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

Not at all :) If the design isnt up to the job then its not up to the job. Criticism is how all novices get to a good stage...just like RC haha. I have the CAD files for all th parts I need now, with a number of them provided by Jazz...so the rest is just making a new frame which can still utilise these parts :)

The one thing i really dont understand is how something has to be so strong and precise but are hand welded most often by people who have a mig and no jigs...

Is it possible that the design I have done can be beefed up? with no real change to the machine? Just to save going back and forth...I dont mind, I have taken everyones advice onboard...I know it looks like im rushing into everything but I hav been doing allot of studying on the forums with build logs, The questions I ask which seem similar to other peoples questions are because I want to put it in my own words so I understand it better...

The criticism is good though, didnt show as unfriendly at all, no worries there :)

Good man...things can be welded by novices (me!!!) but still precise as long as either some form of adjustment (shimming the top rail on the x axis for exampke) or the epoxy method is employed.

Yes you can still use your design if you look into the gantry a bit more. Bracing is your friend as wood needs to be cut quick so on full chat your machine will be dancing about (how you fixing it down by the way?) all over the place...

The best way to get stiffness in your gantry is to keep the sides as short as possible with the best being to mount it directly on the linear carriages...this needs raised rails.....

Sent from my Galaxy S5

Quote:

---

Originally Posted by FlyHighRC

The one thing i really dont understand is how something has to be so strong and precise but are hand welded most often by people who have a mig and no jigs...

---

The trick is to be creative about how you get the accuracy. Making a reasonably accurate (+/- 1-2mm) bed is possible with a reasonable amount of skill and care. If this is properly designed it can then be made very accurate using epoxy to level the important faces. On to those now accurate faces you bolt bought in accurate components (the linear rails) and so on.

So, you don't have to worry (too much) about accuracy of the welded frame but you do have to consider its stiffness - it would be worth you reading up on some basic engineering - in particular stiffness of beams. This is not just a 3d design problem, it's also an engineering problem and although you don't need to be able to do accurate stiffness calculations for each part of the machine, you do need to appreciate how the material responds to stresses.

Thats good then. If i can still use this general design that would be nice. In terms of the gantry, where the two braces are connecting the two sides, is that where I need to brace? I,e Add vertical members between the two?

Are you a Solidworks user Neil?

Yes you can still use your design if you look into the gantry a bit more. Bracing is your friend as wood needs to be cut quick so on full chat your machine will be dancing about (how you fixing it down by the way?) all over the place...


Sent from my Galaxy S5[/QUOTE]

I didnt include it on the solidworks model since i didnt feel the needs but there will be 6 tabs around the bottom of the machine with m6 bolts to bolt it to my other workbench, depending on how much vibration there is i might replace m6 vibration isolation...i use m3 equivalents for quadcopters

maybe if this all works out well i can make it an early last year uni project :D

Quote:

---

Originally Posted by FlyHighRC

Thats good then. If i can still use this general design that would be nice. In terms of the gantry, where the two braces are connecting the two sides, is that where I need to brace? I,e Add vertical members between the two?

Are you a Solidworks user Neil?

---

Got SW but not installed it...I've got Geomagic Design on my PC at work so trying to get used to that at the moment so won't install it as I don't want to confuse myself (easily done!!)

You need to put in vertical and also cross bracing....like you would in a CF quad arm...

Neils bluntness was required here because if you are looking at other builds, esp those built from steel then your not paying attention because you'd have seen NONE are built flimsy regards the Gantry and 99% are done using timing belts and pulleys for connecting to motors. Your design at best is medium duty wood use only.!
Reason for Pulleys is Resonance handling, I'm not getting into how's and why's because it's been covered in many threads so go check some out, But will say it's an important thing on steel framed machines so wouldn't skip it.

Gantry needs to be much more beefed up and braced with plenty of thought to rail alignment and Ballscrew Fixing/adjustment/access. Your design severly lacks here, esp on the Twin axis area. Ballscrew alignment and fixing needs to be strong and easy to setup, if the ballscrew end bearing mounts are flimsy then the screw will resonate which can cause whip and affect performance. If bad then it will also have the similar affect has backlash and causing inaccurecys so design these areas strong and adjustable.
Just remember you have two goals Strength and accurecy, and you can't have one without the other. Build in has much adjustment as possible to allow for the DIY factor, this is KEY to successfull DIY build and stress free build.

Don't be afraid of weight it won't be a problem at this size and will actually give positive affect not negative.

One more note.? . . You've drawn the Gantry bearing plate as Angle iron, this won't be strong enough or accurate enough unless Thick material and prep'd. What you have drawn is has much use as tissue paper.!! . . . The Profiled linear bearings are very intolerent of uneven surfaces and will cause binding if fastened to one. Angle iron isn't flat so it would need surfacing and it will need to be thick material to reduce resonance. Just Fastening to the bench won't stop resonance from weak frame material.
Also beaware it's not a good idea to bolt the machine to the bench unless the bench is perfectly flat and true other wise you'll pull the frame into twist and force the rails out of plane. To be honest this machine won't be light so you won't need to fasten down anyway just level bench and put some buffers to stop machine rocking off the bench.

Jazz - I know you seem to favour single X motor and belt drive to two screws (for this general size machine, anyway). For a steel machine where resonance might be a problem, is it still better to have a single motor, or two motors with their own drivers to give the drivers the best chance of controlling resonance - and still having motor/screw coupling via belts?

Quote:

---

Originally Posted by Neale

Jazz - I know you seem to favour single X motor and belt drive to two screws (for this general size machine, anyway). For a steel machine where resonance might be a problem, is it still better to have a single motor, or two motors with their own drivers to give the drivers the best chance of controlling resonance - and still having motor/screw coupling via belts?

---

Actually Neale the reason I like Single motor connecting two screws has nothing to do with resonance really just I prefer because of keeping screws in sync.
In either case single or Twin motors belts to motors are best on steel machines and to be honest I fit them on machines I build from aluminium profile and Ali plate for same reason plus they allow flexabilty with ratios if required.

One thing I will say thou is that the newer Digital drives are getting so good at handling resonance or allowing it to be tuned out that it's becoming less important but I still prefer to have them for reasons above.

Edit: Just realised I didn't actually answer your question did I.? Not a politician honest. . Lol

Well one motor is best really due to not having to worry about sync and each drive keeping perfectly together or resonace just affecting one motor/drive not the other etc plus less to go wrong etc. . . .BUT. . . here's the crux with one motor.! . . . It means larger motor is needed and often this throws you into Nema 34 country and with that comes other side affect's like Larger PSU required more expensive drives etc. In reality It doesn't often work out any cheaper than 2 motors/drives really but it does give piece of mind the machine won't tie it's self in knots if one motor stalls and it NEVER Racks or one side loses position if motors are tuned in-correctly.!

Again thou modern Digital drives are helping here with Stall detect and fault signals to help E-stop the system if happens so not so big an issue really. I only Use Digital drives know and Motion control cards which really do help with good fast quality pulse signals so because of this I'm using more 2 motor setups for simplicity sake and avoiding long belt runs.!

Many thanks for the comments. I'm still building the frame for my new machine, and haven't reached the stage where I have to worry about single or double motors. It brings up a bunch of related issues, so I'll stop hijacking this thread and start another one.

One last vaguely relevant-to-the-thread point, though - I started out a couple of years ago with my MDF JGRO machine in order to make an engraved presentation plaque (immediate need) and the thought that I would be able to cut things like wing ribs, fuselage sides, etc. Haven't done any of the latter yet, but found lots of other uses for the machine that push the (admittedly very modest) limits of what I have. Spent the last couple of days profile-cutting 6mm and 9mm ply components for a slot-together architectural model, and watching the wheels go rounds at a mind-blowing max cutting speed of 700mm/min was tedious. The moral of the story is that if you are happy to cut modest amounts of material at slow speeds, you can get away with a real junk machine. But if you want to be able to do a good job at proper speeds, do it properly first time!

Well for me right from the outset I wanted a mchine that is future proof for me (Hence why these guys are pushing so hard ;) ) so that way i can cut plastics,woods and do so ali work in the future and have a machine capable for it all. I was first introduced into CNC at school and was quickly dissapointed with the machines they had... werent very good. Then I got to start using a Roland MDX-540 as part of a young engineers thing.. that machine was better and could chew through plastics...struggled with carbon fiber and did aluminium at 400mm/min, not sure if it could go quicker or the company didnt want to push it but it was annoying to watch frankly.

Would this be a better frame? Following a design more towards what other people are making...
Attachment 12541

Oh building a machine to now combat resonance..meaning pulleys..those pulleys will need a special mount t fix onto the ballscrew right? This is now sounding more and more expensive... Please try to remember my budget :D

Quote:

---

Originally Posted by FlyHighRC

Oh building a machine to now combat resonance..meaning pulleys..those pulleys will need a special mount t fix onto the ballscrew right? This is now sounding more and more expensive... Please try to remember my budget :D

---

Nothing special needed for using pulleys and the only extra expense are the pulleys & belts roughly £15 per axis. Which isn't much more than direct mounting by time you have bought correct couplings.

ahh yeah, just looked at this site, not too bad at all 31T10/25-2 | Ondrives

Im revamping my old design, the frame im happy with...i cant see there being much wrong with it after i strengthen it which was the main critique with it. So I have stiffened that up now. Also the gantry is sitffened up which I will show later, mian change is 2 pieces of 6mm plate welded onto the box section and did away with angle Iron(Which I call L Bar) where the Lbar was is now replaced with 6mm steel welded onto the boxsection

Green is 6mm plate, Note, I will need to extend the bottom 6mmplate awayslightly to make room for the bolts close to the vetical 6mm piece

Attachment 12542

As a matter of interest have you thought about how you will fix the ball nut and bracket to drive the Y axis it seems you are running the screw through the uprights in the gantry. Also there are cheaper places to buy belts and pulleys. ..Clive

Yeah I realised that haha, Did it at half 7, Ill put it to the mrning haha, those uprights in the gantry will be behind the horizontal box sections

Quote:

---

Originally Posted by FlyHighRC

ahh yeah, just looked at this site, not too bad at all [url=http://www.ondrives.com/p/9519/31t1025-2]31T10/25-2 | Ondrives[/url

---

Nah too expensive try these people or go to your local bearing supplier as they are often cheaper and will sell without Vat for cash.
http://www.beltingonline.com/timing-pulleys-bars-272/

Your gantry is looking slightly better but still needs work. Also by the looks of it you haven't given much thought to how the ballscrew end bearings will mount and how the motor will connect.? Has it stands now it can't be done without special machining and using FF/FK bearing blocks.

10mm plate would be better for the bearing plate 6mm is still a bit flimsy.! Can't afford any flex or resonance here.

Quote:

---

Originally Posted by JAZZCNC

Most machines will scratch Aluminiun away but to cut it correctly and accurately requires rigidity and NONE of these have enough to cut aluminium to any decent standard and survive.

---

I must be misreading this, I thought that Jazz and me would see eye to eye about routing aluminium, I did go to Specsavers so it can't be that.

Did I see that the proposed mill only requires 10mm in the Z axis? Was that a typo? People usually allow 6-12 inches on the Z axis just in case they need it in the future thus destroying the machine for what they actually do. I think 2" or 3" in the Z is optimal if you want 10mm.

If you are going to rout aluminium I further suggest you copy the milling machine. On a milling machine the long bed is only supported in the middle because that is where the cutting happens. No point supporting it where the cutting doesn't happen, let it overhang, no problems.

To make your router strong, bolt Hiwin carriage blocks upside down to something solid such as a concrete joist.

Next bolt your gantry to the same joist so the table can slide through underneath it.

Why not make the gantry out of a second concrete joist? You don't have to move it so go wild on the dead weight.

If you don't like this idea, the usual excuse for not going down this route is, "It takes up twice as much space in the X axis and I haven't got room".

If you want to cut aluminium, design in a flood coolant return from the start, anything else is a ghastly fudge.

Quote:

---

Originally Posted by JAZZCNC

Nah too expensive try these people or go to your local bearing supplier as they are often cheaper and will sell without Vat for cash.
Timing Pulleys

Your gantry is looking slightly better but still needs work. Also by the looks of it you haven't given much thought to how the ballscrew end bearings will mount and how the motor will connect.? Has it stands now it can't be done without special machining and using FF/FK bearing blocks.

10mm plate would be better for the bearing plate 6mm is still a bit flimsy.! Can't afford any flex or resonance here.

---

Do you have any SW fil;es for the pulleys? what Pulley width? Im assuming 16mm should do it? That belt must be very strongi guess? When the gantry comes to a sudden stop wont its inertia and the belt cause it to pendulum a little?

Right, I have beefed up the gantry quite a bit now, Though the problem IM having now is difficult access...

As you can see from the pictures below, having mounted the block the ballscrew doesnt extend far enough so i though i could add a coupler than add some 6mm rod which will go on one end of the coupler and the other end, the pulley. Pretty complex way so unless you guys say thats ok ill scrap that idea... would also need to cut a whole in the vertical box section to be able to tighen the coupler

Attachment 12543 Attachment 12544

How's your ball nut on the gantry going to connect to the Z axis? The way youve got it its going to be impossible to connect...

Sent from my Galaxy S5

Yeha i know haha, did that at like 7 right out of bed thus was a design that cant be done haha. They will be behind the horizontal box sections :)

Is the pulley mounting system ok? Or in need of a severe redesign?

It might be easier to use one large piece of box section for the gantry. Something like 150mm by 75mm, that way you have one nice flat surface after expoxy leveling to mount all your rails and ballscrews bearings. You also need a way of tensioning the motor and pully belt, slots in the mount would be fine. You might want your slots in the same axis as the center to center of the motor and pulley?

Another thing, you want to keep the pulley on the ballscrew as close to the bearing as possible.

Another thing, you want to keep the pulley on the ballscrew as close to the bearing as possible.[/QUOTE]

so now coupler as a way to extend? :/ If so that allot of travel lost in the y axis... as i would need to bring everything closer but keep machine dimensions similar..

i think the best thing to do is a gantry re design perhaps??

Check out the other builds and find something similer that works

>as i would need to bring everything closer but keep machine dimensions similar..

Ahh - compromise!

Hey man, reading through the thread I can see you're very keen to get this done, you're having to take a lot on board in a very short space of time and you have the self-belief - all of this is EXCELLENT. You can get something good built for the money you're talking but given that it's your first build, you just can't rush it.

QUICK CHEAP GOOD

You can have two of the above..!

It is possible to over-think things and there does come a time when you have to take the plunge and just start building, but to put things in to perspective - I spent about a year (more, in fact) on my current machine (most of this was going back and forth design wise) and as it stands it weighs over 100KG (without the base) can cut aluminium/brass quite comfortably and accurately (for my needs: ±0.05mm) and has a working area of around 200x200x150mm. I'm happy with it, but there are certainly things I'd do differently if I was making it over again.

What I'm saying is this - if you head down rush into it you may as well buy one of those machines in your original post. It needn't take you as long as what my build took me, but don't expect to have this done within a few weeks.

Have a look at how your grandad works - is he any good? I'll bet he works thoughtfully and steadily if he is...

Just my two cents.

Wal.

Can I just say you clearly haven't been doing much research despite your claims that you have.? Why because if you had been looking carefully you wouldn't have made so many errors regards mounting things and the COCKUP with ballscrew running thru those supports shows this clearly.!

Your just wasting time at the moment because your not looking at the details and it's the details that matter most. Anybody can fudge together an outline of a CNC machine but building one requires certain Details are 100% correct and your not seeing them because you haven't looked/read closely enough at other builds. Asking questions is good an I encourage it but asking questions that are clearly in other threads that you have been advised to read is just lazy or you haven't done what you said you had.!!

I'll give you this one last bit of info which you should have seen from other builds then I'm not giving anymore I see which should have been obvious in other threads pointed out to you.!!

15mm wide timing belts, HTD series 5mm pitch no less than 18T pulleys because there isn't enough material left on boss for decent Grub screws after boring, plus better tooth engagement. I use 20T all the time when 1:1 ratio and 18/36 when 2:1

Now go back to reading other threads and start paying attention to the details.!!

Quote:

---

Originally Posted by Robin Hewitt

I must be misreading this, I thought that Jazz and me would see eye to eye about routing aluminium, I did go to Specsavers so it can't be that.

---

No Robin we don't agree at all. I'm referring to the machines suggested lower down but your still banging on about that old chest nut that Routers can't cut aluminium properly.! . .Which is frankly stupid because there are many of us on here that do it quite nicely thank you.!

Agreed on the concrete machine thou, correct Concrete and epoxy can make a very solid machine.!!

The cockup with the ballscrew going through the vertical box sections was a stupid mistake because i was tired...this is CAD..i can make a silly error when im half awake...

In terms of asking questions and reading through other build logs, I have done reseach, maybe not to you exacting needs but done what i consider an acceptbale amount. Things like puley dimensions, im nt going to go rolling thourhg everything an read 99% of things that are all disimportant to me when in 30 minutes one of you guys who knows what your talking about and are being helpful can let me know... now of course,someone could perceive that as lazy...but then how would forums like this be created without that form of contact between members.

Just want to be able to build a good machine, So far you have all helped significantly...redesigning the gantry at the moment. Im not being all 'I know better' which there are a few people like that on most forums of any sort... I changed the whole table layout, gantry and am now redesigning the gantry all together... Your advice so far Jazz has been top notch stuff so dont be upset with me for recognising you know what your doing thus having me(a complete novice in this field) come to you and other experienced members questions you may have already answered before.

You are very correct, I have always thought about building a CNC though thought that with £1K i could probably just buy one...then apparently not but then you gus reignited that old idea to build one :)

Yeah im trying not to rush but i do know I can build a CNC machine, I have designed,built and flown functional autonomous survey UAVS before...I can build a CNC. ...But i see exactly your point... Design amtters allot here, as does money which is why your all concerned im getting ahead of myself which i often seem to be doing but i know when im going beyond my limit. My grandads work is very nice, but when it comes down to his work away from his machinning then its similar, paper and his steel haha. He is currently building a racing lawnmower which he has raced a few times...crazy fun thing but it depends on that job at hand.

I am getting less done per hour on this than designing a UAV which is simply down to the fact this is all totally new for me hence spending more while geting less done.

Hope i dont seem groggy haha, Just want a solid design so i can get the price list sorted...

Quote:

---

Originally Posted by FlyHighRC

In terms of asking questions and reading through other build logs, I have done reseach, maybe not to you exacting needs but done what i consider an acceptbale amount. Things like puley dimensions, im nt going to go rolling thourhg everything an read 99% of things that are all disimportant to me when in 30 minutes one of you guys who knows what your talking about and are being helpful can let me know... now of course,someone could perceive that as lazy...but then how would forums like this be created without that form of contact between members.

---

Exactly the problem with young people of today.!! . . .They want everything handed on a plate and they want it NOW.

Others and I are not telling you to go away and read other threads because we don't want to answer your questions which we've answered hundreds of times before.! It's about gaining insight and knowledge of what's required and why they choose the parts they did and also the Cockup's they made(if there honest) along learning what makes a GOOD CNC machine tick.
I'm long in the tooth at this game and I've seen and dealt with every type of person/builder and by far the ones who make the best machines with least trouble are those that put in the research and gained the knowledge plus listen before putting pen to paper. Your on a Budget and I can 100% guarantee that if you don't put in the time to research properly how things work and fit together you'll blow that budget out the water.

You are being Lazy by the very fact your asking simple questions that should have been clear if you'd read other threads. It's exacting details that you need to look at because while modelling is great tool for design it's only as good as the detail put into it. When the build begins then the bullshit stops and it's here you realise that because you didn't pay attention to details and model them into your design you suddenly find the Gantry won't move like you thought because you just generlised the part in CAD and now it won't clear this or that.!!

I'm not being high handed here or youth bashing I'm trying to pass on valuable hard learnt experience and Like Wal say's I'll bet your grandad never rushed a single part of his steam engine and I'll also 100% guarantee he did extensive research and took very careful measurements before making any part of it.!!

Id rather make the plate... ;)

I am taking it all on... As he noticed i do want a machine up and running asap but thats because I need a CNC to help develop a Survey UAv that is designed for my fathers 3D visualisation company. Thats why I want it quick, never mentioned that because its never been relevant until someone calls me up on it!

Maybe, But its helped me quicker... The noly thing I havent learnt much about so far is the elctronics but im going to make the decision that they arent all that different from 3D printer elctronics exacept maybe bigger supplies etc...

My main concern in building this machine is the electronics, I hate them. Do my head in! Im studying Aeronautical & Mechnical Engineering at Uni with a Electronics module...swear to god its magic!

Quote:

---

Originally Posted by JAZZCNC

No Robin we don't agree at all.

---

That's a relief. I was just doing the forms to renew my gun/explosive certificates, can't afford anything that might cast doubt on my sanity.

This must be the longest thread on here with nothing achieved :cower: ..Clive

I resent that :P I ahve learnt allot from this bunch, also I reply quick when I want to learn quick..seems most post on this forum are responses 2 weeks apart...

Quote:

---

Originally Posted by FlyHighRC

I resent that :P I ahve learnt allot from this bunch, also I reply quick when I want to learn quick..seems most post on this forum are responses 2 weeks apart...

---

They will be if you don't start reading and stop asking annoying questions you should have learnt. .:stupid:

Wasnt refering to the people on this thread, You ad two others have been superb in terms of replying :P A compliment :P

Oh dear, I think we might be going through another 12 pages of posts on the electronics. Router electronics are to 3D printer electronics as F1 cars are to go-karts. The principles are the same but...