Hello Guest it is March 28, 2024, 01:26:22 PM

Author Topic: Soft stop at limit switches  (Read 21390 times)

0 Members and 1 Guest are viewing this topic.

Soft stop at limit switches
« on: September 09, 2010, 09:19:48 AM »
Hello expert Mach users,
we had a long discussion about that at a german forum, but we did not conclude, so I'd like introduce myself with the following question:

Usually, when an end switch is triggered, Mach stops clock pulse immediately and sets status to emergency stop. Let there be a reason not to stop clock pulse immediately, being equivalent to a hit with a hammer to servo drives, but slowing down with a ramp, similar (but maybe much steeper) to that connected  to the rapid move/motor tuning conditions. Is there a way to achieve that?

I'm sorry if this topic has been discussed here earlier and just my research at this board has not been accurate enough to find it.

Many greetings
Nicolas

Offline Hood

*
  •  25,835 25,835
  • Carnoustie, Scotland
    • View Profile
Re: Soft stop at limit switches
« Reply #1 on: September 09, 2010, 09:28:16 AM »
Set up soft limits and Mach will stop before you hit the physical limits. Physical limits are meant as a last defence and they need to stop instantly.
Hood
Re: Soft stop at limit switches
« Reply #2 on: September 10, 2010, 04:01:37 AM »
The problem is, that when hitting limit trigger, Mach3 stops immediately drive clock. This has some major disadvantages:

1. At some machines, the limit switches which are reported to Mach are not the last security instance because that one is realized in hardware. So when limit trigger is reported to Mach, it would be convenient to slow down with a steep ramp instead of killing drive clock for a more gentle treat of the mechanic components.

2. At some machines with servo drives, namely those where the servo controllers have no separate input for emergency-stop, slowing down with a ramp leads to faster stop than just clock interrupt.

3. Same is at some machines with stepper drivers, where instant interrupt of drive clock leads to lost steps, resulting in a slower slowdown than possible.

At all these cases, slowing down the drives with an adjustable ramp would be beneficial either for security reasons or for a considerable treatment of hardware.

Many greetings
Nicolas



Offline Hood

*
  •  25,835 25,835
  • Carnoustie, Scotland
    • View Profile
Re: Soft stop at limit switches
« Reply #3 on: September 10, 2010, 04:06:46 AM »
Using SoftLimits will avoid all the above situations so why not use them?
Hood
Re: Soft stop at limit switches
« Reply #4 on: September 10, 2010, 04:16:19 AM »
Of course, limit switches become interesting when soft limits either have not applied, e.g. machine has not been referenced or have been switched off intentionally.
Also, points 2 and 3 are interesting not only when limit trigger is applied, but also on any emergency stop condition.

Many greetings
Nicolas






Offline Hood

*
  •  25,835 25,835
  • Carnoustie, Scotland
    • View Profile
Re: Soft stop at limit switches
« Reply #5 on: September 10, 2010, 04:39:37 AM »
If you neglect to reference or you disable the softllimits then that is operator error in my opinion.
I am finding it hard to think of a situation where SoftLimits would need to be disabled, I dont think I have ever had them disabled on any of my mills since I set them up.
If you are in the habit of forgetting to reference your machine then I suggest that you make a new screen up so that you can not get to any controls until you have referenced the machine.

Having the limits ramp down when seen will mean that you lose working travel and is not a situation I would want and I suspect many others, using SoftLimits does not affect the working space.

If you feel strongly enough about this then probably it is best to contact Brian at Artsoft and ask him to write Mach to suit your needs but also make sure that he writes it as an additional option as I suspect most users will be happy with the way things are now, ie softlimits to do exactly what you are wanting hard limits to do.



Hood

Offline stirling

*
  • *
  •  2,188 2,188
  • UK
    • View Profile
    • www.razordance.co.uk
Re: Soft stop at limit switches
« Reply #6 on: September 10, 2010, 06:06:43 AM »
Nicolas - Let's go with your method for a moment. Suppose from max rapid your machine takes 10mm to come to a controlled stop. You'd have to place your limit switch at least 10mm from where the machine absolutely must be stopped. Now suppose for whatever reason the machine finds itself heading for a position at 5mm from end stop. It'll hit the switch at 10mm and come to your controlled stop. Now suppose you go with the soft limit method. Your machine heads for the 5mm position, hits the soft limit at 10mm, slows down, gets to the 5mm position and CARRIES ON with the job. Which is better? Now suppose it's heading to a position beyond the end stop. It goes into the slow zone and then hits the hard limit - but it does this at a speed set to NOT damage the machine. It's win win either way.
Now your E-Stop - you get your arm caught - would you prefer it to stop NOW or let it chew your arm for another 10mm or so in order not to possibly damage the machine?

And now a confession - Many moons ago I used to think the same as you - that was until a younger Hood and others put me right  ;D

Offline Hood

*
  •  25,835 25,835
  • Carnoustie, Scotland
    • View Profile
Re: Soft stop at limit switches
« Reply #7 on: September 10, 2010, 07:13:52 AM »
stirling,
 just a corrections if I may, you said
Now suppose it's heading to a position beyond the end stop. It goes into the slow zone and then hits the hard limit - but it does this at a speed set to NOT damage the machine.
What would actually happen, if Soft Limits are set correctly, is it would go into the slow zone and come to a stop before it actually triggers the Hard Limit. It will allow you to reposition and no Hard Limits will have been hit so no loss of position.
Hood
Re: Soft stop at limit switches
« Reply #8 on: September 10, 2010, 07:30:11 AM »
Hello Stirling,
your argumentation is based on the idea, that there is a possibility to stop NOW. In fact, it always takes time to decelerate any moving mass. So instant stop is impossible. So the closest position to "now" being physically achievable is: "I want to stop as fast as possible!"

Starting from the "as fast as possible" point of view, please allow me an analogy:
When braking your car in front of an obstacle, for some cars it's the the best way to decelerate just making all four wheels block - you use sliding friction for your wheel on the ground. In many other situations, not to make all four wheels block but keep your wheels rolling on ground, use the sticking friction of your wheels, and slow down with a defined traction results in shorter braking distance.  

Back to my servo drives: There exist some servo drives and also stepper drives, where stopping drive clock leads directly to lost steps resulting in much longer deceleration path than stopping with a defined ramp.

Greetings
Nicolas

Edit: Minor typos

« Last Edit: September 10, 2010, 07:32:26 AM by Nicolas S. »

Offline stirling

*
  • *
  •  2,188 2,188
  • UK
    • View Profile
    • www.razordance.co.uk
Re: Soft stop at limit switches
« Reply #9 on: September 10, 2010, 07:52:23 AM »
Hood - understood - thanks.
Nicolas - fair point - I can see in that case why you might want to introduce a brake on E-stop rather than just rely on the pulse train stopping  - (don't these exist for heavy machines that would benefit?). Afraid I still can't see the limit thing though - given soft limits.

Ian