[Mick14]

Hi,

This is my first post on this forum. I will start by saying thankyou to the many people who have posted over the years that have enabled my to learn as much as I have from just reading.

This forum inspired me to buy a mill to convert to CNC. I have started, and have acquired a number of the required components. I have a KFLOP and a KSTEP sitting on the table awaiting me to start wiring, but before I commence I like to have a solid plan. I post this here seeking any advice/feedback on how I have done things, and because I would like to have my lessons available for others. There is currently very limited detailed info available on setting up a basic open loop system with the KSTEP.

I am comfortable with the motor side of the wiring, but have limitations on the circuitry side of things. I have attempted to design a circuit that achieves the following:

1. E Stop physically cuts power to any element moving under motor control (spindle & steppers)
2. Limits wired to stop all axis movement if tripped (currently through KFLOP firmware, I will likely set the KFLOP to kill motors, but will need to insert a manual override to the limits to enable this - or is it better to run this on the same circuit as the E Stop and have physical relays cut power when a limit is tripped?)
3. Independent home switches for each axis.
4. Spindle direction and speed control.

I think I have managed to achieve all of this from the KSTEP, leaving the KFLOP I/O for future additions such as coolant, PDB, tool changer, the sky is the limit. In particular, I am seeking feedback on:

1. Have I made any foolish conceptual mistakes that are going to cost me a fortune?
2. How I have set up the E Stop - is there a better way?
3. Any thoughts on whether the limits should be separate from the E Stop, or in series (or for that matter, independent on each axis).
4. Comments on grounding. My homework tomorrow is reading the siemens document on star grounding for further research posted by Al the Man, but any advice in this are would be great. It seems counter-intuitive to me to ground different DC voltages and AC to the same node, but I have a very basic understanding of electronics.
5. Review of how I have wired this home limit switches. I bought them before I fully comprehended the difference between PNP and NPN. Based on my research it should still work, but I am not fully confident.

I apologise for the amateurish nature of my wiring diagram - this is the first I have ever drawn. I used LibreOffice Draw, but am currently looking for a circuit specific program if anyone has a decent recommendation.

Once I have this circuit finalised and have finished acquiring mechanical bits and pieces I will be doing a full build thread in the benchtop mill forum, as there is limited info on the mill I bought aswell (Titan TM25V).

Thanks in advance for any help.

Cheers,

Mick.

[TomKerekes]

Hi Mick,

Nice wiring diagram. But I see a few issues.

Relay Coil A and Relay Coil B are in series and then connected to an input. Not sure what you were trying to achieve, but the input will not likely be enough to drive those relays. What voltage/current are the relays? Relay coils should have a reverse diode placed across them so that when the path to them opens there is a path for the coil current to recirculate so that the result isn't a huge revers voltage (spark).

Unfortunately the KSTEP Opto inputs were designed to be driven by a circuit that sinks current (acts like a switch to -). PNP type of outputs usually source current (switch to +). So I doubt if that arrangement will work. You might be able to add a pull down resistor (1K) to make them work.

Since you have many unused inputs I would wire the limit switches to separate inputs rather than putting them in series. This would allow KFLOP to allow motion back out of a limit switch.

Regarding your Common Ground (Earth Ground?). I think it is better to define separate Grounds (Earth GND, DC GND, 48V Motor Supply GND, 24V Limit/Relay GND) and keep them separate. Then you may want to leave those separate GNDs isolated or connect them together with a single connection.

HTH
Regards

[Mick14]

Hey Tom,

Thanks for the quick reply. My intention with Relay A and B was to have a hardwired shut off for spindle and axis movement when the E Stop is triggered. I think to achieve this I would be best to use a DPDT E Stop and run the relays on a 24V line seperately to the K Step input. I have not yet bought the relays, so I can be flexible with voltage/current they are, but I agree I am unlikely to find any with a low enough voltage. Do Solid State Relays still need the diode? I read that they didn't, but am happy to be corrected. Do you think that coil relays are better in this application?

I will not bother with the PNP switches, I will just get some NPN switches instead, and chalk the cost up to learning not to get to far ahead of myself.

I was considering the method you mention for Limit switches, but note that a lot of folks on here recommend wiring limits in series to ensure all axes movement is cut if one is triggered as a safety mechanism. I am happy to be corrected if you think it doesn't make a significant difference. I think this would mean that I could use the limits as home switches also until I get new switches?

Thanks for the advice on the Ground. If I use separate GNDs and them correct them together this would still be a star ground correct, or am I getting confused?

Cheers for your help. Once I get a chance after work I will fix up my circuit diagram and repost.

Cheers,

Mick.

[TomKerekes]

Hi Mick,

Solid State Relays should work well and have no inductive coil so no reverse diode is required.

In general you can't usually put devices (loads) that you want to drive in series as the current will be limited by the smallest load and the voltage will be distributed in an unpredictable manner.

Regarding Limits and EStop: It is up to you to determine the approach to take and the level of safety that you need. I'm no expert but there are basically two approaches. They can be implemented directly in hardware or software (or both). A problem with software is that any bug, or incorrect configuration, or no configuration, or disabling can leave things unprotected. Whereas if hard wired it is more difficult to make a mistake and be left unprotected. So putting all the limits in series is an easy hardware technique to kill everything if any limit is interrupted.

You can use Limit Inputs as Homing Inputs but that requires that limits be disabled while homing.

Regarding Grounding:It is a very complex subject. But if you have multiple separate "star" grounds and connect them together with a zero length wire then you end up with single star configuration. It provides the option to keep them isolated or not tightly couple the grounds if you wish.

HTH
Regards

[Mick14]

Hey Tom,

Thanks for the advice.

I certainly don't trust my programmings skills, so I have decided to go the hardware route for E-Stop and limits initially.

I have also ordered new home switches (NC NPN) so hopefully I have the wiring correct this time!

The attached wiring diagram 'should' meet all the points you have mentioned so far, but I am sure I have missed something else. I need a DPDT relay to shut off the spindle and signal the K-step when the EStop is hit, so I have added in diodes. I should have fixed all the basic errors I made, and am keen to buy the last little bits and pieces so I can make some progress with the electronics. I have also added a limit override switch so that the machine can be backed off once a limit is hit.

I have combined the limits with the EStop circuit to ensure that shut off is default. I will start with this conservative option, and decide if I want to change it down the road - it should be easy enough to do.

Cheers,

Mick.

[TomKerekes]

Hi Mick,

Very good. The biggest mistake I see is that the Reverse Diodes across the relay coils are oriented in the forward direction rather than reverse which would short out the supply and be bad. The idea of the reverse diodes is to provide a path for the coil current to continue to flow after the supply has been disconnected. Current flowing in a big coil is similar to the momentum of a freight train when in motion. Stopping it instantly results in a huge negative force (spark). The diode provides a path for the train to coast to a stop from its own friction (coil resistance).

Functionally I thing everything is correct.

Usually the return to a supply is called "+24V Return" or "+24V GND". A label of -24V is usually used for a signal -24V below 0V

The red wire going to the +24V supply is shifted slightly.

Actually I think you would be better off with no earth GND connections on all 3 of these supplies. I can see advantages and disadvantages either way. An extreme example to support my thinking would be: Assume you have 50ft of limit switch wiring snaking all over the machine and wrapped around a 10HP spindle picking up all manner of noise. This wiring is connected to the +24V supply. The noise will probably cause no problem with driving your relays or the 24V optos. But the electrical connections to Earth GND may couple the noise into the +5V supply and cause problems with the +5V logic signals.

HTH
Regards