[metalmayhem]

you'll find that cnc's above hobby level will cut the power on estop , and it's common to cut the power when the limit switches have been hit sa well . Even my tormach machines which are hobby level kill the power to the drivers . It is a standard practice .

[catahoula]

Didn't mean to kick the hornet's nest here, just looking for a little nudge in the right direction

A little meanwell 24v power supply to run the circuit board and a few case fans is like $30, so I'm just gonna get that and put a contactor on the e-stop circuit for the 63VDC power supply and the VFD. May as well follow the norm, not expensive and not much extra work.

[joeavaerage]

Hi,
I agree its standard practice and in fact required to get an electrical certificate in many jurisdictions......but is it necessary?
It might well have been the case many years ago when motion controllers were 'flaky' and the added certainty was justified.

I'm interested if anyone has any experience where a modern and well sorted motion control ignored an Estop or limit or whatever.
Have you?

If its vanishingly improbable than a motion control does not stop when its commanded, then why insist that it depower?.

Second point is that this is my machine, I designed and built it myself for my use....and no one gets to tell me 'I have to do this or that' just because its
standard practice. If I decide that standard practice does not suit the performance of the equipment I actually have then I'll happliy ignore standard practice in favor of a better
suited regeime....and I won't be asking anyones permission either.

Craig

[metalmayhem]

Hi,
I'm interested if anyone has any experience where a modern and well sorted motion control ignored an Estop or limit or whatever.
Have you?

Can't say I've ever seen it with industrial machines because they have all shut down on estop when I had to hit it . I have seen machines do some weird stuff on a number of occasions so I can see it being justified . They are run by computers and computers have a potential of not being trusted

[joeavaerage]

Hi,
well yes, Mach3 could do some weird stuff, especially when jogging but one thing Mach3 never did that I've seen, is fail to stop when you hit <Estop>,
and thats without depowering.

They are run by computers and computers have a potential of not being trusted

Yes, I understand your point of view but in 13 years whenever I've hit <Estop> the machine has stopped immediately without fail.....EVER. Would you call that luck?
The PC running my machine is standalone, and very deliberately kept lean and mean (if a dual core Atom mini-ITX single board PC with no graphics card can ever be called lean and mean)
and I attribute that to my good fortune.

Craig

[metalmayhem]

I wasn't talking about mach , I was talking about industrial machines having done some weird things . While I always liked mach I've seen a lot of weird stuff happen including uncontrollable runaways . As far as mach goes I can't recall ever adding an estop to a single machine I built or retrofit . I only used the escape key , but then they lacked/lack the power to destroy themselves or possibly kill me .
Now , if I hand employees then that would be a different matter and I'd need to have all the safeguards in place , otherwise full liability lays on my shoulders .

The stickers on the machines with chopped off fingers beside a cutting tool and hands jammed in gears seems a bit unnecessary to but the safety standards are in place and there is no arguing against that . Unlike hobby machines industrial machines can be forced to work under harsh conditions . Heat for one can have a major impact on the whole control system which can cause a complete automatic shutdown or instability

You need to keep in mind that your relying on a couple of wires to tell the control it has been estopped . Most machines don't have keyboard shortcuts or a simple mouse click on a pretty interface that has an onscreen button . If one fails the other will pickup the difference . If you ever have to face a huge tool eating a massive part while sparks are flying and the building is shaking then you'd learn to appreciate all the safeguards you can get

[joeavaerage]

Hi,
my machine has 750W servos on all five axes, easily enough to do damage, and I demand a solution where the machine stops if I hit the <Estop>, and
lo and behold.....it does, without the need to depower. What's more its done so for many years without fail. Just because it does not cut the AC power does not preclude the fact
that the machine stops. Surely that's the name of the game.

Craig

[metalmayhem]

maybe you should contact all the manufacturers and tell them they've been doing it wrong for all these years

i've never needed the ebrake on my vehicle but the day may come that i'll appreciate that it is there

[maxspongebob]

This issue is a prime example of the problem of making Safety "Priority One".

Mike Rowe, of Dirty Jobs fame, has a saying that addresses this very issue "Safety Third". You can read about it here. Also if you want a real-world example of the outcome of making safety first here is one.

https://www.ishn.com/articles/93505-...Jwxv-q0hbs12Wk
https://mikerowe.com/2020/03/walk-me-through-this-safety-third-thing/
https://apa.az/en/america/russias-cu...k-times-411719

[Al_The_Man]

The reason I had not responded because it would appear that minds have been made up.
Whether or not the regulated practice of implementing a Machine E-stop that removes all motive power has saved any serious damage to man or machine, is rather a moot point, as all the Industrial machines I have had experience working on, and any that I have put together in a custom fashion such as the Galil you referenced to, have had the required mandatory E-stop circuit applied.
It is impossible to know the result this has had on eliminating damage to machine or personnel !
There are an estimated 600,000 CNC machines in operation world wide, so who knows how many serious accidents could have potentially occurred without the feature, is impossible to know.
Personally I do not see what the problems are, as it is not as though a machine E-stop is in general use on a regular operational/machine-control basis.
Or at least should not be used this way.
And as already pointed out, those creating a machine in a DIY fashion, are able to build it in any form or fashion as they wish, without the regards for safety or industry regulations.

[joeavaerage]

Hi,
I believe manufacturers do so to comply with the rule. Do you suppose the rule is outdated?

Some years ago who would have thought that motor vehicles would ever be allowed to be 'steer by wire' and yet that is exactly what has happened, and approved
by the US safety regulator.

Craig

[metalmayhem]

what difference do you see from lets say a mori seiki mill from the 80's to now that would indicate that the rule is outdated

here's something a bit different but runs near the same in regards to not following standards .
My tormachs are considered safe in the US but not here in Canada . A Canadian distributor does a few simple mods , gets them certified then sells the machines here in Canada to institutions or guys like ourselves . My mills were brought up from the states which do not pass the Canadian safety standards . If one of my mills happens to light my shop up in flames then I may have a tough time in making an insurance claim if the adjuster is on the ball . May seem unlikely to happen but my novakon had nearly lit itself up and fortunately I was there to catch it . Whether it could have torched the place or not who knows but places have been leveled by less .
so now , In a case where things light up within or outside the control panel would it not be beneficial to be able to smack a button and kill all power immediately and avoid possible electrocution or further damage

[joeavaerage]

Hi,
the motion controller of a modern day Mori Seiki is a world apart from the 80's version.

You guys are advocating a depower on Estop as a means of certainty that the machine will stop. I contend that at least my motion control has stopped on every occasion
that its been requested and the extra certainty of depowering is not required. To my knowledge all modern well sorted motion controllers stop instantly when required
by an Estop. It would be a pretty bad failing if it did not. Have you heard of any motion control that does not stop when commanded to do so?

If you want the extra certainty and/or need to comply with the rules for business purposes, well that how most machines are made. I do not have to comply with that particular
rule, which I consider outdated anyway, neither do I estimate that I need the 'extra certainty' that depowering confers over and above what my motion control does already. After all I'm more
likely to suffer a major earthquake than I am for my motion control to go cranky....at least according to recent history!!

A Canadian distributor does a few simple mods , gets them certified then sells the machines here in Canada to institutions or guys like ourselves

Here in New Zealand we follow CE rules and does Canada, ie European rules and all equipment sold is supposed to be CE certified. In the US they have a CSA certification. They are slightly different.
I would not regard US equipment as dangerous, but its not quite the same as the required CE standard.

In CE equipment for instance the required EMI filter at the input of a machine is inductive input type, usually two stage. In CSA equipment the EMI filter is capacitive input, again usually two stage. In truth either work pretty
well but for CE certification you cant use the US EMI filter....you have to swap it out.

I don't know about Canada but here we have an agreement such equipment made and certified in China, I can't remember now the acronym, but maybe NDA or similar, but can be sold in New Zealand
as if it is CE certified. When you actually pull them to bits the actual build standard is at best fair, and commonly much worse, to the extent that I consider them electrically dangerous.
I have little or no problems with US, European or Japanese made equipment, they all have there own rules, but all of them a pretty good....but some of the Chinese stuff is atrocious.
I worked for seven years repairing welding equipment and found that cheap Chinese made inverter welders and plasmas are just a joke.

Craig

[metalmayhem]

Hi,
the motion controller of a modern day Mori Seiki is a world apart from the 80's version.
Craig

What experience do you have with them to back that ? While I'll agree that they have made a number of improvements a lot remains the same . I worked on worn out 1988 mori mv40's that could still outperform a new haas vf . I know this because I've run both . If we are to talk about $1m+ machines then yes there have been numerous advancements since those days but that is well outside the realm of this discussion

[Al_The_Man]

One other fairly modern addition to the Machine safety factor is the recommended use of the Safety Relay in the E-stop circuit, I believe it is generally mandatory in Europe, all the machines I have seen imported from there in the past few decades have them. N.A. seems slow in adopting them so far.

[phomann]

Focusing on whether a software Estop can replace an Estop that removes power from the drives and spindle is only looking at one part of the safety issue.
The purpose of removing power is to ensure that every thing stops no matter what the failure is.

Early on in the Mach3 days, I got called out to look at a Syil X4 (I think) that had a problem with a coolant pump. At that stage they were using Mach3 via a parallel port.
I needed to get to the back of the machine to get to the controller. As the owner of the machine went to swing the computer out of the way, the parallel port cable fell out of the controller socket. The spindle immediately started flat out.
Hitting the Estop turned it off instantly. In this case a software Estop would do nothing.

Sure, there was a design flaw in the controller, using the wrong default state on the breakout board, but that’s why removing the power is the last resort.

The same may happen for servo runaway when an encoder fails. There’s plenty of reports about that. It’s why servo systems feed the limits into the safety contactor.
It’s what my pick and place machine does. It has limit switch controlled by software that puts it into reset, requiring rehoming.
If the software falls to act, there is a 2nd set of limits that trigger the safety contactor removing power from the drives.
The P&P also has door switches that halts motion with opened.


Cheers
Peter


Sent from my iPhone using Tapatalk Pro

[joeavaerage]

Hi,

The same may happen for servo runaway when an encoder fails. There’s plenty of reports about that. It’s why servo systems feed the limits into the safety contactor.
It’s what my pick and place machine does. It has limit switch controlled by software that puts it into reset, requiring rehoming.
If the software falls to act, there is a 2nd set of limits that trigger the safety contactor removing power from the drives.

That is a good idea. I use the <Stop> button whenever I need it, typically a toolpath doing something unintended, or gouging a clamp or whatever. You might call that 'occasional operational
use' rather than an emergency stop. Given my moderate use of the <Stop> button, say once a month or so, I do not want to have to restart the machine from scratch, and I am therefore
reluctant to do so.

When I get around to building a new electrical cabinet (six month time frame) I will have a power panel that exposes two buttons, an ON and an OFF, and that will control the incoming AC contactor, and it will be available
alongside the PC, ie readily accessible. That would give me two buttons, the existing <Stop> button, and just as accessibly a power isolation button.

Craig

[catahoula]

I use the <Stop> button whenever I need it, typically a toolpath doing something unintended, or gouging a clamp or whatever. You might call that 'occasional operational

This is actually one of the ideas that motivated my original post. I was thinking a long the lines that being new to CNC machining it may be beneficial to be able to use the e-stop button fairly liberally when testing new programs, etc, and have the machine not lose steps or other setup parameters, in order for it to be quicker/simpler for it to be back up running afterward. I also considered that the power switch could be the "hard stop" while the e-stop becomes the "quick stop"

Obviously one major reason to not use the e-stop this way is that in an emergency situation you want to have to do as little thinking as possible and e-stop buttons are specifically designed to work well in a panic.

But ultimately it was some googling that turned me- I found a bit in the EdingCNC manual referencing a "Smooth stop" function which is the software level gentle stop which ensures no loss of steps to the motors/drivers but stops in a rapid manner. So that will suffice for my more liberal use and I can reserve the e-stop (with full motive power removal for thoroughness' sake) for actual panic situations.

(Edit: changed to correct feature name "smooth stop")