[timber_MG]

This is a log documenting my first CNC router. Same story, long time lurker and I’d like to let some images do the talking because I lack the CAD skills. Any input would be gladly appreciated.

The machine is to cut Wood, Foam and aluminium in an envelope (X,Y,Z) of 1200mm, 850mm and 200mm respectively. I would be happy with 0.1mm repeatability, but would be prepared to settle for less.

The construction is mostly steel, but I plan on using some aluminium profile on the gantry for reasons of weight and simple straightness.

I found a scrap steel table that was rather straight and seemed a good bet…and that it was. After mapping the X axis rail mating surfaces with a dumpy level and set square I set about to grind it by angle grinder, but will likely cast these surfaces with an infusion epoxy. I really like the dumpy level and want to engrave a plate to allow more repeatable measurements (currently I employ eye-averaging and a spread sheet and statistics and whant to use a 1:10 sloped lines with 1mm graduations to give me 0.1mm steps and perhaps the ability to interpolate)

The X and Y axes will be spring loaded rack&pinion drive and V groove wheel guides. These are to be fashioned after the Mechmate, ground out of steel 70x70x6mm angle iron cut to 70x40x6mm with the 40mm end receiving a V-track profile. The whole process is surprisingly achievable after one has done it but I intend to run a conventional bearing running underneath tensioned by ex-center bushing and bearing to keep the assembly on the rails.

Once I have the initial machine constructed I could always use the bottom of a milled slot in the table as a relative reference to grind more accurate track out of angle iron as I go along (and the epoxy cast base will certainly help in this regard)

Initially the 15 tooth 1 Module R&P will be direct driven until I can build a belt reduction gear box for the axes. My calculations seem to result in a 3-4:1 reduction, but the belt reduction is certainly not cheap by the look of things and for first I will cut light wood.

The gantry uprights will be steel U-channel (100mm x 50mm) milled (and in the case of an over-eager machinist surface ground on one side) and attach to the aluminium gantry by means of T-slot fixtures and an angle plate.

The Gantry will be constructed from 45mm x 90mm aluminium profile (see http://www.cncdirect.co.za/htm/profile.html ), but with steel guides (as above) running on the top and bottom of it to give the structure additional strength and provide a suitable V track for the Z car. This profile is an exceedingly stout beam.

The Z axis is to employ 20mm unsupported shaft and a ground base plate and end-plates (reamed and slotted) to accept the linear shaft. Actuation will be by 10mm pitch ball screw.

The spindle is an Ebay water cooled 2.2 kW unit with VFD.

Lastly the motors will initially be fed by a Hobby CNC Pro 4 axis (slaved X Axis) board and Keling 276oz-in 8 wire motors. I have Geckodrive 203Vs , cncdirect.co.za BOB (excellent unit and support is fantastic) and 1200 oz-in motors for a stalled Mechmate project on hand that may do duty on this router, or get the Geckodrive 540 for this, but if the hobbyCNC can do the job, it might at least be worth a try (lest my 8x20 lathe not escape a gang zoning)

Thanks

Martin

[timber_MG]

I have been practicing some CAD drawing. This is the basic structure of the gantry showing the V-plate bolted to the aluminium extrusion. I will likely have the plate and the drilled and tapped flat bar to bolt it into the Ali extrusion laser cut and countersunk and then use the Mechmate grinding skate to grind on the V-track or have this section machined.

Notice that the gantry extends a fair bit over the bed...this is intentional for mounting a potential future 4th axis or to machine assemblies that don't fit onto the bed.

If I find the construction to lack stiffness I will likely reinforce the Ali extrusion with some steel angle bolted top (or even bottom) to the 45mm sides of the extrusion and perhaps attach the R&P drive to this angle iron.

I hope to receive my Z-slide back from the machinist sometime this week which means that it'll be time to work on stepper mounts soon.

[timber_MG]

Time for an update

My machinist suggested an indexing plate to go with the Z-slide. Lesson learnt, never give a machinist free reign over your project, you will pay dearly!

He did however do excellent work on his well maintained acera and deckel milling machines (huge iron with small throw guarantees fantastic finishes). He is of the old school and lives off hand drawings (which are better looking than most CAD drawings mind), but by now I could have easily paid for a K2 Z slide and more.

The index plate (the plate with many holes) is drilled to fit the linear shaft pillow blocks, nut holder as well as accept the spindle holder in increments of 15 degrees. Everything is 16mm Aluminium except the linear shaft supports that are 20mm thick (recessed 10mm for the bearing block)

[timber_MG]

I have also sold off my Mechmate components, so what remains are my Keling KL23-276-8B motors.

I have decided on a belt-reduction gearbox, but still have an open question regards drive: The gantry, head and spindle will weigh in at at around 45kg (say 50kg). Should I stay with Keling KL23-276 with 5:1 reduction and Gecko G540, move to the larger Keling stepper motors, have a go at the DMM-tech http://www.dmm-tech.com 250W AC servos (tempting) or something else all together?

I can achieve a 5:1 reduction on the steppers, giving me 10Nm at the pinion, where I can only manage a 3:1 reduction on the Servo becasue of its 14mm shaft requiring a larger pulley with a peak 7.18Nm (2.4Nm const) so for low speeds the servo is matched ito cutting forces by the stepper, however at the price of these DMM devices it is mighty tempting and make a speed monster out of this machine.

[timber_MG]

Progress has been slow, but here a small update.

Here is a rendering of the table with the guides I am having made for it. I will fit some 100mm x 100mm square tubing to the existing table and some legs to bring the Table to a workable height (probably add some diagonal flat bar to stiffen the assembly up)

The V track I am having laser cut and then ground out of an abrasion-resistant steel. I will turn a bunch of 10/12mm ex-center bushings on my recently acquired 8x20 lathe that will serve for fine-tuning the height of the rail. Following on my goal of optical alignment I am tending to a laser pointer and dual 0.5mm apertures pointing at a gantry mounted CCD from a web cam sans lens..should make for some interesting alignment options.

[timber_MG]

I have fired up the Z axis and I must say that I am quite impressed by the assembly...

I borrowed my boss' camera to take some pics of the assembly.

With one last modification (making 4 new holes in the spindle mount plate to move the top mount down ~50mm) the slide will have the spindle nose point out 100mm over the bottom of the slide. The mounting holes on the bottom of the slide which will keep the V groove wheels have been spaced to allow me to move the slide down 50 or 100mm.

In the future (read when I want more than 100mm Z) I plan on having a longer plate made up that will double as a swarf cover and give me the full 190-200mm Z depth. For milling aluminium or profiling wood, however, I would rather have the assembly as stiff as possible as am prepared to live with some extra set-up time as I plan to use this machine for very varied uses (not precluding mounting a circular saw to profile wood in 4 axes amongst other ideas that I will spare you for now)

This is a rough video taken with my mobile phone showing the axis rapid moving with the stepper on my table configured at 1A, 50V on a Gecko 203V: [ame="http://www.youtube.com/watch?v=3vp7CgpnOWo"]YouTube - Z axis slide first test[/ame]

[timber_MG]

This build has taken a slightly different direction. I decided to go with some 2510 ball screw from linearmotionbearings on Ebay and these screws are very impressive. 1" Ballscrew and AC servos has a nice ring to it. Chai is good with his shipping and I got these within a week (unbelievable really, considering that he machines the ends too).

I have also ordered the DMM-tech 250W servos and am looking forward to see what they turn out like. Will probably run these 3000/3500 rpm motors on 2.5:1 belt reduction. Should make for a stiff linear motion setup with 10mm pitch screws with almost 1:1 inertia matching.

I have been thinking of the best way to adapt the machine to suit the screws as I don't want to forfeit too much Y travel (for the gantry extending past the table) while protecting the screws. Any good pointers would be really appreciated and any other good pointers would be great.

[harryn]

Hi - nice work.

There is a variation of the moving gantry systems where the moving uprights are replaced by fixed uprights on each side. In other words, the Y gantry is just a beam moving on two elevated rails.

The advantage of this is that you don't have to deal with the flex of the uprights and all of the drive components are above the work - potentially reducing dust into the drive screws.

The downside, is that loading the system is harder, as side access is more limited.

Both designs have been built on the forum here, but of course I cannot find a good example of the one I am describing right now.

Take care,

Harry

[timber_MG]

This project has been lingering for a little while now, but finally things are moving forward.

I would really appreciate some input on my gantry uprights. Originally I wanted to use some angle iron and a seperate plate to mount the V-guide bearings to, but after drawing up all the drive components out of Ali (to be sent for machining shortly), I thought up the following uprights to simplify construction.

What thickness aluminium should I use for this uprights? It would hold the 4 V-bearings and connect to the gantry with a square. Options are 10,12,16,20,25mm. I have even thought about constructing them out of MDF for a quick and dirty build and machining the replacements out of tooling plate or something cast.

[timber_MG]

After reading a lot on the zone and using the Beam Boy application to calculate the stresses, I have decided to build the Gantry side-walls out of 1" aluminium and to add a triangle support to stiffen the upright/gantry joint.

I will hopefully post some drawings later, but I am placing an order for aluminium and hope to have these parts done by next week.

[derekj308]

Hi timber_MG

Could you please post some close ups of the end machining done by linearmotionbearings and your opinion on the quality of the machining.

BTW the z axis looks really good and very smooth motion. I look forward to the rest of this build.

[timber_MG]

The machining certainly looks good, but I don't have a micrometer to tell you exactly how close a fit it is. The fit is very tight though and I will have a machinist fit the end bearings (I decided not to go with their bearing block units after the Z axis and chose rather to machine some housings for dual row angular contact bearing units to save costs).

The Z axis was just tested temporarily, but it is very stout and I really hope to have some success at cutting Aluminium with the machine once done.

[mhasting2004]

Nice work Timber_MG

I'm at that stepper versus servo debate on my machine too due to speed. FYI if you don't already know there is a mach plugin now for the Huanyang VFD over here http://www.machsupport.com/forum/ind...c,14182.0.html

Derek I can second MG's comments about Linear motion .. they are very fast and the quality of the machining was spot on in my books (ie it fits nice). The quality of the actual linear screw I got was ... what I paid for it. In other words I have better ones but they cost considerably more.

I really like the indexing head and had toyed with the idea for awhile but my Z axis design precludes this for the moment.

I think the design that Harryn is refering to is where the X axis rails are elevated in referance to the cutting bed this is the case in mine and to some extent yours already. It does mean that you will be loading from the ends of the machine rather than the sides.

Thumbs up on the build

Cheers

Mark

[timber_MG]

Thanks for the kind words and that link,Mark. I had been looking for the plug-in since I saw it mentioned on the zone.

When I though out the table, I wanted to machine objects placed besides the table (also for a potential later rotary axis) which precluded the raised X linears (a la Solsylva) I did however raise them above table level, as mounting them below the table makes the gantry act as a long arm and increases the spacing needed for a given stiffness.

In hindsight, those BK mounts from linearmotionbearings2008 are not at all expensive. I will have them quote up a set of BK/BF15.

[derekj308]

Thanks timber_MG and Mark for the opinions on linearmotionbearings screws. I saw them on ebay and the prices vs quality looks quite reasonable so hearing good responses from fellow 'zoners is reassuring.

Keep the pictures of your build coming timber_MG

[derekj308]

Thanks guys for the opinions on linearmotionbearings. I saw them on ebay and I thought their prices vs. quality is reasonable. It is reassuring to know that fellow cncers have had good experiences with this seller.

Keep the photos coming of your build timber_MG

[derekj308]

Thanks guys for the opinions on linearmotionbearings. I saw them on ebay and I thought their prices vs. quality is reasonable. It is reassuring to know that fellow cncers have had good experiences with this seller.

Keep the photos coming of your build timber_MG

[derekj308]

Can a moderator please remove posts 15 and 16. Some idiot at my computer keeps posting the same message because they are too impatient and think the post didn't get through.

[timber_MG]

The quote for the BK/BF15 came back quite reasonably at 180 US$, with much of that going down to shipping. Perhaps I should look at this as an opportunity to add some screws and linears for a mini desk-top machine as I have a hobby cnc controller and motors left spare.

Best Regards

Martin

[timber_MG]

I have bought a whole bunch of Aluminium Tooling plate, but the chop-saw with which them seem to cut the stuff was not exactly square.

So I though I could use some time off during the easter weekend and see what the ancient art of scraping is all about (just without the blueing against a reference plate, this was more to square up the cubes that will be cut diagonally on a band saw to hold the gantry square to the uprights.

I took an old file and ground off some teeth (not far back enough) and with my machinists square, starrett ruler and ground steel plate pictured proceeded to square up the pieces and wet-sanding on top of the plate to get the surfaces more flat.

Lessons learnt: a workpiece files/scrapes much more easily when held in a vice and feeler gauges are actually useful in gauging how much more material to remove. I wonder what my machinist will say about the partially square stock.