LumberJocks

Woodsmith CNC Alterations & Additions #1: A few things to make it more convenient and work better

  • Advertise with us
Blog entry by Mike_190930 posted 10-08-2021 12:22 AM 851 reads 0 times favorited 6 comments Add to Favorites Watch
no previous part Part 1 of Woodsmith CNC Alterations & Additions series no next part

After having built the WoodSmith CNC router and run several projects on it, the time came, as it always does with DIY projects, to whittle away at the annoying aspects of the thing and to add a few personal touches. There was a short list of things that bothered me. First, the z-axis did not always move properly, jumping a little either up or down randomly. Second, the electronics made a bunch of electrical noise, so much that my trusty FM radio pulled in mostly static while the CNC was plugged in, much less running, and the Smooth stepper board was picking up random limit trips induced by the electrical noise as well. Third, the electronics box was mounted at the far end of the work surface, per the WoodSmith plans, and that prevented feeding through long boards that were being worked as tiled jobs. I also wanted to float the monitor above the work surface for easier viewing, add a fourth axis for rotary work, and put it all on a proper bench with wheels so I could push it all out of the way when necessary. Quite a list admittedly, but what else did I have to do last January?

The z-axis problem traced to the fact that I had used a segmented shaft coupler between the motor and the threaded shaft. The coupler was stretching on the up travel and compressing on the down travel. In a belt and suspenders approach, I substituted a solid shaft coupler and also placed a thrust bearing and stop collar at the lower end of the threaded rod. The thrust bearing takes the weight of the z-axis carriage and router motor off of the stepper motor bearings. That done, the jumps vanished.

Closeup of the thrust bearing (under the washer) and stop collar before final assembly

I designed a rolling bench with drawers and cabinet space to house everything and support the CNC at a convenient height, low enough that I could put my weight on panels and boards while clamping them to the surface. There is also space to relocate all the electronics into metal boxes underneath, and ventilated cabinet shelf space for the computer. This shows the front of the completed bench, with drawers closed and open.

Front of rolling CNC bench, drawers closed …

... and drawers open. You can see the monitor on its new arm tucked behind the end of the bed in storage position.

Moving the old wooden electrical box out of the way also allows much easier clamp placement from that end of the bed.

I tore the electronics out of the wooden box that had been mounted at the end of the bed, and re-housed them in a pair of steel 12” x 12” junction boxes. The low signal electronics, namely the ESS board, 5 volt power supply, fuses, and router motor relay are housed in one box, and the 60 volt power supply along with the four stepper drivers are mounted in a second 12” x 12” steel box. The strategy is to keep the big electrical noise makers shielded and isolated from the low signal electronics in order to eliminate false limit switch and e-stop signals.

Low Signal Electronics

High Power Electronics

Both boxes are ventilated with 4” fans. Drawing from under the bench has largely kept sawdust from getting into the electronics boxes. I also ripped out the unshielded limit switch wires and replace them with shielded wiring. All shields and grounds are tied to the steel boxes. So far, the false limit trips have gone away, and I can listen to the radio again, so that much has worked. While I was remodeling the stepper motor wiring, I added aircraft style connectors connecting the motors so I can easily swapped motors if needed. The aircraft connectors also add a better layer of shielding to the motor wiring, again cutting down on the electrical noise.

I added a fourth motor for rotary work. This motor is intended to be clamped to the bed for rotary work, and stored when not in use. Since the motor driver instructions caution not to apply power to the driver if there is no motor connected, I looped the DC supply power for this driver through the aircraft plug connecting the stepper motor to the driver. That way, if the rotary motor is unplugged, the DC supply circuit to its driver is also disconnected.

I mounted the computer monitor to a gas charged monitor arm (around $40 from Amazon). The monitor can be positioned easily to view from a variety of angles, and stored at the end of the bed when the CNC is not in use.

Monitor on Its Arm

Someday, when funds allow, I will replace the monitor with a touch screen version and get rid of the mouse. Mice don’t like dusty environments.

An earlier blog post describes a simple modification to the CNC to do horizontal routing, not repeated here.

I am satisfied with these changes and additions. The CNC is easier to work with and to set up jobs, and having it on a wheeled bench lets me move it over to a corner to make room to work on large non-CNC projects. That being said, I find that I am integrating CNC operations into many projects in order to gain more precision or save time on making identical parts, as well as do what was nearly impossible without it. I’ve found uses for it well beyond signs and relief carvings, although those are a lot of fun too. I can see the day coming when every home shop has some degree of CNC capability, whether it be a small laser engraver/cutter, a small CNC router, or even a 3D printer.

-- Huh? Whadaya mean it ain't "measure once cut twice"?



6 comments so far

View Redoak49's profile

Redoak49

5390 posts in 3232 days


#1 posted 10-08-2021 12:59 PM

Well done…...!!!

View Woodnmetal's profile

Woodnmetal

183 posts in 89 days


#2 posted 10-08-2021 01:59 PM

The z-axis problem traced to the fact that I had used a segmented shaft coupler between the motor and the threaded shaft. The coupler was stretching on the up travel and compressing on the down travel. In a belt and suspenders approach, I substituted a solid shaft coupler and also placed a thrust bearing and stop collar at the lower end of the threaded rod. The thrust bearing takes the weight of the z-axis carriage and router motor off of the stepper motor bearings. That done, the jumps vanished.

————————————————————————————————-

You have excellent talent Mike, Nice work.
If you don’t mind,

What amount of Z movement discrepancies (re-tract, R level movement, plunge depths, and vibration) were you seeing with the original set up?

When you mention “stretching”
Could you tell us what material your original shaft coupler and thrust bearing were made from? Was it Teflon, aluminum or brass ?
Could you post a pic of the wear shown?

What material did you make the new couple and thrust bearing with?

Thanks,
Gary

-- I haven't changed... but I know I'm not the same.

View Mike_190930's profile

Mike_190930

51 posts in 751 days


#3 posted 10-08-2021 04:40 PM

————————————————————————————————-

What amount of Z movement discrepancies (re-tract, R level movement, plunge depths, and vibration) were you seeing with the original set up?

When you mention “stretching”
Could you tell us what material your original shaft coupler and thrust bearing were made from? Was it Teflon, aluminum or brass ?
Could you post a pic of the wear shown?

What material did you make the new couple and thrust bearing with?

- Woodnmetal

I was seeing roughly 50 to 100 mil jumps. Never measured it since I knew the performance should be much better than that. The rotational speeds are too low to worry about vibration, maybe 100 rpm tops.

First, I should say that I have very minimal metal working capability in my shop. Hacksaw, drill press, bench grinder, and files are about it. So all the metal parts are sourced commercially.

The original coupler was a commercial item, not sure anymore where I found it. It was machined from a solid rod of aluminum, about 1.5 inches long by 0.75 inch diameter. Each end was bored to accept the 6mm motor shaft on one end and the threaded rod on the other. Each end was fitted with a set screw that locked to the mating shaft. If that was all there was, and that accurately describes the replacement coupler I installed, then all would have been good. But additionally, the material between the shaft bores, about 3/4” long, was cut in a spiral to allow some amount flex to accommodate any offset or angular misalignment between the two shafts. Unfortunately, it also acted much like a slinky spring toy but with less stretch. There weren’t really any worn parts since the problem was an incorrect part.

I did not mention it before, but I used a similar springy coupler on the y-axis (y being the rails on the bridge). That also may have been stretching, but the forces were lower than on the z-axis. I replaced it at the same time anyway. The x-axis had a solid metal coupler from the start.

I added the thrust bearing which was not in the original plans. When I looked at the assembly with the rigid shaft coupler, I realized that all the weight of the carriage and router motor would be hanging from the stepper motor shaft. Not being able to find the stepper motor bearing load specs, I was concerned about wear and tear on the motor. To be safe, I added the lower thrust bearing. It is a ball bearing 0.75 inch outer diameter and 1/2 inch inner diameter, and mounted with the top surface proud of the piece of wood it rests in. The threaded rod passes through the bearing. The stop collar takes the downward force and transfers it to the bearing race. I put an oversized washer between the stop collar and the bearing just to try to keep sawdust from accumulating in the bearing as well as provide a smoother running surface for the bearing. Both the bearing and stop collar are steel.

I have measured the depth accuracy since the changes, not enough to put a fine point on the statistics, but it easily holds +/- 5 mils. That’s tolerable for what I am doing.

-- Huh? Whadaya mean it ain't "measure once cut twice"?

View Woodnmetal's profile

Woodnmetal

183 posts in 89 days


#4 posted 10-08-2021 07:55 PM

Thanks for taking the time to respond Mike.

Having the ability to re-engineer an already engineered machine build does have its advantages as you indicated above.

Although now mostly retired,
My CNC background in metal machining and programming is slightly different, yet similar when it comes to the principals, code and operation. This leads me into a couple more questions haha.

I feel, discussions such as this will be beneficial to many here considering CNC as a necessary tool going forward.

Now,
Since the initial assembly had its downfalls along with incorrect parts supplied, you seem to have limited the inaccuracy to within +/- 5mils which is quite a stretch from the 50-100mils with the supplied parts.
This type of machine may not be a good fit for individuals looking to enter into the CNC side of things.
You did a great job !! Both for user friendliness and rigidity.

Now, having said that,
This leads me to another question if you don’t mind..
Would it be possible that the Z steeper configuration settings were out initially, this giving you such inaccuracy in the Z motions? Leaving X & Y out of the equation for now.

Again,
I appreciate the feedback and hope these questions will assist others, including my need to know personality lol.

Cheers,
Gary

—————————————

I find that I am integrating CNC operations into many projects in order to gain more precision or save time on making identical parts, as well as do what was nearly impossible without it.

I can see the day coming when every home shop has some degree of CNC capability, whether it be a small laser engraver/cutter, a small CNC router, or even a 3D printer.

——————————————————————-
I agree, the need for CNC is increasing, even down to a young hobbyist getting into woodworking.
1 machine can out perform several manual machines in a much smaller footprint.

Cheers to that!!!

-- I haven't changed... but I know I'm not the same.

View Mike_190930's profile

Mike_190930

51 posts in 751 days


#5 posted 10-08-2021 08:13 PM


Since the initial assembly had its downfalls along with incorrect parts supplied, you seem to have limited the inaccuracy to within +/- 5mils which is quite a stretch from the 50-100mils with the supplied parts.
This type of machine may not be a good fit for individuals looking to enter into the CNC side of things.
You did a great job !! Both for user friendliness and rigidity.

Now, having said that,
This leads me to another question if you don t mind..
Would it be possible that the Z steeper configuration settings were out initially, this giving you such inaccuracy in the Z motions? Leaving X & Y out of the equation for now.

I should clarify that this CNC build was not a kit, but rather built from plans and a BOM published in WoodSmith Magazine issue 242 and 243. As such, all the parts were sourced individually commercially by me or in the case of wooden parts, fabricated in my wood shop. I did substitute parts in the BOM for others as availability changed over the course of the build. The shaft coupler in question was one of those substitutions, and entirely my error in choosing the wrong part. So all inaccuracies were mine alone. Others here have built from those plans with grand success. The CNC is a router CNC designed for general woodworking, not metal working.

The z-stepper settings were not changed during the course of this refit, so I am certain that nothing in the software of electronics was at fault.

I have several other blog entries that more fully describe the build and my changes to it.

-- Huh? Whadaya mean it ain't "measure once cut twice"?

View Woodnmetal's profile

Woodnmetal

183 posts in 89 days


#6 posted 10-08-2021 08:36 PM

I must have misunderstood, I was questioning the initial set-up in which you mentioned in your post.

———————————————————
First, the z-axis did not always move properly, jumping a little either up or down randomly.
————————————————————

This is what my question was based on ^^^^, I figured this could have been addressed within the Z stepper config of the software.

Thanks for the quick response, Interesting stuff, I will have a look at your previous blog entries.

Gary

-- I haven't changed... but I know I'm not the same.

Have your say...

You must be signed in to post the comments.

DISCLAIMER: Any posts on LJ are posted by individuals acting in their own right and do not necessarily reflect the views of LJ. LJ will not be held liable for the actions of any user.

Latest Projects | Latest Blog Entries | Latest Forum Topics

HomeRefurbers.com