E-Stop

[ajl6549]

Originally Posted by unterhaus That's the point I'm worried about. Sure, it would be nice to be able to panic and hit the e-stop, then realize nothing was wrong and start right back up again. But on the other hand, I really want to be able to shut the thing down if there is a big problem. So maybe we need a big honkin' red switch for e-stop, and the "I'm just a weenie" sized switch for all the other times.

If you've ever seen an operator, in a panic situation, hunt for the right button to stop the the machine, you'll agree that only one large button is really all that should be avail. in a panic situation

[sdantonio]

Originally Posted by ajl6549 If you've ever seen an operator, in a panic situation, hunt for the right button to stop the the machine, you'll agree that only one large button is really all that should be avail. in a panic situation

The way I'm putting together my power supply is this: Once switch turns on the "power one" power supply. This supply powers up the three gecko's and the BOB, and through the gecko's, the motors. It also supplies power to the router itself. Then there is a little toggle switch that opens current flow from the power one to the gecko's to avoid and startup surges or anything like that.

If I have enough current in the circuit I may also put the computer on the same switch.

So basically, there is one big switch between the whole CNC unit and the wall.

[ajl6549]

That seems good. Is there any chance of a "coasting" situation? I realize that your typical "hobby type" machine may not incorperate a brake for axis motors etc., but...?

[sdantonio]

Originally Posted by ajl6549 That seems good. Is there any chance of a "coasting" situation? I realize that your typical "hobby type" machine may not incorperate a brake for axis motors etc., but...?

I wouldn't think coasting would be a problem. Steppers are pretty much designed to be able to stop on a dime (under power) and have a relatively high inertia (while not under power). So I would think they would stop pretty quickly. But I have been wrong in the past and expect it may happen again. And at70 to 100 ipm I wouldn't quite describe the thing as flying. I would think the motor inertia and the leadscrew inertia would all stop the thing pretty quick.

[ajl6549]

I agree 70 - 100 ipm is'nt "fly'in" by any strech but still...? I guess my thought is in safety first. Do the steppers actually "brake" when powered down or do they "coast"???

[Shotout]

Originally Posted by ghyman ... killing my friend. But in the middle of a two-hour rough turning cycle (on the tenth part in the order), it's not always feasible for the operator to be right next to the control in the unlikely event that there will be an emergency!

I'm sorry about your friend.

I agree with you about not being able to babysit a machine at all times. In our shop I run two Haas TM-1 mills and a Mazak QT250. As a general rule I'm moving between two machines constantly with the remaining one running a longer cycle, I try to plan it that way so I can keep up with parts loading and still program etc.

Our machines stop immediately when you e-stop them, spindle and axis travel both. I've stopped it a couple of times when threading, tapping or moving into a first cut with the lathe tail stock extended, but I've yet to have to e-stop one for my own safety. I did however have to do so once for someone else's. I use to think the mill e-stop was in a poor location before that, but learned it was in a place where it couldn't be blocked by the operator's body if a second person had to come into the picture, or the operator could hit it as he tried to get out of harms way if needed. Guess I have to chock one up for an engineer on that one. From what you've said it seems that there may be more fore thought in how it shuts down a machine these days.

Scott

[sdantonio]

Originally Posted by ajl6549 Do the steppers actually "brake" when powered down or do they "coast"???

I don't know. I was just assuming they would return to their power down state and the inertia would take care of breaking the motor. Of course inertial works both ways as it is just the resistance to change in the state of motion. But the inertia od the spining spindle is much less than that of the magnet. So it would actually be mre of a fixed magnet interaction that would slow it down.

[ajl6549]

Originally Posted by ghyman It's unfortunate that the machine manufacturers at that time (late eighties) chose to make the e-stop shut down EVERYTHING. A broken clamp let a part come loose from a vertical turning center that a friend of mine was running, and he hit the e-stop. After what seemed like minutes, but was probably less than five seconds, the table was still spinning at 200-300 RPM, (the e-stop killed the hydraulics, which provided braking for the table.) and the appx 1.5-ton part came out of the machine, killing my friend.

Sorry to here about your friend
It seems like all to often I hear of situations like this where the E-stop causes unwanted results like these. It's bad enough when it's the fault of a machine tool that has poor maintanece or simply is broken, but when it's the fault of the machine design someone should be held accountable!

[M_Controls]

Some basics about a proper e-stop:

I design industrial machine controls occasionally. E-stop design is one of the most crucial designs, due to the liability & just plain personal safety regardless of the liability; except if you are some foreign manufacturers. Yes, as mentioned above, there are applications where it is more dangerous to remove power from the servo(s), usually due to inertia or gravity issues. In such cases, brands that are the most trusted to bring motion to a stop in such a condition MUST be used. I have went so far as to include a timer to remove power from servos with a high-inertia load after sufficient time has expired to allow the servo to brake the inertia, just to increase the safety level of the e-stop function. Some machines should use brakes to stop axes with inherent drift pressures such as caused by gravity. Any such brake (involved in safety) MUST be of the type that uses energy to remove the braking action, for instance spring-tension-activated braking with electrical release.

It might not be as easy as you think to design a good e-stop:

Unfortunately, not all manufacturers take appropriate measures to guarantee a fail-safe e-stop. When the enable signal is shut-off, most servo or motor drives will only respond by entirely shutting down the outputs. So in many cases these drives must not have the enable shut-off on E-stop condition, in order to bring the inertia to a quick stop. Thus my timer solution as mentioned above, which should be implemented with a simple timer not dependant on a larger computer system, to increase reliability. It still is not a perfect solution. But I think that some companies will throw safety out the window when they realize that they are adding cost for a imperfect safety solution. At least that solution is very safe, at least in my opinion. And I would not trust the drive enable signal for e-stop, I add a contactor to remove the power from the drive unit.

Some possible reasons for the failures to design a good e-stop:

Some manufacturers have designed e-stop systems very poorly. Others may have had a good e-stop design in the early stages of machine design, and then did not re-evaluate the e-stop design when the machine had undergone major changes (in the mechanics or the control). Sometimes manufacturers run out of budget (time or labor) for re-design after changes and refuse to fund time or labor for the re-design. Others may be just sloppy - haphazard. These are the ones that deserve to be sued. I sometimes must compete (in bids) with this.

The real root of the problem:

I think that poor e-stop design is often affected by the safety standards in the country of design, and/or perhaps it is affected by the legal difficulty of suing a foreign company for a safety-related accident. Of course these issues are entirely dependant on the exact country of the manufacturer. In my opinion, this is one of the biggest factors in the decline of machinery manufacturing in the US! Many US manufacturers have gone out of business due to the liability, and the foreign companies can make the machines cheaper since they do not have to add expense for some safety issues; like a good e-stop system for instance. I am not supporting a relaxing of liability for e-stop accidents here, as much as arguing against free-trade agreements (such as NAFTA) with other countries. Obviously, judging by history, telling people to buy american won't be effective; people tend to buy the cheapest regardless. It's all about the bottom line, and some managers do not recognize many safety hazards as a threat to the bottom line when they look to buy a new machine; and some other managers just miss them. After all, most managers do not know what constitutes a good e-stop system, and only a few companies will have written standards in place that prospective machines must comply to. So cheap usually wins. I cannot and would not take such chances.

[ajl6549]

Wow! Thats quite a post for your first time and very well put I might add.

[Al_The_Man]

As I mentioned in an earlier post, there is guide lines laid out in standards like the NFPA79 for industrial machines. But I have found that in general, European machines have been ahead of N.Amer. in the terms of E-stop safety, The use of special Safety Relays has been in effect in Europe for some years and this practice is slowly become normal here. especially seen in the case of imported equipement from that part of the world.
There are many configurations of Safety relays to take in account the controlled machine shut down in case of e-stop.
They are not cheap, hence manufacturers reluctance to use them.
Al.

[kevincnc]

In multi-axis, servo controlled machines that we manufacture, we use STI safety relays with timers to allow for a controlled shutdown. Basically when an e-stop is pressed or an enclosure door is opened, servo drives are sent an immediate stop command, allowing them to decelerate to a stop under power. At the same time contacts on the VFD's control side are opened to decelerate the spindles to a stop. The timer is set to give everything time to stop, then power is removed from the VFD's and Servo drives.

To be a truly safe system, every component must have redundacy. For example, there are two contactors in series that drop out to remove power from the drives. That way, if the contacts weld shut on one of the contactors, the other one should still remove power.

On the control side, true safety relays look for a pair of contacts to be switched simultaneously in the door switches when the door is closed, or when the e-stop is pulled out in order to enable control power, usually a set of N/O and N/C contacts. The reason for this is if only a single contact is used, an open circuit or short to ground could bypass an e-stop function, and you would not know it until you needed it.

This is still not 100% safe, but about as close as you can get.