Weird Limit issue

[Mogal]

Hello All, I'm having a weird issue with my limit switches.
The setup is a UC300ETH-5LPT and a C10 board.

The limit switches are wired for 5 volts, normally closed, X+, X- on pin 10, Y+, Y- on pin 11, Z+, Z- on pin 12. Probe is on pin 13.

All the axis move and home correctly, but looking at the DIAG page, I get this:

- when the X axis limit is triggered, X+, X-, X home, AND Y- light up
- when the Y axis limit is triggered, Y+, Y home and Z- light up (Y- does not light up)
- when the Z axis limit is triggered, Z+, Z home and A- light up (Z- does not light up)

I've been going through my wiring and it all seems good. (But I'm tired now, and maybe missed something)
I am completely lost in why the limits are behaving the way they are. A Axis isn't even configured!!
After all my trouble shooting, I can't seem to find the issue. (Corrupt file maybe?!)

The system moves and homes okay, but I'm just not comfortable with how this is working right now.

Any suggestions?
(please don't mind the wiring, after ripping it apart so many times looking for trouble spots, I've kinda messed it up!)

[dezsoe]

Please, upload your profile (.pro) to check it. Also, which program version do you use?

[Robertspark]

I suspect that your problem could be that you are sharing the same power supply for the UC300ETH and the C10 board.

The C10 appears to require 5V

https://medw.uk/liberty/download/file/966

the problem is what is the purpose of the C10 board?

If it is to provide isolation from the UC300eth to the various items of equipment.

How is that going to happen if you are using the same power supply source as both sides of the optocouplers / optoisolators are from the same 5V supply

In addition, you probably want to consider / reconsider those buck converter power supply voltage regulators (LM2956) (Shown with a red arrow) as they do not provide an isolated voltage supply (they share the same ground connection from input to output).



You are also sharing that same ground with the stepper motor switch mode power supplies.

Also..... how is the other end of your probe wired (is it through the spindle with a PCB type conductive touchoff arrangement or is it via an isolated probe / stand alone touch off height sensor.


https://openbuilds.com/attachments/img_0470-jpg.29237/
https://lh3.googleusercontent.com/proxy ... 2AWvwM5KsE



Although not related to your issue, you may need to wire up the enable signal for the C10 or the outputs won't step.
top RHS of linked document here:
https://medw.uk/liberty/download/file/966

[Mogal]

Please, upload your profile (.pro) to check it. Also, which program version do you use?

Hello,

Attached is my current profile.
The software version I am running is: 1.2047
The firmware is: 1.0230

I'm still using this version until the new 2017 screenset come out.

[Mogal]

I suspect that your problem could be that you are sharing the same power supply for the UC300ETH and the C10 board.

Yes I am using the same power supply. It comes from the 36v to two buck converters. one for 12v and one for 5v.
The 12 goes to the fan, the other to both C10 and UC300ETH-5LPT. I had wall warts originally, but thought these would be OK and clean up the system. Guess not?!
So far in my preliminary runs, there doesn't seem to be a noise issue... but I'm not fully up and running to test.

the problem is what is the purpose of the C10 board?
If it is to provide isolation from the UC300eth to the various items of equipment.
How is that going to happen if you are using the same power supply source as both sides of the optocouplers / optoisolators are from the same 5V supply

Sorry, I'm not understanding this? The purpose of the C10? are you implying that it can be removed and everything plug into the LPT5?
I've shared the PSU in the past without issue... I know, doesn't make it right, but I can't run 3-4 individual power lines to the CNC.

In addition, you probably want to consider / reconsider those buck converter power supply voltage regulators (LM2956) (Shown with a red arrow) as they do not provide an isolated voltage supply (they share the same ground connection from input to output).

So I should put the wall warts back in to provide 5v and 12v? That means the addition of 3 more PSUs. Oh, and I just don't have that many 110vac lines run to the CNC machine to give them individual supplies.
I do have the machine itself grounded to a second location.

You are also sharing that same ground with the stepper motor switch mode power supplies.

Again, I don't have that many 110vac lines running to the CNC machine. I'm not sure how to over come this without rewiring half the house.

Also..... how is the other end of your probe wired (is it through the spindle with a PCB type conductive touchoff arrangement or is it via an isolated probe / stand alone touch off height sensor.

I have the probe wired through the spindle to two touch plates. One stationary and one movable. Its an open circuit until the tool touches the plate.

Although not related to your issue, you may need to wire up the enable signal for the C10 or the outputs won't step.

The system works. I haven't run any code yet, but I can home the machine, move the spindle, turn the spindle on/off. Everything is working, I am just getting weird results when the limits are hit.
I haven't seen any noise in the system yet, but I'm still putting it back together. (I'll be cleaning up the wires again once I get things sorted out)

Want a good laugh? This is the old box I had, and it worked great! No buck converters, but shared a single 110vac line as well as the 5v again (albeit with the wall wart)
I made the new box to incorporate a 4th axis and an A axis (for a future upgrade for dual lead screws on the long axis)

And a pic of the machine itself. This is with the old spindle that self destructed. Both the brushes and bearings failed at the same time. (That's what started this entire upgrade sequence!)
I was able to file down some spare brushes from a different motor, but the bearings where a write-off. (Couldn't source bearings that fit properly - Not worth the repair)

[dezsoe]

I've just tested your profile as you can see in the video. It works fine. Maybe, you could check the diagnostics screen for the LED numbers. I don't remember if there was a problem with it, but I have the latest version of the 2017 screenset and it works fine.

[Mogal]

Okay, thanks for testing it. But now I'm really confused.
I just don't understand how the limits switch around like that. Especially the A axis, as its not wired or configured for it!
Right now, it works, so I'll do a couple more tests, and make a couple practice runs while monitoring the system.

I have to rebuild the mount for the new spindle (Its like 3mm narrower then the old spindle.. sigh)
Once I have the mechanics finished, I'll be back to testing everything else.

I'm not prepared to format the computer for reinstall, but if it comes to that I will.
I'm not sure how to check the LED number as you mentioned, but I'll look into that too.

Thanks!
Chris

[Robertspark]

You won't need to format the pca

[cncdrive]

My advice is to pick up a multimeter and measure the inputs to figure out what is going on.

[Mogal]

Okay, everything is back in order!
Thanks to dezsoe for finding the issue!

Turns out (somehow) the LEDs got misnumbered. I got the latest copy of the screenset and installed and overwrote the old files.
Once that was done, I tested the limits and the all reported correctly!

WOOT!

Thanks everyone for the help. Now to mill out the new mount for the new spindle!