[gjhammer]

Hello everyone,

I am in the process of designing/building my very first CNC router that I plan on using mainly for woodworking. I have visited these forums many times and have found them invaluable but with no experience, have not contributed yet until now. I want my machine to be able route 2' x 4' sheets of wood with about 8" of Z axis travel.

Here is the core components of my machine so far:

Base assembly (X-axis):

• 2040 80/20 extruded aluminum frame
• 2 15mm Hiwin rails
• 4 15mm Hiwin blocks
• 1/2-10 60" single-start acme lead screw
• Delrin nut
• Nema 34 stepper motor (TBD)

Table assembly:

• 2020 80/20 extruded aluminum frame

Gantry Assembly (Y-axis)

• 2040 80/20 extruded aluminum frame
• 2 15mm Hiwin rails
• 2 15mm Hiwin blocks
• 1/2-10 48" aceme lead screw
• Delrin nut
• Nema 34 stepper motor (TBD)

Z-axis:
TBD

I obtained most of the extruded aluminum and Hiwin blocks/rails from:

http://www.automation-overstock.com/
http://www.automation4less.com/

I obtained most of the miscellaneous hardware from:

http://www.mcmaster.com/

I obtained the acme lead screw/nuts from:

http://www.precisiontechmachining.com/



Attached are pictures of my design so far. I will post pictures later of the progress building the machine later tonight.

[fortyonethirty]

Why did you choose to make the work table seperate from the rails that the gantry slides on?

[gjhammer]

First off, I am no expert as this is my first time building a machine.

I did the table separate from the base for several reasons:

• Keeps the sweep of the gantry contained with the base, no worry about outside objects interfering with the sweep.
• Just one of the designs for achieving a fully supported X axis (I referenced the web-site http://www.cncroutersource.com/build...nc-router.html)
• Allows me the flexibility of swapping out tables in the future, i.e. maybe a full t-slot table or other options

Just keep in mind though that the table will be fastened to the base. I'm always open to suggestions or pros/cons to my design.

[gjhammer]

Here are pictures of my build so far. The base is partially done, the gantry is partially done and mounted on the X-axis, and the table is done.

Note that not all the brackets for the gantry have all the screws in them to save time unscrewing them when I do some more work on the gantry.

[gjhammer]

I still need to do the following for the base assembly (X-axis):

1) Mount bearing blocks I am getting from http://www.precisiontechmachining.com/. I will provide radial support at one end and radial/thrust support at the other end where the motor will be mounted

2) Install acme delrin nut into bottom of gantry assembly

3) Install acme lead screw

4) Install motor mounts but I won't do this until I have the motors and couplings in hand. I probably won't do this until near the end of the build.

For the gantry (Y-axis) I still have ways to go.

[FandZ]

Hey man, glad to see your getting a machine together. Hopefully I can give you some good critiques. Inside tracks are generally avoided. With the tracks of your X axis on the inside there's a chance for some binding/chattering. With the linear components you plan on useing I doubt you'll have a problem with any racking but it won't be that hard to flex those X's side rails outwards. Also you can't use any cross supports. If you went for the outside you could use cross beams and really gain some anti racking strength.

Again if you went outside you could simplify the table top and stiffen up the machine. Having fixtures is great but you want to mount them on your table top. Once you get everything squared you don't want to risk ruining your alignment by remounting various table tops.

Other than that, the Y axis stepper mounts seem like a lot of effort and material. I'd try to simplify it but bringing in other materials.

[gjhammer]

Hi FandZ,

Thank you for your critique! Anything I can do now to save myself problems down the road is great.

As far as deciding for an inside track versus an outside, can you help me understand why the outside track design leads to less binding/chattering. I'm not doubting you, just wondering for my own information. Is it because the inside track design would cause the Hiwin block to be pushed toward the rail while the outside track design would cause it to Hiwin block to be pulled away from the rail?

I did notice alot of the commercial CNC routers do have outside tracks or tracks mounted on top of the X-axis. So it is definitely something I will consider now before I progress any further.

As far as the Y-axis stepper mounts, I will definitely look at simplifying this part. Let me see if I can come up with some alternative designs.

[FandZ]

Hi FandZ,

Thank you for your critique! Anything I can do now to save myself problems down the road is great.

As far as deciding for an inside track versus an outside, can you help me understand why the outside track design leads to less binding/chattering. I'm not doubting you, just wondering for my own information. Is it because the inside track design would cause the Hiwin block to be pushed toward the rail while the outside track design would cause it to Hiwin block to be pulled away from the rail?

I did notice alot of the commercial CNC routers do have outside tracks or tracks mounted on top of the X-axis. So it is definitely something I will consider now before I progress any further.

As far as the Y-axis stepper mounts, I will definitely look at simplifying this part. Let me see if I can come up with some alternative designs.

In a nut shell it comes down to flex. When you spindle is machining on one side of your table, it'll act as a lever on the other side. Basically your whole gantry will be trying to twist out of it's railed confinement. That twisting action will cause a binding effect. The end result may mean you have to go slower on your cuts or else your X axis motor may loose a step, or it'll have a chatter effect, or worse may cause ribbed tool marks in your material.


By moving your rails to the top or outside, you free up the inside to run cross support beams. These cross support beams will also serve to support up your table top. So it's a win win situation.


I also saw you plan on using nema 32 steppers. I'd really recommend you look at nema 23 steppers and a gecko g540 driver. To make good use of a nema 32 motor costs a minimum of 2 to 3X more. You really are not going to be pushing around the kinds of weight where nema 32 is needed. Forgive me if you already have all that worked out.

[gjhammer]

Hi FandZ,

As far as side mounted designs go, any advice on the following two options:

1) Mounting rails/blocks on the sides, and installing stepper motors and ACME lead screws on each side to push/pull the gantry assembly

-or-

2) Mounting rails/blocks on the sides, and installing stepper motor in center of X-axis and have it push/pull the gantry assembly from underneath the X-axis

From a rigidity standpoint, I think option 1 is best (fully supported X and Y axes) but it is more expensive and more complicated (two motors driving the X axis). The second option is simpler but less rigid, though I am using 2" x 4" aluminum extrusion which is 64" long for the sides so it does not flex that much (I don't have the deflection calculator in front of me so I can't exactly how much). I am leaning towards option #2.

As far as Nema 34 motors, I have not figured everything out yet. I just felt that I could get motors with higher torque ratings when I went from Nema 23 to Nema 34. and your right, the final weight of the gantry probably will not require as much torque a Nema 34 would provide me. I will look at Nema 23 for sure. I did actually recently look at the Gecko g540. It looks very nice but I saw that it was limited to 3.5A motors (which most Nema 34 motors aren't).

Definitely some good input! Thanks so much.

Geoff

[FandZ]

Hi FandZ,

As far as side mounted designs go, any advice on the following two options:

1) Mounting rails/blocks on the sides, and installing stepper motors and ACME lead screws on each side to push/pull the gantry assembly

-or-

2) Mounting rails/blocks on the sides, and installing stepper motor in center of X-axis and have it push/pull the gantry assembly from underneath the X-axis

From a rigidity standpoint, I think option 1 is best (fully supported X and Y axes) but it is more expensive and more complicated (two motors driving the X axis). The second option is simpler but less rigid, though I am using 2" x 4" aluminum extrusion which is 64" long for the sides so it does not flex that much (I don't have the deflection calculator in front of me so I can't exactly how much). I am leaning towards option #2.

As far as Nema 34 motors, I have not figured everything out yet. I just felt that I could get motors with higher torque ratings when I went from Nema 23 to Nema 34. and your right, the final weight of the gantry probably will not require as much torque a Nema 34 would provide me. I will look at Nema 23 for sure. I did actually recently look at the Gecko g540. It looks very nice but I saw that it was limited to 3.5A motors (which most Nema 34 motors aren't).

Definitely some good input! Thanks so much.

Geoff

If I was you I would design both options and post the m up for suggestions.


In a well designed machine it won't matter much but I like a dual screw even though I haven't built a machine using two screws. To me just not having to worry about racking is worth it. You still are going to want to have a cross support connecting the gantry sides underneath but in my eye's it is worth the trouble.

A multi-start screw is going to be a must for you and like Ahren is saying you may want to look into a R&P setup. I'm using Ahrens parts on a mill build that I'm working on and am completely satisfied with them.


If you haven't had any experiences with stepper motors you'll be surprised just how strong a 280oz to 380oz stepper will be. I'm using 280oz steppers and they are powerful enough to pull me at a low RPM. And that is the thing, steppers loose power at higher RPM's. They are the strongest at a dead stop and drop off rapidly after that. I'm pretty sure once you research it you'll end up with Nema 23's and a g540. Dollar for dollar it's the best you can do and you won't be dissatisfied.

[LEVONCNCDUDE18]

I've make my own stepper driver the interface is DB25 pin>> and my software is mach3..
any one try a usb converter ,do you think is it working ... What software can be converted to usb..

[gjhammer]

Hi LevonCNCDude18,

You might want to post your question to another forum. I am a newbie and I don't think I can answer your question.

[gjhammer]

Test

[gjhammer]

TIDBIT #1: For those people in the design process, you can get 3D models of most of the 80/20 profiles, parts, fasteners, joining plates, etc. at:

http://www.3dcontentcentral.com/3DContentCentral/

They can give you 3D models in many different formats so hopefully you can get them in a format your CAD system will support.

[ahren]

gjhammer,

Welcome to the forum! I would offer one piece of advice on your build so far, and that is to use multi-start ACME screws rather than the single start ones you have now. With a 60" screw, you're going to get whip like crazy at even moderate speeds. This can be mitigated somewhat by using a 5 start screw, since your screw only needs to turn 1/5 the speed for the same linear speed. Really , about 4 feet is the max for most 1/2" screws, which is why I came out with my R&P drive system, although I have seen people get a 60" axis to work with multi-start ACME.

Best regards,

Ahren
www.cncrouterparts.com

[gjhammer]

Hi Ahren,

Thanks for the input. I did consider multi-start ACME lead screws. It seemed that across the board people are split between single and multi start. Though maybe they are dealing with a smaller X-axis than mine. I will look into it some more and may have to switch to a multi-start screw.

Thanks for your input,

Geoff

[DIYaholic]

gjhammer,

You can drive 2 screws with one motor. There are two ways to accomplish this:
1) Center mount the motor and drive both screws with belt & pulley.
2) Direct drive one motor and use belt & pulley to drive 2nd screw.

Just something to think about (not that there isn't enough to sort out already).

Randy,

[CarveOne]

Hi Ahren,

Thanks for the input. I did consider multi-start ACME lead screws. It seemed that across the board people are split between single and multi start. Though maybe they are dealing with a smaller X-axis than mine. I will look into it some more and may have to switch to a multi-start screw.

Thanks for your input,

Geoff

Ahren is right about the multi-start screws. The extra cost is well worth it. I started with cheap single start Acme screws on my first machine.

The second machine had 5 start Acme screws initially. When Ahren came out with the pinion drive I was worried that the 1/2-10 5 start on the 6' X axis would whip so I bought two pinion drives for the X axis. Later, I bought a pinion drive for the 57" Y axis. The 5 start screws will be re-used to replace the single start screws on my first machine, so they are not wasted money. Then I'll have two very useful machines.

I think the only people who are split about which screws to use are those who haven't built a machine yet. Once they use 5 start screws they won't even consider the single start for another project.

CarveOne

[stevespo]

I like hemsworthlad's design with the screw behind the X axis, and shifting the Z back to keep it mostly centered over the trucks for the linear rails. This was something I did in my own build, and important to help minimize torquing on the uprights. Those gantry uprights are short, rigid and nicely milled. Very clean and compact.

I'm a little curious about the complexity of gjhammer's gantry. How are you going to attach the linear rail trucks to the 8020? They are typically attached to a piece of plate (1/2", 3/4", etc) and then the plate is attached to the 8020. Maybe I'm just missing something, but it seems like the short bits of 8020 and all the additional brackets/joining plates could be replaced with a piece of AL plate with a few strategic holes.

Also, whenever I see 8020 bolted to the back of the uprights like that it makes me a little nervous. I'd like to see it sandwiched between the uprights and not relying on a host of braces and joining plates for rigidity. It seems to me that you'd get much better triangulation on those joints (and a stiffer frame) with better piece to piece contact and fewer bits holding them all together.

I can't (and won't) claim that there is any real evidence to back up my gut feeling, but when I think about woodworking joinery I want mortise and tenon, dovetail, etc - over a screwed on butt joint. The comparison may not be valid, but that's my way of thinking. Naturally, there are many ways to approach this and there may be an equally compelling reason why my thinking is wrong.

Steve (My own 8020 build)

[gjhammer]

Hello everyone,

Based on everyone's suggestions, here is my latest design. It contains the following changes:

1) Y-axis screw between two cross members in the gantry
2) Simplified attachment of gantry to carriage blocks with a custom bracket made of 3/8" thick aluminum. I will also through bolt gantry uprights to carriage blocks.
3) Shifted the Y axis a little back

I think it is much more elegant and more rigid.