Motion Control and Motor Drives

cancel
Showing results for 
Search instead for 
Did you mean: 

Reverse limit switch fail?

I am using a linear stage with 9512+7610 combo. The stage have two limit switches (as usual) and a linear encoder. I have an issue with one of the limit switches, which is on the negative direction of the encoder, so I call it reverse limit. The problem is very strange: the 9512 unit ignores the switch when the motor is enabled. This means, when I turn it with hands and hit the switch, the controller responses correctly; but when it drives the motor toward the switch, it go all the way to the hell until I disable the driver. As soon as the driver is disabled, it realizes the switch is on, and hold the motor. At the same time, the forward switch works as expected, stops the motor as soon as it is active. The engineer of the stage ensures me that both switches should act the same electronically, only the reverse switch is higher in mechanical precision. Both switches are connected to a common ground, so I used sourcing output setting for both on the 9512. 

 

To further diagnose the issue, I measured the voltage on the switch. Strangely, both switches seems to be working as expected: 0 V when not active, 11.6 V when active. However, for the limit switch, even I can see the voltage jump from 0 to 11.6 when it becomes active, the controller just totally ignores it: even the 'limit' LED on the unit does not light up. I also tried to switch the forward and reverse switch on the wiring, and still, the forward (working one) still works on the reverse port, and the reverse (not working one) act the same on the forward port.

 

I know this is a very confusing problem, so any suggestion or speculation is welcome. Thanks!

0 Kudos
Message 1 of 13
(5,281 Views)

Hi fineabcyu,

 

Thanks for posting in the forums! It seems this may be a compatibility issue with your linear stage and switches. Have you checked to make sure they are compatible? Has your system ever worked before or has this problem always occurred? 

Tina K.
Applications Engineering
National Instruments
ni.com/support
0 Kudos
Message 2 of 13
(5,241 Views)

Hi Catatina,

 

Thanks for the reply. This is a new system, and I have been in contact with engineers from both the stage manufacturer and NI to make sure they are compatible. The stage engineer says it is just a normal mechanical switch with one end connected to the ground wire (also connected to the motor body, but isolated from the motor actuator). All the voltage and ohm tests support these statements. However, it still seems to be different from just a mechanical switch, since I tried to disconnect the wire as the motor is moving and the driver correctly stops the motor. I feel that more tests are needed to find the cause of the issue, but I don't know what more I can do. Any suggestion is welcome!

 

Thanks,

Zhihao Yu

0 Kudos
Message 3 of 13
(5,239 Views)

 I tried to monitor the voltage of the switch output. When the switch is active and the motor is enabled, I see weird waveform, as attached. The reverse signal has a deeper vally, which I guess why it is not detected. However I am still not sure where does this signal come from. It is not there when the motor is disabled or the switch is not connected. Anyone can identify where are these signals from?

Download All
0 Kudos
Message 4 of 13
(5,229 Views)

Hold on a second.  Are you saying one side of the switch is connected to earth ground?  Please don't say that is so.  0V, yes.  Earth, no.  Assuming this is a linear servo motor, you are going to have all kinds of noise.

0 Kudos
Message 5 of 13
(5,220 Views)

It's a stepper motor, but anyway, you are right - all the noise seems to be from the motor. The ground connected to the motor body is not my wiring, it's built in the stage, damnit. 

 

So given this situation, any suggestion?

 

Thanks!

0 Kudos
Message 6 of 13
(5,213 Views)

If you have to keep the switch wired to earth, then make sure the motor has a heavy ground wire on it going to earth.  The larger the surface area the better.  Make sure the drive is well grounded also.  Is the motor cable shielded, and if so, at both ends?  Goes against convention, but shielding motor cable at both ends is proper for high freqency.  Whose stage is it?

0 Kudos
Message 7 of 13
(5,210 Views)

As you suggested, I checked the grounding of the motor and the driver. Since the entire setup is on a grounded optical table, everything seems to be grounded pretty well. I also tried to connect every gound port to the table with a wire but nothing seems to help at all. 

 

The motor and encoder cables are 15 pin d-sub cables with free leads connecting to the NI 37-pin terminal block and 7610 driver.

 

The stage is made by Thorlabs (LNR50SE)

 

Thanks!

0 Kudos
Message 8 of 13
(5,205 Views)

Is the motor cable shielded?

0 Kudos
Message 9 of 13
(5,198 Views)
Yes, they are all shielded d-sub cables.
0 Kudos
Message 10 of 13
(5,195 Views)