Im new, am a 3D designer. Need a New cnc machine and need advice and guidance

Quote:

---

Originally Posted by FlyHighRC

Do you have a link to these linear rails?

---

No I don't but just use the Hi-win rails for the model has the spec is virtually the same, height, width etc. If you PM me an email address then I can email you SW files for them.
If you want exact dimensions to model them let me know and I can measure some I have here.

I would like to add that in terms of precision welds, I am most likely going to be taking the parts to a machine shop where they will have all the nessesary jigs and fixtures to avoid distortion if i were to tackle this myself. The ideally make the holes for the rails ever so slightly larger than required to give some wiggle room :)

Quote:

---

I would like to add that in terms of precision welds, I am most likely going to be taking the parts to a machine shop where they will have all the nessesary jigs and fixtures to avoid distortion if i were to tackle this myself.

---

you dont need to be to exreem about the wields, as long as its sort of level on the top, and measures square ( you will see any twists when cutting the steel into parts, its easy to avoid any visible distortions, just use the best bits for the top ). I did the same as your suggesting though, i got a firm to do the welding, after i had cut and marked the pieces.

Epoxy does most of the rest,

Quote:

---

Originally Posted by george uk

you dont need to be to exreem about the wields, as long as its sort of level on the top, and measures square ( you will see any twists when cutting the steel into parts, its easy to avoid any visible distortions, just use the best bits for the top ). I did the same as your suggesting though, i got a firm to do the welding, after i had cut and marked the pieces.

Epoxy does most of the rest,

---

Sounds good :) hopefully i can make a start on the design soon...Just need to pick the rails i want as i can then design the machine around these. 1000 by 600/700 linear rails...know where I can get me some? :D

I can't really recommend an epoxy as I haven't used this particular technique, but as I understand it does need certain properties for it to work..

Firstly, to be clear - I'm not talking about the method used on this build:

My First Router, Built in Steel - Page 17

The method I'm referring to is the one where once the frame is assembled/welded, a couple of channels are built onto the rail bearing surfaces (the channels don't need a bottom as the surface of the frame will serve that purpose) these channels are joined by a bridge (probably more than one, which will - of course - need a bottom). Once the frame is reasonably level and the epoxy is poured in, the bridge(s) ensure that the epoxy on both sides of the frame sets at the same height. Once everything's hard, take away the channels/bridges, scrape away the meniscus and voila - parallel, same height rail mounting surfaces.

Attachment 12529

A few folk on here have used this method and will be able to elucidate further.

Wal.

This is my frame in my works...you can see the epoxy on the top...my problem us that I was a numpty and was rushing and got the mix wrong so its not hard enough.....

http://img.tapatalk.com/d/14/06/04/vy3yqequ.jpg

http://img.tapatalk.com/d/14/06/04/esarubuz.jpg

Sent from my Galaxy S5

Cool guys, How deep are those channels? Epoxy can be very expensive...and sueply you'd only need a couple mm max!

Can you guys please point me in the direction of linear guider off 1000mm x 600/700mm please :)

Seems not many people know where to find em

>Cool guys, How deep are those channels?

They don't need to be as deep as I've illustrated 'em - I think 2mm would be under-doing it a bit - mix too little (or if the frame's level is slightly off) and you're maybe gonna end up with bald-spots... I'd calculate for a 5mm thickness of epoxy and don't forget to make the channel wide enough so that the resulting strip of epoxy supports the full width of the rail.

That's about the extent of my knowledge regarding this technique - someone will be along soon to better advise!

Wal.

Hey guys, are these compatible?

HGR20 Rail

Hiwin HGH20 Square Carriage

Wont be buying them from here, but nice to know i can if needed for whatever reason

Quote:

---

Originally Posted by FlyHighRC

Hey guys, are these compatible?

---

Yep, should be fine - they appear in each others related items but before buying anything check compatibility via brochures and then double check with the vendor. It's pretty head spinning when you first start looking at the catalogues etc, but after a while drilling down through the relevant info becomes a lot more straightforward. Probably best to ask questions here about the physical characteristics (pitches of screws etc.) of the hardware you want to acquire and whether it'll suit your needs but don't rely on opinions regarding compatibility - although the guys on here are amongst the most helpful and knowledgeable in the community, it's always best to double check yourself..!

Wal.

Quote:

---

Originally Posted by Wal

>Cool guys, How deep are those channels?

They don't need to be as deep as I've illustrated 'em - I think 2mm would be under-doing it a bit - mix too little (or if the frame's level is slightly off) and you're maybe gonna end up with bald-spots... I'd calculate for a 5mm thickness of epoxy and don't forget to make the channel wide enough so that the resulting strip of epoxy supports the full width of the rail.

That's about the extent of my knowledge regarding this technique - someone will be along soon to better advise!

Wal.

---

If you look through the build logs for mine, EddyCurrents and others you'll see that we calculated how much epoxy we'd need. Eddy and I went for approx 5mm.

The width needs to be about 38 mm to give enough either side for the miniscus not to come into play. Also important is to make the bridge wide enough so the epoxy flows ok.

Sent from my Galaxy S5

Ideal, Yeah allot of measurements, just googled it to get the dimesnions f 20mm rail and the carriage to pu into CAD. once i made up the rails etc i can start the design....

Any design aspects i should need to know that people often do? Save me going back and forth as much :)

Quote:

---

Originally Posted by FlyHighRC

Ideal, Yeah allot of measurements, just googled it to get the dimesnions f 20mm rail and the carriage to pu into CAD. once i made up the rails etc i can start the design....

Any design aspects i should need to know that people often do? Save me going back and forth as much :)

---

Been busy today but I'll email the SW models in next 10 mins.

Thanks man, What do you think of the rails and Carriage I posted on the previous page?

Quote:

---

Originally Posted by FlyHighRC

Thanks man, What do you think of the rails and Carriage I posted on the previous page?

---

Yes these are the type of thing you need and Hi-win are good quality linear rails just expensive. I've sent email and put a little about rails in it.

Quote:

---

Originally Posted by JAZZCNC

Yes these are the type of thing you need and Hi-win are good quality linear rails just expensive. I've sent email and put a little about rails in it.

---

Hey, thanks, I have seen the email...allot of time saved with using these files! much appreciated..

When you sau adjust to suite do you mean extend the rails or shorten the rails in SW, per manufactured one...no point me making one 846mm right.

Ill get to work on a design now then, Any prefernace on something i have seen dont but to me there is no added benefit... Is it better to put the rails on the side of the machine or on top? The one benefit i see is that if its on top you get much more freedom in terms of fine tuning the machine.

Quote:

---

Originally Posted by FlyHighRC

Hey, thanks, I have seen the email...allot of time saved with using these files! much appreciated..

When you sau adjust to suite do you mean extend the rails or shorten the rails in SW, per manufactured one...no point me making one 846mm right.

---

Yes just lengthen or shorten to suit your needs. That particular rail is just one I grabbed from a machine design and each machine design will have it's own length version. The Bearing is standard across all machines so only need one instance which you don't change.

Ideal, Spent most of last night getting the basic design on paper then spent a couple hours on SW going through a few differnt configurations with this one being the easiest and simplest to make :)

What do you think? There is a piece connecting the rail section underneath going across the X axis. Oh and the board would be MDF...SW doesnt seem to have that
Attachment 12530

That baseboard will bounce and sound like a drum!! That is unless there are lorts of supports underneath it, but I can't see any bolt holes apart from the ones down the outside? Why have you put the rails below the base, all the swarf will drop onto them! How are you going to mount the ballscrews and stepper motors? You're not doing a simgle ballscrew for the X axis are you? What is the gantry going to look like?

Have you looked at the build threads? Most (I do realise that you don't have to go along the same lines as most other people...however there is normally a reason why most of the builds look pretty much the same!!) look like mine below...

Attachment 12531Attachment 12532

there is one support under the table though I could easily add another :) The table is above the rails to give more rigidity which ill show from the underneath. The, so bare in mind that its not at all done..still fiddling with the design. I see your point about crap going onto the rails but that would happen in most machines...yours even...rather than have the stuff been blown onto them after running alone your board mine will do the same but the machines rubbish will blow off the board and possibly land on the rails...but either way..Im going to add 3mm poly between the gap of the table and the rail carrages...so efficiently having a box where the machined rubbish(Any jargon here) will stay and be easily cleaned up afterwards :)

Ahh, I was planning on ussing a single ballscrew down the middle? Is that a big no no?

Also, the rails that the gantry run on...is there a good distance to keep them from eachother? Assuming that the closer they are the more slop you could possibly get?

My gantrry is 1 vertical piece with a large 40x40x4, 45deg support

I had planned on using one ballstrew for each axis... waht you see would have 100mm legs on each corner and would then have a piece connecting each side of the gantry that can be bolted onto the gantry sides to make assembly that bit easier
Attachment 12533

You are weakening the machine having that base like that and it's completely not needed. You need to keep the gantry sides height to a minimum and your creating 60mm more than needed and every bit helps if your wanting to cut aluminium.
If you want the best strength then having raised sides with rails on top with gantry sat directly on bearings is massively strong and much much more suited to an all round machine. If your wanting to cut aluminium then you WILL need 2 ballscrews on the axis moving gantry. This is not just an opinion it's hard learnt and tested thru building many differant machine designs trying to find the best design for an all round machine without being silly about design.

The machine your building with single screws will be fine for mostly wood use and occasional light duty Aluminium work but have no doubts it won't handle cutting aluminium correctly and give quality finish. At a minimum you'll need 2 ballscrews moving gantry to stand any chance and very beefy gantry.

Edit: Oh and the bed base needs far more supports than just one in middle.

So loose the second level of boxsection with the board ontop?

Keep gantry height to a minimum

Design in two ballscrews now rather than later

add more support for bed..

got it :) The general gantry design.. Is that ok? with a vertical piece and a 45 member to give some strength... How far should i keep the bearing carriages apart? from the faces that would touch..

Quote:

---

Originally Posted by FlyHighRC

So loose the second level of boxsection with the board ontop?

Keep gantry height to a minimum

Design in two ballscrews now rather than later

add more support for bed..

got it :) The general gantry design.. Is that ok? with a vertical piece and a 45 member to give some strength... How far should i keep the bearing carriages apart? from the faces that would touch..

---

IF you've gone through the build logs then you'll see a pattern in the design of the routers built recently which are 90% (this one isn't...http://www.mycncuk.com/forums/gantry...ercnc_mk3.html) all raised rails above the bed like the pictures I posted previously. They'll also answer most of the questions you're asking about designing them. You don't have to go re-inventing the wheel! Most people on here want to cut the same things that you do (I do as I want to make parts for balsa and ply planes, carbon fibre heli frames and quad frames, ali parts for my helis and quad etc...) so their routers have been designed as a best compromise to achieve this. Re-read all the build threads to see how designs have changed along the build. Mine's a very good example of just this. I started with a similar design as the one you're doing and has evolved as I've read more build logs and asked questions...

The 'routercnc_MK3' is my machine and I'll explain why it looks like that.

It was much easier to build it like that based on a simple evolution of the previous version. It is only intended to cut balsa wood, liteply, and plywood and it works very well. I've been using it recently for cutting some 20mm thick aluminium upgrade parts, but that is not the main use.

If I do cut aluminium on my machine (6mm carbide 2 flute cutter) I can run at 1.0mm depth of cut and about 800mm/min feed. Any deeper and I get chatter.

If you want to cut a fair amount of aluminium the raised X axis designs are far superior and probably the best DIY CNC arrangement without simply converting a bridgeport mill to CNC. If you are starting from scratch there is no reason not to really.

Quote:

---

Originally Posted by njhussey

IF you've gone through the build logs then you'll see a pattern in the design of the routers built recently which are 90% (this one isn't...http://www.mycncuk.com/forums/gantry...ercnc_mk3.html) all raised rails above the bed like the pictures I posted previously. They'll also answer most of the questions you're asking about designing them. You don't have to go re-inventing the wheel! Most people on here want to cut the same things that you do (I do as I want to make parts for balsa and ply planes, carbon fibre heli frames and quad frames, ali parts for my helis and quad etc...) so their routers have been designed as a best compromise to achieve this. Re-read all the build threads to see how designs have changed along the build. Mine's a very good example of just this. I started with a similar design as the one you're doing and has evolved as I've read more build logs and asked questions...

---

Not trying to reinvent any wheel, I have lowered the bed...I am still going to keep the sides now as they are good for more strength and will also save me using 3mm poly as a way to stop rubbish going onto the rails...Im quite happy with this though it may need more support on the underneath... I need to make bracket on either end and either side of the machine to hold the motors and ballscrews...

My one concern is the accuracy in welding...if i take the parts to a welding shop will they know how precise this had to be? Dont want them to weld it up and realise its off by a few mm

Attachment 12534

Quote:

---

Originally Posted by FlyHighRC

My one concern is the accuracy in welding...if i take the parts to a welding shop will they know how precise this had to be? Dont want them to weld it up and realise its off by a few mm

---

Don't waste your money because they won't be able to stop distortion only lessen it and in any case you will still need to either use shims or Epoxy method.
Only thing you want to pay special attention to is Twist. If the base frame is twisted then the rails are not on the same plane, Epoxy method is good for sorting this but if you pay carefull attention when welding up then it's possible to get away with just light shimming of the rails. . . BUT with your design the Epoxy method will be much easier and more accurate than your capabiltys to measure rails are on same plane.!

Squareness of base frame isn't critical either square to 1 or 2mm is fine, obviously less is preferable. The important area is the gantry is square to the rails.
The base the cutting bed sits on needs to be resonably flat but again it's not critical as you'll surface the bed before using and this will bring it flat parallel plane.

Like Neil says you really need to go reading some of the build threads more closely and pick up the tips and you'll see whats needed and where your design Lacks or CAN be improved.

I have spent allot of time on the build logs so far. Im seeing allot of parts that require a mill and to me that dsrives up cost and complexity. Im sure it increases strength but above my computer I have K.I.S.S which is proving handy :D

A couple questions where I cant seem to find the answer..

When the epoxy is applied to the steel and sets, to bolt on the rail im assuming you just drill and tap? otherwise that must be a very fiddly job getting all them nuts down there haha

carriage space seperation, Is there an optimum distance between the bearings? im assuming the further apart the better but that obviously reduces work area so i guess there is a happy medium somewhere, right?

Quote:

---

Originally Posted by FlyHighRC

I have spent allot of time on the build logs so far. Im seeing allot of parts that require a mill and to me that dsrives up cost and complexity. Im sure it increases strength but above my computer I have K.I.S.S which is proving handy :D

---

If you made the gantry out of steel or Ali extrusion you wouldn't need a mill.

Quote:

---

A couple questions where I cant seem to find the answer..

When the epoxy is applied to the steel and sets, to bolt on the rail im assuming you just drill and tap? otherwise that must be a very fiddly job getting all them nuts down there haha

---

Depends on the thickness of the steel. If you've read Eddys build log you'll see he glued in 6mm flat and drilled and tapped into that...

Quote:

---

carriage space seperation, Is there an optimum distance between the bearings? im assuming the further apart the better but that obviously reduces work area so i guess there is a happy medium somewhere, right?

---

Pass...i settled on 200mm as a good compromise between losing cutting area and strength...



Sent from my Galaxy S5

Thanks, I prefer the epoxy method thatn gluing some flat bar onto the frame... Purely because it sounds simpler :) Thanks on that! I was thinking nuts inside the box section would be very tedious...any special taps Id need for steel? I have a set of taps but arent made for steel...

In terms of what i said about a Mill, I was talking more along the lines of motor mounts, Even saw one guy with a very nice machine who had his whole gantry cut by laser...

The flat bar was as well as epoxy...best sort of taps are spiral flute taps...buy a load on ebay from China they're cheap as chips. Most taps are made for steel, even the cheap black spur ones they sell in pound shops. If you've a good printer, a decent centre punch and a hand drill you can make a decent gantry. Sure a mill helps but ultimately care and attention to details and hand tools are enough if that's all you've got and you want to make a CNC router ;)

Sent from my Galaxy S5

Anyone got a SW file for the bit of a ballscrew that is bolted to the gantry? 16mm...I googled the dimension to make one myself but each picture seems to show slightly different dimensions...Why arent all thee bits generic across the board -.-

Its not that milling would be a problem, My taid (Wlesh for grandfather) has a workshop for model engineering...he builds those little steam engine that pull 20 people... Also I dont mind getting a few parts done on a laser either... my budget can go up a couple hundred which i can use some for parts to be made...allot of company quotes to come haha. If they all turb out to be stupid in price then ill do it myself.

Quote:

---

Originally Posted by FlyHighRC

When the epoxy is applied to the steel and sets, to bolt on the rail im assuming you just drill and tap? otherwise that must be a very fiddly job getting all them nuts down there haha

---

I used 5mm box section and drilled and tapped through the epoxy post 6 here:- http://www.mycncuk.com/forums/gantry...st-router.html is the epoxy being poured. Eddycurrent has used a similar system but he glued a flat plate with gorilla glue inside the box section to give more of a bite for the threads. ..Clive

Quote:

---

Originally Posted by FlyHighRC

.. How far should i keep the bearing carriages apart? ..

---

With a single ballscrew in the centre then racking becomes a problem whereby the gantry can rotate or twist if you are cutting anything towards the edge of the gantry. To minimise this you can space out the bearings further, but there is a compromise on travel.

With twin ballscrews all these problems go away, which is why it is highly recommended. The gantry cannot rack as a ballscrew holds it at each end. Your considerations then come down to placing the bearings so that the centre of gravity of the gantry (which will be somewhere behind the spindle centre line) and the cutting tool centre line both lie within the region between each bearing, as viewed from the side. If you need to deviate from these positions then don't worry as the profile linear bearings will take quite a load.

As I guide I would say between 100-200mm is about right, probably going closer the 200mm end of the range, giving you maximum travel.

for a good machine what is recommended, A router or spindle? cant see any bvious advantage or disadvantages to either... Yet to see anyone give a preference on any of the build logs

Also bear in mind that the deflection of the axis, due to the linear bearings, for a given cutting force is inversely proportional to the square of the bearing spacing. A relatively small increase in the spacing you choose can therefore make a big difference, assuming the stiffness of the system isn't dominated by another weaker part. The reason for this is that the force on each bearing decreases with their spacing and the deflection of the tool also decreases for a given bearing deflection if their spacing is increased, so overall you get a quadratic relationship.

Quote:

---

Originally Posted by FlyHighRC

for a good machine what is recommended, A router or spindle? cant see any bvious advantage or disadvantages to either... Yet to see anyone give a preference on any of the build logs

---

Simple answer is get a proper spindle. But please read the build logs more - people have already linked to some good ones in this thread, yet all the questions you've asked so far are already answered in the logs.

Quote:

---

Originally Posted by FlyHighRC

for a good machine what is recommended, A router or spindle? cant see any bvious advantage or disadvantages to either... Yet to see anyone give a preference on any of the build logs

---

I don't think you will find many routers on here being used. They are very noisy. The popular spindle is the 2.2Kw water cooled they run very quiet and will run at speeds up to 24,000 rpm. ..Clive

So I have nearly completed the first draft of my design, Im very comfortable with being able to make this machine with the tools and equipment I have access too. The gantry im happy with and the way in which the gantry is mounted to the rails. Im also happy with the location of the rails, steppers and mounts etc... the ball screw support blocks have yet to be mounted to a plate(next job) then onto the z axis :)

The frame is mostly 60x60x3mm box section with the gantries vertical section and support section being 40x40x5mm, Hopefully you guys approve this design...in which case I start my budget plan to make sure it can be built within my budget. Then progress to the parts list and DXF files for the laser cut stepper mounts and then go from there :)

Added some colour to give it a bit of life...

Attachment 12535 Attachment 12536

OK Mike, sorry I'm going to be blunt here. Jazz has said to go through the build logs as have myself, Jonathan and the others I still think you need to go back and do some reading as the design you have put here has been done before (or designs nearly identical to this one) and commented on.

Again being blunt:

Quote:

---

Originally Posted by FlyHighRC

Do you have a link to these linear rails?

I dont know much about CNC but i know how to incorprate strength into a design. If there is one thing Im good at its designing...dont get me wrong...i have seen builds on this forum which to me seem fine but people who have been there and done that have pointed out aspcts of designs that didnt show up to me...but all their critisims and expertise has been written down on my Pukkapad to bear in mind when designing my machine, which will be posted so others can tell me what needs improvong and whats ok etc etc.

Really appreciate the help :)

---

The design you posted is anything but strong...people will have pointed out things in other builds which to you don't show up because they have experience of building the machines, normally by making the same mistakes and regretting it. My brief comments on "first draft"

• Bed does not have enough support and will vibrate and bounce giving poor finish.
• X axis bearing mount to gantry is too thin and flexible.
• Mount the steppers with belts to reduce resonance, not essential but preferable.
• Gantry has no bracing or support so will bend/bounce/twist giving poor finish.
• 40x40 box (for the gantry) isn't strong enough as Jazz said before.

The above assumes that you will be cutting some aluminium at some stage.

Sorry my reply might come accross as being unfriendly and harsh but I can see you going down a route as Jazz said that will leave you dissapointed if you're not careful and if you're not careful you'll waste money prodcing a machine that won't cut what you want to cut.