PROJECT: DIY stringing machine tensioner

Discussion in 'Badminton Stringing Techniques & Tools' started by kwun, Apr 25, 2013.

  1. kwun

    kwun Administrator

    Joined:
    Apr 24, 2002
    Messages:
    41,043
    Likes Received:
    2,066
    Occupation:
    BC Janitor
    Location:
    Santa Clara, CA, USA
    let's design our own tensioner.

    the WISE is affordable and all but hardly the best when it comes to quality. it is slow, noisy and the controls are 1980's and limited.

    wouldn't it be awesome if we can design and build our own, using as much off-the-shelf component as we can.

    there are many design aspects and choices that needs to be made, now here are the ideas i have so far:

    drive system:


    this is by far the soul of the whole design.


    • ball screw - the gold standard that is used by many stringing machines including the ES5Pro's is a tension head mounted on a ball screw. this is well tested system and we can buy ball screw off ebay or wherever. there are different quality of ball screws we just need to find the suitable one.
    • linear belt drive - this is an alternative that could be interesting, all we need is to have a belt (easy to find automotive ones) pulley on one side, and the motor driving the other one on the other side. the tensioner will be driven by the belt. i am not sure if any tensioner uses it today, but it would be very very quiet and smooth.
    • others? (no rail guns please. ;) )

    sensor/control:

    this is slightly more difficult and something that i am not completely sure about. one can buy servo controller boards that controls a DC motor with a encoder on it to maintain constant speed or location. what we have though is we need to maintain constant tension/force. the WISE has a strain gauge right at the tensioner body to measure force. we can do the same thing so sensory shouldn't be a problem. the only problem is how to use it to feed back into the control side.

    motor:

    DC motor seems to be the way to go. careful choice must be made to choose the right size and speed. we need to make one that is strong and fast, no point having the same slowness as the WISE.

    gripper:

    gripper isn't the most high tech component but we need to solve it anyway. we can either build our own or somehow find one that is ready to be purchased. not sure what the solution. and surely anyone who had a WISE would have their machine's original gripper sitting in a box somewhere.

    interface/UI:

    this is where we can go a bit more fancy and into the 21st century. i envision most of the motor control will be done on a motor control board with its own microcontroller. and all we need is to interface with this controller. well, this is 2013 we should be able to do something fancy. with touch screen phones in it 3rd/4th generation we probably have an old iphone/android phone lying around. so how about write a little app and have it be the UI and it will talk to the motor controller processor via USB interface? such contraption/adaptor already exist and we can just leverage from it.
     
  2. AZbadman

    AZbadman Regular Member

    Joined:
    Dec 28, 2007
    Messages:
    363
    Likes Received:
    18
    Occupation:
    Sales and Marketing
    Location:
    United States
    Sounds good! I'm in :p
     
  3. kwun

    kwun Administrator

    Joined:
    Apr 24, 2002
    Messages:
    41,043
    Likes Received:
    2,066
    Occupation:
    BC Janitor
    Location:
    Santa Clara, CA, USA
  4. AZbadman

    AZbadman Regular Member

    Joined:
    Dec 28, 2007
    Messages:
    363
    Likes Received:
    18
    Occupation:
    Sales and Marketing
    Location:
    United States
    that's cool! look down at the seller's chart, he has different sizes available. Which one would we need?
     
  5. kwun

    kwun Administrator

    Joined:
    Apr 24, 2002
    Messages:
    41,043
    Likes Received:
    2,066
    Occupation:
    BC Janitor
    Location:
    Santa Clara, CA, USA
    my guess is around / at least 300mm long. as for which diameter. hard to say. need to dig out the spec sheet to take a look.
     
  6. kwun

    kwun Administrator

    Joined:
    Apr 24, 2002
    Messages:
    41,043
    Likes Received:
    2,066
    Occupation:
    BC Janitor
    Location:
    Santa Clara, CA, USA
  7. Mark A

    Mark A Regular Member

    Joined:
    Sep 2, 2005
    Messages:
    7,170
    Likes Received:
    695
    Location:
    St Helens, UK
    Dang - if I'd known we were going to do this, I would have made an electronic tensioner for my Final Project at university! Has elements of everything - electrical, electronic and programming.

    And it would have been free!
     
  8. kwun

    kwun Administrator

    Joined:
    Apr 24, 2002
    Messages:
    41,043
    Likes Received:
    2,066
    Occupation:
    BC Janitor
    Location:
    Santa Clara, CA, USA
    anyone understand ball screw calculation?

    if we want the machine to handle say, 80lbs linear force, what torque rating do we need for the motor? also want to be able to do that with decent speed as well, does 2 second end to end sounds too ambitious?
     
  9. Mark A

    Mark A Regular Member

    Joined:
    Sep 2, 2005
    Messages:
    7,170
    Likes Received:
    695
    Location:
    St Helens, UK
    I think the required torque would depend on the thread pitch...
     
  10. johnlowe88

    johnlowe88 Regular Member

    Joined:
    Jan 26, 2011
    Messages:
    207
    Likes Received:
    17
    Occupation:
    IT Consultant
    Location:
    Sydney, Australia
    I think a stepper motor would be more suitable than a DC motor. Why? Generally ball screws are used in conjunction with stepper motors for cnc control. For this particular project, we are not actually going for positioning, but for tension as measured by a load cell or strain gauge. This makes stepper motor controllers very suitable because of its microstepping ability.

    If we use dc motors, we will still need to use a speed controller to control the tension. DC motors have brushes that tend to wear out over time - I wonder if this is likely to be a problem though?

    I have been playing with a number of Eagnas Plus 8000 tensioners, i.e. for repair. This is using a 1/4" pitch chain drive with what appears to be a geared dc motor.
     
  11. Fidget

    Fidget Regular Member

    Joined:
    Mar 4, 2008
    Messages:
    3,664
    Likes Received:
    330
    Location:
    Canada
    I have no technical knowledge to contribute, but I just had to say that you guys are AWESOME to be doing this. :cool: ( It's like badminton-nerd Nirvana!)
     
  12. DinkAlot

    DinkAlot dcbadminton
    Brand Representative

    Joined:
    Feb 20, 2005
    Messages:
    12,682
    Likes Received:
    290
    Occupation:
    Social Distancing Specialist
    Location:
    Southern California
    Dink will go home tonight and check, he should have his old gripper plates and can donate them for badminton "science". :D Also can give you Dink's Diablo as he doesn't use it.

    Kwun: do you have a machinist? If not, there's a good one down the street from Dink's office but he's a bit pricey.


     
  13. johnlowe88

    johnlowe88 Regular Member

    Joined:
    Jan 26, 2011
    Messages:
    207
    Likes Received:
    17
    Occupation:
    IT Consultant
    Location:
    Sydney, Australia
    Regarding sensors - we should use a load cell, which is basically a device with strain gauges on it. Depending on the maximum tension we want to go to, we could use a load cell with a higher rating, i.e. we want to use within its sort-of linear range. Since we are designing our tension head, I would suggest that we do a test calibration of each load cell, and download the calibration into the processor that is running the tension head. If you want to go all the way, definitely add a temperature sensor so that we can adjust for load cell temperature as well and we can have a super accurate tension head, no need for adjustments - that would be fantastic. I might just order a load cell to play with, i.e. 60kg should be sufficient.
     
  14. MikeYeap

    MikeYeap Regular Member

    Joined:
    Nov 27, 2012
    Messages:
    143
    Likes Received:
    3
    Occupation:
    Engineer
    Location:
    Malaysia, Nilai
    using stepper motor will cause the movements to be slower isn't it ?
    DC motor may move faster, using internal sensors with stopper where it senses current peaks and stop the motor. Let me check in detail which motor I am referring to..( little blur here..sorry )

    But for parts, I could recommend this:
    Motors = Oriental motors
    Ball screw/linear bushings = MISUMI

    this are industrial part suppliers with complete detailed catalog. So no fuss on the specs of size, torque, bla bla bla....
    best of all, MISUMI's warehouse is located at Singapore itself...
     
  15. johnlowe88

    johnlowe88 Regular Member

    Joined:
    Jan 26, 2011
    Messages:
    207
    Likes Received:
    17
    Occupation:
    IT Consultant
    Location:
    Sydney, Australia
    Some stepper motors used with ball screws in CNC machines can move 10 inches a second - more than fast enough. Servo motors would be faster, but DC motors again are also fine. I just prefer steppers because that is what I am looking at using on my cnc conversions.
     
  16. johnlowe88

    johnlowe88 Regular Member

    Joined:
    Jan 26, 2011
    Messages:
    207
    Likes Received:
    17
    Occupation:
    IT Consultant
    Location:
    Sydney, Australia
    A couple more details. We need limit switches, i.e. home position, then max travel limit causes return to home position. How much travel should we need - 6 inches, 10 inches? With stepper motors we can set the speed, and have multiple speeds, i.e. slow, medium & fast, but what does this actually mean in terms of inches per second?

    Another question is pre-stretch, how long to hold at pre-stretch tension before returning to required tension?
     
  17. R20190

    R20190 Regular Member

    Joined:
    Jun 11, 2004
    Messages:
    2,458
    Likes Received:
    414
    Occupation:
    Chartered Civil Engineer
    Location:
    London, UK
    I admire the idea and the enthusiasm but I can't help but think this would be too much work for your average badminton nut to put together in their garage - even if we managed to get a prototype to work.

    Trust me when I say I love these ideas and love building things myself (so I'd be pleased if proved wrong here) but this one may be one mountain too high, at least for me.

    I'd be more inclined to go with a DIY manual tensioner which although is a bit stoneage in comparison, I think it can be designed and built reasonably well from simple parts. Also, most people who build their own machines do it for the satisfaction and perhaps to save a bit of money. I'm guessing it may not be cheap to put an electronic tensioner together?

    Presumably, with a DIY electronic tensioner you would also need some knowledge of electronics to buy the right components and to be able to put them together without ending up frying yourself.
     
  18. johnlowe88

    johnlowe88 Regular Member

    Joined:
    Jan 26, 2011
    Messages:
    207
    Likes Received:
    17
    Occupation:
    IT Consultant
    Location:
    Sydney, Australia
    But isn't this a great idea to pass the time with?

    To answer some earlier questions - from what I can come up with...

    RM1605 ball screw to move 5 inches per second would require a screw rpm of 1524.

    Torgue required to move say against a 100lb force, would be 0.37Nm or 3.3lbs*in or 524oz*in - stepper motors around 640oz*in would be Nema 34 in size and cost about $80 or so. One in particular is 1.8 degrees per step, so 200 steps per rotation. To get 5 inches/sec would need 304800 steps per minute or 5080 steps per second. This is eminently achievable. The only thing is whether this will give sufficient resolution on the load cell. If not, then we need to half-step or microstep.

    Another thing I envisage is a calibrate menu item maybe under utilities with a pin number. Allows the tensioner to pull at specific points within your usual tension, say from 18 to 30 lbs - pulls at one tension like 18, then allows you to key in what your tension meter says, then do 20 etc. After doing this for the required range - ask if you want to save this and replace the current calibration. Assuming we have an accurate tension meter.
     
  19. johnlowe88

    johnlowe88 Regular Member

    Joined:
    Jan 26, 2011
    Messages:
    207
    Likes Received:
    17
    Occupation:
    IT Consultant
    Location:
    Sydney, Australia
    My earlier calculations were for direct drive where the stepper motor is directly coupled to the ball screw. If we use a gear or pulley system, we can reduce the torgue requirements of the motor allowing us to use a smaller one.
     
  20. R20190

    R20190 Regular Member

    Joined:
    Jun 11, 2004
    Messages:
    2,458
    Likes Received:
    414
    Occupation:
    Chartered Civil Engineer
    Location:
    London, UK
    Yes and no I guess.

    Yes in that this is a great challenge/project and no in that I'd probably prefer to pass the time hitting some shuttles.

    Don't let me put you off though, if this is something you are determined to do, you certainly have my support! But if this is intended as a blueprint for others with little electronics knowledge to DIY at home, imho it's probably cheaper and far less hassle to pay someone like Kwun to string it for them! :)
     

Share This Page