[mactec54]

I interpret the answers then as 2 mechanical limit switches per axis screw, (8 total), that are tied to the E-stop circuit to prevent a malfunctioning motor from causing damage and 1 inductive switch per axis for homing, 1 for x, one for y and 1 for z. The y one can be on either side, but does not need to be both.

Are the e-stop limit switches wired to the controller individually?
Should the e-stop be NC so if there is a broken wire the circuit would be open?

You only need (1) Micro switch or proximity sensor per axis just as this video shows, most modern machines where the controls have soft limits, you only need the (1) Micro Switch / Proximity sensor per axis

If you need Gantry Squaring, then you would use ( 2 ) on that axis.

Proximity Sensors are the best to use as they have better noise immunity than Micro Switches

Power EMI Filters are always needed for a system like yours, this is a Filter I recommend for new machine builds, TDK Lambda RSEN-2030L just a heads up not all Power Filters are created equal

You will not need a Breaking Resistor installed for any of your Axis Drives

Here is a video for you to look at, that may help with your wiring, is using the same control as you are using

https://www.youtube.com/watch?v=DcgBTxemPlA&t=13s

[TeeTomTerrific]

You only need (1) Micro switch or proximity sensor per axis just as this video shows, most modern machines where the controls have soft limits, you only need the (1) Micro Switch / Proximity sensor per axis

If you need Gantry Squaring, then you would use ( 2 ) on that axis.

Proximity Sensors are the best to use as they have better noise immunity than Micro Switches

Power EMI Filters are always needed for a system like yours, this is a Filter I recommend for new machine builds, TDK Lambda RSEN-2030L just a heads up not all Power Filters are created equal

You will not need a Breaking Resistor installed for any of your Axis Drives

Here is a video for you to look at, that may help with your wiring, is using the same control as you are using

Honestly I'm still a bit confused. There are two ends of an axis that needs to have a limit. A runaway servo motor could be traveling in either direction. I can see where a combo Home and Limit can be used on one end, but do you wire another proximity switch in the same loop on the other end? I don't think I will need gantry squaring so I assume the two limit switches will only be on one side of the two Y axis?

[ihavenofish]

Honestly I'm still a bit confused. There are two ends of an axis that needs to have a limit. A runaway servo motor could be traveling in either direction. I can see where a combo Home and Limit can be used on one end, but do you wire another proximity switch in the same loop on the other end? I don't think I will need gantry squaring so I assume the two limit switches will only be on one side of the two Y axis?

He is suggesting the second scenario discussed earlier. NO hard limit switches, and only a home switch, allowing the software to handle safety. I personally wouldn't do it this way on a fast servo machine, but many (most) hobby machines work like this.

Top break it down since its gotten a bit mixed up:

option 1

- A single inductive switch on each drive axis (you have 4). This switch acts when the machine homes. The machine has NO hard safety limits, instead using only the controls soft limits to set travel extents with zero safety backup.

option 2

- A single inductive switch acting as home AND hard limit switch. The machine homes and then moves off the switch. The hard limit switch triggers at both ends and runs only to the control, and will trigger e stop through the control. In theory this can be safer than option 1, but that depends on the control.

option 3

- A separate inductive home switch, and a mechanical hard limit switch. This would behave the same as the above, the only difference is the switches are independent. This makes it a little more flexible as you don't trigger any limits when sitting on home and you can set home in a different location, but otherwise identical.

option 4

- A separate inductive home switch, and a mechanical hard limit switch. This looks the same same as above, EXCEPT you would wire the limit switch into the E stop chain so it will kill the drives in any accident. This is how any basic industrial machine will be done for guaranteed safety. New machines can have much more complicated setups where the switch kills both power and tell the drive and the control what is happening. This may or may not be easy to do with your chosen control and drives but the basic e stop chain method will always work.

option 5

- Everything is separate. One switch for home, one for + hard limit, one for - hard limit, with the e stop chain, the control, and the drives all talking. This is what your new fancy pants machine will have, for both safety and diagnostics. This is *probably* beyond anything you would need or want.



As for the hard limit switches, weather you use one triggered at both ends, or a separate one on each end wired together, or into separate inputs does not matter. You do what is convenient. the end result (hard limit kills machine) is the same for all of them.

[dazp1976]

You only need (1) Micro switch or proximity sensor per axis just as this video shows, most modern machines where the controls have soft limits, you only need the (1) Micro Switch / Proximity sensor per axis

I have issue with this too. Why?. Well:

- say put limit on x+.
- you're cutting away and,
- it misses some steps in the x- direction,
- you don't notice,
- what this has done is technically made x- end plate nearer than what is was (should be) which the machine won't know.
- now it's possible to move x- past the end (crash).
- there is no limit switch on x- end to prevent it.

It's a flawed design.

Soft limits won't make a difference in this scenario because they will follow the software dro not the physical axis.

[mactec54]

Honestly I'm still a bit confused. There are two ends of an axis that needs to have a limit. A runaway servo motor could be traveling in either direction. I can see where a combo Home and Limit can be used on one end, but do you wire another proximity switch in the same loop on the other end? I don't think I will need gantry squaring so I assume the two limit switches will only be on one side of the two Y axis?

This is how all modern machine manufacturers do it, your hobby machine should not need to be any different, you can of cause add another Micro Switch or Proximity switch for each axis if you want it, just for peace of mind, they would connect into the same circuit, Servo runaway is unheard these days unless you have a very poor-quality servo system or very old system, the servo drives have safety systems that prevent this from happening, the servo drive will fault if there is a problem.

Soft limits work very well, even Mach3 from yesteryear has soft limits, if your system is wired correctly, the servos won't have any Power until the Machine Control is fully powered on, and you have pressed the RESET Button, then the servo motors will be active

[mactec54]

I have issue with this too. Why?. Well:

- say put limit on x+.
- you're cutting away and,
- it misses some steps in the x- direction,
- you don't notice,
- what this has done is technically made x- end plate nearer than what is was (should be) which the machine won't know.
- now it's possible to move x- past the end (crash).
- there is no limit switch on x- end to prevent it.

It's a flawed design.

Soft limits won't make a difference in this scenario because they will follow the software dro not the physical axis.

CNC Machine manufacturers have been doing it like this for a number of years, and I have never seen a problem, just insecurity on your part and not trusting your control to do is job it's designed to do

In your case, you would set your Soft Limits to accommodate any possible over travel

Soft Limits will work just the same even if you lose some steps, if you were to lose enough steps to affect the soft limit position, then you would of had a crash into your part long before you got to the soft limit position

[TeeTomTerrific]

OK then. Soft limits it is including one at each end and e-stop. I assume I can have a second soft limit input for the opposite end from the Home/Limit on the other end? Or is it better to have all the limits on a single input?

[dazp1976]

CNC Machine manufacturers have been doing it like this for a number of years, and I have never seen a problem, just insecurity on your part and not trusting your control to do is job it's designed to do
In your case, you would set your Soft Limits to accommodate any possible over travel
Soft Limits will work just the same even if you lose some steps, if you were to lose enough steps to affect the soft limit position, then you would of had a crash into your part long before you got to the soft limit position

I'm using soft limits atm without physical switches.
I want to put some switches into the new setup.
This is why I'm also debating which way to go with implementing them in.
Its been working fine, software works properly, it's my cam abilities that are sometimes questionable.

[mactec54]

OK then. Soft limits it is including one at each end and e-stop. I assume I can have a second soft limit input for the opposite end from the Home/Limit on the other end? Or is it better to have all the limits on a single input?

Yes, you set a soft limit position for each end of the axis travel, this also lets you move straight off the soft limit without having to do an override like you have to do, when you hit a hard limit

[mactec54]

I'm using soft limits atm without physical switches.
I want to put some switches into the new setup.
This is why I'm also debating which way to go with implementing them in.
Its been working fine, software works properly, it's my cam abilities that are sometimes questionable.

It has nothing to do with Cam, unless your part is bigger than your safe machining area, and how you chose to setup your X and Y G54 work-offset

[dazp1976]

It has nothing to do with Cam, unless your part is bigger than your safe machining area, and how you chose to setup your X and Y G54 work-offset

Setting up the geometry isn't a problem.
It was more a case of trying to push the speed too much and causing a spindle stall.
Resulted in stepper jump, cutter snap, and once the x axis table nearly ran off the end (one side open ended with no end plate) when I left it alone.
Yes... It had skipped steps that much!.


I'm more Conservative these days, learn from mistakes.

[ihavenofish]

OK then. Soft limits it is including one at each end and e-stop. I assume I can have a second soft limit input for the opposite end from the Home/Limit on the other end? Or is it better to have all the limits on a single input?

Soft limits are software, they have no relation to any physical switches. In a hypothetical machine with no switches whatsoever, you can still home manually and set limits in the software and in all likelihood never have an issue.

The only downside is you have no hard safety backup. That's really all it will come down to, do you want/need a hardware safety limit to prevent crashing? If yes, separate home and limits. If no, one switch per drive axis.

As I said, most hobby stepper machine are done with no hard limits, but they are also not very fast and run steppers. Steppers "bounce" when they hit an obstacle, servos just ramp up torque, so in the event of a crash they can cause much more damage. The servos will of course cut out milliseconds after the crash on their own.

[ihavenofish]

Setting up the geometry isn't a problem.
It was more a case of trying to push the speed too much and causing a spindle stall.
Resulted in stepper jump, cutter snap, and once the x axis table nearly ran off the end (one side open ended with no end plate) when I left it alone.
Yes... It had skipped steps that much!.


I'm more Conservative these days, learn from mistakes.

It's always a risk not to have hard limits.
The arguments against them is that its easier. Very little else of benefit.

The important thing here is there is no accepted right answer. Do what you need to do to make your machine operate safely.

[Okir123]

Hallo ich habe 1 Problem mit meiner bridgeport interact 412 und zwar macht die Maschine keine Referenzfahrt fahrt mehr kan mir jemand helfen

[dazp1976]

It's always a risk not to have hard limits.
The arguments against them is that its easier. Very little else of benefit.

The important thing here is there is no accepted right answer. Do what you need to do to make your machine operate safely.

I think I've come up with an idea for mine.
I've added a wiring diagram. Brief description:
So, entire stop circuit is NC. When estop opens/active the red light on tower lights up.
Limits:
3 limit switches in series on - end of axis tied into estop.
3 switches on + side of axis for homing wired individually.
Homing switches act as trigger for estop via opto module but then can't home without causing stop so...

Homing:
Override NO self release push button added to stop circuit, closes estop when HELD in and lights tower orange as warning via opto. Can now home without triggering e stop, if homing overtravels just release button and it opens/activates estop because switches will be lit from overtravel.
Notice push button has to be physically held while homing for it to work.

Servo spindle:
Estop tied to servo enable. Servo enable on manual button. If not enabled estop is active so MC cannot work. When enabled, estop is off and tower is green light for ready.
If servo disables itself it activates estop.
Stop also disrupts servo main power via contactor.

Other things still need to be added as I go no doubt.

See what you guys think, and mention any tweeks if you think it's crap!.

[joeavaerage]

Hi,
just a comment about the way you have combined the opto's.

The output circuit of an optoisolator is a phototransistor, so what you have done is put a whole bunch of phototransistor collector-emitter circuit in series.
Phototransistors often do not saturate particularly well, that is to say they may have a V_CEsat of 0.5V, maybe more. Thus if you series four of them then the voltage across all four will
be 2V, maybe more, and that is well above the low threshold voltage of 5V logic.

Compare that to a typical small signal BJT in saturation, they have a V_CEsat of 0.1V to 0.2V

Craig

[mactec54]

I think I've come up with an idea for mine.
I've added a wiring diagram. Brief description:
So, entire stop circuit is NC. When estop opens/active the red light on tower lights up.
Limits:
3 limit switches in series on - end of axis tied into estop.
3 switches on + side of axis for homing wired individually.
Homing switches act as trigger for estop via opto module but then can't home without causing stop so...

Homing:
Override NO self release push button added to stop circuit, closes estop when HELD in and lights tower orange as warning via opto. Can now home without triggering e stop, if homing overtravels just release button and it opens/activates estop because switches will be lit from overtravel.
Notice push button has to be physically held while homing for it to work.

Servo spindle:
Estop tied to servo enable. Servo enable on manual button. If not enabled estop is active so MC cannot work. When enabled, estop is off and tower is green light for ready.
If servo disables itself it activates estop.
Stop also disrupts servo main power via contactor.

Other things still need to be added as I go no doubt.

See what you guys think, and mention any tweeks if you think it's crap!.

If the Servo Drive has a Fault output that should be used and not the Enable

[dazp1976]

Hi,
just a comment about the way you have combined the opto's.

The output circuit of an optoisolator is a phototransistor, so what you have done is put a whole bunch of phototransistor collector-emitter circuit in series.
Phototransistors often do not saturate particularly well, that is to say they may have a V_CEsat of 0.5V, maybe more. Thus if you series four of them then the voltage across all four will
be 2V, maybe more, and that is well above the low threshold voltage of 5V logic.

Compare that to a typical small signal BJT in saturation, they have a V_CEsat of 0.1V to 0.2V

Craig

I'll bench test that one and check ASAP before diving in the box.
But... The entire thing is running at 24v. The only 5v logic is on the collector side of the bottom 4 optos which is my UCBB control board.
The small opto diagram top left which has the 3 in series on the stop side is the issue, correct?. Due to being a NC circuit they will all be on, it's 24v circuit.
Opto is 24v in & 24v out.

[dazp1976]

If the Servo Drive has a Fault output that should be used and not the Enable

Will do. I've just ordered some of those HF-41F type relays and cartridges to go on din rail with contactor and breakers.

[joeavaerage]

Hi,
if running at 24V then even a 2V-3V saturation voltage is no issue. If the same combination is presented as an input to 5V logic then yes its an issue.

Craig