584,842 active members*
4,196 visitors online*
Register for free
Login
Page 1 of 2 12
Results 1 to 20 of 39
  1. #1
    Join Date
    Jan 2017
    Posts
    31

    Force on the ball/lead screw

    I've got this feeds and speeds app on my phone, it gives all kinds of data. It gives a calculation on "cutting force". If the direction of cut lined up with the an axis / ball screw then that cutting force is directly pushing on the ball nut and screw right?


  2. #2
    Join Date
    Nov 2013
    Posts
    4280

    Re: Force on the ball/lead screw

    Hi,

    It gives a calculation on "cutting force". If the direction of cut lined up with the an axis / ball screw then that cutting force is directly pushing on the ball nut and screw right?
    Yes that is correct, note that the force is transmitted through the bearing block also, and is for that reason that quality angular contact bearing blocks are so very expensive, but they just have to be good.

    In addition to the cutting forces there is significant forces associated with accelerating the axes. Commonly the accelerating forces will be many times the cutting forces.

    Craig

  3. #3
    Join Date
    Jan 2017
    Posts
    31

    Re: Force on the ball/lead screw

    Thanks Craig. I wasn't even thinking about acceleration, thanks for the tip. That's just another variable when trying to work out torque needed for the stepper. I need to refer to the stepper torque curve to make a guestimate on a given stepper at a given speed to see if it's suitable for the job.

    The above screen shot shows "torque" needed to make the cut and a "safe torque" for the requirements of the spindle motor which is roughly 3 times greater. Do I need to add that much safety factor (x3) to the steppers?

  4. #4
    Join Date
    Nov 2013
    Posts
    4280

    Re: Force on the ball/lead screw

    Hi,
    high acceleration is very VERY important to toolpath following. If the toolpath makes a change in direction then the faster the axis can accelerate the better
    it will follow the programmed path.

    If your machine has ballscrews then you might also be surprised to know that it is often the rotational inertia of the ballscrew and the armature of the axis motor which dominates the inertia equation.

    As an example my new build mill has 32mm diameter 5mm pitch ballscrews 700mm long. The axis beds are 115kg of cast iron. You'd think the 115kg chunk of iron would dominate the inertia
    wouldn't you? But no, 80% of the inertia is the ballscrew, 12.5% the inertia of the armature of my servo and only 7.5% is the inertia of the 115kg axis.

    Craig

  5. #5
    Join Date
    Jan 2017
    Posts
    31

    Re: Force on the ball/lead screw

    Lol, you're right I am surprised by that. That seems like the most likely point when steps get missed.

    Do closed loop steppers help with missed steps or do they just do catch up after the fact?

  6. #6
    Join Date
    Nov 2013
    Posts
    4280

    Re: Force on the ball/lead screw

    Hi,

    Do closed loop steppers help with missed steps or do they just do catch up after the fact?
    They try to with varying degrees of success.

    Manufacturers of closed loop steppers tell you they are faster, more powerful, never miss steps....all BS. ALL steppers lose torque the faster they go,
    and closed loop steppers are no different. If a stepper is marginally overloaded, that is too say the torque required by the load is somewhat more
    than the stepper can produce at that speed, then it misses a step. The drive of a closed loop stepper will insert an extra step, but guess what....the stepper is already
    marginally overloaded so the extra step will be missed as well.

    Closed loop steppers do gain increased resolution (depending on the encoder fitted to the stepper) but you are paying quite a premium for a small advantage.

    If you want genuine closed loop performance buy AC servos. The real advantage of servos is that they have about three to four times rated torque as overload, and they
    retain their torque at full speed. The overload capacity makes them behave like much larger motors than the numbers would suggest. They'll eat any stepper ever
    made.

    Look at Delta servos, a Taiwanese brand made in China, or DMM servos, a Canadian brand made in China being great quality, great support, good documentation
    and most importantly free setup and tuning software, at fair prices.

    Craig

  7. #7
    Join Date
    Jan 2017
    Posts
    31

    Re: Force on the ball/lead screw

    Aah yes I've read the closed loop servo is the shiznit. I was thinking their biggest benefit is at high speed which I didn't think was too important to me since I'm learning. Acceleration seems to be a key point I missed. I'll reconsider them and check my $ comfort level.

    Thanks for the insight

  8. #8
    Join Date
    Jan 2017
    Posts
    31

    Re: Force on the ball/lead screw

    This got me wondering, is there a rough cut off point that steppers are more suitable than servos e.g. size / rigidity if a machine or is it more aligned with what material is being cut e.g. wood vs steel ??

    Or do servos reign supreme in all cases?

  9. #9
    Join Date
    Nov 2013
    Posts
    4280

    Re: Force on the ball/lead screw

    Hi,
    steppers are very torque dense, even more so than servos of the same size.

    For instance a 24 size stepper 100mm long might have a rated torque of 450oz.in or 2.3Nm. A 400W servo about the same length (and weight) will have a rated torque of 1.3Nm.
    So the stepper is better....right? Well not quite the servo will have an overload rating of 4Nm or more, so for short durations the servo can more than match a stepper.

    The real issue is that steppers lose torque at speed, and its very common for steppers to have less than 10% of their torque at 1000rpm. As a rough approximation most
    steppers will have say 50% at 400rpm. Thus the stepper is as good as the servo, at least in terms of torque, up to 400rpm. The servo will have its rated torque out to 3000rpm,
    and of course has that huge overload potential that a stepper does not have.

    At low speed steppers compare with servos for torque but still have poor resolution compared to servos. While some may like to latch onto those few situations where steppers
    perform well.....overall servos eat steppers alive under 99.9% of situations.

    In order to have good available thrust to counter cutting forces and getting the highest acceleration for a given torque you want lowest pitch ballscrews, say 3mm, 4mm or more commonly 5mm.
    The downside of a fine pitch screw is that to get good highspeed traverses then the ballscrew needs to spin very fast. That could with a very thin ballscrew cause the ballscrew to 'whip', and
    also the servo or stepper need to spin very fast, or be geared to do so. Servos tend to spin fast anyway, so often no extra gearing is required however steppers we know are very poor at speed.

    To obtain high axis speeds without 'whip' or requiring really fast steppers/servos you want higher pitch, say 10mm or 20mm.

    You can see the two design goals are at odds.

    Lets say you choose a ballscrew of 5mm pitch, then your stepper which we guess maxes out at around 400rpm will have a G0 axis speed of 5x400= 2000mm/min, pretty damn slow.
    If you used a 400W servo rated to 3000rpm then your G0 speed would be 5x3000=15000mm/min.....much more respectable.

    If however you chose a 10mm then the steppers would get to 4000m/min, which may well be usable for you.

    For most hobbyists machines I would suggest that 5000mm/min is a useful and practical lower limit.

    My new build mill has 5mm ballscrews and the servos are rated to 3000 rpm but can in fact go up to 5000rpm, and I have tuned the G0 rapids to do this which means that I get 25m/min rapids!
    Its scary fast, and I slow the machine down to about 10-12m/min just to save my heart rate.

    Your choice of steppers OR servos will be largely about cost, if you can afford servos get them. If you can't afford them then steppers will have to do....and just accept the performance limitations
    that result. There are tens of thousands of stepper driven machines out there, and with intelligent design and use will perform just fine. If you have a specific need for speed and/or overload capability
    for production purposes or the highest possible resolution and accuracy then servos are the answer.

    Craig

  10. #10
    Join Date
    Jan 2017
    Posts
    31

    Re: Force on the ball/lead screw

    I feel like I have a better understanding of the differences between stepper and servo now.

    I know I'm new to this but the feed rates you mention seem awfully high...?? Is that for milling wood? The speed/feed calc I posted is for aluminium and indicates a feed rate of 300mm/min

  11. #11
    Join Date
    Nov 2013
    Posts
    4280

    Re: Force on the ball/lead screw

    Hi,

    I know I'm new to this but the feed rates you mention seem awfully high...?? Is that for milling wood?
    No. The rates I mentioned are G0 rates, that is the rapid traverse rate, not cutting rates. My mill has travels of 350mm x 350mm x 350mm, and is largely for metal milling,
    although I use it extensively for small metal and plastic parts for automotive instruments. I also use for isolation routing of circuit boards.

    I mentioned that I have tuned my servos to do 5000rpm, but that is somewhat reduced torque (about 1.7Nm) from its rated torque of 2.4Nm at 3000 rpm.
    That is to say that I get full torque and therefore the greatest thrust and acceleration at 15m/min.....not that I cut at that speed. I typically cut aluminium
    with small diameter tools and anywhere between 300mm/min up to about 2000mm/min. Note that cutting moves are done with a G1 Fnnn, where F is the feedrate.

    G0, or Rapid traverse moves require torque for acceleration but do not encounter cutting forces.

    Typical production mills run at 30m/min right on up to 100m/min. Some, like Makino claim 'Chip to Chip Time' of 3 seconds. That is to say the machine is cutting
    and making chips at t0 when it does a tool change. The spindle decelerates to a stop, the machine rapid traverses to the tool change location, releases
    the old tool, inserts and clamps the new tool, spins up the spindle and rapid traverses to the location of the next machining operation and start making chips with the new
    tool in 3 seconds!!! Machines that can do that have accelerations of 10g and better and rapid traverses of 60+m/min.

    That is more than I need, and hugely outside my budget!

    Most hobbyists report rapids of 100 inch/min (2500mm/min) and higher, and most find that acceptable but 5000mm/min should be where you aim as a minimum.
    Believe me you'll want it. My first mini-mill had 10:1 planetary reduction boxes and so the steppers could only manage 1200mm/min rapids. Given that the machine had
    travels of 180mm x 180mm x 180mm it wasn't too bad....but I determined I was going to do much MUCH better next time. My new mill goes that fast it scares me.

    Craig

  12. #12
    Join Date
    Jan 2017
    Posts
    31

    Re: Force on the ball/lead screw

    I think I'm missing something here so forgive me for sounding ignorant but why would I want super fast rapid speeds that will only scare me...??

  13. #13
    Join Date
    Jul 2018
    Posts
    6248

    Re: Force on the ball/lead screw

    G'Day PinkP from Oz - Its usual to design a machine in steps. But up front you need to decide what speeds you want to achieve as a design spec (and accelerations if possible as these will be the ultimate determining factor for motor size eg maker level machines are about 0.1g, commercial fast mills can be 1.0g or more) These specs will determine your required motor sizes and their speed range. As noted steppers run out of steam at say 600rpm and servos are good to 3000rpm or more. So if you use steppers with a 10mm pitch you are limited to 6000mm/min. If you want to go faster then that you need to gear the motor (which will impact low speed performance) use servos or use belts. For instance one of my machines used belts and I regularly ran it at 20m/min for fine finishing timber moulds. If you had to finish at 1000mm/min it would take days to do the job... To acheive 20m/min with a screw 20000/10=2000rpm then you have to use servos. So if you have a big machine and your doing lots of air passes then you need fast rapids to minimise cycle times to make money. If your a hobbyist who can let the machine run thru the weekend then who cares... 10mm or 20mm pitch screws work quite well on hobby machines (for XY I'd go 5mm on Z) with a medium or large N23 or N24 stepper. if the budget is there go asian servos as at some point once you learn the machine you will want it to be fast to get the jobs done, why take 1hr when you can do it in 15mins.... money well spent as motors are the heart of the machine... Plus these days say you use Fusion 360 and you use a high speed machining (HSM) strategy you will be up against the machines speed limits and torque limits immediately trying to take advantage of these things and steppers will definitely stall if you push with HSM. Peter

    also note that most commercial mills are cutting aluminium at 300meters/min these days so maker routers cutting at 1 or 2m/min are slow. I'd aim at 5m/min min for cutting timber these days on small machines...

  14. #14
    Join Date
    Jan 2017
    Posts
    31

    Re: Force on the ball/lead screw

    Yeh I can understand when trying to make money then speed is key. I'm definitely a hobbyist. Not looking to make/ build a machine but wondering how suitable the stock steppers are on my recent (few months ago) purchase and get a feel for it's limitations without finding out the hard way. Also if I were to upgrade what would be suitable. I thought machine rigidity would come in to play at some point but I guess not if it's more about the rapids...

  15. #15
    Join Date
    Mar 2014
    Posts
    1131

    Re: Force on the ball/lead screw

    Hi,
    I am thinking on installing an app, also willing to purchase if I can find a good one. What is the name you have? Recommending?
    Thanks

    Sent from my MI 5s Plus using Tapatalk

  16. #16
    Join Date
    Nov 2013
    Posts
    4280

    Re: Force on the ball/lead screw

    Hi,

    why would I want super fast rapid speeds that will only scare me...??
    It is an often asked question....may I suggest you build and use a machine. After a few months ask yourself 'would this machine be better if it
    weren't so ff....gg slow?'. It will happen. Most hobby machines have rapids in the range of 5000mm/min up to 10000mm/min. I run my machine,
    for general use at 12500mm/min and if I'm doing a repetitive task with a well proven toolpath I push it out to 25000mm/min.
    Using the rated torque only of my servos I have an acceleration of 0.27g. If I tune them to dip into the overload capacity I could have up to 0.75g,
    although how realistic that would be in practice I don't know. Even at 0.25g my 800kg machine starts to lurch around the workshop.

    There are two characteristics that determine how quickly a job is completed, and it varies quite widely with the type of toolpath.
    For many woodworking jobs where there are long straight lines in the toolpath having fast axis speeds is the key to reducing cycle time.
    HSM toolpaths on the other hand require very high accelerations to achieve minimum cycle times. I would suggest you need to achieve a
    balance, and I would guess that accels of 0.2g and rapids of 5000mm/min would be a good goal.

    Acceleration and max velocity are related, but by and large for a given axis motor torque the higher the acceleration the slower will be the max velocity
    and vice versa....so it is a balance. Often the choice is 'I can have more acceleration (speed) by having gear/belt reduction' or be happy with the acceleration (speed)
    I can get but WITHOUT all that shagging around with gears/belts.'

    Craig

  17. #17
    Join Date
    Nov 2013
    Posts
    4280

    Re: Force on the ball/lead screw

    Hi,

    I thought machine rigidity would come in to play at some point but I guess not if it's more about the rapids...
    Rigidity is ALLWAYS the highest concern, it affects the machines accuracy and that is largely independent of speed. Lack of rigidity is likely to be found out
    by high accelerations and high cutting forces. Just because you intend to run the machine slowly does not lessen the need for rigidity...ever. Making a rigid machine
    is the most expensive, time consuming, material intensive part of any machine design, whereas speed and acceleration are really just a matter of choice, you can, within limits,
    bolt in any ballscrew and motor combination you like to get the balance you seek.

    Craig

  18. #18
    Join Date
    Jul 2018
    Posts
    6248

    Re: Force on the ball/lead screw

    Hi PinkP - You mention machine rigidity. I mentioned machines are designed in steps. Rigidity is the first step. Its the DNA and backbone of the machine. Machine rigidity is difficult to improve in the future, it sets the possibility of the machine's performance. Motors electronics etc can be upgraded but the machines basic rigidity can't be upgraded easily. So usually establishing and refining the machine structure is the first step in a machine design, then once you know that the motion system can be designed then you can pick the electronics and accessories including the nail polish... Machines can never by too stiff. Peter

  19. #19
    Join Date
    Jan 2017
    Posts
    31

    Re: Force on the ball/lead screw

    Thanks for all that explanation Craig.

    I think I'll use this machine as is for a little longer and see how I feel about it after a few months. I don't really want to spend money on it now only to have my confidence and ability outgrow it later on.

  20. #20
    Join Date
    Jan 2017
    Posts
    31

    Re: Force on the ball/lead screw

    Peteeng (Pete the engineer..?) How does one test rigidity? If that is the factor that determines what motors/electronics then there must be a way of gauging how stiff = what motors...?

Page 1 of 2 12

Similar Threads

  1. Adding ball screw to lathe for Electronic Lead Screw
    By harrzack in forum Hobbycnc (Products)
    Replies: 0
    Last Post: 12-28-2018, 03:28 PM
  2. Lead vs Pitch of ball screw
    By mit_ball_screw in forum Linear and Rotary Motion
    Replies: 1
    Last Post: 02-26-2016, 09:13 AM
  3. New technology.... old debate.....ball screw vs lead screw
    By trubleshtr in forum Linear and Rotary Motion
    Replies: 1
    Last Post: 08-16-2005, 02:42 AM
  4. Ball Screw or Lead Screw? Which is Better?
    By ljoe1969 in forum DIY CNC Router Table Machines
    Replies: 4
    Last Post: 07-28-2004, 02:49 PM
  5. Ball Screw and Lead Screw
    By ljoe1969 in forum DIY CNC Router Table Machines
    Replies: 9
    Last Post: 11-21-2003, 03:06 PM

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
  •