VSDEPI, Mach3 and limit switches

[stevespo]

I'm getting ready to switch over my CNC router to some DC brushless servos and VSD-E drives. I've got things hooked up on the bench, and the motors are under Mach control. It's a really nice setup (Thanks Jason at ZealCNC).

I have the EStop and Home switches configured just fine. The Home switch for each axis (XYZ) are wired through the encoder inputs to pin 4.

What I'm wondering about are the limit switches. On my Gecko setup, I have my limits wired in series and then sent to pin 15 on my break out board. Since I'm not using the VSDEPI "A_CMD" input for anything, can I simply wire my limit switches to the 16 pin harness that would normally be used for the 4th drive? Will that simply pass thru?

So, what pins (on the VSDEPI, A_CMD) are tied to pin 15 and GND? Perhaps I can poke around and discover this myself. Hope this makes sense.

Thanks,

Steve

[Jason3]

Hey Steve,

Hopefully Tero will drop by and confirm regarding using the A_CMD, but I think the answer is yes you can use the unused pins for home switches. Let me see if I have the pinout here and I'll get back to you.

Best regards,

Jason

[Jason3]

Hi Steve,

The A_CMD pin 3 goes to pin 9 on the VSDEPI DB25 output, pin 5 goes to pin 8 according to the manuals. Tero did also mention to me once that you can use pins 9 and 10 which are unused in single axis mode, but that some manual wiring would be required.

Best regards,

Jason.

[stevespo]

Hey Jason, as always - thanks for the help. I couldn't quite remember what strategy you suggested the last time we discussed this. I can see the pinouts in the VSD-E manual, and it does make sense.

Pin 1 (NC)
Pin 2 (IO_COM) Common voltage for optoisolated I/O
Pin 7 (IO_VCC) Voltage supply for optoisolated I/O
etc...

So, I would think that Pin 9 (IN4) or 10 (IN3) look right. They're both "medium speed optoisolated inputs). Pin 10 is labled SPI MOSI (Master Output/Slave Input) so that seems like the likely candidate. As long as it passes back through to pin 15 on the parallel port, that's all I need. I should be able to tell in just a few minutes.

Thanks,

Steve

[stevespo]

Actually, it looks like it's pin 12 on A_CMD. As expected, I get a 5V reading on pin 7 (IO_VCC) and 4.3V on pin 12 (OUT1, medium speed, general purpose output).

If I configure parallel port pin 15 as X++ limit switch and then short pin 12 to pin 2 (IO_COM), Mach registers that as the limit switch being triggered. Thanks again for the help - I think I'm all set.

Steve

[Jason3]

Excellent, nice work. I'll store that away as a handy thing to know

Best regards,

Jason

[Xerxes]

I published the VSDEPI schematics, hope it helps:
http://www.students.tut.fi/~kontkant...schematics.pdf

A_CMD pins 3, 5 and 12 can be easily used as additional I/O as they go straight to parallel port.

The other end of switch should be connected to one of GND pins and a pull-up resistor (1-5kOhm) from input pin to VCC is needed to make switch work.

[stevespo]

Xerxes, thanks for your help. This is one of the last things I need to do. Without a VSD-E in place to perform filtering, do you think it's necessary to add a small capacitor and create an RC filter? Pin 12 should be opto-isolated via the VSDEPI, but any harm in filtering?

Steve

[Xerxes]

Originally Posted by stevespo Xerxes, thanks for your help. This is one of the last things I need to do. Without a VSD-E in place to perform filtering, do you think it's necessary to add a small capacitor and create an RC filter? Pin 12 should be opto-isolated via the VSDEPI, but any harm in filtering?

I think there is filtering options in Mach3 settings. Passive filtering like in the picture surely won't hurt if you encounter problems.

[stevespo]

Mach has "debounce" settings that are useful. In the past, when I had problems with noise on my inputs I had better luck with a small cap as pictured. I'll try it without and see what happens.

Steve

[stevespo]

Xerxes, can you look at my little diagram above and see if anything is incorrect? I am still seeing noise issues on those limit switches and when I use the circuit above it keeps the input high regardless of the state of the switch.

My understanding of this setup is that when the switch is closed the current will flow to ground and the signal on pin 12 should be low. When the switch is open, the current will go to pin 5, and it will read high.

I'm actually not sure what role the capacitor is playing with the diagram drawn as above. Is a low pass filter the correct choice here? Can you enlighten me or at least suggest a filtering option that I could try?

Steve

BTW, I'm using a 4.7K ohm resistor, and I've tried both .01uF (103) and .1uF (104) capacitors.
The other thing I've noticed is that pin 13 on the VSDEPI port A doesn't appear to go to ground.
I suspect that is typically supplied by the (missing) VSD-E? So, I've gone right to my grounding block.

[Xerxes]

It really seems to have a flaw, sorry for not checking it earlier.

Switch can't pull Vout to GND thru a capacitor (because capacitor acts like an open circuit to DC currents). To fix this, connect the switch in parallel with the capacitor (instead of series connection like in the diagram).