Easy(?) Inlays again... deja vu, or just another rabbit hole?

dkj4linux · June 25, 2025, 11:22pm

Recently, I’ve been testing my latest machine build – MPR&P – to see just what it can do. I’ve been quite surprised with the spindle testing I’ve done… and am completely shocked that these small 775 motors, used as spindles on many cheap desktop CNC machines, are as capable as they are. They are not toys, as I previously thought, and make IMO a very nice accessory for MPR&P.

I’ve done some faux “drag racing”… patterning some of my testing after the 'CNC drag races" thread that Ryan instigated several years ago. Rather than “unlimited” as those drag races were, however, I limited myself to using just the 775 spindle motors and shrunk the V1 test logo by 50% in all three dimensions. I also used some white pine scraps I had on hand rather than using the standard MDF used in the real drag races… and attained some pretty solid cuts with 6mm DOC and a feedrate of 1000 mm/min with a 3.175mm single-flute end mill. But rather than pushing it further, I’ve decided to move on to other things.

And V-carved inlays have been a topic of discussion recently with the early releases of MillMage. So, with laser cut inlays experience under my belt, I’ve been looking into how accomplish them with MPR&P. I ran across this very well done video put out by IDG Woodcraft and took advantage of the offer of a free instruction kit (and a bunch of emails offering other stuff) for the feather inlay… including the gcode files for the Longmill CNC used in the video.

I uploaded the 1/8" end-mill (similar to bit used on the Longmill) roughing gcode for the feather inlay piece straight into MPR&P… completely unaltered. A first attempt was running beautifully but ended with a USB disconnect between UGS and GRBL about 5000 lines into the 5400+ line job. The hole in the catalpa board (and my new work surface) is the result of me working through the mid-job disconnect problem. I’m now using Jeff’s V1Pi and CNC.js once again…

In action and a bit noisy… but it is a ROUGHING cut. Please turn down your volume…

I don’t have a proper 1/4" V-bit (I have one on order from IDC…) to finish the full-sized feather project but have since moved on see if I can process my own inlays. I’m not about to spend the $$$ on the expensive V-Carve software that is most often used for such things… and being a Linux user, the pickins’ seemed pretty slim. I’ve played a bit with F-Engrave, which apparently can do V-carving and inlays, but got bogged down pretty much in the differences in terminology used. I’ve gotten both pocket and inlay gcode files… but nothing close to a proper fit.

Then I stumbled on to this video which quite nicely explains the major issue I was having…

He’s using Carbide Create to do the inlays here and I had no idea it could do that. But, of course, Carbide Create (which I though was just another Easel or somesuch) isn’t a Linux program. But there’s this “new” thing called Bottles for Linux users… that allows running M$-stuff, if you hafta, on your Linux box. I had recently loaded Bottles on my Debian laptop because I was pretty sure I was gonna hafta resort to using Estlcam… just to keep up with @Tokoloshe and his impressive inlay work.

Carbide Create is a free download so I downloaded it and installed it thru Bottles. It ran great… but, turns out, with CC7 and later they don’t allow gcode output without your $$$ for PRO and a subscription. Fortunately, turns out, CC6 is still available for download and DOES ALLOW gcode output… and it’ll do all the stuff the guy does in the video! And it runs quite nicely in Bottles (though the font in quite small but mostly readable)… and has a GRBL postprocessor built in. Another nicety in CC6, it has a pretty extensive tool database… and when selecting the tools and setting in the toolpaths, they provide recommended feed rates and cutting parameters. I didn’t have to guess… and fairly good feed rates magically appeared in the gcode.

I’m waiting on the glue to dry. The inlay may look like crap when I cut it open but for a first real attempt… I’m very encouraged. I’m also having to deal with tool changes and probing for the first time… so it really is a rabbit hole I’m stumbling down right now.

But I’m having a blast with MPR&P… more so than with any machine I’ve ever built.

Gotta stop for now… later.

– David

Jonathjon · June 25, 2025, 11:30pm

I must say… That thing is QUITE IMPRESSIVE!!! It has been a joy to see this progress and how capable it is.

I didn’t notice any real deflection in it either with that much Z travel which is even more impressive.

Cant wait to see how your inlay turns out when its complete!

Looks like the only thing you are missing now is some dust collection

vicious1 · June 26, 2025, 12:02am

yes!!!

Tokoloshe · June 26, 2025, 4:50am

Looks really great, hope it turns out well! Also, @Jonathjon needs to feel a bit of peer pressure so he can finally do his inlay…

dkj4linux · June 26, 2025, 5:43pm

As expected there’s a good deal of chipout where thin fingers in soft pine and/or catalpa couldn’t stand up to the rigors of being closely shaved. And maybe I was going too fast since the gcode auto-magically used the recommended speeds and feeds from similar, but different, bits than I used. And just maybe MPR&P isn’t up to the task of doing small inlays in crappy woods. And maybe my simple clamping method allowed for too much material vibration while cutting…

And maybe it’s just me… not knowing what I’m doing!

But, still, from a “proper” viewing distance, it’s almost recognizable as a feather…

Jonathjon · June 26, 2025, 6:51pm

That doesn’t look bad at all. I bet if you slow down that gcode a bit it would look even better.

Tokoloshe · June 26, 2025, 7:59pm

Looks great for a first inlay. I mean, you chose a pretty complex thing to cut…

dkj4linux · June 26, 2025, 8:56pm

Thanks, guys, for the kind words. I’m pretty happy with this first attempt. I just need to start playing more with the different bits and cutting parameters… and use some better hardwoods to do real inlay work. One of the real differences in laser inlays and these is that without tool forces on the wood… lasering is not nearly so picky about the woods being used.

There’s obviously a lot to learn but I find playing with all this new stuff great fun… and always have!

dkj4linux · June 29, 2025, 5:06am

Been wondering about that myself. The big stuff is no big deal for such a small machine… but it’s the fine dust you don’t see that’s starting to show up everywhere.

On a very early machine, I added dust collection and really didn’t know enough to struggle with the design; i.e. shopvac on the floor, pipe clamp sticking straight up to clamp the hose to, and let the hose just drape over to where the router was. I discovered then that MDF cut/routed so nicely (but was so dusty!), I thought every other material should cut/rout as easily also…

Well now I have one of these…

and a 3d printer… and wondering what I might do with those two long metal tubes that run right down next to that spinney thing…

Now I guess I oughta talk to Tux about the “greater good” and get him to move over and give me room to manufacture a manifold to access those two boreholes next to him…

maybe I… erm, we… can come up with something that’ll help the cause. I’m sure it won’t be the best dust collection setup but it’ll be fun… and maybe it’ll help a bit.

Later.

dkj4linux · June 30, 2025, 10:21pm

Getting closer… 1st SWAG at dust collection.

Britt · July 1, 2025, 4:07pm

A great solution Dave. It has been both fun and a learning experience following along with you on this adventure.

dkj4linux · July 1, 2025, 5:46pm

Thanks, Britt, for the kind words. We’ll find out how “great” it is when I get off my butt and start actually doing stuff with the machine…

dkeller · July 2, 2025, 12:06pm

Very cool!

But what does MR&P stand for?

MakerJim · July 2, 2025, 12:38pm

Bartman · July 2, 2025, 4:13pm

R&P stands for Rack and Pinion. The build above has been dubbed MPR&P (Mostly Printed Rack & Pinion), where gears are used instead of belts and lead screws.

dkj4linux · July 9, 2025, 7:48pm

All,

Upfront, this is gonna be very difficult to write because I know some/many of you are doing v-carve inlays very happily and successfully… and I don’t want to disparage the method or enjoyment you receive from it. PLEASE keep this in mind if you choose to read further.

Also, please know that I’m speaking from my own personal recent experiences in this post… and I’m gonna admit I’ve really struggled with the v-carve method so far. It’s probably a combination of factors however… since I apparently refuse to give myself the best chance of success by my choice of machines and materials. But I’m quite probably the only person here who has used BOTH my lasered “easy inlay” method AND actually tried to v-carve inlays as well. So, this is to share my experiences with both techniques and contrast the results.

Here are the results of several days of trying both methods… with several attempts at the v-carve feather (top) and only one attempt with the lasered version (below)…

I actually thought (before glue-up) the v-carved version above stood a good chance of being the best I’ve done. For comparison purposes, I also went into Lightburn and processed the same feather image and lasered it using the same cherry and pine…

The v-carved version was disappointing… and not even as good as the first feather attempt I documented earlier in this thread.

So, here’s what I’ve learned (but been unable to successfully do!) about v-carve inlays – and why I admire what Philipp and others here are able to do.

Use good close grained woods that machine well. Maple and walnut are classic inlay woods… common pine and catalpa are not.
Surface the material first if it’s not flat… critical for correct depths of cut and proper glue and surface gaps.
Use quality bits that are in good shape… and take light-ish cuts.
Adjust feeds and speeds to avoid chatter and tearout. I’ve not been successful at this and MPR&P is unproven yet… so, still suspect.
Be patient, take your time, and be tolerant of noise, sawdust, bit changes, and often lengthy runs for flattening, v-carving, and clean-out.
Know and understand the settings in your v-carve program to attain proper feeds, speeds, depths, and gaps.

Lasered “easy inlays” by contrast is simply much more forgiving. Almost any wood and any laser machine (20W-30W laser power is probably best) can be used. Since there’s no physical contact between the “tool”/beam and the material being “machined” there’s little chance of tear-out, regardless of the wood used, and modest machine rigidity should be sufficient. The wood doesn’t need to be perfectly flat (“eyeball flat” is good enough) and no special care is needed for DOC or gaps. There are no bit changes and the relatively straight-sided lasered walls allow for as much or little “engagement” as your laser’s beam power and feedrate has provided… and whatever clamping pressure you bring to bear. Any “engagement” from 1mm to 2mm down to “bottoming out” is close enough to get a nice fit.

Regarding time and effort…

With the v-carve method, a minimum of a couple of hours has been required for the relatively small inlays I’ve been attempting to do. I’ve attentively sat at the noisy machine, done the required bit changes, and initiated separate lengthy runs for flattening, vcarving, and clean-out for EACH of the pocket and inlay pieces. Dust collection helps but there’s still been some house-keeping required. And all with no guarantees of a good inlay… many ways to go wrong.

Even with needing to refresh my memory, I fired up Lightburn and imported the feather SVG into a Lightburn file I had previously used. I processed it using my “easy inlay” method and created both the pocket and inlay pieces… maybe 15 minutes of work. I loaded the pocket wood plank (cherry straight off the bandsaw) into the laser machine and from my recliner (next to the machine) ran the raster burn for the pocket… another 15-20 minutes. Loaded in the inlay wood plank (pine from a common 2x4) and ran the raster burn for the inlay piece… another 15-20 minutes. A little bit of noise from the inline exhaust fan – but not enough to drown out the TV – and all the waste/debris goes out as smoke and soot. A total of maybe 45 minutes… certainly less than an hour.

If the pocket and inlay pieces even lightly try to engage when properly aligned… the result is almost guaranteed to be good.

Again, this is not to disparage the v-carve inlay work that so many folks find immense enjoyment in engaging in. Please know that I admire anyone with the patience and skill to turn out beautiful v-carved inlays.

But, as far as “EASY” goes, which is the question mark in the thread title, I’ll personally have to stick with doing them with the laser.

And, of course… YMMV.

– David

Tokoloshe · July 9, 2025, 7:57pm

No worries, it takes a while to get into it and as you might remember, I was sooo close to just burn the stupid chess board, because nothing would fit and I forced it together with my 500kg force book press… and even then I had to redo quite a bit, because I was too stingy for the 6mm endmill and took a 3.175 that flexed…
By far the most important thing: 0 has to be exact. You can‘t eyeball 0, especially not with those pointy endmill where you can‘t see whether the tip is 0.3mm below the surface or not. I have a 30° endmill where the tip broke off and I always wasn‘t sure whether that had really happened, because it is sooo hard to see with the naked eye, but I finally bought a new one… I am missing 0.5mm or so, which means the inlays didn‘t fit because, top and bottom together, were 1mm off. I did sand the inlay down and it worked, but that‘s just one thing where you have to be extra careful. It took me a while to leave the „eh, good enough it‘s probably going to fit“ attitude behind when doing inlays…

dkj4linux · July 9, 2025, 8:14pm

TBH I really don’t begrudge the time I’ve spent playing with this. I’ve learned a lot and I’m sure with practice I could do far better than I am now. But I spent several days messing with planks straight off the bandsaw before I finally decided to surface one of them using Jamie’s surfacing test pattern generator. “Eyeball flat” was always good enough when I did the inlaid gift boxes for family at Christmas and birthdays… but there was probably a full 1mm or more variation over the surface of a plank. But then, when I started trying to set glue and surface gaps… it all became quite clear that flattening each plank was critical. With the straight-sided walls off the laser, I never had to worry about flatness or cut depths and gaps… I could just jam it together until it feels like it has bottomed out.