Vectrac CNC Knee mill refit from Fanuc 0M to Camsoft

Karl_T

H-M Supporter - Sustaining Member
H-M Platinum Supporter ($50)
Joined
Oct 14, 2014
Messages
1,306
I've had a really nice Vectrac CNC knee mill with a Fanuc 0M control for quite some time. Great machine, but the old control is kludgy and impossible to maintain. Time to upgrade it.

This will be my fifth machine refit to Camsoft controls. My goal is to make this one the best yet. The real strength of the Camsoft control is open sourcing of hardware and the ability to fully configure virtually an unlimited amount of I/O. (I have heard it said the biggest strength of Camsoft is everything is configurable. And the biggest weakness is everything is configurable) Camsoft is NOT a control that a person without a lot of machine control refit experience should attempt.

The plan will be to first get the machine running with a default control setup, then add a great many upgrades one at a time. The first and most important one will be a complete operator panel like seen on professional machines. This is the biggest difference between a hobby grade control and a pro control. Anther will be encoder feedback on the spindle to allow features like rigid tapping. A fourth axis will be added. This is a knee mill, a feature will be added to use the knee to do all the tool length offsets. This concept has already been accomplished on my Excello knee mill and really eliminates the big weakness of a knee mill - too small a Z travel. Got a few other things planned.


OK, here's before pics




287046287047287048
 

Boswell

Hobby Machinist since 2010
H-M Platinum Supporter ($50)
Joined
Feb 27, 2014
Messages
870
looking forward to your rebuild
 

Karl_T

H-M Supporter - Sustaining Member
H-M Platinum Supporter ($50)
Joined
Oct 14, 2014
Messages
1,306
Day one was rip all the old control out. had it down to an empty cabinet with only sixteen bundles of connectors. See pic.

I then spent time figuring out what they all are. See the second pic for results.

I had marked five cable bunches with letter labels for function as I tore things down. THEN I lost the notes on these.
so I got more wires to trace someday.

I did find all the limit and home switches, Learned something, Fanuc combines forward and reverse limits into one input. camsoft splits them out. Going to be a problem on the Z axis as On one wire pair was run up to the head. So, I got to pull a new wire pair here.. See pic of Z axis

Found the wires for the spindle - it has five wires plus ground. That's two more than need for a three phase motor. I will have to take the motor cover off to find what's going on here. Also got knee and coolant pump motor wires IDed.

I do have one question. there is a pair of wires that is the spindle brake. How do I determine what voltage is needed to energize this? something I should have checked before having the whole Fanuc control out of there.

Anyway, found everything needed to run the new control, the other things can be done later.


287115287116287117
 

Karl_T

H-M Supporter - Sustaining Member
H-M Platinum Supporter ($50)
Joined
Oct 14, 2014
Messages
1,306
WOW, on rare occasion I am truly amazed. This is one of those times.

The Camsoft vendor contacted me by email late the same day I started this thread asking for details on what all is planned for this refit. I was happy to supply this, glad they were interested.

Today, I got an EXTREMELY COMPLETE outline of how to do the installation. It even included details of how to wire the I/O to work with a default (.cbk) file to run the machine. this was no small task for them. TOTALLY unexpected and much appreciated on my part.

I only bought this software back in 2002, guess its still warranty work


pic of all the files attached.

information packet.JPG
 

Karl_T

H-M Supporter - Sustaining Member
H-M Platinum Supporter ($50)
Joined
Oct 14, 2014
Messages
1,306
OK, got to work on the power wiring. Installed a fuse bank, motor contactors, and three VFDs - spindle, power knee, and flood coolant.


I took the connector cover off the Spindle motor and do not know how it should be connected. Its is a two speed 220 volts motor with six wires. Here is the data from the name plate and wiring terminals

JIN SHIN Motor
3phase induction

Type AEVF
Poles 2/4
5 HP
3.7 KW
Volts 220
Hertz 60
Amps 6/5.6
RPM 3400
RPM 1700
Design CNS - C4088
Bearings 6205 6206
SN# 501306


6 terminals on motor and wire color

Z6-black X4-red Y5-white

??-brown V2-blue W? - orange

(note terminals VERY hard to read, some guessing here)


I would like to connect this motor at low speed to the VFD. How should it be done?


287474287475287476287477
 

Karl_T

H-M Supporter - Sustaining Member
H-M Platinum Supporter ($50)
Joined
Oct 14, 2014
Messages
1,306
No pics today, forgot the camera :(

Got a lot of power wiring completed, traced out the rest of the I/O from the old control. wired up all the 110 and 220 devices to 24 V DC relays, put in several extra for future upgrades. Got the computer and opto 22 boards mounted in the door. Got a real challenge putting two pound of stuff in a one pound cabinet so I am using the door, both sides and the bottom of the cabinet. It does make the wiring harder and its a challenge to mount components so they don't hit the computer with the door closed.


Set up an E-stop circuit and powered the contactors. THEY BUZZ SO BAD! I can't work around that. I seem to remember reading about a simple mod to stop buzzing in 110 VAC contactor coils. Anybody point it out to me?
 

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
8,080
Pics or it didn't happen :grin: I have used the door and the cabinet sides when needed. It seems there is never enough panel real estate.

Normally if you clean them so the armature firmly seats on the iron core they will quit buzzing.
 

Karl_T

H-M Supporter - Sustaining Member
H-M Platinum Supporter ($50)
Joined
Oct 14, 2014
Messages
1,306
Ok here's pics of the current state of things.

2 steps forward 1 back. Thanks to Jim, I researched rebuilding contactors. Took them all out and went through every one. Then put them all back in. At the end of the day its in the same spot with one IMPORTANT difference. Can hardly tell if the contactors are energized. I am also now a contactor refurbish expert.

I did have time to check out the spindle. As Jim suspected, the first three wires run the motor slow - 4 pole. the second three run it fast - 2 pole. I will leave it on 4 pole for more low end torque. Can get the same max speed by just running the VFD at 120 Hz. In fact, experience has taught me 10% overspeed is not a problem so I'll set max Hz to 135. In operation there will be no need to ever adjust the variable speed belt. On rare occasion when doing something like drilling with large bits, the machine can be switched to back gear. With this concept S codes can be used to set spindle RPM

Also checked out the knee - The VFD runs it up/down just fine. Did not get to the coolant pump. First on the list tomorrow. In fact got to spend a bit of time running it. the sump has a nice layer of dirty oil. The wires for the Servos lay too near this to avoid dragging them through it during installation. I'm going to find a way to jury rig the belt skimmer from the CNC lathe to get all the oil out. After its cleaned as much as possible, put a whole jug dawn or similar in there with a bunch of water. FUN job. NOT


287850287851287852287853
 

Karl_T

H-M Supporter - Sustaining Member
H-M Platinum Supporter ($50)
Joined
Oct 14, 2014
Messages
1,306
Got all the devices installed so far checked out and running. Not much to show, but a fair bit of labor. learned I still have the capacity for stupid mistakes. Could NOT get the optos to fire the coils - they would latch on after first energize. I looked for all sorts of complex problems. gave up and went to bed. In the morning I noticed I had installed AC optos and I now know they will not work with DC 24 volt. DOH!

Got computer software installed and running. used it to checkout the 12 outputs installed so far. See the pic. The control has a dignostic screen with all the I/O. I labeled each output. I had to fake in a mouse but when its hovered over an output the label pops up. Click on the out and it will energize or turn off - toggle. this screen is HUGE for maintenance and debug.

I build my own controls with a major emphasis on future easy maintenance. Wire # on all wires, each device labeled. Instead of keeping an electrical print, I find a spreadsheet of wires and function to be most helpful. See the pic. For each wire, the connection on each end is noted along with its function. I just write it down as it is installed. I just brought up the spread sheet from the excello install in 2004 and could see immediately how something was wired.

288052288053288054
 

Karl_T

H-M Supporter - Sustaining Member
H-M Platinum Supporter ($50)
Joined
Oct 14, 2014
Messages
1,306
This is a direct question to Jim Dawson. I'll put it here for future records.

I have little confidence that I'm wiring the SSR for amp enable correctly. So I quick dummied one connection up. See the pic.

Red wire ampenX pin 40 to +3A1 on Cydrom
green ground to 4/A2 on Cydrom

green Pin1 DYN4 to +1/L1 on Cydrom
white Pin15 DYN4 to 2/T1 on Cydrom

Amo enable relay.jpg

DDM servo amp to Gall   Dawson controls .jpg
 

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
8,080
Hi Karl,

Looks like you have the wiring on the ICM-1900 side correct. Confirm this by issuing a SHX command and see if the LED on the Crydom comes on. If the logic is inverted, then exchange the 7407 chip for a 7406 to invert the logic. Normally these chips are shipped with the the ICM-1900 so you should have some kicking around.

On the DYN4 side it's a little more complicated.

If you are using the internal 14V supply, you need to connect pin 4 to pin 17. Your connections to the Crydom are correct
 

Karl_T

H-M Supporter - Sustaining Member
H-M Platinum Supporter ($50)
Joined
Oct 14, 2014
Messages
1,306
WooHoo! Life in the new servos. Guess you cannot see motion but it is running in front of the computer in the pic.
Don't think I've mentioned DMM 1.8 KW servos and DYN4 amps were selected for all three axis. http://www.dmm-tech.com/
Quite a bit of wiring was done in the bottom of the cabinet for the Galil 1900 connect module to the DMM servos.
Next, I'll test if Galil commands can at least move the motors. After this works, the servos can be mounted permanently.

The amp enable circuit is not working - not sure of the problem. Looks like I can proceed anyway as the amps can be set to always be enabled. Will have to get back to this in the future.

288718288719
 

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
8,080
That's a lot of motor. Not going to have to worry about running out of power. :grin:

I couldn't get the AEN to work the way I wanted it to either. I just wired my enable to an output.
 

Karl_T

H-M Supporter - Sustaining Member
H-M Platinum Supporter ($50)
Joined
Oct 14, 2014
Messages
1,306
That's a good suggestion. First, I got to find where the heck I put the logic probe. One more search tomorrow, then buy another if it don't show. Then it will be sure to turn up . Bet I'm the only fella on H-M that misplaces his tools :)
 

Karl_T

H-M Supporter - Sustaining Member
H-M Platinum Supporter ($50)
Joined
Oct 14, 2014
Messages
1,306
I haven't kept up this thread. Sorry, life gets in the way sometimes.

I had an ABSOLUTELY AWFUL time getting the servos installed and running correctly. Would have given up on the whole project without the assisstance of Jim Dawson. I will bet he is tired of my constant requests.
The issue was run aways. NOT good with big powerful servos. Had more than one problem so fixing one thing didn't help. I'd put it back then try the other, so that trial did not work either. Just for a quick summary:

All the DMM servos want to run what I call backward. When finished, every axis has either reversed encoder or reversed axis direction in the Galil software setup.

The used Galil ICM was defective on the axis I selected to try first. had to replace it. Shopping for parts on eBay does have its disadvantages.
I installed an update to the Galil card and could not get it to respond after the update, so i went to a very old card. Never did get this one to run. luckily, another one came up on eBay. ( I sent the first card to Jim and he could find nothing wrong - then it worked for me when returned) The machine is still running with card number three.

There is a parameter titled max speed. I had set this to a low value for testing. The title of this is extremely poor, it is actually the center of the voltage dead band for servo drift. A value of 65 is for 0 volts.

Biggest issue was with my misunderstanding of how Galil enable/disable should work. I thought is was a 5 volt output. it is acutally an open collector. Wasted a huge amount of time here. I promise to never make this particular mistake again. For DMM servos, connect the sevo enable input though a 2K resistor to the Galil enable output if using external 24 volt to power the servo's enable input. Then remove the RP1 resistor pack on the ICM. this results in backward function. So exchange the 7406 IC for a 7407.

With all this finally resolved, the servos were tuned first under DMM. then again with Galil WSDK.
All the limit switches and home swithes were connected and debugged. An input from the Estop was connected to the galil card. outputs to the spindle knee and coolant were connected and debugged.
Camsoft was kind enough to provide a basic .cbk for this machine. A couple days making minor program edits to this .cbk has the machine up and running WOOHOO!

The machine could certainly be run as is. A great many stop at this point and go to work. But that's not the final goal for this machine. When complete the control will be better than even the newest commercial controls. I can say that because it will work exactly as I want it to.
Screen shots of the DMM servo setups are attached, mostly as a personal backup.
 

Attachments

Last edited:

Karl_T

H-M Supporter - Sustaining Member
H-M Platinum Supporter ($50)
Joined
Oct 14, 2014
Messages
1,306
I'm working on my control panel button layout and have a question...

First pics of what the control panel looks like on my lathe and mill.CHNC lathe panel.jpgmill panel.jpg

Here's a preliminary layout for the Vectrax

panel layout preliminary.jpg

I took a piece of scrap and populated with the parts I have. Help to be sure its right before cutting on the enclosure.

new mill preliminary panel.jpg

Now for my question. I would like to order a printed sign with the text for each button out of PVC or another durable plastic. Bond it to the enclose and then drill all the holes for each component. Who could do this sign and what exactly what should I ask for? I see sign companies for things like political adds, etc. but found no one that would take a .dxf file for exact placement of the letters.
 

Karl_T

H-M Supporter - Sustaining Member
H-M Platinum Supporter ($50)
Joined
Oct 14, 2014
Messages
1,306
Thanks, Jim. Dunno what I'd do without you on this project. I'll keep that engraving plastic in mind for the next job. This particular job is getting a bit over schedule.
 

rgray

Active User
H-M Supporter - Gold Member ($25)
Joined
Nov 26, 2012
Messages
1,228
Why not engrave it with the cnc?
Could paint or powder coat first to make the lettering stand out better.
 

Karl_T

H-M Supporter - Sustaining Member
H-M Platinum Supporter ($50)
Joined
Oct 14, 2014
Messages
1,306
Yep, once the vectrax runs, I can engrave. Got to get it going first. The old excello just breaks small endmills and engraving bits.

I suppose if the quote comes back out a this world I will re think it.
 

Karl_T

H-M Supporter - Sustaining Member
H-M Platinum Supporter ($50)
Joined
Oct 14, 2014
Messages
1,306
Spent the last several days mounting an encoder to the spindle. had to get real clever here. Only got about 1/4" of the top of the hollow spindle exposed and it can't get in the way of the tool changer. The pics show the parts and the final completed project.

A spindle encoder really increases capability. It allows the spindle to be treated as a galil axis to control the spindle VFD. S codes for programming spindle speed are now a piece of cake. Rigid tapping can now be done. if a 4rth axis is added, it can be slaved to the spindle to hob gears. By comparing analog command voltage (Galil TT) to spindle speed, it is possible to determine where the adjustable speed pulleys are set.spindle encoder tool changer  parts.jpgSpindle encoder pulley.jpgspindle encoder complete.jpg
 

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
8,080
That is very creative, I may copy that one.
 

Karl_T

H-M Supporter - Sustaining Member
H-M Platinum Supporter ($50)
Joined
Oct 14, 2014
Messages
1,306
I see I neglected to explain the concept here. The bushing at far right in the parts pic just slides inside the hollow spindle. I drilled tapped for two #6-32 screws on the end of the spindle. two of the four screws in the bushing fit here. The other two are tapped and function as jack screws for disassembly.

The red bottle in the pic is bearing retaining compound. the lower part of the bushing is coated with this and slid into the top of the spindle. two screws hold it in place. If it ever must be removed it will need to be heated to break the bond. I put a note to this effect on the tool changer. doubt I ever tear it down. The note is for some poor soul in the future.

The pulley in the pic fits on the top of this bushing. The two tapped holes also function to hold the pulley on. I did not use retaining compound here.
 

Karl_T

H-M Supporter - Sustaining Member
H-M Platinum Supporter ($50)
Joined
Oct 14, 2014
Messages
1,306
current work has been the machine operator panel. IMHO, the biggest difference between a profession quality and a hobby quality refit is a great control panel. A lot of thought went into this one to make it exactly as I want. I like to be able to run the machine from the control panel but not have it look like a 747 cockpit airplane control. Some things are better done on the PC keyboard and function hot keys.

This one needs fifty conductors for all the I/O, see the pic of the inside got pretty crowded in there. I also placed it high and ahead of the spindle to keep flying swarf away. It lined up so that I needed to extend the backgear lever. this made it easy to add an input for backgear with a limit switch, see pic.

control panel.jpgcontrol panel inside.jpgback gear input switch.jpg
 

matthewsx

H-M Supporter - Silver Member
H-M Supporter - Silver Member ($10)
Joined
Jan 2, 2019
Messages
832
I feel like I am standing at the base of a very tall and steep mountain, looking up at your project....
 

Karl_T

H-M Supporter - Sustaining Member
H-M Platinum Supporter ($50)
Joined
Oct 14, 2014
Messages
1,306
Thanks for the compliment. This is my fifth Camsoft install. I did a couple AHHA (DOS stepper control) and a Mach 2 before that. Things do get better with practice.
 

Karl_T

H-M Supporter - Sustaining Member
H-M Platinum Supporter ($50)
Joined
Oct 14, 2014
Messages
1,306
ARGH - a step backward.

I lost the eight inputs on the second Galil card. Had two working, one was the estop input. Now all eight are dead. I double checked by booting up Galil WSDK. the first eight on 1900 ICM one are fine, second eight on 1900 ICM are all dead. FWIW, have +24 for signal, DC common connected to INCOM.

I double checked manual, see attached pic, they are all opto isolated so 24 volt should be no problem.

<Jim> See anything wrong? What component is most likely zorched?

karl
Capture.PNG
 

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
8,080
First I would swap the cables on ICM 1900-1 and ICM 1900-2 to see if it's the card or the ICM. I have also had this problem when the 100 pin cables are not well seated.
 
Top