that even moderate speeds may be a challenge with those steppers, that controller and that power supply. There is a good chance you will also fight mid-band resonance.

    Dave
     
  3. Joe Santarsiero

    Joe Santarsiero OB addict
    Staff Member Moderator Builder

    Joined:
    Oct 30, 2014
    Messages:
    902
    Likes Received:
    196
    Hey thanks Steelspinner.

    "First, the steppers you have chosen are pretty strong at VERY low speeds...... "

    Ya know. That did occurr to me and was a concern. I guess there were quite a few things going on in my head. Cutting aluminum, changing pulleys, potentially switching from timing belt drive to ball-screws (or even torqsplines) on the x and y once I have the ability to cut brackets, worrying about snapping timing belt or skipping teeth, and whatever I missed.
    Did I pull the trigger on the high torque motors? Probably, but I'm impatient and couldn't determine what torque curves and speeds were available for all the 23's I looked at. I notice you just posted a stepper motor calculator, but it doesn't open for me. Could be a problem on my end. Though, you might want to check on it. I'll have to roll with the 425's for now and will consider smaller sizes once I have something up and running. The nice thing I noticed about these electronics is that they retain a fair amount of resale value, so I don't think I'll lose to much dough if I decide to sell later on. That includes the TinyG if need be.

    That KFLOP is pretty slick.

    I plan on eventually building a more industrialized machine once my ankles are wet enough. So, I can put rigid tapping aside and invest in a cheap drill press and tapping attachment for the time being. Still yet, being able to control spindle on, off, and at least having an rpm read-out would be my next goal.
    Any suggestions for this route would be great. I have been eying the quiet cut spindles on inventables, but don't know if going with a slower RPM spindle is a good decision if I choose to do 3D work. Plus, they don't really give much specs on those things which is a turn off. I could always (and have contemplated a multi-spindle machine) go with one of the RC motor mods for high speed 3D and choose a torquey spindle with all of the control I desire as mentioned previously for slower, but heavier work.

    As far as spending a ton of money for a rigid tapping solution that could do 40 in-lbs: Couldn't a nema23 that does around 320 in-oz do the job? Maybe with a custom algorithm that considers stepper rotation decel and accel at the bottom of a hole in conjunction with the same z-axis stepper characteristics? Maybe I'm thinking of this wrong or on the wrong track.

    I have read the gecko link prior to my post and it is a great primer. Thanks for the heads up on what I'll be encountering as far as mid-band resonance. I may eventually have to replace these steppers to get this machine to do what I intend it to. This isn't necessarily a bad thing considering it is part of the cost of learning.

    Any responses received are taken into great consideration and I appreciate the time you spent with your reply Steelspinner.

    I'll be starting a build section soon to document my experiences so that others after me can learn of any mistakes not to make. I have seen these 425 steppers driven by a tinyG on only one video and the performance seemed to be what I was expecting as far as RPM goes. then again my inexperience could not be telling me what I'm missing so I'll be sure to share the results here with the community.

    Thanks again for the time spent.
     
  4. steelspinner

    Builder

    Joined:
    May 2, 2014
    Messages:
    24
    Likes Received:
    12
    I am pretty sure you'll be able to make it work with the big motors. I would say definitely go with a higher voltage power supply. Drivers that can handle 48v are pretty common and cheap (not recommending cheap, but....). You may want to give some thought to the belt size you are planning. Torque like those motors have will be hard on smaller belts and may be more trouble than they are worth. Ball screw, acme screw or torque spline would probably be better. You may consider larger belts or even chain. The problem with screws is going to be motor speed to movement speed, they are going to be slow to get any kind of axis resolution. Belts or chain with give better speed but resolution is going to be very poor (from a machinist point of view).

    I have done a lot of thinking about spindles. I am really intrigued by using a brushless motor, either RC or industrial. I was considering an RC spindle with a 2:1 belt reduction or industrial with a 2:1 belt step up. Price is close to the same if you do some shopping. The RC looks a little more bang for the buck but the industrial looks a little more robust. :banghead: Controlling spindle speed isn't too complicated, depending on your hardware. I am not very knowledgeable about TinyG or other Arduino based motion controllers. I find them limiting due to processor speed for real time activities. They are WAY better than cheap china TB based all in one usb boards, not as good as a good FPGA card or a real time operating system on a more robust computer.

    I hadn't thought about using a stepper as a spindle. I would thing a heafty one would be able to tap without to much problem. You won't get any kind of cutting speed for EM's or drills though. You could probably do a bolt on Nema 23/34 spindle just for tapping. Direct drive (I would key or pin) would work just fine. You could run it as a axis (C axis if you are politically correct). I like the idea.

    Spindle speed control can be accomplished with a PID loop and an indexer. By indexer I mean an encoder on the motor or spindle shaft or as simple as an disk with a slot at the edge and a light gate. You need a variable speed control such as a VFD or PWM board. This topic is something for another thread, there is a LOT to discuss and you need hardware specifics.

    Sorry about not being able to open the calculator. I made it with Libreoffice. I am too cheap to buy Microsoft office. Solidworks, yes, office no. :D
     
  5. Tweakie

    Tweakie OpenBuilds Team
    Moderator

    Joined:
    Jan 18, 2014
    Messages:
    784
    Likes Received:
    326
    If it's any help, here is Dave's calculator in Excel format.

    Tweakie.
     

    Attached Files:

  6. Joe Santarsiero

    Joe Santarsiero OB addict
    Staff Member Moderator Builder

    Joined:
    Oct 30, 2014
    Messages:
    902
    Likes Received:
    196
    Thanks Dave. You've given me some things to ponder. I have seen a bit of evidence of the 425 oz-in steppers running at decent speeds. I've calculated 385ipm rapids which I'm satisfied with 300. I'm sure I could tone down the size a lot, but I'll roll with them for now and maybe consider 48V drivers for future snappiness. The PID with encoder was a good idea. Something to think about. Mine as well go servo though. I do think I'm competent enough in my math and G-skills that I could run a stepper on a rotational axis for tapping, though I haven't given it anymore thought than that. Z could feed and A could rotate tap (with some math in between). Getting it back out could be an issue or maybe not. Might even need to figure out a macro for all of it. Meh. Someday.. I probably should be worrying about other things ahead of this. Sourcing GT3 pulleys with 1/4 (6.35mm) bores. Gates powergrip catalog is wretched. One must have a degree in it specifically to figure out. Good info for a weekend seminar, but too much clutter for one whom chooses to be a layman on the subject.

    Thanks @Tweakie . The xls worked great. Any chance of getting GT3 pulleys with a 6.35mm bore back in the parts store anytime soon?
     
  7. Tweakie

    Tweakie OpenBuilds Team
    Moderator

    Joined:
    Jan 18, 2014
    Messages:
    784
    Likes Received:
    326
    Hi Joe,

    I am pleased the .xls version of the calculator worked OK for you. :)
    The stock situation of the Parts Store is not within my control - you could try contacting them directly, asking the question about the GT3 pulleys with imperial sized bore, using the contact form http://openbuildspartstore.com/about-us/

    Tweakie.
     
  8. steelspinner

    Builder

    Joined:
    May 2, 2014
    Messages:
    24
    Likes Received:
    12
    Joe,
    I am pretty sure you'll be ok with your setup. The PID controller can be made pretty easily with an arduino and a something to get an index pulse from the motor. There are lots of arduino PID examples out there. The g code for tapping with an axis is pretty easy, final z depth/(1/thread pitch)=number of 360deg rotations of the axis rotating. Just get or make a floating tap head and no broken taps if you miss a little. Try this link for pulleys http://us.misumi-ec.com/vona2/detail/110302193560/?Inch=0&CategorySpec=00000029797::ag .

    Dave
     
    #8 steelspinner, Nov 5, 2014
    Last edited: Nov 5, 2014
  9. steelspinner

    Builder

    Joined:
    May 2, 2014
    Messages:
    24
    Likes Received:
    12
    Here is a code example for taping with an axis:

    G1 Z-1.0 C 7200.0 F3000.0
    G4 P2000
    G1 Z.1 C-7200.0 F3000.0

    This is for a 20 pitch thread. The feed rate looks very high but generally rotational axis is in degrees (some use radians) per minute, depends on the controller. If there is a single move rotation limit on the axis you can break it up into smaller chunks with the same results.
    Dave
     

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice