My Huge (new to me) CNC Mill

A little update:

The USB floppy drive worked and I got the software to load properly. After I repaired the machine door so the limit switches would work I was able to do some basic operations. The machine homes and jogs properly, and I managed to release the tool from the spindle.

But..... I have to say I'm rapidly getting frustrated with the Fanuc emulator software. I really want to use the machine, not try to learn Fanuc software. So far I have not been able to get it to load a tool, it seems that there is no button that allows manual tool changes, so I assume it has to be done programmatically using a M6 tool change routine. No reference in the book for manual tool changes.

I'll play with it a bit more before I rip out the controls and install my own system. :mad:
 
Karl_T said:
After designing and building your own control, old fanucs just don't get it.
Click to expand...

Ya got that right. I'm used to doing everything with a mouse click on well labeled buttons, not having to deal with something that was designed to run on an 8086 processor, pre DOS, and deal with cryptic graphics on push buttons. My hat is off to the original programers, I'm impressed that they were able to run a machine at all. But there is no reason I should have to take a giant step backwards into the computer stone age and learn 30 year old software just to run this cute little machine. :rolleyes: I'm too old for this ****.

The machine is on a pallet sitting in the corner for now until I get time to play with it again in a couple of weeks. I have some paying work to get out the door, gotta support my tool habit. :grin:
 
Last edited:
A little update.

I could use the machine as it is, but I don't want to expend the energy to learn obsolete Fanuc software, I upgraded my lathe just because of this. 30 year old Fanuc software in that case.

So here is what I want to do: Since everything is in working order, I want to use the existing ACC and write my own software to replace the EMCO WinNc software. I can write the CNC software, no problem there, what I don't have is any clue as to what the command strings and structure looks like across the ethernet connection.

So before I sit down and write a port sniffer, I thought I would ask if anyone has the command documentation.

Or do I just rip out the ACC and install a new controller and my software?
 
Jim, your talents never cease to amaze me. Writing a port sniffer and reverse engineering a command protocol is not a trivial task. To answer your question though. Why rewrite or port your software to the existing system. Replacing the controller so you can use your software drive standardization across your shop which can only be a good thing. Regardless of what path you choose, I look forward to reading about it here.
 
Boswell said:
Jim, your talents never cease to amaze me. Writing a port sniffer and reverse engineering a command protocol is not a trivial task. To answer your question though. Why rewrite or port your software to the existing system. Replacing the controller so you can use your software drive standardization across your shop which can only be a good thing. Regardless of what path you choose, I look forward to reading about it here.
Click to expand...

Thank you for the kind words. :)

I think you're right. Installing my own system would be the path of least resistance. Pretty much plug & play, and no learning curve. I'm going to open up the panels today and see what the electrics look like.
 
Karl_T said:
Especially if you care about upgrades and ease of maintenance, rip 'er out.
Click to expand...

I started the process this afternoon. Opened up the electrical cabinet and looked at the wiring and compared to the electrical documentation. I actually have the original electrical documentation. The documentation is very good and is accurate per the machine. This machine is a perfect candidate for a DMC-4060, but am thinking about using a Dynomotion Kflop board. Haven't decided yet.

The panel is well laid out and is wired to European standards, typical ISO 9000.
1574474109927.png

Closeup of the original computer box. This has an actual off-the-shelf micro PC in it, but interestly, no hard drive or memory. Has a proprietary PCI board that appariently has the firmware and memory onboard. That PCI board is ported out to the connectnections you see here.

1574474194249.png

X, Y, Z stepper drive and spindle VFD. The stepper drives are 3 phase output. Haven't seen that before. But they take a standard step & direction signal so no reason to change them.
1574474326328.png

I'm thinking about doing something like this. I grabbed this example picture from Ebay, this DMC-4080 has a proprietary interface board attached that just plugs into the standard Galil D-sub connectors. It would be pretty easy to design a similar interface board to be compatible with connections on the computer box above. I may just buy this DMC-4080, still thinking about it.

The reason for building a compatible interface board is to not disturb the original wiring and thus allow easy return to the original computer system in case a future buyer wants the stock system, that way the value of the machine is maintained.
1574475555295.png

Then I need to decide what computer to use. I'm thinking a micro PC so it will fit in the existing enclosure. It is deep enough that I can stack a micro PC and a Galil or a Kflop and put it all in the same footprint as the original unit. Then just bring out the monitor and USB connections and hang a keyboard/monitor mount on the side of the machine. Would be a nice compact unit. I just need to find a micro PC that will run Fusion 360 so I can have that on the machine computer.

So at the moment the real decision is Dynomotion Kflop or Galil DMC-4080. My software will work with either and both are good equipment, but there would be a learning curve with the Kflop, where the Galil is almost plug & play for me. The Kflop is a lot less money, $249, vs. a used Galil >$1000. Is the difference worth the couple hundred hours of learning/programming time?

One other option is to use hardware I already have on the shelf and use Galil PCI cards. I have several 4 axis cards around and could use 2 of them to run everything. The down side of that is the wiring is going to get messy, and they require a full tower PC to stuff them into. So I lose the nice compact system.

Decisions decisions. :faint:

I may worry about this project after I finish up the the C axis upgrade on my lathe, just finishing up the software for that one, but still need to buy a servo motor and build the clutch hardware. Then there is the CNC surface grinder project that I have been putting off for the last 2 years. I have all of the parts on the shelf for that one also.
 
JimDawson said:
I just need to find a micro PC that will run Fusion 360 so I can have that on the machine computer.
Click to expand...

I'm using a Zotac Zbox micro PC on my mill and it runs F360 just fine. I've installed 2 others on plasmas and they also run just fine. They're not the cheapest you can find, but no fussing around and they also offer fanless version to keep the debris out if that's important for your installation.

It runs F360, UCCNC, and internet youtube all at the same time while machining - not habitually, of course, but just for 'testing' purposes. ;)

The onboard HD600 graphics work just fine, and I've got an i7 w/8G of ram. F360 doesn't lag except when I have a very, very complicated model with tons of inter-related sketches. And that could be fixed with better drawing & parameter management than I'm capable of.

Iv'e found that if you want proper fast F360, then you'll need to step up to a dedicated graphics card computer. And those seem to be double the price of any onboard graphics.

Following your work on this as I'm still in the spit-balling phase about the future lathe project. I narrowed it down to Kflop, Linux, and your software on top of Galil hardware - thanks again for the help playing with it!

-R
 
You must log in or register to reply here.
Back
Top