Converting a Brother TC215 to LinuxCNC

Hello everyone,

I’ve been lucky enough to be the recipient of a Brother TC215 midway through a LinuxCNC conversion. I’ve always wanted to try my hand at CNC work, as I’ve done a little bit of manual milling and a lot of fabrication work over the years, and I’m always curious to learn and build new things. It’s been a pretty steep learning curve, as I’m new to basically all of this, but I’ve leaned a lot via search and previous forum threads and I think I’m at the point where starting a thread is worthwhile. 

On the hardware side: 

• Mill is a Brother TC215
• Dyn4 AC Servo Drivers/motors
• Mesa 7i92 / 7i76 / 7i77D control boards
• HP Elite 8300 USDT running LinuxCNC / ProbeBasic

I'm currently planning on running some of the older Brother boards for things like coolant, safety switches, limit switches, etc, just to keep things simple for now, as I already have breakout boards for all the connectors. I can clean everything up later if it gets that far.  My friend validated that the mill’s mechanical functions are in good shape and made it about halfway through the wiring - but my current priority is to wire everything up such that I can get XYZ servo movement first, and then I will add additional functionality. So I’ll try and keep this thread focused on the help I need to get things moving under their own power. 

What I’ve got so far is: The hardware is mounted in the mill, I roughly understand how things are laid out. I feel like i have a broad sense of how this should go together, but the details are oftentimes a little more opaque than I’d like.  What I’ll be doing when I have a minute :

1. Test all my servo controllers and drives (easy to do, just need to get my windows laptop out to the mill)
2. Power up the Mesa board (pretty easy, just need to spend some time looking at all the wiring and make sure I don’t fry anything)

The thing that’s blocking me right now is that I don’t really know enough to confirm that I’m wiring the servo drivers in a reasonable way. My understanding is: I should be able to wire my X / Y / Z servos in the following way, to do analog +10v control as a simple starting point. I need to set it up so I’m wiring: 

• 7i77D +24v from field power to Dyn4 JP4-15 (Servo Enable) 
• 7i77D TB5 (Analog Drive Interface) To Dyn4 JP4-13 (Analog +10v) & JP4-25 (Analog GND)
• 7i77D TB3 (Encoder 0-2) to Dyn4 JP5-2 through 8 (Encoder outputs)

My plan was to disconnect all the motors from the mill and test them on the bench, but running through LinuxCNC, so I can confirm my understanding on how LinuxCNC operates. Once the motors are confirmed to work on the bench, then I would start on wiring up the Estop, limit switches, and other things I would need to get the mill safely moving, and then follow that with wiring in the spindle.


Here's a picture of how things it sits:


Any and all feedback appreciated, as well as suggestions about if I should approach this another way. I've learned a lot here and I hope that documenting this in the long run helps someone else

I had no idea Brother made CNC machines!

I thought they just made printers.

forum.linuxcnc.org/30-cnc-machines/43456...-215-retrofit-advice

Yeah, I was amused to discover the same!

Thanks for the link - I was hoping he'd have some wiring guidance, but he never ended up getting the mill.

What I really need to proceed is to be confident that I'm not going to blow something up by wiring the Mesa board to the servo in the described fashion - it appears from reading the manual that it's fine to wire it as "Analog Circuit D", on Page 18 to the Mesa driver pins, and I'm assuming that I can then also wire up the JP5 Encoder output to the Mesa Encoder pins as a "Line Drive / Reciever" as on page 20. Once that's done, it feels like I could set up a basic LinuxCNC HAL / INI / PnCconf setup to power and move the servos.

Manual link here for ease of reference:
www.dmm-tech.com/_files/ugd/ac1ef8_924ca...fce9d.pdf?index=true

I can also call my friend who has his setup and get him to walk me through his setup and emulate that, but I like to know how the system actually works vs just emulating other people's configurations

Looking at something existing that works is a great way to learn.

A lot of the stuff in that picture is totally unconnected.

I would remove those, at least temporarily, so you can focus on what's actually in use first.

I suppose those AC servo drives probably connect to those parallel and serial port plugs; connect them up and see, perhaps it works as-is.

Though it appears 2 of the drives aren't connected to power, so fix that first.

The AC servo drives won't work because the wiring currently is only halfway complete, and I want to have the system documented in the intended state before I power anything.

There's no config loaded on to the Mesa cards yet (or a config between the Mesa cards and LinuxCNC), so after talking to my friend who kicked the project off, I've decided to take a slightly different approach, which is I'm going to set up and configure the EStop switch for LinuxCNC first, so I can verify that each of my components work. I will: Document the wiring, wire the switch, confirm the connections, verify the Mesa card functions, establish the appropriate configuration in LinuxCNC for the EStop switch, and use Halscan to verify it works. Once I've confirmed I can get a functional switch in LinuxCNC, I can start taking on something the servo wiring problems. 

That way I'll know the base of the software / hardware stack works well, and I can proceed confidently with wiring the servo drives. It also hopefully gives me a chance to do more research on if my wiring setup will work / get some feedback on my proposed plan of Analog +10v control with the encoder wiring as documented above. Once my documentation is a little more cleaned up, I'll share that too.

Edit: Oh, and I appreciate the advice to focus on what's in use, but I like to see things in the context they exist in, it's helpful for me to conceptualize the whole system - I've already roughly documented all of the existing connections and what they're used for, so it's mostly about trying to figure out how the systems fit together and deciding on the pattern that I want to wire things in to make it so it's easy for me to understand it if I have to work on it in the future.