[cspearsall1]

QUick Update: These particular drivers have 24v/5v switches. After switching to 5v I have motion!!!! Still a work in progress, but one step closer. Thanks to all for your input and help.

[maxspongebob]

My $0.02 for what it is worth.

Why are you not wiring the stepper enable signals. There is plenty of I/O left over to hook those up. When you press enable on the computer it will lock the steppers in place thus letting you adjust the position of the axis when disabled.

Wiring un-fused power in a cable is not a good idea. If a cable is cut or crushed somehow, you don't want your power to be shorted such that it takes out the whole system. It also helps to reduce EMI noise in the system if your power stays inside the controller box. With regard to limit switches, if you are using NO/NC switches, connect the switch between ground and the input signal. Then use a NC switch so that when you reach the limit the circuit is opened allowing the pullup resister on the ESS to activate the limit signal. This also has the added benefit of activating the limit signal if the cable is disconnected. If you are using PNP/NPN sensors for the limits, at least add a fuse to the power wire and take advantage of the same logic if the cable is disconnected.

With regard to the E-Stop circuit, just cutting power to the system is not the best way to do it. Consider this. The spindle VFD has an internal brake. When the enable rotation signal is cleared the brake is applied to stop the motor fast. If you just pull the power cable the motor will have to coast to a stop. Those seconds will seem like hours with your shirt sleeve wound around the cutter. The goal for the E-Stop is to stop motion. Wire the E-Stop into the BOB so the ESS can do what it is supposed to do to stop the system. Also, make sure to apply the cable disconnected logic to the E-Stop. Breaking the cable should stop the system.

[cspearsall1]

Your $0.02 is noted and valuable. It will take a bit of time to change the wiring but I can do it. As far as e-stop I don't have a vfd but the advice I believe is still valid and will come in handy when I actually do have one. Thank you again.

[joeavaerage]

Hi,
I do not cut the power on Estop either. I rather suspect that in many countries around the world it is mandatory that an operator may stop the machine and depower it,
and as a consequence manufacturers do it that way....and it has become the norm because its legally required, not necessarily the best and safest way.

The other thing I don't like about Estop is that you lose reference. With Mach, in fact all CNC software solutions running on Windows platforms, all use buffering. That is to say
that Mach sends numeric instructions to the ESS in a buffer. The default buffer length is 180ms. So if the buffer is full and you Estop, the contents of the buffer is aborted and that
trajectory data is lost. Once you Reset the machine it no longer knows exactly where it is and you need to re-Home to go back to the job, a PITA.

If you want to stop the machine but WITHOUT losing reference use <Feed Hold>. <Feed Hold> will consume the contents (trajectory data) of the buffer and THEN stop. So potentially as long as 180ms of movement AFTER
you hit the <Feed Hold>. But 180ms is scarcely the blink of the eye....so I don't use <Estop>, I use <Feed Hold>, it stops the machine 'after the blink of an eye' and does not lose reference.
So you see the pic of my machine with the big red <Estop> button (on the right) but it is in fact a <Feed Hold> because that's the way I like it. Note that in Mach4 there is a setting (Mach4 Control plugin)
that allows you to choose whether the spindle shuts down on <Feed Hold>. I chose not to have the spindle shut down....but Max (spongeBob) has a point....it would be safer if the spindle did shutdown.
Note that the rate at which the spindle decelerates is determined by the programming of the VFD, not Mach at all. Normally my spindle just coasts to a stop, that is the deceleration I programmed into the VFD.
So the spindle coasts to a stop no matter what I do in Mach.

I should note that while the button is <Feed Hold> if any of the axis limit switches operates or any of the servos alarm....that Estops the machine. In the two years I've been operating my machine I've had two Estops,
both due to Limit events where I had forgotten to turn on Soft Limits....dummy!!! Estops are very rare, while me hitting the button because some bad s*****t is about to go down, like milling into that bloody clamp,
or the workpiece moving in the vise, or I just broke the tool, or the coolant has got blocked up and I need to stop and fix it before the tool turns hot hot.......those happen all the time.

Craig

[Al_The_Man]

It is mandatory in N.A. on an E-stop to remove all motive power to all motorized parts of the M/C, there are however 3 categories of E-Stop, from removing all power immediately, to allowing a controlled removal where in, doing so immediately can cause damage to machine or personnel.
The power can however, remain on the control and all PLC etc, inputs.
Most industrial M/C's and those such as Galil, will still record any axis movement at or during any such E-stop. Therefore, on release of the stop, the M/C does not have to be re-referenced.

[joeavaerage]

Hi Al,

It is mandatory in N.A. on an E-stop to remove all motive power to all motorized parts of the M/C,

But who enforces it? I can well imagine that to get a US Electrical Safety cert, whatever that is called, that you would have to depower the machine, at least partly.
But if, like me, you design and build your own machine at home then who is to stop you? It's certainly not illegal either.

I've read the Crimes Act and the Summary Offenses Act and neither of them specify that you have to depower your machine. No doubt the Health and Safety in Employment Act
may well specify such a thing, and it would be an offense against that Act if you ignored it.......but here's the thing....they can't jail you for it. They might be able to fine you
if it were being used by an employee....but if it's your own machine you can stand on your property and tell the nosy inspector to 'F*****K OFF!', he may not step onto the property
unless you invite him. Even the Police can't do so because they can't get a warrant because no Crime (under the Crimes Act) or Offense (under the Summary Offenses Act) is being
committed.

For my purposes I find that an Estop that cause loss of reference a PITA, so I use FeedHold. In the eleven years I've been using Mach never have I ever had the machine fail to stop if I've commanded it
to do so, so I personally regard de-powering the machine unnecessary.

Craig

[Al_The_Man]

Anyone doing DIY for their own purpose can essentially do what ever they want! At their own risk of course.
It is someone like manufacturers of systems or, such as myself doing commercial work, where a system has to conform to NFPA79, "Electrical Standards for Industrial Machinery", Of Which conforms to both NEC & CEC.
I have never really found any issue or negative results & effects by conforming to such, recommendations.
If I were to disregard these standards completely and then someone gets seriously hurt or worse, then there is always the threat of those seeking recourse.
.

[joeavaerage]

Hi Al,

If I were to disregard these standards completely and then someone gets seriously hurt or worse, then there is always the threat of those seeking recourse.

I 100% agree. I do not wish to design, build and operate a dangerous machine, it is I whom would most likely be injured after all.

I personally am of the opinion that the regulatory standard is lagging behind the technological standard.

For instance in the earliest days of Ethercat and Profibus neither bus were permitted to carry safety data like Estop, door interlocks and that sort of thing. The regulations
prevented it. After several years the regulations caught up and now both Ethercat and Profibus can and do carry safety data.

Another example is in automotive electronics. Systems like Anti Skid Brakes and Airbag Controllers are deemed safety critical and there is a quite rigorous process for
qualifying electronic devices for this service, part related to the reliability of the components but also the deign......is 'realtime' absolutely deterministic? For automotive safety it must
be deterministic and there is a very rigorous process for proving it.

In both cases it is the case that the regulatory environment had to catch up to the technological state of the art.

As I use Mach which runs on a Windows platform and that it has a trajectory buffer I find a hard Estop a pain, so rather use FeedHold. It satisfies the need to stop the machine but without the complication
of depowering. This does rely on the Windows PC behaving as expected. The prospect of relying on a Windows PC for safety would give most people nightmares and I rather doubt anyone could make
a case that Windows platforms are safe from a regulatory point of view. Having said that the little PC (Dual core Atom single board PC) in my machine has never failed to respond to either an Estop or a
FeedHold in the eleven years I've been using it. That does not constitute proof that its safe but is just one data point. If it had missed a sfatey event of the type (Estop or FeedHold) even only once
then I would insist that Estop depower the machine also.....but it has no, eleven years and counting.

Craig

[Al_The_Man]

Another thing NA has been lagging behind in adopting is the Safety Relay, it has been fitted to European machines for some time now and is a regulation in order to replace the simple E-stop string relay.

[cspearsall1]

@averagejoe - Craig, what size catch diode do I need for the relay on the 110v 15amp 3.25 HP Porter Cable router?

[joeavaerage]

Hi,
the catch diode is for the coil current of the relay, the size of the load that the relay switches is immaterial.

I would suggest, unless you've used an out-size relay, then a 1N4148 or any equivalent signal diode would be fine.
Ordinary 1N4007's (or equivalents) which are standard rectifier diodes work too, but are somewhat slow....but otherwise OK.

Either of these diodes are less than 10 cents each at 100 of quantities. I buy them by the thousand.

Craig

[maxspongebob]

Here is a quick resource for calculating the diode value. Your power contactor may have a recommended value so start there.

BTW, I have always heard of it referred to as a Flyback Diode. Anyhow it is the diode that protects the circuit from the reverse voltage generated by the magnetic field collapse of the relay coil.

https://www.plantengineering.com/art...de-and-rating/

[maxspongebob]

BTW, I am just now looking, closely, at your schematic. On the contactor, don't you want to turn it on and off by using one of the outputs from the MB3? Take a look at page 21 of the MB3 manual for wiring. Also, if you were to use a solid state relay for the spindle power, you would not need the flyback diode.

If your contactor is controlled with AC and you cant find a 24VDC controlled contactor, you can use one of the MB3 relay outputs to switch AC contactor control.

I would have at least 2 circuit breakers maybe 3 at the AC input of the box. A really small one for the control circuits one for the spindle, and maybe a separate one for the stepper power supply. Match the circuit breaker to the power supply max rating.

Don't forget a fan for the box.

[cspearsall1]

Just to clarify. The schematic might be a little outdated now. I am using two contactors, one for the main power and one fed through the e-stop buttons for spindle and motor power (this was discussed a few posts ago in this thread and is not necessarily recommended. When I hit the estop the board remains powered but it kills spindle and motor controller power. I believe I will be changing the estop circuit to just alert the controller so that the router/spindle break can engage). The router is controlled through a solid-state relay using 24V from one of the output pins on the MB3 (Y301).

As I understood Craig, I can use a 1N4007 diode (which I have), but I am using a solid state relay 24V coli controlling 110V for the router, does this mean as @maxspongebob (I'm loving these names BTW) stated that I do not need a flyback/catch diode?

[joeavaerage]

Hi,
if using a solid state relay, then no catch/fly back diode is required. It is only traditional relays that have a coil that operate the relay contacts, they require a catch/fly back diode.

Craig

[cspearsall1]

Thank you Craig and @everyone who has helped me so far. I am just getting started and I am very grateful for people like you guys who are willing and able to help/

[Al_The_Man]

Incidentally, for a while now there has been fast versions of the 1N4003/7 etc line, they are prefixed UF.
e.g.UF4007.

[joeavaerage]

Hi,
yes, I use fast and ultra fast diodes where I have to, but for a small relay with a 24V coil and 20mA why bother using a costly fast diode like a UF4007 when I can use a surface mount
1N4148 at 4 cents each? Signal diodes like 1N4148 and 1N914's have reverse recovery times of 4ns and better, superb for low voltage/low current situations.

Craig

[Al_The_Man]

Well US$0.50c is not really going to put a halt to the project!

[joeavaerage]

Hi Al,
no, you are right 50cents would not stop me. I have one circuit I make quite frequently for a Nissan speedo. It contains not only the speedo but the drive for the odometer stepper motor.
It requires eight diodes, two Zeners and one rectifier diode. Currently it cost 32 cents for the diodes, I'd rather not inflate that to $4.00.

Craig