Safey issue with motors & VFDs

BobB said:

The advice I have from my drives manufacturer is "DO NOT USE A CONTACTOR IN THE OUTPUT LINES FROM THE DRIVE!! IT IS NOT GOOD FOR THE DRIVE!!!"

Click to expand...

It may not be a good idea to open an output contactor while a drive is energized (running), but I won't have a piece of equipment without output isolation contactors, for both AC and DC drives.

My normal sequence is to have a two-mode safety relay, with a 2 to 3 second drop out delay between instant and delayed. The instant contact forces a fast stop, the delayed contact removes a drives enable, and drops out the isolation contactors.

After all, in an emergency stop, I want the equipment to get to a zero energy state as soon as possible, but if it can't, or something else is wrong, I want to remove all sources of power to driven devices.

And as a nod to human stupidity, Lockout/Tagout procedures are NOT always followed, no matter what. Without an Isolation contactor on a drive output, even if it is de-energized, there will generally be a high floating voltage to ground. Physically opening the outputs fixes that.

To re-iterate what has already been said; use braked motors with output contactors on the VFDs and interlocked guarding. Opening the guard stops the motor and disconnects and disables the VFD. If the guards are to be opened frequently (i.e more that 6 per hour typically) then keep the VFD powered up and use the contactors on the VFD outputs otherwise the frequent on/off of the power will damage the VFD. If there is a significant danger from the brushes then you may need redundant contactors and solenoid locking on the guards as well. This machine is likely to need safety control category 3 which requires redundant devices so that the failure of 1 device does not cause a hazard and the the failure can be detected. STI or Pilz will give you more information on control categories.

Why not do what they always do? Wait for someone to get killed or hurt seriously, then put in the necessary safety.

Our transportation department has been using this method for stop lights for at least the last 60 years.











No, don't do it. It is called "irony."

I use the exact wiring as described in the original post...except...I use a timed off delay contact in the main safety relay to drop out an auxilliary safety relay which opens the contactors between the drives and motors. The PLC monitors the main safety relay and sets the drive command to zero speed. The time delay in the relay is set slightly longer than the decel rate programmed in the drives. Mine are usually about 2-3 seconds. That's enough time, in my applications, to brake the load using the VFD. It functinos like this: The gate/curtain is entered, the drives immediately ramp to zero speed and stop, then the contactors open. This way, the motion is stopped as soon as possible and a PLC/drive hiccup can't cause unexpected motion while the safety circuit is broken. Some high inertia loads require mechanical braking or dynamic braking to augment the drive capabilities to meet the stop time requirement of the risk assessment. I've never had a drive that suffered any ill consequences due to this arrangement. I've done it with a lot of servo drives too without problem. I would not recommend opening the contactor frequently while the drive is enabled and running. Always command the drive to stop first if possible.

(safety relay example p/n ALLEN-BRADLEY P/N 440R-G23029)

Just as a side note:

If you go with redundant monitored input devices to your safety relay (e-stop buttons, guard switches, etc) do you also go with redundant monitored output devices (drive contactors, control power supply contactors, etc). I don't see the utility of one without the other. If we trust a single contactor between the drive and motor to open we should also trust a single e-stop channel and and a standard relay (albeit high quality) for our e-stop relay.

The reason I bring this up is I see all sorts of machines that use 'safety relays' and dual e-stop channels. But very few machines use output side redundancy and monitoring, especially on the power devices.
Keith

Thanks for everyone's responses thus far.

I think the best solution would be a brake motor or some kind of mechanical brake, as was suggested, but we just don't have the room. I don't have a problem with using timed locks on the guard doors, but I have no doubt that they would eventually be removed at management's request, after continual griping by the operators. (Unfortunately it seems that keeping everyone happy is sometimes more important than safety.)

There doesn't seem to be a consensus on whether having a mechanical contactor between the motor and drive is a good idea. I have read some drive manuals that warn against it, although my Delta book does not. One post mentioned that dropping the enable line on their brand of VFDs with a safety relay is an acceptable e-stop method per the documentation. My Delta book doesn't say anything about that but I left a message with our Delta rep to find out.

I added some braking resistors to the drives this morning and found that the motors could ramp down in as little as 1-2 seconds (even with the heavier brushes), which would be acceptable. So if I could safely remove the contactors from the system, I think my problem would be solved. We are using a Banner PicoGuard safety controller to handle the E-stop buttons and optical guard door switches. We could easily tie the safety relay in to the enable lines on the drives and... Poof! problem solved... Or is it?

To answer a few questions:
- Yes they're only fractional motors but still very dangerous. If we leave the system the way it is, it's only a matter of time before somebody would lose a finger. This I am sure of.
- The operators do need to get inside the machine, maybe up to several times an hour for setups.
- You can't rely on training. It's a nice idea, but from a safety standpoint, not worth the risk. My last machine project had a number of razor-sharp cutters inside. I went to great lengths to guard the machine and make it safe. One operator still had a hospital visit because he removed a guard (with the machine off) and accidentally slammed his hand into one of the non-moving cutters. So we had to make a special cover that the operators are supposed to place over the cutters whenever they're working inside the machine. The last time I was over there, the cover was hanging on the wall with about 1/2" of dust on it.

Eric great sign. I cannot believe they had the guts to put it on.
I wonder if it was worth the time and cost. I am very skeptical on signs by the way.

The last time I was over there, the cover was hanging on the wall with about 1/2" of dust on it.

Unfortunately it seems that keeping everyone happy is sometimes more important than safety.)

The above two statements probably summarize the reasons I got out of the safety business.

Please don't get me wrong I am all for protecting people from hazards and good equipment good guards and basically making sure everyone goes home at the end of the day. I am just burned out on all the politics and useless OSHA paper and being a slave to the paper god at a desk when I should have been out on the job with the people doing the work.

Dan Bentler

keep the contactors

If your drives will stop the load in less than 3 seconds, I would recommend that you keep the contactors, just keep them closed for 3 seconds following the e-stop via timed contact on a safety relay. I have seem drives "lose their marbles" and run when they aren't supposed to. I don't trust a drive digital input with operator safety. I trust a contactor a lot more, especially if it's correctly controlled and monitored.

Check to see if your PicoGuard has a timed contact that can be used to operate the contactors, then your problem will be solved without keeping an unbroken connection between the drives and motors.

Paul C.

I think the best solution would be a brake motor or some kind of mechanical brake, as was suggested, but we just don't have the room.

Click to expand...

Really ? A brake will usually be installed on the shaft where the impeller for the ventilation is installed. The overall length of the motor will only increase with approx 15%. The motor diameter will not be affected.
NB: Obviously the motor has to be made for the brake option. You cant retrofit the brake onto a motor where it wasnt designed into it initially. The shaft just isnt long enough.

Instead of a timer (that could be said to "guess" when the motor has stopped) AB makes a Back EMF Monitor that works in combination with a safety relay. It detects stopped motion "with single or three-phase induction motors by measuring the drive voltage and the back electro-magnetic field of the motor." You could then drop the contactor on the VFD output after confirming its fully stopped and allow the door to open.

The part number and link is 440R-S3500x.

I hadn't seen this product before and will be sure to use it in the future.

Your link didn't work for me (blank window), but I found the PDF of that catalog page HERE in case anyone else has difficulties.

The sample diagram shows it connected to a motor started across the line. I wonder if this will work properly with a VFD controller motor?...



-Eric

Eric

Spec sheet says not to use it on VFD

Looks to be a right handy unit for rolls and other equipment.

Dan Bentler

A possibility would be to use one of the drives digital inputs to make the drive rapidly decel the motor and use the injection brake to help bring it to a stop when the gate is opened.

Another would be to have a really LOUD alarm go off when the gate is opened and the drives have not sent a zero speed signal to the PLC.

SEW EURODRIVE, Electromechanical Braking, it means retro fitting the motors on the heavy belts.

Hi,
2 issues. Having a 3 switch or contactor in between the output of the drive and motor is not recomended. Altho not as big as issue when configured as V/hz the significant issue is when in torque control. As the contactor opens and current arcs across the contacts the current drops but the drive feeds more to maintain the current and torque.

How many times does the operator enter the machine and how long does it take to clear and get back to speed. This equals X minutes per day
This gives the machine an effeciency value. Rather than spend $$ making the machine safe for all these cases. Consider spending the loss production money and stop the faults at the source. You could give each failure a rating based on freq and possible cause of harm. and work on the bigger issues.
When it comes to finding the fault ask yourself 5 times why it happens.
ie
Why does operator enter machine - belt of brush
Why belt of brush - Product stuck in brush
Why product stuck in brush - etc

Find the source of the errors -loss of production and minimise the need to enter an unsafe area.

Cheers