Tree Journeyman 425 CNC milling machine control retrofit

slodat

H-M Supporter - Diamond Member
H-M Lifetime Diamond Member
Joined
Jan 20, 2021
Messages
296
Thought I’d share my my cnc milling machine build. This is copied over from another forum. I just found Hobby Machinist a few weeks ago.. Forgive me, some of the writeup may be past tense, considering the mill is up and running now.

I’ve wanted a cnc milling machine (and lathe) for a long time. After converting my router to Centroid Acorn, I knew I wanted my other machines to run their control as well. Mill was 150 miles away. Weighs 6000 pounds, CAT40 spindle taper, oiler, coolant, Baldor DC servos, 7.5 HP spindle. Turned out to be a perfect candidate for control retrofit and will be a great first cnc mill for my shop. Seller needed it out of his shop. I got it for scrap value or a little under.

I got a pdf of the original control schematics from ZPS. They own the remaining Tree CNC stuff. They have been great for a few items I needed.

Original plan was Centroid AllInOneDC, Ended up going with Centroid Acorn. Mostly due to cost, and I wanted new servos instead of messing with the old DC servos. I am limited to single phase supply into a VFD for the spindle.

ef004b791a1cdaa44c37c2dce94b82b5.jpg


9785670f15814a13dbbf60d83f1bce79.jpg


389f18f1486853e825116e9c564df4f9.jpg


241a98ec0710343d1fead215b662283a.jpg


ff30e5417d9963f39f9dc608af0c9e4c.jpg


Made it to my shop without incident.

2206cff712c50a4ca17ad199876ed442.jpg


Delta 20 control.

006b503c7ee3ddf06919664c739fb2bf.plist


The seller had a 4k forklift. It took everything it had to get the mill loaded on the trailer.

62e4c4e922c43285f76086a343da9e46.jpg


Picked from the top with both forks together. Two straps and a clevis mostly leveled the load.
97b817992e0b46547aee8a694b9205f5.jpg


Lifting it to get the steel bar cribbing out was the limit of the forklift.
91c4ebad09c46ab64e308437f1945d87.jpg


Load was in the sweet spot for the trailer.
d9546f86605f4d30f5575e46fc1fcbc1.jpg


Tied down secure and ready to roll.
4255bdaf88877084476b8d944229c2df.jpg


11f085ef853f2851cf6af2c53d7873cc.jpg


Stopped to check things out. No issues. Drove between 30 and 55 mph on two lane backroads the whole way home.

9b0774a3b0309dd69086692c0ea26b30.jpg


Made it home just as the sun set. Twelve hours round trip.
46f9252682cdc987b8b4f61635ecaf4b.jpg


Rented a 10k telehandler to unload.
ac4f7c64fc85ec7ea6651e781e1100cd.jpg


Same rigging as the day before to pick it.
78fbd6cec1312f7b42febd6a0350101d.jpg

3f36c03d8972c4ca20fde8e45dd222d0.jpg

d0cbb3251a5e50f11a6f4934d7e66105.jpg

d90616a52ee89988655193ed34e423f7.jpg


Coming up to the shop door.
c84fbd0a27cc952b86d90719367536ef.jpg


We set it back on the 3” square tubing it was on at the sellers shop and picked it up from the bottom and drove it in the shop. It’s sitting on the tubing now. I’ll lower it off the tubing onto round bar with a toe jack and roll it into place on the bar. Once it’s in place the toe jacks lift and set it on the ground.

Looks like good bones. Needs a thorough cleaning and a complete electrical rebuild. Lots of useable parts in the original control. PO said spindle VFD has failed.

c60f60a5488d083312e33051ae8d5d56.jpg


b884e203c7a2ee2093c98abb4a28ff3a.jpg


ab263f7927a3d99299c6ddb9672104c1.jpg


Spent several hours and a gallon of cleaner scrubbing the coolant and grime off the mill today. Will need another pass of cleaning. After getting a closer look, I’m still happy with the machine. Should work well for my needs.

e0ebd8b3f2741deecefa2bf8d1fb6197.jpg


523dac19ec33ef45eb8d89dba1838eb2.jpg


05013bcc9bffee9745c304c3cee45164.jpg


63f1bd0b32610be8ae2079c934f4827f.jpg


62f0bd4694160c4263b7e37ce03adcd0.jpg


Not sure what this motor is on the X axis. It’s not the same as the Baldor on the Y.

0b48c19e2a3a5b60d2093a905efef4cd.jpg


Cleaning that dirty old beast was no easy task!

Schematics in hand I set out to test the control. It powered up, all faults cleared and I was able to reference the machine.

de5b6bbee5f0d8978b628b2a8edd0f20.jpg


The Dynapath Delta 20 control worked. The spindle drive had failed. I was able to sell the original control for about twice what I paid for the mill.

I tested the spindle on an inexpensive 10hp HY VFD. It did fine for testing, but I wouldn't run this big spindle motor on one and expect it to last.

I wonder what the old drive’s life was like..

d8332a6ffefa6f1f8c414a9947f5ba5d.jpg


a2693601b05fb653dc65b5500681d869.jpg
 
Removed the Dynapath Delta 20 Control and servo drivers. The cabinets were like the rest of the machine. Absolutely filthy. A little better after some cleaning.

I originally thought I could put the new control in one cabinet. Once things started going together I was really happy to have both cabinets. Plenty of room to work and no electrical noise/interference issues.

6ccfa0cda0623548ce1fed5d4d1a52b0.jpg


f744e23c3f7cf04f6031dc2d06d827f3.jpg


First step is to do a bench test of everything to make sure all is working prior to installation. Really helps when troubleshooting later on. All went well. Next step was to blow this apart and start the build out. Since I know people love photos..

(It’s bench testing. Not final work. I used chunks of wire I keep in a box for stuff like this. I know it’s a rats nest ;)

c79cdc51550ccc2c071d3dd33ae4643a.jpg


cd3b0b6f30accbbd04f9776fe8467187.jpg


2cdcab8eb23a7346d0999f23bea9da19.jpg


Had a bit of drama with one of the motor pulleys. Found a used one at ZPS. They have been awesome for info and parts.

Next came panel layout. I really like the way it’s setup with the VFD and input power in one and control in the other. There’s tons of room, existing venting for the VFD heat sink, etc.

e014149624fd8b60ae6a599b04f18fb7.plist


149b8e41551f96f2eb5430efdcfdf3ac.plist


AC power section taking shape:
f06b7075daa4a35e6016621fdd2cd394.jpg


Progress on the controls side:

c88a95d2ea7aa62ed909addb827ace14.plist


Cut the slot for the cable pass through. This is a Centroid item. Really slick.

7aa193e2822d19325d66f397795c72f2.jpg


10ea685a2faf1077a207062ca05fda51.jpg


Starting to look like something.

aae2954b219050609bf338e12b11292e.jpg


Start of the control panel. Used a chunk of 1/4 aluminum I got from the scrap yard. Was some sort of sign. It’s turning out how I had it pictured.

I need to weld the lcd bracket standoffs to the front plate, engrave the lettering and sort out the mounting holes.

7cdfb35be1155e8c1aec9892b1a8bf58.plist


c17e7abdd19729e3b08801f69d25c76d.plist


353cbb4356adcbb88027f55ee8abcc06.plist


Engraving the letters went well. Metabo burnisher did a great job on the scrap aluminum.

It’s mounted.

593106f104e101b32858eb96ececa255.plist


e22b9fa25691dc3a40a9aa2a5c34b4fb.jpg


Used sharpie markers to color in the letters. Worked great!
02a980dffc6723fac374f4e56676e908.jpg


Powered up the control for the first time. Control panel is in and working. Lots left. Nice to see the screen up and running.

0d126078306c946a844bb045b84f6f81.plist


The Intel NUC front panel is accessible from the side away from the spindle. The opening was existing.

63d0373c307c0760785a98ae58b23cc6.plist


2e9b373729b52b2d9ed4cdc92822a3ab.plist


e433f1b2d8ed14147bc001e864dfca4c.plist


dc01d673b7ed3a84c14bfb81330041c7.plist


b0310d1795d837bd0c57643b7c9f519a.plist


Received the Hitachi SJ-P1 drive for the spindle. This is a very nice drive. And, they are rated for single phase supply to very high HP levels. I bought the P1-00800-LFUF 15kW model rated for 31a output continuous when powered from single phase. Drive comes with bracket for wall or through wall mounting which places the heat sink and fans outside the enclosure. This is how the original drive was mounted when it was new. I had to open up the hole a little.

bcf5f1c2e06c820bdc74387f0cb68af7.jpg


b96ffa02e0aeb22f2f6caf9621ec56f6.jpg


61b055e15d2ace842d5e217753f85e7c.jpg


The spindle motor has an encoder and Hitachi has an encoder interface option for the P1 drive. I think I’m going to give closed loop sensor vector a try and see what that’s like. I’ll also have a chance to run the thermistor wire to the drive when I run the encoder wire and replace the motor main leads with modern VFD cable.

I wanted the spindle fan to run for a few minutes after the spindle stops, so I ordered an off delay timer. I replicated the spindle FWD and REV outputs to drive the spindle fan contactor with an interposing relay.

The machine is getting close. I’ll get the probe receptacle and spindle load meter wired and I’ll be pretty much done with the conversion. Very happy with the machine thus far.

Short video of the spindle and drive doing their thing.


Getting close to completing the mill build. Still a few details. Got the way covers cut out of rubber sheet. Looks like it will do the job. Touch probe receptacle is in and the probe works great with the built in probing cycles.

This was a bore probing cycle on a 1.0000 Mitutoyo ring gauge.

44247177ad6275867d329ecd6702ad56.jpg


Plan is to make a keyboard tray/mount like so:
e3053bd1ce330f1dee132ba26f4a9f32.jpg


Starting to look the part. Next up is getting the table chip enclosure back on.

e7596a597883b9bbccd9950dadfd27de.jpg


For the most part, the mill is ready to use. As I type that I realize I have some tidying up to do with the air stuff and the wire trough cover.

Used it for a real job tonight. The Y axis cover needed clearance for the position of the new motor pulley. There is a knockout in the cover. Used the probe to locate the existing hole. Mind blowing how well this works! I’ll make an appropriate cover for the opening.

bdb7375019e06f9b2095b1889f8a954b.jpg


As it sits now with chip shields in place. The table is really high when close to the quill. 48” in these photos.

0f2f38bdddd3ff0cb2db5e3e74c05a81.jpg


c23fb7d972bce2954fe9ccbf7e971bec.jpg


1c97e8cc4b29ec94917b57c945a40154.jpg


fb60540a29c9b261d3bd247c3512f56e.jpg


591c5d6cafd782cc50241feeda34dc8d.jpg


Started making some chips. Looks like I need to dial in the turns/inch a bit better. I may have forgotten to check it and dial it in. Things are looking good otherwise.

fcd8603af15f2b567f0084659a0bea3a.plist


Sorted out a keyboard tray. This was a nice upgrade over it getting moved around a lot. And, cleaned the chip guards up and reinstalled. They make a big difference.

971c6276968a40e79abadffa035fb3f6.jpg


Hopefully I’ll get it dialed in a little better tomorrow and make some more chips!

Had an issue where I thought my Z axis was accumulating error. After the suggestion from the Centroid forum, I checked if I was losing steps between Acorn and DYN4. I had been through this before on my router. I incorrectly assumed this would get fixed based on my experience a year ago. Nonetheless, it was short work to modify the DMM DYN4 -> Acorn cables and switch to the Acorn step/direction signals on the DB25. Once I did that all was well. No more lost steps between Acorn and DYN4.

At that point I needed to go through the axis calibration and backlash measurement. That went well. Lastly, I dialed in my Hallmark ITTP again now that the machine was moving as the control commanded. I'm happy to say I made the part I've been using to test this all out. Part turned out good. Two sided part using two WCS. Fusion 360 for CAD and CAM. Really happy with Fusion for CAM on the mill.

3822972fb20f5de0dfe551c269b1edbc.jpg


1493a23c9f11b4b734758a1f80f9b14a.jpg


8319f0a2e09d8596b5f03abd58817e90.jpg


Some of the various pieces of scrap I made diagnosing issues and dialing it in:

6f750025a5e048f419c3be5304196dae.jpg


Installed:

478db62861f11a8bc1021067d7b4651d.jpg


The side of the mill control panel has an existing opening with a small cover. I mounted the new control pc behind this opening. This leaves the front panel of the NUC CNC PC accessible. It's on the far side from the spindle, so no coolant or chips to worry about. I sold the original Dynapath 20 stuff on ebay. All in all, I am very happy with the mill, the retrofit, etc. I highly recommend the Hitachi SJ-P1 drives!
c836c9a3bd2aa599588bc268081e8fdd.jpg


The cover:
c0b5ad6384318fab48ae5d6bc37e3329.jpg


Some measuring and drawing:
8fb8857b4aa7bb6a1a737a34ff84e902.jpg


Point and click:
f8928d7f0ca728b6f3d55eedef12536d.jpg


d401e63d4e3dfbcbc72e7bf5cd7bf50b.jpg
 
The mill has been done for a year now. It has been completely reliable and is a pleasure to use. I have kept the CNC12 software updated as the releases come out. Centroid is very active in development and listening to their customers. I'm very happy with the machine and highly recommend this sort of thing if it's what you are into.

I have a Hardinge lathe with an Accuslide CNC conversion that I'm working on a very similar conversion. I'll post a thread about it soon.

I welcome comments, questions, etc.
 
Last edited:
Very impressive work. I'm still waiting to dip my little toe into CNC with my plasma cutting table, but have to finish up the shop first. Mike
 
So, as a retrofit of an existing CNC, you didn't have to replace the ball screws, just the servos? If you don't mind my asking, how much did the conversion end up costing?
 
So, as a retrofit of an existing CNC, you didn't have to replace the ball screws, just the servos? If you don't mind my asking, how much did the conversion end up costing?
Correct. I did not replace the ball screws. I replaced almost everything electrical with a few exceptions - spindle motor, limit switches, pneumatic solenoid and pressure switch, etc. I believe I spent somewhere around $8k all told on the control retrofit. The spindle drive can run the motor full load indefinitely on 240v single phase supply. That drive was a bit more money because of the upsizing needed for the single phase supply.

Retrofitting a machine like this results in a very good machine for around the same money as the lowest end Tormach or a retrofitted benchtop machine. Let me know if you have more questions. It's a lot to dive into. I'm happy to share what I've learned.
 
Impressive work. I have built a lot of electrical controls and equipment from scratch. Of course a lot of it was relays and such over CNC as it was years ago. I really enjoyed going through this thread and seeing the forethought and execution of the retro. Very nice work.

I am looking forward to having my shop again when I retire next year. Been playing with CNC controls on some Stepper Motors as a learning process. The goal will be to build my own Plasma table so hoping this learning I am doing will help prevent a lot of costly mistakes/
 
Impressive work. I have built a lot of electrical controls and equipment from scratch. Of course a lot of it was relays and such over CNC as it was years ago. I really enjoyed going through this thread and seeing the forethought and execution of the retro. Very nice work.

I am looking forward to having my shop again when I retire next year. Been playing with CNC controls on some Stepper Motors as a learning process. The goal will be to build my own Plasma table so hoping this learning I am doing will help prevent a lot of costly mistakes/
We aren't terribly far from one another. Let me know if you're ever over on the other side of the state. I'm a high voltage protective relay craftsman by trade, so that helps with some of this stuff. Thank you for the kind words. These machines are a lot of fun to retrofit. And, they are MUCH better than a benchtop or Tormach if you're so inclined and have the space. A CNC plasma is on my want list. I think I'm going to buy one already done though. I don't have the equipment to machine the frame the way I would want. Sparq Robotics makes a nice machine with linear rail motion. Cheers!
 
The mill would be a lot better if I had chip shield extensions. I’ve thought of a few ideas.. came up with this today. It slips over the top edge of the existing chip shield and has two press-fit nuts on the inside. Now I can screw flat material to them to contain a lot more of the chips.

3d printer again for the win!

tree01.jpegtree02.jpegtree03.jpeg

Made the prototype of the mill chip enclosure extension out of scrap mdf. Clips on to the existing (red) chip shield with the 3d printed brackets. Works really well. The Fogbuster does a great job of only getting coolant on the tool and workpiece. The mdf will probably last a long time. Quick on and off. Once I've got the kinks worked out, I'll cut the doors and probably the front left and right out of clear acrylic. Next up is to make some chips and see how it does with keeping them contained.

tree04.jpeg

tree05.jpeg

The enclosure extension is about as good as I'm going to get on this style machine. Significantly less chips on the floor! The Fogbuster is so much nicer with the loc-line setup.

tree06.jpeg
 
These Fogbuster mounts leave a lot to be desired. Always falling, etc.

tree07.jpeg

James Clough (clough42) did a video some time ago on adapting Loc-Line to the Fogbuster nozzles. I modified his parts so they would bolt up to the mill using existing mounting holes on each side.

tree08.jpeg

There's a captive nut on the back side of the mount. This allows the clamp to cinch down tight on the Loc-Line.

tree09.jpeg

So much better.

tree10.jpeg

The other side.

tree11.jpeg

tree12.jpeg

This is where a 3d printer really shines. They are just crazy useful in the shop, on the shop.
 
Back
Top