Drives and Stopping Safely.

We had an application were 5 drives were moving 8000lb at 72 ips. In case of fault when running at speed, what is responsible for safely and quickly ramping down the drive? If there is a fault and the motion controller outputs 0 volts and turns off the drive enable, should the drive be able to ramp down the motor by itself. During this time the motor will act as a generator and power the DC bus and there should be braking resistors to aid in slowing the system down. It works but not very quickly. I would like to see the drive stop and just under a foot. Instead the drive will coast for many feet and possibly crash into something before stopping. The motion controller could activiely ramp the drive/motor down by outputing a reverse signal like it normally does during decelerating to a stop. ( the system is in torque mode ). It looks to me like the best answer is for the motion controller to actively ramping down the motor.

Hydraulics don't have this problem. All one needs to do is to shut off the hydraulics but motors will coast too far even with the braking resistors. Some form of active control is required during a fault for shortest stops. What is responsible for this, the motion controller or the drive?

What if there is a feed back error? Then the motion controller can't ramp down the motor. What then? It doesn't seem like the drive is capable ( smart enough ) to quickly ramp down the motor by itself.

What if power is lost to the motion controllers and drives? Then what stops the motor quickly when there is no active control?

Sounds to me like a need for an electro-mechanical brake that engages anytime power is lost to the motor or a fault occurs. I have used several that can be adjusted so there will be a slow stop process. I am sure you are familiar with Warner and others. I prefer the type that is energized when running, its easier to set them to engage on power loss or a fault.

Most newer drives have the ability to select the kind of stopping that occurs on a failure. Many have multiple discrete inputs, so you can have an e-stop work one way and another fault work another way. The options can include coast to stop, normal deceleration with normal braking, and regenerative braking.

You may not be able to find a mechanical brake to decellerate 4 tons quickly, in less than a foot from 72 ips.

Tom Jenkins said:

You may not be able to find a mechanical brake to decellerate 4 tons quickly, in less than a foot from 72 ips.

Click to expand...

Although we don't deal with the sort of speed involved with Peters application, we do have to stop heavy weights very quickly in the event of an emergency stop application.

The main hoist on our ship to shore container cranes use these brakes , the speed is a lot slower but the weight will be up to 72 tonnes. On our boom rope drum (weight of boom is approx 100 tonnes) we use these brakes in the event of an emergency stop.

Bubenzer might well have a brake that could handle Peters application.

Just a thought

Paul

Let`s Stop

Peter how long does it take you to get to 72 ips? Maybe I should ask
how far does it take to get to 72 ips. If it`s 1 foot then the
drives with the right resistors will stop it in 1 foot. You still
don`t have any stopping power if you loss power. The mechanical brake all mentioned is the only sure way to stop. Depending on the drives
to stop anything without power is like depending on your hydrolics to stop something and a line brakes. If you are using 5 drives to run
this thing you may have to use a mechanical brake on each of these rather than 1 big one.

We had an application were 5 drives were moving 8000lb at 72 ips. In case of fault when running at speed, what is responsible for safely and quickly ramping down the drive?

Click to expand...

I would say WHO IS instead of what is. This would be the man driving the train. banghead
(I`m sure you already know all this just practicing typing TS)

These are the answers I was looking for.

I like Ron's answer. Basically what you guys are saying is that a fail safe mechanical brake is required. That is good because no one said the drive or the motion controller were responsible for the stopping incase of a fault.

BTW, the system accelerated to 72 inches in about 31 inches or at 84 inches/second^2

(Velocity^2)/(2*Aceeleration) = Dist

72^2/(2*84)=30.85

The acceleration rate is 84 inches per second.
The current system takes about 8 feet to stop.

I agree that stopping in 1 foot is tough, but it is only a little more than .55 Gs.

This question came up recently and we didn't have any good answer for what the motion controller could do besides signal an error and set the output to 0. We couldn't see what the drive could do either even though we aren't responsible for the drive. We normally don't get involved in the mechanical design, but in this case it looks like something was overlooked ( 5 mecahnical brakes). So far there have been no problems but nothing runs perfectly forever.

Thanks Guys.

Ok so in "elevator terms" (my life) this machine is traveling at 360 fpm (72 inches per second = 360 fpm, right??) with a 8000 lb load; and you want it to stop in 1 foot???? Thats a real neck snapper Peter. Sounds like it would be hard on the equipment too. I understand that catapults run on steam.....

Anyway, I would think that the machine would be designed for providing runby, and slowdown limits that would over-ride the control system to prevent a crash. (also an elevator feature by code).

At any rate, a mechanical brake is sure in order. These brakes , (again for elevators, but I've seen them on other equipment) are very simple, and highly reliable. They close on deactavation of the coil. Usually the brake is picked with a higher voltage, which is then lowered to enable the brake to drop faster when deactivated. Important feature is that the two shoes act independtly of each other, providing some braking in case of a mechanical bind of some sort on one shoe or the other.

Mike

Not quite right

Divide FPM by 5 to get IPS.


There are no people on these 8000 lb loads so stopping them quickly is limited only by the mechanics or the drive.

I appreciate the links. We will forward them to the mechanical people.

The key piece of information here, Peter, is that drives (at least all that I have ever come into contact with) de-energize the output to the motor on fault. This almost has to be because output shorts, lost phases, and motor problems all cause faults. As soon as the drive releases the motor, all you have left from a drive standpoint is coasting to stop--clearly unacceptable in your application.

Ron's suggestion for a mechanical safety brake is the standard approach to this condition. In your case where braking torque is quite high, you must size your power train to handle whichever is greater, motoring or braking torque. Also, don't be tempted to put a large brake on the backside of a relatively smaller motor even if the motor has a brake mounting face. Broken motor shafts are the usual result of braking torque being much higher than motoring torque. Instead, put the braking module in front of the motor, between the motor and the rest of the power train.

I think that Tom Jenkins is referring to Run Inhibit or Run Enable control inputs when he mentions selecting which decel ramps, etc. that the drive responds to. Those control signals leave the drive in control of the motor as opposes to faulting which doesn't.

Can the motors be disengaged from the conveyer in case of a fault?
That way the motors can slow down at their normal speed, and you can stop the conveyer without any damage to said motors. This way you might even be able to stop the conveyer in under 12"....
Just an idea... don't know if it's feasible, but it might be a viable option.

Peter,

"Divide FPM by 5 to get IPS": So 360 / 5 = 72

Anyway, DickDV is absolutly right about the brake placement (With such large mass; never on a motor, always on the machine). If you have it on a gearbox the brake drum/disk usually should be placed on the input shaft of the gearbox.

The more I think about it the more I'm amazed at the desired stopping distance and speed. I think were looking at a total 100+ HP drive system, which means even more mass (motors, gearboxes etc).. However if I understand it right there are 5 drives moving one 8000 lb load ??? So if each drive gets a brake then you have a chance of getting the desired results. But if this is 5 seperate peices of equipment each with one drive moving its own 8000 lb load, I think your going to be in hard luck finding a single brake to handle the job. ($$$$$)

So let's presume that there are 5 drives on one system. You will need feedback from all the brakes to set the coil voltage, and know that all the brakes are indeed picked before running the drives. So the sequence goes something like this:

1. power drives (zero speed)
2. pick brake (high/pick volts)
3. accelerate load, and lower brake voltage
4. stop load (normal stop with drive)
5. drop brake (or on E stop goto line 5)

Anyway you need a switch on each brake to send a signal to the control system to check if the brake is indeed picked, (all brakes must be picked before moving load), and to lower the applied brake coil voltage (aka holding volts) so that when released the brake coil will collapse quickly. Incidently, each brake will have it's own isolated driver. Like drives and motors, you cant really share brake circuits among brakes.

In short, dont leave it all up to the mechanical people; this is a control problem too..

Have fun, Mike.

Yeah Fun.

I am only envoled in this because we sold the motion controller. The motion controller is really just a very small part of this machine and we don't not have system responsibility. DickDV has also confirmed what I wanted to hear when he says the output is de-energized. I knew that drives had an enable. Our controller has an enable output, but I can see this is not enough if power is lost. I just need to make sure that we are doing what we can. It is up to the others to figure out how to stop safely. Those brakes look expensive.

Russ, there is no conveyor. There are five separate saws moved by motors that move the saws back and forth along a track at 72 ips (360fpm). I have graphs of the saws moving at 80 ips (400fpm) Each saw and motor weighs 8000lbs. There is another motor for turning the big circular saw. All five saws could be moving at once to different locations going different directions. The key is that there can only be one saw in a particular location at a time or bang. Since the saws are on a track one saw can not get past another.

Mike, this is like having 5 evelators in the same shaft!

Now for some sawmill speak.

A log from the log yard is moved end wise by a chain into the 'bucking system'. The log is 'scanned' and 'optimized' while it is moving. The optimizer, a high power PC or maybe many PCs, makes a decision about how best to cut the log into smaller logs. The solution is then transfer to the PLC, a Control Logix, and the Control Logix sends the saw positions to by Ethernet to a motion controller that controls the motion of saws toward their 'sets' or command position. When the saws are in position and the log is also stopped, the big circular saws swing across the log and cut in to smaller logs. This permits a long log to be cut into smaller 8' to 20' lengths so the small logs will fit in the primary break down machines. I thought I would describe this because many have not seen a lumber mill. They are really quite automated now. I did not see this mill, but I have seen others like this.

This is the first time someone has asked us what happens when power is lost and the motors coast.

Again, thanks for the info. It will help.

Re: Yeah Fun.

Peter Nachtwey said:

Mike, this is like having 5 evelators in the same shaft!

Click to expand...

Woowhaaa! Way toooo scary for me! But I think you may be famous. I saw (no punn inteneded) something like this on Modern Marvels, or some show like that, no too long ago. Truly amazing.

So I gota ask; does the motor/drive ride along with the saw? or is the saw carriage pulled along by a set of cables? Just wondering where the brakes gona go??? More and more mass...... Lots of fun..

Re: Re: Yeah Fun.

Originally posted by elevmike
Woowhaaa! Way toooo scary for me!

Click to expand...

I thought you could appreciate that. We just sold the motion controller. We did not design this. We are not the integrator or the mechanical designers. I think think it will work well though. The motion looks real good on the graphs.

Originally posted by elevmike
So I gota ask; does the motor/drive ride along with the saw?

Click to expand...

I think so. That is why the system weighs so much. I could ask our guy that was actually there. I have seen this kind of thing with flying shears but at half the mass and speed, but a much higher repitition rate.

STOP

The current system takes about 8 feet to stop.

Click to expand...

Peter
is this a controlled stop or is this under a fault condition?
Would you have to know the decel time as well as distance to figure
decel?