Welcome to Our Community

Some features disabled for guests. Register Today.

TB6600, Steppers, and Micro-stepping

Discussion in 'Motors' started by Paruk, Jul 26, 2015.

  1. Paruk

    Paruk Journeyman
    Builder

    Joined:
    Jul 13, 2014
    Messages:
    397
    Likes Received:
    149
    For those who want to know more about choosing stepper motors, I've posted a resource information about "Micro stepping myths and realities" here http://www.openbuilds.com/resources/micro-stepping-myths-and-realities.158/

    It also could give an explanation why the cheap TB6600 drivers give problems with missing steps and being erratic in operation. That problem might be solved by setting the micro steps at 1/4 or 1/8, hence giving the motors more torque and less chance of loosing steps.

    I tried yesterday evening my Buffalo at 1/4 micro stepping and Oh Dear! It's more loud than before but also nearly 4 times as strong! It also seems to work faster than at 1/16 settings. Didn't do any cutting yet and right now contemplating on setting it first to 1/8 to do some testing of heavy cutting. see how that goes.
     
    #1 Paruk, Jul 26, 2015
    Last edited: Jul 26, 2015
    Mark Carew likes this.
  2. Jonny Norris

    Jonny Norris Well-Known
    Builder

    Joined:
    Aug 17, 2014
    Messages:
    495
    Likes Received:
    95
    You do know that tb drivers dont actually output the current they say they do? Your motors are noisy because they are under powered hense why you are loosing steps, Decreasing the number of micro-steps will increase torque. but if your motors are rated over 2.5a which by the sounds of it they are, then they are underpowered and in which case changing the micro steps will not help.
     
    #2 Jonny Norris, Jul 26, 2015
    Last edited: Jul 26, 2015
  3. Paruk

    Paruk Journeyman
    Builder

    Joined:
    Jul 13, 2014
    Messages:
    397
    Likes Received:
    149
    I suggest that you re-read my posts before starting to blow of the tower. After that you can read the resource document about Micro stepping myths and realities, so you might understand where that is about. I'm sure you then will understand my remark about the TB600 stepper motors.;) It had nothing to do with the output current of those steppers.
     
  4. Jonny Norris

    Jonny Norris Well-Known
    Builder

    Joined:
    Aug 17, 2014
    Messages:
    495
    Likes Received:
    95
    yeah ive read that, i know what your saying, microstepping vs torque. Just the fact you said the noisy motor and tb6600, they should be quiet. fix that you fix the problem. use whatever microstep you like run correctly

    i have two nema 23's 3nm 1/10 micro steps moving a 200kg gantry hense torque is not your issue.
     
    #4 Jonny Norris, Jul 26, 2015
    Last edited: Jul 26, 2015
  5. Paruk

    Paruk Journeyman
    Builder

    Joined:
    Jul 13, 2014
    Messages:
    397
    Likes Received:
    149
    The first set of TB6600 drivers I had installed, were only 15Khz and simply couldn't cope with the amount of instructions that the controller was sending. Lets say that I was running the cutting at 2000mm/min with my stepper driver set to 1/16 micro stepping. That comes to 33.33mm/sec x 53.33 SPU = 1,777 dir/step pulses per second. A 15Khz driver cannot handle that amount of instructions.

    Now, if you decrease the amount of micro stepping from 1/16 to 1/4 the calculation is different. Than we have 33.33mm/sec x 13.33 SPU = 444 dir/step pulses per second and that fits better with the 15Khz driver.

    Given the same amount of power (AMP), be it less than specified or not, it becomes clear out of the above that a driver at 15Khz cannot cope with large amounts of instructions per second and therefore is going to loose some of those instructions, resulting in missing steps. If one would decrease the amount of SPU to 1/4 or less, the driver receives less instructions and is less likely to loose steps and as a bonus the stepper motor has more torque with the given power (AMP).

    The other solution would be to have the 1/16 SPU set but run the machine at considerable lower speeds, hence reducing the amount of instructions per second. But then the motors will have less torque and are likely to loose steps because of that.

    To conclusion, a TB6600 driver at 15Khz can be used with low micro step settings (1, 1/2 or 1/4) to run the machine at reasonable feed speed and DOC and more available torque. A trade off would be a micro step setting of 1/8, but 1/16 would most likely result in loosing steps because of low available torque but gives a smooth sound. So it depends what you want, nice sound and loosing steps (defying the principle of precision CNC routers) or raw stepper motor sound but more power for cutting?

    The new TB6600 drivers I have are maximum 200Khz, and it is evident that they can handle the large amount of instructions, since they didn't miss a beat in the last half year. I was trying them out on lower micro stepping to see what the result was. I did some testing today at 1/8 which gave clearly improved power and relatively smooth noise. But I think I'll give it another try at 1/4 and have 40% of maximum torque available so I might cut deeper at same speeds. Maybe I can reduce the noise of the steppers at 1/4 by accelerating faster and run faster. If not..well power and speed is everything in CNC routers(providing you've taken care of the precision ;)).

    And Jonny, read that resource. Maybe it can help even you to improve on the feed speed and DOC of your machine. After all, CNC is SCIENCE! :)
     
    #5 Paruk, Jul 26, 2015
    Last edited: Jul 26, 2015
  6. Paruk

    Paruk Journeyman
    Builder

    Joined:
    Jul 13, 2014
    Messages:
    397
    Likes Received:
    149
    Where do you get the idea from that stepper motors should be quiet?
     
  7. Paruk

    Paruk Journeyman
    Builder

    Joined:
    Jul 13, 2014
    Messages:
    397
    Likes Received:
    149
    While I'm at it, if I set my drivers to 1/4 micro stepping it will result in a resolution of nearly 340 DPI (1mm/(60/(200*4))=13.33 x 25.4 = 338.6 steps per inch). I'm not sure if this assumption is correct, but if so, then 3D work can be done at glossy magazine resolutions? Not bad for a Buffalo.:)
     
  8. Jonny Norris

    Jonny Norris Well-Known
    Builder

    Joined:
    Aug 17, 2014
    Messages:
    495
    Likes Received:
    95
    A video I made comparing effects of current and motor noise, if you skip to the end the motors are powered correctly, I'd class that as quiet. (for anyone that would like to know they are hexfet drivers)


    Doc is a matter of rigidity more than anything so barking up the wrong tree there.

    Proof is in the pudding mate, if you think you can improve on 3m/min at 3 mm plunge on a large format ox be my guest, I'd love to know how you get on.
     
    #8 Jonny Norris, Jul 26, 2015
    Last edited: Jul 26, 2015
  9. Jonny Norris

    Jonny Norris Well-Known
    Builder

    Joined:
    Aug 17, 2014
    Messages:
    495
    Likes Received:
    95
    I think its fair to say that if people were to stop using tb6600, technical support posts would drop 75% just leaving the rest of the pwm chip types and the misbranded for cnc 3d printing gear such as the xpro, pibot etc etc.
     
    C-CNC and BNMaker like this.
  10. Paruk

    Paruk Journeyman
    Builder

    Joined:
    Jul 13, 2014
    Messages:
    397
    Likes Received:
    149
    You're right about barking up the wrong tree. You seem to keep missing the point of what I trying to explain. Never mind, we say here in Thailand.

    I'm here to learn and share with others what I've learned so they can benefit of it too. It's the purpose of this forum, I guess. Sorry to say, but I don't learn anything of what you're posting. Your hexfets or mosfets are meaningless to me, since you fail to say anything about them or even fail to post the links for where to get them or read more about them. Neither did I ever seen any technical documentation of them posted here. Or did I miss that?
     
    #10 Paruk, Jul 26, 2015
    Last edited: Jul 27, 2015
  11. Jonny Norris

    Jonny Norris Well-Known
    Builder

    Joined:
    Aug 17, 2014
    Messages:
    495
    Likes Received:
    95
    yes it has been mentioned once or twice but its a long thread. it was just an example of quiet praux, but a search for "mosfet dsp microstepping driver" will point you in the right direction, dsp being the more expensive version but the same manufacturer will most likely sell the std mosfet ones too.

    I do know what your trying to explain, i once too had tb6600 drivers and have been through all the notions you are going through, so funny enough i can even relate. But to speed you up on your way.. your motors are not short of torque, upto 1/20 pos. more micro stepping should be fine, kernel speed yes 14khz is rubbish but should still operate your motor smoothly enough without noise. Your drivers are 2.5a max, this allows you upto 3a motors max. Good Luck.
     
    #11 Jonny Norris, Jul 27, 2015
    Last edited: Jul 27, 2015
  12. Paruk

    Paruk Journeyman
    Builder

    Joined:
    Jul 13, 2014
    Messages:
    397
    Likes Received:
    149
    Ok, I found an interesting problem with my TB6600 drivers. (This is where Jonny jumps up!;)). Somehow in the beginning I had the little current adjusters set too low. somewhere in the range of 1.5 Amp. So, while I was on it with changing the micro stepping, I tuned it up to see how far it would go. Just before the indicated position of 3.5 Amp I got about 1.5 Amp on each phase (looks you're right about that one, Jonny). So, these adjusting screws are crap and practically useless. Strange though, that the thing was running pretty smooth at lower current with 1/16 micro stepping.

    But if I cranked it up to its maximum setting, the drivers went in error and stopped. I then tuned it down little by little, but it only worked just above the 3.5 Amp setting. Still not getting more than 1.5 Amp.

    At 24V this translates in a power of 36Watt, which is not a lot, I guess. So to compensate the lack of power, my solution is to increase the voltage to 48V so the steppers can deliver 72Watt. That should do the trick.

    Weird enough, I was doing some test cutting today at 1/4 micro stepping, at lower feed speeds and plunge rates than usual, and the thing ended up with 5mm (!) in lost steps. Beats me. It should be stronger, but looses more steps than with less power?

    ****, everything was going well before, why did I had to start fumbling around with the stuff?
     
  13. Jonny Norris

    Jonny Norris Well-Known
    Builder

    Joined:
    Aug 17, 2014
    Messages:
    495
    Likes Received:
    95
    Well Think of it this way, at least you found fault before you plunged into an expensive piece of stock with an expensive bit, not mid job. Both stock and bit can cost double all the drivers put together, so a light escape in the scheme of things ;)

    The reason I expect that happened is that they are already optimised in that manner, they have large value resisters built in already which suggests the chips would fail without them even at 24v. Which ultimately is the cause of why they don't do what they say on the tin.
     
    #13 Jonny Norris, Jul 28, 2015
    Last edited: Jul 28, 2015
  14. Julius

    Julius Well-Known
    Builder

    Joined:
    Nov 13, 2015
    Messages:
    193
    Likes Received:
    61
    I know this is a year old, but I hope you guys know all Nema 23's are not made the same, the longer they are the stronger (higher torque) they are. My 175oz-in can cut through 1/2" of MDF in a single pass at 120ipm, 1/8 usteps. Theres also a difference between 24, 36, and 48V supplies.
     

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