3 Axis CNC router

first option is quite difficult , my first build i did that ...

second option is way easier, especially if you use epoxy to level.

both my machines are steel welded . i guess using aluminum extruded will be easier for option 1

The first option is, as Jazz says, like a cornflake packet with the ends removed, just think how stiff that is! If you move the rails nearer the face you'll reduce the flop. I've done this on my gantry and it seems plenty stiff to me. Second option puts the router further out from the face. Both will work...

Quote:

---

Originally Posted by njhussey

The first option is, as Jazz says, like a cornflake packet with the ends removed, just think how stiff that is! If you move the rails nearer the face you'll reduce the flop. I've done this on my gantry and it seems plenty stiff to me. Second option puts the router further out from the face. Both will work...

---

Hold on here Lads this is Jo's thread so lets not hi-jack it hey.!! . . . . Suggest Start a new thread if required.!

Edit: Used your Quote Neil to high light point being last post on it not having a go at you mate. . :thumsup:

I don't mind the highjack interesting to read... Although I think the cornflake packet is where I'm headed! Jazz when you build your L gantry machines which option do you use? I thought I had seen you use rails top and bottom?

Quote:

---

Originally Posted by JoeHarris

I don't mind the highjack interesting to read... Although I think the cornflake packet is where I'm headed! Jazz when you build your L gantry machines which option do you use? I thought I had seen you use rails top and bottom?

---

Yes top and Bottom like this.

Attachment 16100

Or This without tall sides.

Attachment 16101

Hi Jazz,

So your small design above doesn't have a back piece on the Z axis assembly, but has side panels to give strength to the joints?

Looking again at rail location, would one rail on the bottom and one on the front be a good combination? Like this machine:


http://www.mycncuk.com/attachment.ph...id=16103&stc=1

Quote:

---

Originally Posted by Tenson

Hi Jazz,

So your small design above doesn't have a back piece on the Z axis assembly, but has side panels to give strength to the joints?

---

Yes doesn't have back piece because it add's no strength just weight. The Panels you refer to are not for strength, thou they do slightly add some. It's purely a Cover to protect the whole Z axis assembly from dust and crap. It's very affective and neat.

The 90Deg joints are perfectly strong enough for any type of work upto and including machining aluminium.!

Regards rail location then it's a personal choice and also partly machine design related. Me personally I prefer to keep things stout and compact and gain maximum real estate. But if you need clearence for ballscrews etc at the front then putting rails on the front allows this and keeps it stiff. Each way has it's own Plus's & minus's. Both work equally well so take your pick.!!

Bit of an update. I've bought some plate and started to make some of the aluminium parts. Tried out my new counter bores too which were an absolute joy to use [emoji3]Attachment 16896Attachment 16897Attachment 16898Attachment 16899Attachment 16900

Oh and the model has changed again too! Attachment 16901Attachment 16902

Looking good Joe coming along nicely

How do the two gantry profile sections affix together? I can see a ali bar in the T-slot but I'm not sure still.

Tenson- I tried to explain this in a previous post on this log - http://www.mycncuk.com/showpost.php?p=72346

Attachment 16927milling aluminium with a router was fairly successful. More successful than I expected anyway... Attachment 16928Attachment 16929Attachment 16930Attachment 16931Attachment 16932 just need to wait for my m12 counterbore and some m12 capscrews of a sensible length to arrive! and the gantry will be well on the way.

Just goes to show what can be done with simple tools, very nice

Well done Joe. My first machine was hand cut with router so know how scary cutting ali by hand is for first time. . :thumsup:

Looking good Joe. :)

Great work Joe. Solid looking machine in the making . . .

:thumsup:

Other side cut... Wishing I had more space! Attachment 17033

Looking good Joe, coming on nicely!

Quote:

---

Originally Posted by JoeHarris

Other side cut... Wishing I had more space! Attachment 17033

---

Hi Joe,

I have read all of your build and all of the helpful comments with great interest. You are building more or less exactly the machine that I have pondered about for a long time.
Do you have an idea of what the final cost will be when done?
Thanks
Pete

A lot! I think the build will be the around 1500 to 2k? But to be honest I've kind of lost track! The biggest cost so far by miles is all the tools I have bought for the project. I keep realising I need another bit of equipment. From drill bits to taps and tap wrenches, centre punches, transfer punches, counterbores, a router, drill press, cut off saw, band saw, lathe, welder, callipers, dial indicator, computer, software etc etc the list goes on - so who knows. What I do know is I've learnt a huge amount and loved the experience so far... Look forward to seeing your build.

Attachment 17145 on the subject of buying stuff my new m12 counterbore has arrived. Time to finish the gantry. Attachment 17146

I love the sight of shiny Aluminium.....really looking forward to seeing this move Joe :thumsup:

Fabulous job Joe !

Even more so that you are making it by 'hand' without another cnc machine to help cut bits. Keep ticking away at it and before you know it you'll will be rewarded with a really solid machine.

Hello,

I'm very exiting your cnc design. Maybe you have video how cnc works? maybe can get drawings from you?

Quote:

---

Originally Posted by trigus

Hello,

I'm very exiting your cnc design. Maybe you have video how cnc works? maybe can get drawings from you?

---

Hello, whilst I'm not at all opposed to sharing, in my experience the design process is essential if you want to understand what you are building and why. I would encourage you to read a lot on this site and others then get drawing!

Attachment 17272

I had decided to go with slaved motors but I really want to use a CSMIO/IP-M Motion Controller and they don't like slaved motors. Early on I had toyed with a belt drive and after reading posts on biketrialdave's build log I'm heading back that way I think. The belt will need to be made as it is over the size limit for standard belts but the cost of a custom one does not seem extortionate! The basic idea is as the image above... which would not require much modification from the existing setup. As usual, any comments are gratefully received!

Quote:

---

Originally Posted by JoeHarris

Attachment 17272

I had decided to go with slaved motors but I really want to use a CSMIO/IP-M Motion Controller and they don't like slaved motors. Early on I had toyed with a belt drive and after reading posts on biketrialdave's build log I'm heading back that way I think. The belt will need to be made as it is over the size limit for standard belts but the cost of a custom one does not seem extortionate! The basic idea is as the image above... which would not require much modification from the existing setup. As usual, any comments are gratefully received!

---

Hi Joe,
I wonder if that long unsupported run of belt would flap and vibrate around. Do you think it would be in perfect syncrony on each ballscrew? Also, a single motor would need to be more powerful than 2 singles.

On my new machine (not CSMIO) I am going with 2 motors and 2 short belts, then one long slave belt between them which mostly does nothing unless there is a problem. Would CSMIO let you feed step/dir signals into 2 drives? Is that acceptable? I know you can't 'slave' home in the traditional sense with that suggestion, but with a link belt you wouldn't have to.

I was going to up the motor to a nema 34 12nm beast. This is a similar size belt to Dean's first machine I think? He had an idler part way along not at one end. If I get flapping I could add one in the same way. I understood that the ip-m just did not allow proper slaving...Attachment 17273

Quote:

---

Originally Posted by JAZZCNC

Hi Joe,

Sorry not replyed to some of your previous post's where you asked me directly, I've had eye trouble so time on computer has been limited hence me not being around much.!

Others have answered the belt Q's regards width, pitch etc but there are things you need to be made aware about with this size machine.?
One of the things I found when I built my machine was that long belts where not available in 5 x 15mm HTD so depending on the X axis screw centres you may have the same trouble.?
The X screw centre's on my machine are 1300mm using 30T pulleys so needed belt length over 2750mm the longest HTD 5mm pitch 15mm wide I could find was around 2400-2500mm.
The way I got around it was to have them special made costing around £20 each. I bought 2 keeping one for spare, that was over 3yrs ago and I've only recently swapped for the spare and that was because I left a pulley lose after changing ratios from 1:2 to 1:1 for high resolution job and it fell off killing the belt in the process.!!

Regards mounting pulleys you will need the ballscrew ends machining longer than standard so the shaft goes all the way thru the pulley giving better purchase for the grub screws. Also grind a small flat on the shaft for the grub screw to grip and stop the pulley spinning.
Regards idlers and tension then You don't need lots of tension on the belt and infact if you put them under too much tension they wear quicker and it also puts tension on the ballscrew bearings and if Ott can cause binding and possible missed steps.? Just enough tension to stop the belts flapping around and jumping teeth is all thats needed and lightly tensioned idlers is all thats required. (I just use 2 bearings with large penny repair washers for guides)

I'm posting the pics of my well used setup and While my setup looks ugly and is roughly thrown together it's simple, works very very good and accurate with high reliabilty even thou I run it in some of the worst conditions for running belts.!!
I dont have any guards or shields, the machine mostly 98% of time cuts Aluminium with chips flying every where and it has never failed or lost position either thru skipped teeth or missed steps. While It's capable of max rapids around 11mtr/min with 5mm pitch screw 1:2 ratio(effective 10mm pitch) 6Nm Nema 34 motor running 75Vdc I detune it to 7mtr/min which is more than enough for my needs and also why It's NEVER, NOT WOUNCE missed steps or lost position or sync under cutting conditions or rapid movement.

To give you an idea of the resolution and what 1:2(10mm pitch) belts can achive then I've attached the Aztec calender pic with a link to video see the detail better.

Hope this helps and inspires.!

http://www.youtube.com/watch?v=wjDvi...ature=youtu.be

Attachment 5950Attachment 5951

---

Quote:

---

Originally Posted by routercnc

Hi Joe,
I wonder if that long unsupported run of belt would flap and vibrate around. Do you think it would be in perfect syncrony on each ballscrew? Also, a single motor would need to be more powerful than 2 singles.

---

Yes and No. Yes it needs a some support in the middle to stop flap. Does it keep perfect sync.? Yes it does.

Quote:

---

Originally Posted by routercnc

On my new machine (not CSMIO) I am going with 2 motors and 2 short belts, then one long slave belt between them which mostly does nothing unless there is a problem.

---

Not so sure that will work has good you expect.? While in theory it will help keep screws in sync I'm not sure about the conflicts you might get if one motor starts missing steps. You'll have one motor fighting the other and because the Motors are connected directly to the screw I think you'll still get racking because the motors torque will force racking or affect the other motor and cause stalling etc.?
Also Don't see the point in taking the conflict risks when you still have the same belt setup.?

Edit: Then you have the affect of extra inertia of 4 pulleys on screws. It will affect performance.!

One thing for sure if you do this then make sure you have good drives because it will do funky things to the resonance and the drives will need to able to deal with it.!

Quote:

---

Originally Posted by routercnc

Would CSMIO let you feed step/dir signals into 2 drives? Is that acceptable? I know you can't 'slave' home in the traditional sense with that suggestion, but with a link belt you wouldn't have to.

---

Bad idea with any controller but this isn't needed anyway the IP-M does allow slaving just doesn't independantly home each Axis.

Quote:

---

Originally Posted by JoeHarris

I was going to up the motor to a nema 34 12nm beast. This is a similar size belt to Dean's first machine I think? He had an idler part way along not at one end. If I get flapping I could add one in the same way. I understood that the ip-m just did not allow proper slaving...

---

12Nm is way to Big Joe and will affect performance. 6 or 8Nm will be more than enough. Try to find motor with lowest inductance possible and use plenty of volts.

Attachment 17540

The revised plan for the twin ball screws is to drive them from one motor with a long belt as shown in the image. The plan is two use two idler pulleys one of which will be fixed. I have been advised by one of the guys at beltingonline to support the long length of belt on sheets of UHMWPE which will not wear and is self lubricating. I will fix this to some lengths of steel angle I have left over. I should get the recommended 5 tooth engagement on all pulleys. I have allowed long enough slots for the motor mounts to use a smaller pulley if i want to change to 1:1 ratio. What do you think?

Joe, I'd put a plate over the square steel section your rails are moumted on, move the top pulley to the right and move the other pulley next to it. That way you'll have more teeth in contact with the ball screw pulley.

Quote:

---

Originally Posted by njhussey

Joe, I'd put a plate over the square steel section your rails are moumted on, move the top pulley to the right and move the other pulley next to it. That way you'll have more teeth in contact with the ball screw pulley.

---

Yeah that's quite a nice idea. I think I might be too tight for space to get a good depth to tap for the idlers. I could certainly pull the lower left one up though which will improve things. This was a quick and dirty to test. I still need to draw a nema 34 as this is still based on the smaller nema 23 frame. Going from 56 to 86mm will change the setting out quite a bit...

Also any ideas why the SY85STH118-6004B Nema 34 Stepper motor from Zapp costs £94 and the cnc4you 86HS115-4208 8.7Nm Nema 34 Stepper Motor costs £71? Same size and torque and Inductance is the same...

Attachment 17549Attachment 17548
More or less complete design for the single motor, long belt method.
The idlers are both fixed and made from standard skate bearings spaced in the middle with a washer and edged with two penny washers to provide flanges - these will be mounted on a M8x40 bolt taped into the aluminum plate and stood off with a stack of washers. I can adjust the height of the whole motor assembly if necessary and fine adjustment can be made by raising or lowering the motor on 10mm slotted mounts, this was the max I could achieve with the large circular projection (which I assume houses the bearings) at the center of the motor.
Attachment 17550
Both lengths of the long belt (top and bottom) are supported on UHMWPE (blue stips on the image below) which will be bolted down to steel angle which in turn will be welded or bolted to the frame with strips of 90x6mm flat bar. This support framework can then double up as the framework for a belt cover.

Attachment 17547

The motor this is all based on is the cnc4you 86HS115-4208 8.7Nm Nema 34 Stepper Motor as highlighted on the data sheet below:

Attachment 17551

I'd still put the top idler on the end of the square section (with a suitable thick plate welded/bolted in) and move the lower idler to the first idler's position.....:whistle:

Quote:

---

Originally Posted by njhussey

I'd still put the top idler on the end of the square section (with a suitable thick plate welded/bolted in) and move the lower idler to the first idler's position.....:whistle:

---

Like the blue lines in this?: Attachment 17552

I like the idea but it's a pain to implement. I only have a few mm of space: Attachment 17553

Like this....excuse the freehand scribble!!

Attachment 17554

Edit...Hah just seen the lines you drew....need to open my eyes!! Yes like that, will give much more teeth in contact, I know 5 is the min but I always think the more the merrier!!

[QUOTE=njhussey;76714]Like this....excuse the freehand scribble!!

Yes, sorry that's what the really faint blue lines were supposed to show!!!