Reply with QuoteSorry, I don't remember printing any limit switch covers for the mill but that's a good idea.
Hoss
Mounting your limit switches show and tell in one thread for G704
So lets see how you folks mounted and utilized the various different types of limit switches for your G704 cnc conversions.
I know Hoss has a wonderful youtube video of his, however not all of us have a cool 3d Printer to make those water tight enclosures.
Similar Threads:
Sorry, I don't remember printing any limit switch covers for the mill but that's a good idea.
Hoss
Gosh, you've... really got some nice toys here. - Roy Batty -- [URL]http://www.g0704.com[/URL]
Sorry Hoss, I got your water tight limit switch project mixed up.
132 view and only1 reply! Lots of interest though.
Anyways. I'm going to use the existing holes on the front of the mill where the handles for the X lock handles were to mount either a strip of flat metal or ABS plastic which will then hold two mechnical limit switches using the existing X table stops.
The Y axis will be similar and using those lock handle holes as well to hold the strip of metal or plastic.
Yup, lots of interest from me tooI'm working on installing my proximity switches, I'll machine a cover box and both the X and Z axis are simple. The Y axis is slightly more complicated to package, no real flat surfaces to mount and measure from!
Not my machine, but a very neat looking optical sensors, some good ideas.... http://www.einfach-cnc.de/bf20_cnc-umbau.html
CR
Had to use google chrome to translate that page. Thanks for the info. Not much detail spent on the sensors. but it gives another option on how to mount them under the table versus externally.
Those look like proximity sensors to me, not optical. I should have had my prox sensors hooked mounted by now, but have been too busy with other stuff to work on the mill lately. I have a plan but am curious as to what others have done. I can always change direction.Originally Posted by ny_racer_xxx
Those are definitely prox sensors, like these:
http://www.ebay.com/itm/5PCS-Inducti...item51a6e43611
Regards,
Ray L.
Those look to be Normally Open contact, I think what you want is a Normally Closed contact like this:
http://www.ebay.com/itm/DC-3-wire-6-...item35c390f505
The reason being is that is if the wire breaks with a NO contact, the machine will crash before you know there's a problem. If it's a NC contact, if the wire breaks, it will stop.
CR
I was just get ready to start a thread on limit and homing switches. I was looking to use proxy switches, is it best to use 2 per axis?
They are not "contacts", they are open-collector outputs. You can wire them all to the same BOB input, with a pull-up resistor, and the input will be normally high, and will go low when you reach a limit. This requires that each switch comes close to some ferrous metal at the limit, and not at any other time. If you use normally close switches, and wire them all to a single input, then you have to ensure each switch is close to ferrous metal all the time, EXCEPT when at a limit. Prox switches cannot be wired in series like mechanical switches, and since they are about equally likely to fail in either state, there really is no benefit to either method - it's more which is physically easier to implement on your machine. They will be FAR reliable long-term than mechanical switches in any case. Wiring failures should not be a significant concern if the wiring is done correctly.
Regards,
Ray L.
There is no "best". Whether you use one or two per axis depends on the switches you buy, and how you choose to do the installation on your machine.
Regards,
Ray L.
I have two mechanical limit switches for both X, Y and Z. Emulating Hoss's Z adjustable limit switch position.
My BF20 is similar to the G0704 so I thought I'd add my two cents worth......I'm using CNC4PC limit sw which is basically six NO/NC contact switchs wired in series with an eithernet jack for ease of use. .....However this configuration may cause (on a rear occasion) a crash with Mach3......and what I mean is that if for any reason one limit sw is activated (i.e. triggered) the way it comes wired each axis shares the three states "home,- & +" on only one pin.....and therefore when a limit is triggered and "reset" but NOT released (i.e. the axis has NOT been moved off the limit switch) when another axis reaches it's limit and triggers its own sw Mach will not register this limit (as another limit is triggered) and the machine will come to a crash stop........However with this scenario as I mentioned above with correct measurements inputted in Mach's "homing and limits" page and inputting a slow to stop (for example I have the last 10mm of axis travel) this sort of crash can be avoided.....
And I've also made limit mounts myself to reduce cost......and if anyone is interested in making their own too the pics below may be of use too...
Eoin
I guess I forgot to "show" mine as well.
These are the mechanical microswitches.
http://www.youtube.com/watch?feature...v=uA0B46tvX-4#!
http://www.youtube.com/watch?feature...&v=RGjuVftscoQ
And this is the little optical switch.
http://www.youtube.com/watch?feature...&v=4B0tII8A_yU
Hoss
sorry for the links, apparently we can't embed more than 1 video per post now.
Gosh, you've... really got some nice toys here. - Roy Batty -- [URL]http://www.g0704.com[/URL]
What is "best" is debatable. as Himmy has noted, it depends on how you set it up. You can set it up so that a single switch can detect both ends of the mechanical travel of an axis. If you ask some people would say that 3 switches per axis is the ideal, one home and a limit on each end. You can go from that extreme to no switches and just use the software limits. My planned setup is to use one prox sensor on each axis that serves as a combined home and limit for one end of travel. The other end of travel will be handled by software limits. I have found that the software limits alone work very well for me and I am adding the switches mainly for homing, but since they are there they will also serve as a limit switch for that and of the travel. I am going to run a little experiment when I can and if it works I am hoping to use a single prox switch for both ends of the axis travel. If it works I will post the results.
Thanks for the reply's. My initial plans were to use 2 proxy switches per axis. Was also thinking wiring these through a relay for more or less of a way to kill power to drives besides software softlimits.
I'm currently on a single switch for homing wired in an NC loop. That may change as my knowledge level and use of the machine change. I think positioning them to so that they can limit travel in both directions may be where I end up to keep the number of switches and the amount of wire down to a minimum.
Ray, thanks for the education, I was thinking exactly like they are contacts.... So if they were all connected to the same pin on the BOB how would the machine know where it is? I'm probably going to build my machine with a Geko G540, there's separate pins for each axis...
CR
With Mach3, if you have only one LIMIT switch input, it assumes that when a limit trip occurs, ALL axes in motion have hit their limit. I use the same logic on both of my machines with the KFlop, and it works just fine. When you hit a limit, you want ALL motion to stop ASAP, so it really doesn't matter whether one axis hit, or more than one. You'll know which one hit, and can simply back off on that one axis, re-reference, and pickup where you left off. I can't see any significant functional advantage to having multiple limit inputs.
My machines both use normally-open NPN prox sensors. They are all connected to a single input, with a pull-up resistor. In the normal state, there is no ferrous metal near any of the sensors, so all are effectively "open", and the input is HIGH, due to the pull-up. When any axis approaches a limit, that one sensor will approach a piece of ferrous metal. Once it gets close enough, that sensor will pull the limit input LOW.
You can also use normally-closed NPN prox sensors, and make sure they are all near ferrous metal EXCEPT at a limit. These will also behave as described above, so you can also use NO on some axes, and NC on others, and still wire them all to the same input. I prefer the normally open, because it makes it easy to make the limit position adjustable, by simply having adjustable "flags" of ferrous metal, with slotted mounting holes, or mounted in T-slots.
Regards,
Ray L.
Posting Permissions
