[Dodgy Geezer]

Hi - I'm a new member, and joined because I'm aiming to make a simple and small CNC router for cutting 1/16" balsa sheet.

I've read through the FAQs - interesting and useful, but they tend to dwell more on the sophisticated and expensive side of CNC work. I'm expecting my first endeavour to be cheap - ideally, only using parts from my bits box. I have stepper motors from printers, some aluminium angle, MDF and some studding thread, which should be enough for a first attempt - but I still need to source some drive electronics.

I see some very cheap (under 10GBP) single axis drive boards from China, and it might be worth getting one of those to experiment with. But I would welcome comments about the feasibility of what I am trying to do, and advice about pitfalls to avoid...

[louieatienza]

Hi - I'm a new member, and joined because I'm aiming to make a simple and small CNC router for cutting 1/16" balsa sheet.

I've read through the FAQs - interesting and useful, but they tend to dwell more on the sophisticated and expensive side of CNC work. I'm expecting my first endeavour to be cheap - ideally, only using parts from my bits box. I have stepper motors from printers, some aluminium angle, MDF and some studding thread, which should be enough for a first attempt - but I still need to source some drive electronics.

I see some very cheap (under 10GBP) single axis drive boards from China, and it might be worth getting one of those to experiment with. But I would welcome comments about the feasibility of what I am trying to do, and advice about pitfalls to avoid...

If I remember correctly, a lot of these printer steppers have low step counts per revolution. Even then not sure how much power they generate and you have to gear them down, which will really slow things up. I see many good deals on steppers under $20 each. You can buy a LiniStepper or SlamStepper kit for very little and it will blow away any of those cheap Chinese drives, but will require assembly. The good ones like Leadshine cost a lot more. I've seen so many problems here with those damn TB6560 drives I wonder if people even research before buying them. And if less people bought them, maybe they'd try improving them.

While it is great to see first movement, that feeling leaves fast when you find your machine underperforming, so I suggest you read up on builds here and see what people are making with them. There is a recent MDF build here called MDF Madness and the guy's already making nice stuff with it.

[Drools]

How big of a cutting area do you need?

[vtx1029]

Microcarve has some nice machines. His BBX design could easily be built with a drill and table saw.

[Drools]

I had the Microcarve in mind when I asked about the cutting area size. The A4 might work fine with salvaged steppers.

[Dodgy Geezer]

How big of a cutting area do you need?

Thanks for all the responses!

An A4 cutting table would be heaps big enough - I was thinking of something about 1/3 that size. Though perhaps a bit longer on the long axis would be nice...

I run a small web site about the Keil Kraft EeZeBilt series of 1950s model boat kits. These are simple starter kits, die cut out of 1/16" balsa. Each sheet is about 10" x 3", and since I have drawn up all the lines I thought it would be possible to make up some pieces by machine rather than by manual cutting. The site is here if you want to see the plans... http://modelboats.hobby-site.com/

I don't expect that 1/16" balsa will require much in the way of cutting power, and was hoping to run the CNC cutter from batteries. Do you have a reference for this BBX? I'm still finding my way around the forum...

A typical stepper I have is the AIRPAX 57L048 - 7.5 deg per step. So each step might move roughly one thou with direct drive. I would be happy (initially!) with 10 thou total accuracy...

DG

[microcarve]

One thing that's sometimes overlooked with model building
with thin woods is that the bits used can't make perfectly
square inside edges in corners. The bits are round and you're
limited to what size diameter cutter for those corners.

Some things lend themselves to being laser cut better where
you can get those crisp sharp square edges.

If you want to DIY something quick and cheap, some of the drawer
slide machines can work fine...but tolerances will depend on the
quality of the slides...(and assembly)....

I'm pretty doubtful of the motors being useful.....but, strange things
have worked for others before. Your electronics and motors will
matter a lot if you want your parts to fit together reliably...and predictably.


John

[mmcp42]

I wonder if some sort of trailing knife blade would do the job for you?

[Dodgy Geezer]

I wonder if some sort of trailing knife blade would do the job for you?

I had thought about a knife blade mounted on a flat-bed plotter, but wondered if it would cut effectively across the grain. Plotters have very little extra power. So I suspect I will need to build a table. For cutting I was expecting to use a 12v model motor with a small diameter dentist burr.

Microcarve - a small diameter curve would be fine in many cases - it would help to strengthen the corners which can be prone to splitting in small balsa sheets. A laser sounds ideal, but it also sounds expensive, hard to source and dangerous.

Most draw slide assemblies I know have about 1/10" sideways motion - a bit too much, even for me. So I am currently thinking of V blocks of nylon running in a V section of aluminium. But it is the motors and associated electronics that I will have difficulty with. I may well not be able to get enough torque to drive home-made rails reliably, even with a very light load...

[microcarve]

You can outsource laser cut stuff to places like this...

3D printing, laser cutting

there are several of them that can be fairly reasonable...

The drawer slides can work if they're preloaded against each
other. I've seen some do Very Nice work....check for videos on
youtube. That's a great place to get ideas.

A Hobbycnc driver kit is about as cheap as they come for what
it can do for you. It can run mid-sized machines very well.
Motors...those are best bought with what you want from them
in mind. There are thousands of variations, but most of them
aren't anywhere near ideal for DIY cnc work.


John

[Dodgy Geezer]

You can outsource laser cut stuff to places like this...

3D printing, laser cutting

there are several of them that can be fairly reasonable...

The drawer slides can work if they're preloaded against each
other. I've seen some do Very Nice work....check for videos on
youtube. That's a great place to get ideas.

A Hobbycnc driver kit is about as cheap as they come for what
it can do for you. It can run mid-sized machines very well.
Motors...those are best bought with what you want from them
in mind. There are thousands of variations, but most of them
aren't anywhere near ideal for DIY cnc work.


John

Since I will be prototyping a lot and experimenting with materials I didn't really want to outsource the cutting. Some of those drawslides actually look very good - if you can get big ones and tune them up. They might be an idea for a Y axis, where I really only need 6" or so. But it might be nice to get an x axis running to 24", which is a bit big for the drawer slides...

The HobbyCNC driver boards look fine - but remember that what is $60 in the US turns into about $150 over in the UK, with shipping, VAT and admin charges. I really need UK sourcing...

[louieatienza]

Thanks for all the responses!

An A4 cutting table would be heaps big enough - I was thinking of something about 1/3 that size. Though perhaps a bit longer on the long axis would be nice...

I run a small web site about the Keil Kraft EeZeBilt series of 1950s model boat kits. These are simple starter kits, die cut out of 1/16" balsa. Each sheet is about 10" x 3", and since I have drawn up all the lines I thought it would be possible to make up some pieces by machine rather than by manual cutting. The site is here if you want to see the plans... http://modelboats.hobby-site.com/

I don't expect that 1/16" balsa will require much in the way of cutting power, and was hoping to run the CNC cutter from batteries. Do you have a reference for this BBX? I'm still finding my way around the forum...

A typical stepper I have is the AIRPAX 57L048 - 7.5 deg per step. So each step might move roughly one thou with direct drive. I would be happy (initially!) with 10 thou total accuracy...

I think most people would be happy with that accuracy. I don't know if drawerslides are made to that precision however. You're only as accurate as the surfaces of the drawerslides are machined and if the play in them (after preloading) is less than the accuracy you need, meaning of the play in the drawerslides are less than .001"... under load. It isn't a trivival thing to do even on a commerical CNC. I would strive to make your machine as sturdy as you can, and aim for good repeatability.
DG[/QUOTE]

[Dodgy Geezer]

I think most people would be happy with that accuracy. I don't know if drawerslides are made to that precision however. You're only as accurate as the surfaces of the drawerslides are machined and if the play in them (after preloading) is less than the accuracy you need, meaning of the play in the drawerslides are less than .001"... under load. It isn't a trivival thing to do even on a commerical CNC. I would strive to make your machine as sturdy as you can, and aim for good repeatability.
DG

[/QUOTE]

I'm sure that drawerslides aren't made to a great precision - they actually need a fair bit of play, otherwise you could only close a drawer by pushing it precisely at 90 deg! But I was impressed by microcarve's comments - his standards seem much higher than mine, and if he says that it's possible....

Since I need a small system, sturdy should not be a problem. But an ideal size for this would be a work area 4" x 36" - to take a standard length of balsa sheet. 4" is no problem, but I don't expect to do 36"; particularly not with draw runners!

At the moment I need to sort out a way to control the steppers from start to finish - at low or zero cost! I've loaded the trial version of KCam, Gcoded one of my DXFs, and run a test with an RS232 interface box on the parallel port so that I could see the output. I do get some, though I'm not sure what actual signal is going down each line!

I can probably stretch to one of the cheap chinese single-axis drivers advertised on ebay for about 5GBP. At that price, getting the box to turn up is probably as much as I can expect, but I hope that I can connect it up and make a small stepper turn. That would be enough to give me some idea of torque and hence the sort of power I have to play with. If it doesn't do anything, that will be a good learning experience, finding out why it doesn't work...

[microcarve]

Yep, what I had in mind was pretty small for the drawer slides....

Your electronics/motors are important because with an ultra-cheap
first attempt at building the mechanics, there'll likely be some
error and you'd want at least the power to overcome those
mechanical glitches. Weak electronics will bind in places and
it can be very frustrating because you can't be sure if it's the
driver, power supply, or some mis-alignment.

At least with a good driver, you can narrow problems down a
lot easier. And, once you see it move, you'll want to begin
the inevitable upgrade path...mechanics and electronics.
(already thinking of that 36" version...ehh....

Check out instructables for some super cheap designs...

Technology - Cnc - How to Make Instructables

We all begin somewhere....


John

[Dodgy Geezer]

...At least with a good driver, you can narrow problems down a
lot easier.....


John

My main concern at the moment is not understanding enough about how to run the motors. I don't, for instance, know if a breakout board is an essential, or a nice-to-have. Though I understand the theory, I don't understand what the connections do on the ones I can see on ebay. And I don't know where the price/performance break points ought to be. Until I know what's essential and what a good price is, I would rather buy cheap and learn by finding the shortcomings. But only if I can buy VERY cheap...

[microcarve]

It's Very possible to build a good, very precise small machine from
pure junk. It may take 2-3-7 times to get it right enough, but
it's entirely do-able.

I doubt it's possible to run it underpowered, though. If it's too
slow and weakly powered, your high speed spinning bit will just
burn the wood. You will need at least some useful force to drive
it. Stalling and ruining good stock gets expensive...

Many have bought the cheap import drivers and with persistence,
got them to work fine. Overall though, they're generally viewed as
more trouble than they're worth. Can't hurt to get one to play around
with, though...

Nope, you don't necessarily need a breakout board.

It all takes a lot of homework. But, it's worth it in the end....


John

[louieatienza]

My main concern at the moment is not understanding enough about how to run the motors. I don't, for instance, know if a breakout board is an essential, or a nice-to-have. Though I understand the theory, I don't understand what the connections do on the ones I can see on ebay. And I don't know where the price/performance break points ought to be. Until I know what's essential and what a good price is, I would rather buy cheap and learn by finding the shortcomings. But only if I can buy VERY cheap...

You can buy a basic breakout board for $10. Instead of soldering (and possibly resoldering and resoldering) you can make connections via screw terminals. In the long run it may be a wise investment. Buying cheap can end up costing if it means having to rebuy or hours or days of frustration.

If you really want to go cheap there are many open-source stepper driver plans avavilable. But what you save in money you end up spending in time; the point being you have to decide whether building a CNC is a means to an end, not the other way around.

The great thing in this age of almost instant information is that by just a little bit of research, you can minimize trial-and-error, sinply because it's been done time and time again. There are many designs, such as John's, or the many skate-bearing designs, that are proven to work. So while there may be pitfalls, by doing due diligence researching here, and asking questions, hopwfully those pitfalls will be minimized.

[Dodgy Geezer]

It's Very possible to build a good, very precise small machine from pure junk. It may take 2-3-7 times to get it right enough, but it's entirely do-able.

Thanks for that - that's good to hear...

I doubt it's possible to run it underpowered, though. If it's too slow and weakly powered, your high speed spinning bit will just burn the wood. You will need at least some useful force to drive it. ...
John

Hmm... I understand that milling 1/2" MDF requires a good sideways push if steps are not to be missed. But I'm talking about 1/16" balsa. I'm hoping that a decent thin cutter will move through that as if there is nothing there...

[louieatienza]

Hmm... I understand that milling 1/2" MDF requires a good sideways push if steps are not to be missed. But I'm talking about 1/16" balsa. I'm hoping that a decent thin cutter will move through that as if there is nothing there...

The thing is, cheap drive, coupled with underpowered steppers and inexpensive leadscrews will equal slow feedrates, and for wood you'd need higher feedrates to avoid burning the wood and dulling bits. I use a .024" endmill on ebony and rosewood, infinitely harder than balsa, taking .024" passes at over 60ipm. I'd think you'd need at least that for cutting 1/16" balsa which you should do in one pass. Also the slower your machine, the more a chance for your battery powered Dremel to lose its charge.

This is not to discourage you, I mean this more as to what to expect. Just because you use a 10tpi leadscrew and have steppers that have .72degree step resolution doesn't mean that you all the sudden have .0002" accuracy. Far from it. As to the steppers, many of the .72deg. steppers I've seen are 5-phase, so you may have to do more research before purchasing a drive.

[Dodgy Geezer]

The thing is, cheap drive, coupled with underpowered steppers and inexpensive leadscrews will equal slow feedrates, and for wood you'd need higher feedrates to avoid burning the wood and dulling bits. I use a .024" endmill on ebony and rosewood, infinitely harder than balsa, taking .024" passes at over 60ipm. I'd think you'd need at least that for cutting 1/16" balsa which you should do in one pass.

Remember that balsa is very soft and open-grain. In cutting terms, it's probably like expanded foam. I was expecting to use a dental burr of about the size you mention, but I wasn't expecting to get burning, no matter how slow the tracking. One pass, as you say, and, of course, it would be nice if a 100 step part could be completed in a matter of minutes.

Also the slower your machine, the more a chance for your battery powered Dremel to lose its charge.

Oh, I wasn't expecting to use something as bulky as a Dremmel. I suspect that the Y axis stepper could hardly move such a weight! I was expecting to use a small 12v model motor - perhaps 100g/4oz. The whole thing powered from a 12v/24v car battery or similar supply...

As to the steppers, many of the .72deg. steppers I've seen are 5-phase, so you may have to do more research before purchasing a drive.

Indeed. This discussion is helping me to understand what questions I ought to be asking. The motors/drivers are going to be key. If I use ex-printer ones, we're talking about 7.5deg steps. They seem to have either 4 or 6 wires, but I don't know what phase that implies...