E-Stop Best Practice for Spindle

All things related to Centroid Oak, Allin1DC, MPU11 and Legacy products

Moderator: cnckeith

Post Reply
Talyrond
Posts: 5
Joined: Thu Sep 02, 2021 10:23 am
Acorn CNC Controller: No
Allin1DC CNC Controller: No
Oak CNC controller: Yes
MPU11 & GPIO4D -w/ 3rd Party Drives: No
DC3IOB: No
CNC12: Yes
CNC11: No
CPU10 or CPU7: No
CNC Control System Serial Number: none

E-Stop Best Practice for Spindle

Post by Talyrond »

Hi Guys, what is the expected behaviour when the E-Stop is pressed while the spindle is running. Should it decelerate to a stop as quickly as possible? This is what I have seen on all machines I have used. However once stopped should the power then be removed?
All the inputs to the drive – Servo enable and EMG cause the motor to freewheel to a stop when removed. So I can only kill the 0-10v to bring to an immediate stop, should I put that signal through the E-stop contactor?
I could be completely on the wrong path here, so any input would be much appreciated!

Control: Oak
Spindle: Chinese Servo Drive and Motor - AASD generic type. 0-10v determines the speed.

Thanks

Julian
cncsnw
Posts: 2414
Joined: Wed Mar 24, 2010 5:48 pm

Re: E-Stop Best Practice for Spindle

Post by cncsnw »

Any answer you get to this question is going to be qualified with "you have to do your own risk assessment" or similar sentiments.
Answers also depend very much on what type of spindle drive and emergency stop relay is installed in a particular system.

For many years, typical Centroid installations would only interrupt the run signal when emergency stop was pressed. This results in a ramped deceleration, but does not guarantee that a spindle drive malfunction might not cause motion even with emergency stop pressed.

The only method that ensures the spindle cannot start, restart, or continue running in the case of a drive failure (e.g. short in the command circuits) is to remove line power when emergency stop is pressed, via a suitably-sized contactor; or to use a drive with a true "safe torque off" or similar inhibit.

Many drives will ramp to a stop, using latent charge in the bus capacitors and regenerative energy to keep the logic circuits alive, in the event that line power is removed. It sounds like your Chinese drive does not do that, though.

Many modern drives have a "safe torque off" inhibit, that allows you to reliably inhibit the drive by switching a low-voltage control circuit. The main advantage to this is that you do not have to use a contactor large enough to break line power to the drive, and refraining from cycling line power increases the life span of the drive's bus-charging circuitry. Unfortunately, opening a safe torque off circuit while the spindle is running generally results in a free coast to stop: STO does, after all, prohibit torque.

In all normal conditions (no failure of the drive, no failure or error in your PLC software, no failure of your PLC output relays, etc.), when you press Emergency Stop, Centroid's PLC program will remove the spindle run command, and -- if you have your drive configured in a rational way -- the spindle should decelerate to a stop.

My preference in recent years has been to use a safety relay with off-delay contacts to control the spindle drive's "safe torque off" contacts. This gives the normal control sequences time to decelerate the spindle to a stop, but then reliably applies the safe torque off inhibit to ensure that the spindle cannot restart or continue to run.
https://www.automationdirect.com/adc/sh ... 92-61-24-5
Talyrond
Posts: 5
Joined: Thu Sep 02, 2021 10:23 am
Acorn CNC Controller: No
Allin1DC CNC Controller: No
Oak CNC controller: Yes
MPU11 & GPIO4D -w/ 3rd Party Drives: No
DC3IOB: No
CNC12: Yes
CNC11: No
CPU10 or CPU7: No
CNC Control System Serial Number: none

Re: E-Stop Best Practice for Spindle

Post by Talyrond »

Thank you for the comprehensive reply, that helps a lot and understand all advice given is just that.

The EMG signal is described so: prohibits drive motor, cutting motor currents.
That sounds like STO to me, what do you think?

The manual recommends a contactor is used for the main power, so I guess worse case I can use a safety relay with off-delay contacts to remove power via the main contactor after the spindle has stopped?

Thanks

Julian
cncsnw
Posts: 2414
Joined: Wed Mar 24, 2010 5:48 pm

Re: E-Stop Best Practice for Spindle

Post by cncsnw »

The EMG signal is described so: prohibits drive motor, cutting motor currents.
That sounds like STO to me, what do you think?
It does kind of sound that way. I am not sure I would fully trust it, though.

Does the drive manufacturer say anything about a "Safety Integrity Level" (SIL) or IEC 62061? If not, then I would stick with interrupting the line power.
The manual recommends a contactor is used for the main power, so I guess worse case I can use a safety relay with off-delay contacts to remove power via the main contactor after the spindle has stopped?
Yes, that is a reasonable solution, which I have used on multiple systems when a Safe Torque Off function was not available.
Talyrond
Posts: 5
Joined: Thu Sep 02, 2021 10:23 am
Acorn CNC Controller: No
Allin1DC CNC Controller: No
Oak CNC controller: Yes
MPU11 & GPIO4D -w/ 3rd Party Drives: No
DC3IOB: No
CNC12: Yes
CNC11: No
CPU10 or CPU7: No
CNC Control System Serial Number: none

Re: E-Stop Best Practice for Spindle

Post by Talyrond »

No mention of "Safety Integrity Level" (SIL) or IEC 62061, so I'll go with the main contactor solution.

Thank you very much for the help.

Julian
ashesman
Posts: 306
Joined: Thu Dec 03, 2020 4:54 am
Acorn CNC Controller: No
Allin1DC CNC Controller: No
Oak CNC controller: Yes
MPU11 & GPIO4D -w/ 3rd Party Drives: No
DC3IOB: No
CNC12: Yes
CNC11: No
CPU10 or CPU7: No
CNC Control System Serial Number: none

Re: E-Stop Best Practice for Spindle

Post by ashesman »

If you are going to cut the power to the VFD, consider the inrush current when enabling your contactor. There can be up to hundreds of amps for a short period of time while the VFD chargers its capacitors. This is drawn while the contactor is closing and possibly still bouncing. High end VFDs have inrush current regulation and current waveform shaping. Basic VFDs (most used on smaller machines, even good brands), basically just connect the capacitors through a rectifier directly to the mains. Its not an issue, just make sure your contactor is appropriately rated.

Also, If the VFD doesn't have a backup power supply, then it will appear to the controller as in fault state while the E-Stop is pressed. You may need to modify your PLC code to have a delay between releasing E-Stop and checking the VFD fault.

EMG signal is usually a digital input that can be configured to act as emergency stop. This signal goes through the VFDs CPU. It is not the same as STO. STO is a special pair of signals that most must be independently in the correct states and are hard wired directly to the disable of the output power driver of the VFD. STO is required to have two isolated inputs by regulation and a big section in the manual dedicated to STO functionality. The drive will have STO labelled inputs if it supports proper STO.

Note that STO and a lot of emergency stop systems coast the motor to a stop.

An alternate option is to put a contactor after the VFD in the motor cables. That leaves the VFD powered but ensures the motor can not be driven or electrically unsafe.
martyscncgarage
Posts: 9441
Joined: Tue Mar 28, 2017 12:01 pm
Acorn CNC Controller: Yes
Allin1DC CNC Controller: Yes
Oak CNC controller: No
MPU11 & GPIO4D -w/ 3rd Party Drives: No
DC3IOB: No
CNC12: Yes
CNC11: Yes
CPU10 or CPU7: Yes
CNC Control System Serial Number: none
Location: Mesa, AZ

Re: E-Stop Best Practice for Spindle

Post by martyscncgarage »

ashesman wrote: Wed Dec 29, 2021 2:48 am
An alternate option is to put a contactor after the VFD in the motor cables. That leaves the VFD powered but ensures the motor can not be driven or electrically unsafe.
Ashesman,
I was always told it was not good on a VFD to switch its load AFTER the VFD.
If a VFD got powered down on the line side, there should be no inrush as the control would also be faulted. When the control is reset to its ready state, the VFD would be powered up and ready to go, but not turn the motor until commanded by the control.

The downside to putting a contactor ahead of the VFD, (cncsnw could you chime in on this) is that the VFD would have to boot up, and its fault line may force the control to a SpindleFault until it came on line, in which case the Estop button would have to be cycled.

(Sake of discussion here)
Marty
Reminder, for support please follow this post: viewtopic.php?f=20&t=383
We can't "SEE" what you see...
Mesa, AZ
ashesman
Posts: 306
Joined: Thu Dec 03, 2020 4:54 am
Acorn CNC Controller: No
Allin1DC CNC Controller: No
Oak CNC controller: Yes
MPU11 & GPIO4D -w/ 3rd Party Drives: No
DC3IOB: No
CNC12: Yes
CNC11: No
CPU10 or CPU7: No
CNC Control System Serial Number: none

Re: E-Stop Best Practice for Spindle

Post by ashesman »

martyscncgarage wrote: Thu Dec 30, 2021 11:00 am
ashesman wrote: Wed Dec 29, 2021 2:48 am
An alternate option is to put a contactor after the VFD in the motor cables. That leaves the VFD powered but ensures the motor can not be driven or electrically unsafe.
Ashesman,
I was always told it was not good on a VFD to switch its load AFTER the VFD.
If a VFD got powered down on the line side, there should be no inrush as the control would also be faulted. When the control is reset to its ready state, the VFD would be powered up and ready to go, but not turn the motor until commanded by the control.

The downside to putting a contactor ahead of the VFD, (cncsnw could you chime in on this) is that the VFD would have to boot up, and its fault line may force the control to a SpindleFault until it came on line, in which case the Estop button would have to be cycled.

(Sake of discussion here)
Marty
Inrush current occurs whenever the VFD is powered up. Its DC bus capacitors are charged immediately. It happens very quickly and can see very high peak currents. Most contractors have very high short circuit current ratings often exceeding 1500 Amps and are more than capable of switching on loads with a high inrush current. I just dont like the idea of powering up and down the VFD on every E-Stop unnecessarily. For inrush and the startup time as you have mentioned.

I have seen plenty of VFD installs with isolator switches on the motor side. Especially where the motor is not hard wired. I guess it would be a matter of referring to the VFD manual to find out if interrupting the motor is an issue.

STO is designed exactly for this job. Although a negative of STO is that it will often latch the fault requiring a power cycle to clear so you need a contractor anyway!
Post Reply