Help with Limit Switches please

[Imagineering]

Another new thread in my seemingly unlimitless quest for information . . .

I have NC Limit switches at each end of all 3 Axes.
Each pair of switches for each Axis is wired in series.
One end of each series'd pair is commoned to pin 19.
The other ends are;
X-Axis to pin 10, Y-Axis to pin 11, Z-Axis to pin 12.
E-Stop is NO and connected between pins 13 & 15.

All the limit switches are set to active high as pins 10, 11,12 are shorted to pin 19 when they are not hit by the respective Axis movement.

When the Limit Switches are set enabled & active in the 'Config Menu', I try to run a program and fail due to a 'A limit Switch is active' message, even though all limit switches are closed. (I also tried 'active low' just in case my reasoning was wrong).

Pins 10,11,12 are low when the NC limit switches are not hit, ie 'switch is closed therefore pins 10,11,12 are directly connected to pin 19. Am I right in thinking that the pins 10,11,12 must be taken high when the limit switch is opened by a pull up resistor or associated circuit, or am I missing something yet again??
If so why would I get the 'Limit Switch is active' message when they aren't??

When I enable the E-Stop in the Config Menu I get the same problem ' E-Stop is active' even though I 've configured the E-Stop to active Low and the Switch is not connected. Do I need to pull the up to prevent this perhaps?

Sorry . . . a bit rambling but I can't get my head round the seemingly illogical problem.

[rustyolddo]

Note, pin 11 is backwards in the software so you will need to change the setting opposite of the others. See my post down the board as well regarding limit switches, "G28 Home Function..." I needed bypass caps because the motor noise was getting into the switch lines causing false levels. Also, set the limit switch debounce to 50-100. I suggest working on one limit switch line at a time, disable all the others until you get one working. Chances are it will solve the problems with the rest.

[Imagineering]

Originally Posted by rustyolddo Note, pin 11 is backwards in the software so you will need to change the setting opposite of the others. See my post down the board as well regarding limit switches. I need bypass caps because the motor noise was getting into the switch lines causing false levels. Also, set the limit switch debounce to 50-100.

Do you mean that I need to configure pins 10 & 12 as Active High and Pin 11 Active Low ??

If so, how do I solve the E-Stop problem which defies my logic ??

TurboCNC reports that the E-Stop is activated when it isn't. BTW This switch is NO.

[Evodyne]

Imagineering,

Hi! Typically in industry e-stop switches are normally closed and open when you hit them. The reasoning is that if the e-stop circuit loses power it will shut down the protected equipment. As such it is fail-safe.

Your "E-stop Active" message seems to me to be saying that the software expects you to be using a normally closed switch; to it your normally open unit looks like a tripped e-stop circuit.

Does this make any sense?

Lance

[Imagineering]

Originally Posted by Evodyne Imagineering, Hi! Typically in industry e-stop switches are normally closed and open when you hit them. The reasoning is that if the e-stop circuit loses power it will shut down the protected equipment. As such it is fail-safe. Your "E-stop Active" message seems to me to be saying that the software expects you to be using a normally closed switch; to it your normally open unit looks like a tripped e-stop circuit. Does this make any sense? Lance

All six Limit Switches are NC for your reasons above. Rustyolddo has suggested that there is a Software problem with pin 11 which may solve the Limit Switch problem.
However, I've tried setting the E-Stop activation in the Config Section to both 'Enable High' and 'Enable Low' and both settings have the same effect.
Rustyolddo has also suggested in one of his previous posts that Electrical Noise in the Switch Wiring could cause the problem I'm having.
I would prefer an NC Switch for an E-Stop and will fit one when I get to buy a "Big Red Mushroom" type of Switch.

[Imagineering]

Rustyolddo,
Would .01 uf caps from pins 10,11,12 to a ground pin at the breakout board do the trick, or have you positioned your caps' elswhere??
I also have one of my Limit Switch Cables running with the Motor Drive Cable, Hmmm it just happens to be the Z-Axis that I'm having trouble with too . . .
However, I think that I may have tracked that particular problem down as I've run programs for 1 1/4 hour without the Spindle running and have had no glitches.

We should all get onto Skype to discuss this as a conference call instead of chasing each others Emails around the place . . .

[rustyolddo]

Originally Posted by Imagineering Rustyolddo, Would .01 uf caps from pins 10,11,12 to a ground pin at the breakout board do the trick?.

Yes, that's what I did.

I also have one of my Limit Switch Cables running with the Motor Drive Cable, Hmmm it just happens to be the Z-Axis that I'm having trouble with too . . . However, I think that I may have tracked that particular problem down as I've run programs for 1 1/4 hour without the Spindle running and have had no glitches. We should all get onto Skype to discuss this as a conference call instead of chasing each others Emails around the place . . .

You'll definately get noise induced into your limit switch inputs from the stepper motor drive. If you had an o'scope you could see what was going on. I wonder if you could get a torroid type noise filter for the dremel. You've got two strikes against you with it, the electronic RPM control and the noisy brushes.

When you shield stuff with braid, make sure the ground point is the same for all, what is known as a 'Star' ground. You can negate your work and create ground loops if you have multiple ground points.

[Imagineering]

Originally Posted by rustyolddo When you shield stuff with braid, make sure the ground point is the same for all, what is known as a 'Star' ground. You can negate your work and create ground loops if you have multiple ground points.

I've got what I would loosely call a 'Snowflake' grounding system ie. each cable within each area has a braid around it and is then connected from 1 end, (only), of the braid to a common wire which is then in turn connected to my common point or earth.

Cable Braids
------------------------)
------------------------)_________)
------------------------) ) Common Earth Point
------------------------) )__________________>
)
------------------------) )
------------------------)_________)
------------------------)
------------------------)


Having drawn that, I just realised that my 'Common Earth Point' is actually my 'Common +Ve for the Power Transistors, so that all the braids and Resistor Heatsinks are Common to +Ve and not 'Earthed' or 'Grounded' at all.

There is no electrical connection to the Frame of the Machine, so I'll double check that, and then take my +Ve to the Frame and Ground that. Trouble is, my PSU is separate from the Machine, so no ~230Volt mains goes any where near it - so, no Earthing Line to connect it to. 8-(

I also can't disconnect the braids from the +Ve and earth them and run a separate +Ve feed wire, as the braids from the Steppers are the Common wire which carries the +Ve to them, (4 wire braided cable). 8-(

[Imagineering]

I had a look at my 'ASCII Art' in my last post and it didn't view very well, hopefully you can figure it out from the drawing and my description.

[Imagineering]

Back again, I've managed to get the Limit Switches working, thank you Rustyolddo, It was the reversed High/Low on Pin 11 that was causing the problems.
I've now run the machine through a number of programs & cycles without the spindle running - the Z-Axis has returned to its original position every time,so it's definitely the Dremel causing the interference.

Rustyolddo, I've read your previous posts regarding your RFI interference problem and I've tried a few tests on my Machine. Firstly, My main Frame/Chassis has absolutely NO electrical connection with anything at all, not even the Mains Earth, (as my PSU is separate.
I've done resistance checks from all possible wiring/component/cabling/steppers to the Frame/Chassis and there is infinite resistance between these. I put my digital Multimeter between the Resistor Heatsink, (which is also connected to the +16Ve Stepper feeds), and the Frame/Chassis and have a Voltage between 5.5 - 6.5 Volts DC occuring, (Chassis is Negative). There is only one possible source for this, which is RFI from somewhere. I'll try a few small capacitors between the Chassis & Heatsink tomorrow just to see if it goes away. I'll also try connecting the Chassis & Heatsink directly and observe. I've still yet to put the .01uf caps across the Limit Switch inputs, so I'll also do that tomorrow.

Your comments will be eagerly read if you come back on line later.

Murray.

[rustyolddo]

I need to read your post a little closer, been out in the heat all day & can't think straight. I presume that you don't have access to an oscilloscope which would make it a bit easier. The dremel has no ground so it is going to radiate noise everwhere. Make sure that you don't have any of your control cables near the Dremel, esp. parallel to the Dremel power cord. A ferrite bobbon on the power cord might help or a power strip that has a noise filter in it. You have to have a data ground that the computer uses, I would tie your shields to that. Let me think about it a bit more when my head quits throbbing.

[Imagineering]

Rustyolddo,
You are quite right, I don't have access to a 'scope unfortunately.
I've had a bit of a play around today and modified a few things.
I have built up a filter box for the Dremel and also placed a Ferrite Toroid on the Power Lead where it enters the Dremel. The Filter Box consists of a Filter Socket, (like the ones in a Computer PSU), connected to another Filter package, (one of those Alloy Tubular thingys about 50mm dia), which is then connected to my Control Relay and a Mains Socket for the Dremel. All of this is packaged up in a Steel Box which is Earthed to the main Frame. This should stop any Radiated Noise in the cabling at least. I've also placed an earthed sheet of steel on the Frame, between the Dremel and the Cabling.
I haven't had a chance to test it yet, but tomorrow is another day.

Regarding the Data Ground, this is where I have all my shields attached. I've also placed .01uf Capacitors across every input to earth which should help.

Hope the head gets better - Sun???? what's that?????