[Lathe] Fanuc OT to DC_CNC conversion

JimDawson

Global Moderator
Staff member
H-M Lifetime Diamond Member
Joined
Feb 8, 2014
Messages
11,138
As many of you know we picked up a Hardinge Conquest 42 CNC lathe. It has a Fanuc OT controller, about a 1988 vintage.

1517946281413.png

It lacks a lot of capabilities that we would like, has some odd quirks, and the biggest problem is the lack of memory for loading G code programs from modern CAM software. Also we are not able to work on CAD/CAM on the machine computer. So the logical answer is to just simply upgrade the computer system to something more modern. So we will be installing a Core i7 computer, a Galil motion controller, a 27 inch monitor, and my CNC software. If one were to buy all of the hardware new, the total cost would be ~$7000 for the upgrade. The good news is that I have the parts on the shelf that I normally paid very little for so the total out-of-pocket will be <$1000

The machine is a 3 axis, X, Z, and Spindle, plus the live tooling drive and the 10 position tool turret. So that means a total of 5 axes to control. I was surprised to find the the turret is also servo controlled, I thought it was hydraulic rotated. The hydraulics only operates the locking mechanism.

With minimal documentation this is going to be a bit of a challenge, but doable. The first order of business is to identify all of the I/O points that are needed to run the machine and create an I/O map to be able to determine the requirements of the I/O.

We expect that we'll be making chips again in about a week.

So here are the before pictures.

Computer
1517946499855.png


Cable connector block
1517946546923.png

A little wider view
1517946586383.png

Existing operator panel
1517946634619.png

The main electrical cabinet
1517946700577.png

And the I/O blocks on the door
1517946755369.png

So the challenge is to figure out what each I/O point is connected to and decide what is actually needed and what needs to connect back to the new computer. Most of the controlled devices (contactors, valves, etc.) are 100VAC (yes, 100VAC) and are switched by the relays. The other board is the inputs from limit switches and sensors that are 24VDC.

Gonna be a fun project :grin: .....Stay tuned.......
.
.
 
yeah! and within a week? you're a machine!
 
Spent the day removing the original computer and a whole bunch of cables and configuring the new computer.

Here are a few pictures of the progress

The terminal unit has been removed
1518058284404.png

Operator panel, and computer removed
1518058408813.png

An empty computer cabinet. New board mounting plate installed on the door. Needed parts ordered today, they should be here Friday if everything goes well.
1518058562757.png

Had to do a little surgery on the new computer case, was just a bit big to fit the way we wanted.
1518058650503.png

And the new computer. It was still updating last time I looked.:rolleyes: I'll get the needed software, Galil cards, and drivers installed tonight. Galil cards, plural because I only had 4 and 1 axis cards on hand, so to drive 5 axes, I need to install 2 Galil cards. I've done this before so I know it works. I only need coordinated motion for 3 axes, so I can get away with a orphan card.

1518059009019.png

Still have a few cables hanging out of the electrical cabinet, I'll remove the unneeded ones when I get it all figured out.

1518059750320.png

As it turns out, it seems that the motor encoders do not connect to the servo drives, they went straight to the computer. This is a good thing, that means they will connect directly to the Galil card which is my prefered method. What I can't exactly figure out is how the servos are commutated without any feedback, must be using some type of sensorless vector control. The input the Fanuc drives is just a normal +/- 10V analog command signal.

The new software is coming along, I'll post a screenshot in a few days.

More later........ :)
.
.
 
I'm anxious to see this up and running. Should be a nice improvement.
 
Are you able to re use the Fanuc servo drives???

I have a Vectrax CNC mill with a Fanuc 0M control that is flakey at best. If I don't have to buy new servo drives, I'd drop in a new control in a heart beat. If you do not remember, I also use Galil cards. The only way to fly if you want a great machine.
 
Are you able to re use the Fanuc servo drives???

I have a Vectrax CNC mill with a Fanuc 0M control that is flakey at best. If I don't have to buy new servo drives, I'd drop in a new control in a heart beat. If you do not remember, I also use Galil cards. The only way to fly if you want a great machine.

That is a really good question. As far as I can tell, the drives take a standard +/- 10V analog input, and the encoders connect to the motion controller. The drives themselves do not seem to require any encoder feedback. The encoders are standard RS422 5V output with index pulse, so connecting to the Galil should be straight forward. Then it's just a matter of tuning......I hope.

The exception to this seems to be the spindle drive which has a dual encoders, one of which is external and is connected back to the drive for rotational orientation. In this case my plan is to use the existing external spindle encoder for the feedback to the Galil. One odd thing is that the spindle drive seems to be controlled by a 0-10V signal and the direction is controlled by an input, but according to the documentation that I can find, the drive will accept a +/- 10V command signal.

I should have a good handle on it by Monday because my plan it to have this running by Wednesday. :cautious:
 
Great, you are on the bleeding edge here. Yep, i can get mine running by Wed. also, just a Wed. in 2019. :)

I will follow extremely close here. When you get there, i got a great Galil JOG routine. Also a beautiful two line G76 written in Camosft but fairly easy to replace Camosft with native Galil.
 
AMAZING stuff.... The Kid and I are eyeballing this one close too. We keep looking to get a CNC lathe... And I do find deals listed as 'does not run- bad-controller'. This would be perfect for us. Thanks for the awesome work you do! Wednesday huh. A few by Thomas Edison came to mind and I think many of us know this already!

Genius is one percent inspiration and ninety-nine percent perspiration.

Many of life's failures are people who did not realize how close they were to success when they gave up.
 
Well things are not going according to plan :mad:

It turns out that 2 of the servo drives require a PWM sine wave input for the speed and position control. This is a Fanuc proprietary thing and the drive MUST be connected to a Fanuc controller for this command signal to be generated and I haven't been able to find a work around. :rolleyes:

So....we have decided to rip out the Fanuc drives and motors and repower the whole machine. Right now, it looks like DMM servos are the top candidate. http://www.dmm-tech.com These will work for the axis and turret drives.

The spindle drive is a whole 'nuther problem, 7.5 KW, that's an expensive servo. I think first I'm going to hang a sensorless vector VFD w/encoder feedback on the spindle motor and see if it will run it. I submitted a tech support request to Automation Direct asking their opinion about using one of their GS3 VFDs on a servo motor, I haven't heard back yet. This would be stupidly simple if we didn't need spindle indexing, but from what I saw when the lathe was running the spindle couldn't figure out where it was anyway.

I've done a little research on running Permanent Magnet AC motors on VFDs and it seems to be somewhat common. Both Baldor and Marathon make standard frame PM 3 phase motors for applications requiring higher efficiency. As it stands right now I am not 100% sure the spindle motor is a PM motor. I'll figure that out for sure in the morning. If it's not, then I know a VFD will run it.

I haven't even thought about the live tooling drive motor, but that's just a 3 HP motor. No problem there, and worst case, it's drive takes a standard analog input for the motor command.

Anyway, it's obvious that it's not running yet and I've missed my target date by a mile. :confused:

More later....
 
Back
Top