[billsnogo]

I have mine setup using a c10 BOB set for pull down, and have mach3 set for active low, and my limit/home switches are NC. I have 5vdc at the BOB unless a limit is tripped, so of a wire breaks there is no voltage present at the BOB. So, wouldn't it be less safe if I set mine to active high? I don't understand why active high is safer, I am new to this and am missing something..... :drowning:

thanks
bill (likes being safer than sorry)

[Al_The_Man]

Example: a Estop is usually active low, IOW for normal operation a high is expected, E-stop if low.
Which also means that apart from a legitimate E-stop, in the event of no power the E-stop will activate.
Al.

[RCaffin]

I run active-HI for BoB outputs. No power, or Mach stopped, or something else: outputs are low or dead.
I run Active-LO for BoB inputs mainly because that way I can optically isolate EVERY input easily. Collector pull-down.

What works for me may not work for someone else.

Cheers

[vmax549]

When you set it up as active LOW the switch HAS to make(complete the circuit) in order for it to work. SO if you had a broken wire the switch could never make and you could crash the machine .

IF you set it up as active HIGH then the switch has to break the circuit in order to work. SO if you had a broken wire the switch would always SHOW as tripped and the machine not run.

(;-) TP

[Al_The_Man]

When you set it up as active LOW the switch HAS to make(complete the circuit) in order for it to work. SO if you had a broken wire the switch could never make and you could crash the machine .

IF you set it up as active HIGH then the switch has to break the circuit in order to work. SO if you had a broken wire the switch would always SHOW as tripped and the machine not run.

(;-) TP

All commercial machines I have been involved, Fanuc, Mitsubishi et-al, all use active low for E-stop or overtravel, this means that the 'Activated function' state is Low, so in order for the machine to run, the e-stop has to be high as well as the overtravel.
In the event of a power fail or wire breakage etc, the circuit fails safe.
Al.

[RCaffin]

When you set it up as active LOW the switch HAS to make(complete the circuit) in order for it to work. SO if you had a broken wire the switch could never make and you could crash the machine .
IF you set it up as active HIGH then the switch has to break the circuit in order to work. SO if you had a broken wire the switch would always SHOW as tripped and the machine not run.

Sorry to disagree, but I can design very simple circuits which work exactly the opposite of this.
The BoBs I use (Homann Designs MB-02) have bias resistors at the inputs which can be wired to pull up or down. This lets you run any convention you like in EXACTLY equivalent logic terms.

There is a bit of a convention that eStop is Active-LO, but that is just a convention for convenience. Mach does allow you to run Active-HI for eStop if you want.

The one exception to all this is when something goes rather wrong and you lose the power supply to the peripherals. When this happens almsot every input is likely to go low, and it does make sense to stop the machine when this happens. It is not hugely likely, but the possibility often tips the scales.

The one thing I would advise is to decide what convention you are going to follow, write it down, and then to quite rigorously follow that convention. It tends to preserve sanity when debugging.

Cheers

[Al_The_Man]

There is a bit of a convention that eStop is Active-LO, but that is just a convention for convenience.

Although more for safety than convenience.
Al.

[xjdubber]

Sorry to disagree, but I can design very simple circuits which work exactly the opposite of this.
The BoBs I use (Homann Designs MB-02) have bias resistors at the inputs which can be wired to pull up or down. This lets you run any convention you like in EXACTLY equivalent logic terms.

There is a bit of a convention that eStop is Active-LO, but that is just a convention for convenience. Mach does allow you to run Active-HI for eStop if you want.

Cheers

The nice big red push pull or twist to release buttons are emergency stops, not matter what when or how, if you do not wire them using NC contacts you have defeated their purpose. Limit switches should also be wired in a NC configuration, if not their function has been defeated. Pretty much anything else, do what you want....

edit:

As i posted this another thing came to mind, with the use of mach 3 in mind, many users wire their e-stops only to their breakout boards. At this point and time the PC is relied upon actually recognizing the e-stop command and stopping all motion. However this is not the way it should be the estop circuit should be physically wired so that the e-stop de-energizes all circuits that could injure or damage persons or machinery. This can be achieved with a simple circuit containing and Emergency stop button (red with yellow back ground indicating its an emergency stop) a "system enable button" to energize a relay coil witch latches itself trough the e-stop. Contactors can be used with NO contacts from the relay to de-energize heavier loads(ie spindle motor/drive, stepper drives, servo drives etc.).

going back to the original topic, why are active high configured limit switches safer? it all depends on how your system is setup, but use NC switches and you'll be safe.

[Al_The_Man]

The problem as I see it is a Mach system is cobbled together from different devices from different manufacturers, Mach S/W included.
Every supplier will issue instructions on how his particular equipment should be implemented, so it is up to the builder, often uninitiated to some degree, to figure it all out.
When I implement a system, I use methods based on industrial norms NFPA79 etc,, and tailor or incorporate all and any single manuf device into the system.
Someone mentioned a hard wired E-stop system using a control relay, this IMO, is the minimum that should be implemented to turn off all motive power in the case of an
E-stop.
If the machine in question is a DIY machine, often corners can be cut in terms of safety etc.
But I have got into the habit of implementing systems that are considered normal practice.
Al.

[RCaffin]

with the use of mach 3 in mind, many users wire their e-stops only to their breakout boards. At this point and time the PC is relied upon actually recognizing the e-stop command and stopping all motion.

One could argue quite a bit about this, going back to historical reasons why things were done a certain way with huge boards of TTL logic before we had adequate PCs and SW like Mach, all of which might miss the point a bit. If Mach crashes, then it stops issuing movement pulses, so the axes will stop moving anyhow. If the spindle is driven by PWM then that too will stop. If Mach misses seeing the eStop signal ... yeah, well, I don't think I have ever heard of this happening.

However ... eStop on my system does drop out the axis drivers and the spindle PS as well. But it does it in a couth manner without risk to the electronics. All my drivers have Disable inputs which bring things to a halt cleanly. I am a bit reluctant to stick relays in the paths of the axis motors: the mfrs warn that can blow them up.

I would like to be able to say that what really matters is that the designer stops and THINKS about what is needed, but that fails when you have novices putting systems together with little or no knowledge of electronics, power systems, or CNC. In that case, big red buttons and KISS are useful.

Cheers

[Al_The_Man]

In Industrial machines, particularly in Europe, the norm is for Safety Relays to be fitted, it won't be long before they are mandatory in N.A. also.
Practically all modern machines imported from Europe are fitted with them.
The replace the old single hard wired control relay method.
http://literature.rockwellautomation...d556_-en-p.pdf
Al.

[RCaffin]

In Industrial machines, particularly in Europe, the norm is for Safety Relays to be fitted,

Probably necessary, otherwise commercial mfrs would go for the minimum cost 'solution', which usually means the customer loses out on safety. Such is commercial reality.

However, I do not think that such a requirement is the only way to ensure safety. And those mechanical safety relays are horrendously complex, and probably rather expensive too. They rely entirely on mechanical features with no recourse to equally reliable modern electronics. Yes, one could get into endless arguments about mechanics vs electronics: I don't intend to.

Please don't get me wrong: I have put more safety features on my machinery than most.

Cheers

[xjdubber]

safety relays are completely electronic.

and yes horrendously expensive LG5925-48-61-24

[Al_The_Man]

safety relays are completely electronic.

Not quite completely electronic, mechanical force guided contacts.
Al.

[vmax549]

I should have said the switch should be NC and the simple solution is Logic as active high. That requires ZERO extra hardware. IF the circuit fails(broken wire) the function(limits) would be always tripped and Fail Safe.

Sure you can design it backwards and it still work but WHY go around the block just to walk next door? The results would be the same either way one is just much easier to implement and understand.

(;-) TP

[hub]

There's no difference in a semi-conductor active low or active high (NPN vs PNP). If it fails/breaks, there's no difference -> both types won't output a signal.
If you have mechanical/electromechanical switches like an E-stop, relay then get I would choose a NC type an connect in series. If they fail, it will break the circuit and turn the input/output off. Still no difference if the logic is active high or low.

Hub

[RCaffin]

Blind adherence or reliance on any set of rules will get you killed.

Which makes it a real problem when trying to advise a novice who has no knowledge of electrical stuff.

Cheers
Roger

[Al_The_Man]

There's no difference in a semi-conductor active low or active high (NPN vs PNP). If it fails/breaks, there's no difference -> both types won't output a signal.

The issue is the fail safe (active low) should equate to power loss, if a condition is active high it will not register in a power loss situation!!!
This is the very reasoning behind all commercial practice, no matter whether SS logic or PLC/PMC logic.
Al.

[hub]

The issue is the fail safe (active low) should equate to power loss, if a condition is active high it will not register in a power loss situation!!!
This is the very reasoning behind all commercial practice, no matter whether SS logic or PLC/PMC logic.
Al.

At power loss/fail, for example a photocell won't "output" anything, no matter if its NPN or PNP.
I would wire a relay for example to "monitor" power. Power on = relay on; power off = relay off.

Hub

[Al_The_Man]

At power loss/fail, for example a photocell won't "output" anything, no matter if its NPN or PNP.
I would wire a relay for example to "monitor" power. Power on = relay on; power off = relay off.

Hub

Exactly, we are not discussing arbitrary logic, but devices used or connected as a safety devices, then they would be wired active low, IOW at Hi, system normal, if low then fault, which would also co-inside with power failure, otherwise it would be futile to connect them any other way.
This is the reasoning behind the industry norm for E-stop and O.T. etc.
Al.