Encoder for stepper motor - Page 3


Page 3 of 3 FirstFirst 123
Results 41 to 52 of 52

Thread: Encoder for stepper motor

  1. #41
    Member TomKerekes's Avatar
    Join Date
    May 2006
    Location
    USA
    Posts
    4043
    Downloads
    0
    Uploads
    0

    Default Re: Encoder for stepper motor

    Hi ssjantonio,

    Thanks for all the good info.

    3.7V will not cause damage to KFLOP inputs.

    Because the Encoder can source -20ma you should be able to use IO 16-21 that have 150 ohm pull down resistors for the Index pulses. But you can not have 220 ohm series resistors because the voltage will be divided to too low of a level. A minimum of 2.5V is required at the pin. 47ohms should work.

    I don't understand your statement about ReadBit(24) always reads true. How would you know that? it should be very simple. If the voltage at the pin is < 0.5V then the Digital IO Screen should show off, and ReadBit should return false. If those are not all correct then you are making a mistake somewhere. With noise the ReadBit may sometimes detect an invalid state, but not always.

    Note:your diagram doesn't show any connection to Bit 24. It also is missing one of the 5 input connections.

    Regards

    Regards
    TK http://dynomotion.com


  2. #42
    Registered
    Join Date
    May 2013
    Location
    México
    Posts
    31
    Downloads
    0
    Uploads
    0

    Default Re: Encoder for stepper motor

    Hi Tom,

    Quote Originally Posted by TomKerekes View Post
    Hi ssjantonio,

    Thanks for all the good info.

    3.7V will not cause damage to KFLOP inputs.

    Because the Encoder can source -20ma you should be able to use IO 16-21 that have 150 ohm pull down resistors for the Index pulses. But you can not have 220 ohm series resistors because the voltage will be divided to too low of a level. A minimum of 2.5V is required at the pin. 47ohms should work.
    Ok, I will to connect with 47ohm to the pins 16-21 and with 0.1uF capacitor, that it is what I have.

    I don't understand your statement about ReadBit(24) always reads true. How would you know that? it should be very simple. If the voltage at the pin is < 0.5V then the Digital IO Screen should show off, and ReadBit should return false. If those are not all correct then you are making a mistake somewhere. With noise the ReadBit may sometimes detect an invalid state, but not always.
    I know because I'm running this code:
    Jog(AXIS,100);
    while(!ReadBit(25));
    Jog(AXIS,0);
    And the motor, simply does not move, but it is history, I will to test with the rc filter, I have attached dthe diagram, can you help me to see if its ok?

    Note:your diagram doesn't show any connection to Bit 24. It also is missing one of the 5 input connections.
    Yes I'm sorry, it was because I was probing with pin 16-21. Thank you, I have added the missing connection .


    What about the GND connections of the last diagram? It is ok?


    I have tried to connect the shield of the encoders to earth ground instead of GND in KFlop, together to the shield of the cables, and the noise reduces almost 30% with out the filter, but I dont know if that is ok or it can damage the encoders?


    Thank you very much for your help and your patience!

    Encoder for stepper motor-diagrama_rc_filter-png
    Encoder for stepper motor-board_rc_filter-png



  3. #43
    Member TomKerekes's Avatar
    Join Date
    May 2006
    Location
    USA
    Posts
    4043
    Downloads
    0
    Uploads
    0

    Default Re: Encoder for stepper motor

    Ok, I will to connect with 47ohm to the pins 16-21 and with 0.1uF capacitor, that it is what I have.
    good

    What about the GND connections of the last diagram? It is ok?
    You didn't label the signals. I assume pin6 of the encoders is the shield? If yes, then yes, connecting to KFLOP GND Lug should be a good choice.

    I have tried to connect the shield of the encoders to earth ground instead of GND in KFlop, together to the shield of the cables, and the noise reduces almost 30% with out the filter, but I dont know if that is ok or it can damage the encoders?
    I doubt if it would cause damage to Earth Ground the shields. But I'm surprised Earth GND is quieter than KFLOP GND. It may depend how you are connecting/grounding the scope. The noise relative to KFLOP GND is what matters be cause the input is to KFLOP. I would ground the encoder shields to KFLOP GND.

    I will to test with the rc filter, I have attached dthe diagram, can you help me to see if its ok?
    The circuit with 8 RC Filters looks good to me.

    Regards

    Regards
    TK http://dynomotion.com


  4. #44
    Registered
    Join Date
    May 2013
    Location
    México
    Posts
    31
    Downloads
    0
    Uploads
    0

    Default Re: Encoder for stepper motor

    Hi Tom, it is working now, perfectly. But I have another problem :S.
    When the motores are working with closed loop, they heath a lot more than working without CL, and when the motors are stoped they oscilate between +/- 7 micro steps, maybe that is causing the overheating?
    The motors are stepper Nema34 1600oz-in, the driver is DQ860MA configured for half current, 5.6A Peak and 800 pulses/rev. The maximum. velocity of the motor is 4500 pulses/s. The power supply is 60VDC.
    Do you think that the oscilation is also caused by the noise?

    The PID is:
    P = 0
    I = 0.1
    D = 0

    The encoder is 1024pulses/rev.

    Thank you!
    Best Regards!



  5. #45
    Member TomKerekes's Avatar
    Join Date
    May 2006
    Location
    USA
    Posts
    4043
    Downloads
    0
    Uploads
    0

    Default Re: Encoder for stepper motor

    Hi ssjantonio,

    Oscillation should not cause the motors to run hotter unless you are counting on the motors shutting down when idle which doesn't happen when oscillating. In this case a long running job would overheat your motors anyway.

    Any closed loop servo will attempt to drive the error to zero and tend to constantly oscillate. It is probably caused by mechanical dynamics rather than noise. You might read this from our wiki.

    Servo Dither/Hunting - (oscillation around zero error)


    You didn't include the rest of your settings, the filters, or a plot. Reducing the I Gain may reduce the severity of the oscillation. There is also a Dead Band Range and Gain settings that can sometimes get the system to sit still. For example a range of 7 counts with zero gain will cause the servo to not make any corrections when the error is less than 7 counts. Of course this may increase your total errors by a small amount.

    There are also some "anti-dither" C Program Examples that can watch for no motion and small errors and change the gain settings as an attempt to get the axis to sit still.


    HTH
    Regards

    Regards
    TK http://dynomotion.com


  6. #46
    Registered
    Join Date
    May 2013
    Location
    México
    Posts
    31
    Downloads
    0
    Uploads
    0

    Default Re: Encoder for stepper motor

    Quote Originally Posted by TomKerekes View Post
    Hi ssjantonio,

    Oscillation should not cause the motors to run hotter unless you are counting on the motors shutting down when idle which doesn't happen when oscillating. In this case a long running job would overheat your motors anyway.

    Any closed loop servo will attempt to drive the error to zero and tend to constantly oscillate. It is probably caused by mechanical dynamics rather than noise. You might read this from our wiki.

    Servo Dither/Hunting - (oscillation around zero error)


    You didn't include the rest of your settings, the filters, or a plot. Reducing the I Gain may reduce the severity of the oscillation. There is also a Dead Band Range and Gain settings that can sometimes get the system to sit still. For example a range of 7 counts with zero gain will cause the servo to not make any corrections when the error is less than 7 counts. Of course this may increase your total errors by a small amount.

    There are also some "anti-dither" C Program Examples that can watch for no motion and small errors and change the gain settings as an attempt to get the axis to sit still.


    HTH
    Regards
    Hi Tom, the over heating was fixed with the program anti dither, now is working fine. I have a trifasic async motor controled with a standar VFD driver with 0-10V input.
    I'm testing now the example KonnectPWMToAnalog, it works fine, but the voltaje at the output of the konnect oscillates always +/- 0.2V, The velocity oscillates +/-3rpm (measured every 0.05s with a encoder), do yo think that is enough for ridgid tapping? Or there is a way to have a better control on the speed with this method? I was looking for a driver with Pulse Train input as second option.

    Thank you very much!



  7. #47
    Member TomKerekes's Avatar
    Join Date
    May 2006
    Location
    USA
    Posts
    4043
    Downloads
    0
    Uploads
    0

    Default Re: Encoder for stepper motor

    Hi ssjantonio,

    Strange that the Analog Output would oscillate that much. In our setup we only measured ~ 10mv peak-peak. It also seems odd that the RPM would vary that rapidly? Maybe you explain in more detail what you are seeing.

    But it is likely to be ok for rigid tapping as the Z axis should follow the Spindle motion regardless of the Speed.

    Pulse Train would be another option as long the result is closed loop in a manner that the spindle position is properly tracked without drift that would break the tap.

    HTH
    Regards

    Regards
    TK http://dynomotion.com


  8. #48
    Registered
    Join Date
    May 2013
    Location
    México
    Posts
    31
    Downloads
    0
    Uploads
    0

    Default Re: Encoder for stepper motor

    Quote Originally Posted by TomKerekes View Post
    Hi ssjantonio,

    Strange that the Analog Output would oscillate that much. In our setup we only measured ~ 10mv peak-peak. It also seems odd that the RPM would vary that rapidly? Maybe you explain in more detail what you are seeing.

    But it is likely to be ok for rigid tapping as the Z axis should follow the Spindle motion regardless of the Speed.

    Pulse Train would be another option as long the result is closed loop in a manner that the spindle position is properly tracked without drift that would break the tap.

    HTH
    Regards
    Hi Tom, now is working fine, I wrote a C program to control the Vc with PID using the encoder, and it works very good, it only oscilates 0.5rpm. It was a problem with noise in the 12V power source.

    But now, I have another problem.

    I have 3 siemens contactors (3RTU1035 coil of 24VDC), the coil has a reversed diode, the negative is connected to A2 of the coil, and the A1 is connected to the Konnect board.
    When the coil is activated multiple times (between 3 and 10 times) Kmotioncnc freeze with error "Read Failed Auto Disconnect".
    I need to kill the process of KmotionCnc and reconnect the USB from Kflop.
    It looks like Kflop does not reset because all the programs continue working.

    If I run a long gcode from KmotionCnc it can be working for hours without problems, only if I never operates the contactor.

    Te homing with index is working fine, the Closed Loop, the coordinated motion, etc.

    -The USB of KFlop is connected with an USB Isolator from olimex.
    -KFlop is connected to a power source of 5V, the earth ground is connected.
    -The GND Hole of Kflop is only connected to the Shield of the encoders and to the GND Hole of the Konnect board.
    -There is not connection between Earth ground and DC ground in any component (I haved checked it with a ohmetro).
    -The computer is connected to a no-break, and the power source of kflop is connected directly to the line, using an EMI Filter.

    I have noticed that when it happends in the board menu of kmotion says "0x0 0xFFFFFFF" instead of display nothing.

    As you can see in the pictures I have 4 bifasic 220V fans, when I connect them there is a continuity between the phase line 1 and Earth Ground (700-900ohm), and the same with the phase line 2, its normall?

    I have attached some pictures of the installation.

    I'm sorry for my English! :S

    I really appreciated your help. Thank you very much!!

    Encoder for stepper motor-20161209_201009-1008x756-jpg
    Encoder for stepper motor-20161209_201023-1008x756-jpg
    Encoder for stepper motor-20161209_201132-756x1008-jpg
    Encoder for stepper motor-20161209_201154-1008x756-jpg

    EDIT: I forgget to mention that sometimes the PC goes to blue screen with error "KERNEL SECURITY CHECK FAILURE"

    Last edited by ssjantonio; 12-10-2016 at 02:57 PM.


  9. #49
    Member TomKerekes's Avatar
    Join Date
    May 2006
    Location
    USA
    Posts
    4043
    Downloads
    0
    Uploads
    0

    Default Re: Encoder for stepper motor

    Hi ssjantonio,

    Hmmm. Here are some thoughts:

    The USB cable seems like it is excessively long and coiled up.

    The USB cable doesn't seem to have a Ferrite ring

    The USB cable seems very thin? You might check the shield continuity from one end to the other (should be less than 1 ohm)

    You might check the USB shield connections as it goes through your panel. The shield might not have connectivity from both sides of the panel and/or it may be connected to the panel which may not be desirable.

    The diode across the Relay coils must be fast switching

    What power supply are you using for the relay coils? Is it is isolated from everything else?

    Consider using Solid State Relays to possibly avoid the issue.

    Try disconnecting whatever the relays are controlling to determine if it is the coils causing the problem or the device the relay is controlling.

    Is the PC earth grounded?

    Regards

    Regards
    TK http://dynomotion.com


  10. #50
    Registered
    Join Date
    May 2013
    Location
    México
    Posts
    31
    Downloads
    0
    Uploads
    0

    Default Re: Encoder for stepper motor

    Quote Originally Posted by TomKerekes View Post
    Hi ssjantonio,

    Hmmm. Here are some thoughts:

    The USB cable seems like it is excessively long and coiled up.

    The USB cable doesn't seem to have a Ferrite ring
    I have already changed it for a shorter cable with ferrite ring , thank you for the recommendation.

    The USB cable seems very thin? You might check the shield continuity from one end to the other (should be less than 1 ohm)
    I have checked it with the new cable and it has continuity, 0.1ohm

    You might check the USB shield connections as it goes through your panel. The shield might not have connectivity from both sides of the panel and/or it may be connected to the panel which may not be desirable.
    From computer to the USB Isolator it has continuity with the ground Earth, then, from de USB Isolator to KFLOP it has continuity with DC Ground hole of KFLOP but does not have continuity with earth ground.

    The diode across the Relay coils must be fast switching
    Ok, I have UF5408 and NTE569, wich of them do you recommend? Or another model?

    What power supply are you using for the relay coils? Is it is isolated from everything else?
    I have a 24V Delta power supply, it is used only for the external buttons and the coils.

    Consider using Solid State Relays to possibly avoid the issue.
    Ok, I will to replace them.

    Try disconnecting whatever the relays are controlling to determine if it is the coils causing the problem or the device the relay is controlling.
    After changing the USB cable the problem is still present.

    The three contactors that you can see in the picture are controlling:
    1.- A trifasic async motor of 1/8HP (Screw conveyor transporter)
    2.- A monofasic 120V motor of 1/8HP (Coolant pump)
    3.- A monofasic 120V motor of 80W motor (Oil pump)

    I have disconnected them, and the problem is gone, so it is not problem of peak voltaje in the coil, right?


    Is the PC earth grounded?
    Yes, it is connected to earth ground.

    Best Regards!



  11. #51
    Member TomKerekes's Avatar
    Join Date
    May 2006
    Location
    USA
    Posts
    4043
    Downloads
    0
    Uploads
    0

    Default Re: Encoder for stepper motor

    Hi ssjantonio,


    I have checked it with the new cable and it has continuity, 0.1ohm
    Did you measure through the shield (shell to shell)? Or through the GND wire?

    Ok, I have UF5408 and NTE569, wich of them do you recommend? Or another model?
    I'm not an expert but I think those are a little slow (50ns and 250ns). I've seen references that show the voltage spike rising in 200~300ns. Something like a 1N4148 should start conducting in 4ns before the voltage has time to rise much.

    The three contactors that you can see in the picture are controlling:
    1.- A trifasic async motor of 1/8HP (Screw conveyor transporter)
    2.- A monofasic 120V motor of 1/8HP (Coolant pump)
    3.- A monofasic 120V motor of 80W motor (Oil pump)

    I have disconnected them, and the problem is gone, so it is not problem of peak voltaje in the coil, right?
    I think that is probably correct. However there might be a possibility the coil sparking is radiated through the wiring. Disconnecting the wiring would then cause the problem to go away. Did you leave the wiring intact to the relay but disconnect the power? Or did you disconnect the wiring from the relay contacts. You might also try them one at a time to see if it is one that causes the problem.

    Regards

    Regards
    TK http://dynomotion.com


  12. #52
    Registered
    Join Date
    May 2013
    Location
    México
    Posts
    31
    Downloads
    0
    Uploads
    0

    Default Re: Encoder for stepper motor

    Hi tom, I have changed the coil relays for solid state relays, and the problem was gone, I post my comments for reference of other users. Thank you very much!

    Quote Originally Posted by TomKerekes View Post
    Hi ssjantonio,
    Did you measure through the shield (shell to shell)? Or through the GND wire?
    I checked Shell to Shell and GND wire too, both of them has near to 0 ohm (I say near to 0 because my ohmetro shows 0 when the resistance is below 1ohm).

    I'm not an expert but I think those are a little slow (50ns and 250ns). I've seen references that show the voltage spike rising in 200~300ns. Something like a 1N4148 should start conducting in 4ns before the voltage has time to rise much.
    Yes you are right, those models was slower, I changed the diodes for 1N4148, and the problem still persist after that.

    I think that is probably correct. However there might be a possibility the coil sparking is radiated through the wiring. Disconnecting the wiring would then cause the problem to go away. Did you leave the wiring intact to the relay but disconnect the power? Or did you disconnect the wiring from the relay contacts. You might also try them one at a time to see if it is one that causes the problem.
    With the monophasic motors the problems was much less frequently, maybe 1 in 50, but with the trifasic the problem happends 1 in 10, or sometimes more frequently, I think the theory of the coil sparking is correct.


    Thank you very much for your help!

    Regards.



Page 3 of 3 FirstFirst 123

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
  •  


About CNCzone.com

    We are the largest and most active discussion forum for manufacturing industry. The site is 100% free to join and use, so join today!

Follow us on


Our Brands

Encoder for stepper motor

Encoder for stepper motor