Mister T

Hello,

thank you for the additional information.

Some more questions:

• What version of the engineering software (TIA V17?) do you use?
• What firmware version of the S7-controller (S7-1214) do you use?
• How fast do you drive your axis?

I can install with tia v13, the cpu could not be upgraded, the servo works at max 50khz, not a very fast machine

Mister T

Hello,

thank you for the additional information.

Some more questions:

• What version of the engineering software (TIA V17?) do you use?
• What firmware version of the S7-controller (S7-1214) do you use?
• How fast do you drive your axis?

I use mc_moverelative to drive, once I give a pulse, the position is completed. My aim is to stop the position when I see the photocell on the way to the position, but mc_halt does not stop it.

andolsun

I have 6ES7-214-1AE30-0XB0 series product I have SINAMICS s110 cu 305 dp with this product I cut bags with this product I chose from the technology catalog I drive a pulse with mc_moverelative when I see the photocell it needs to stop the position and cut I do it with mc_halt but it doesn't work what should I do

Hello.

Here is the secret for your CPU:
Use MC_Velocity where the speed is set to 0. 
Use this 0 speed instruction instead of the halt instruction.

Does this work?

I haven't tried it yet, will it cut off when the sensor sees it?

Mister T

Hello,

some more points to check from my side...

• Please check your program, if the call of MC_Halt is done in every cycle of the program (OB 1). You have to call the function block (every motion control function block) in every program cycle, even if it is called with execute=false.
  
  
• Do you really generate a rising edge at the execute input of the motion control function block MC_Halt? To start the function, you need a signal change at the execute input from false to true (rising edge). Otherwise the function will not start...
  
  While the function is working, you could leave the execute=true. In this case, you could see the function block reactions at the output signals, like busy, commandAborted, done or error.
  If execute=false while the function block is working, you only get the output signals for one program cycle.
  
  
• Are you sure, that your drive is following the pulse output of the S7-1200 controller correctly?
  
  If you want to stop with MC_Halt, the pulses will go down with the parameterized deceleration in the settings of the technology object (TO) or at the function block MC_Halt. If you use a ramp generator in the drive, it could be, that the drive will not stop yet, if the pulse input of the drive will stop...

Maybe these points are helpful for you, to solve the problem. Please check the settings an the program. Otherwise, please ask again and tell us, what is going on...

 Hello,

Does using MC_Stop will have adverse effect or will it work same as MC_Halt.

Because i noticed that when axis is 95% at stopping MC_Halt has no effect.

mister DP how are you?

I wish I had paid more attention to your response at the time.

Basically what this program does si to use mc_moveabsolute from -1.0E+7 waiting for the home input signal to position at 0.0, then do home and from there on, the mc_moveabsolute must stop at the programmed X distance. I didn't know this trick can be done and is absolutely mind blowing.

So, what i want to ask is this: what if there are some dirt that makes the sensor trigger false inputs? how can i reduce the measurement size of the home input signal?

can i change the MIN_POSITION to somethig in the size of 1.3*X and put some code to start the homing only when i am reaching maybe 90% of X? do i need to put this in a cyclic OB to maximize accuracy?

thanks in advance!

EDIT: mr DP has left the forum... he was such a nice guy and always tried to help us. i hope he is ok.
I will miss him 

Walter Jacquet

mister DP how are you?

I wish I had paid more attention to your response at the time.

Basically what this program does si to use mc_moveabsolute from -1.0E+7 waiting for the home input signal to position at 0.0, then do home and from there on, the mc_moveabsolute must stop at the programmed X distance. I didn't know this trick can be done and is absolutely mind blowing.

So, what i want to ask is this: what if there are some dirt that makes the sensor trigger false inputs? how can i reduce the measurement size of the home input signal?

 

can i change the MIN_POSITION to somethig in the size of 1.3*X and put some code to start the homing only when i am reaching maybe 90% of X? do i need to put this in a cyclic OB to maximize accuracy?

 

thanks in advance!

EDIT: mr DP has left the forum... he was such a nice guy and always tried to help us. i hope he is ok.
I will miss him 

1) The distance between your reference sensor and the standstill position, should be more than the distance that the movement takes to come to a halt (max speed -> zero speed).

Hi Mr Qwazee.

i have issues trying to cuote or reply you since this web is asking me to switch to the old version every time i want to reply to someone.

So Basically...

this is exactly what i need.

Some questions in mind. I do traces of the speed and it appears to be a bell ( i use Jerk limit). The deceleration is done from the middle of the bell onwards, right? or just i need to avoid it when the speed is reaching zero ? If this is true, then i must trigger the home before the mid of the bell.

another question i have. I read some siemens file where it says the commands need to be placed inside a Cyclic OB, is this true? or its jus the execute commands that needs to be placed there?

and the last question, does this home/moveabsolute trick works if i use Motion Control with analogic output and encoder input? Thanks in advance!