Design ready! But anxious about whether it's truly ready to build...

So, I squeezed in a few hours every night for the last 5-6 days to design this one, using experience from my tiny first machine (moving-bed style for wood) as well as reading lots of threads here.

Pretty proud of the effort as a beginner, but not so confident (read: nervous) about how it will go in reality, so wanted to ask some questions (see further below) and maybe get some feedback!

Below are pictures of the best design that I could manage to create based on:

• my dimensional requirements/constraints
• the goal of cutting aluminum
• taking advantage of at least some components like rails and bearings that I already had lying around

Attachment 365946

Attachment 365948

For the Y-axis, I do plan to have a separator/wall on each side of the work-area to prevent dust from flying into the ballscrews+linear-rails.

Here are questions/concerns that I have (Q1, Q2, etc. are marked in red on the first picture above):

• (Q1) Do I absolutely need a support-bearing block for the Z-axis ballscrew? I did this for X and Y, but the Z-axis construction is really tight on space, so I'd prefer to leave it out if that's fine.
  
• (Q2) I presently have two Hiwin HGH-15 bearing blocks for Z-axis; ordering more would add cost and more importantly time due to re-shipment, not to mention the reduced Z travel. Is two bearing blocks good enough?
  
• (Q3) Any improvements I can make on these plates? I used fairly thick plates (15mm), but I'm more concerned about the shape and positioning.
  
• (Q4) Steppers: 350 oz-in. Are these going to struggle with moving the amount of weight all the parts add up to? I am mostly concerned about Y, even though I have dual-drive (i.e., two steppers) on that axis. I don't need too much speed even on rapids, as this is mostly for occasional personal use. I do want the router to run without skipped steps obviously.
  
• (Q5) Any other general feedback on overall structure/geometry? I'm excited to begin assembling this, and see where it goes.

1) Yes. The bearings in a stepper motor are not designed for axial loads. And you'll have backlash, as the stepper shaft will have some axial play in it.

2) Good enough is subjective. I would say no, but you may find that it works for you. Two bearings per rail distributes the loads better, and adds a lot of rigidity,

3) not really.

4) Depends how fast you want to go. The 350oz of torque is when they are not spinning. The faster they spin, the less torque they have. At some point, they'll stall, or lose steps. Stay below that speed, and you'll be fine. What that speed is would depend on the drives, power supply voltage, and acceleration rate.

5) My preference would be to mount the rails on the sides, with the screws below them, getting everything off the table, for easier loading, unloading, and general use.

To add to Gerry's good comments:

1 - Since the travels are short, you only need the fixed bearing block. Though you could just get the flanged version, and use or design a riser for the stepper, this way you don't have to redesign your Z.
2 - Depends on the overall Z travel. You could just order long blocks which would help. I personally prefer dual blocks, and of course there always is a trade-off between compactness and rigidity.
3 - It seems the support blocks for the gantry uprights are superfluous. You can use a taller plate and fix it to the linear block plate and the plate above the ballnut carrier. Would have a cleaner look as well.
4 - Since you're using the 10mm pitch screws, I'd probably size the stepper up a little, though as Gerry points out a lot depends on the steppers' torque curve and PSU and drive used.

Appreciate the very specific and well thought out feedback, Louie and Gerry.

I tried my best to merge some of the changes that you guys suggested as well as some of my own geometric refinements.

So, here is the 2nd version of the design...

Front view: http://i.imgur.com/Ik9s7uV.jpg
Attachment 366286

Back view: http://i.imgur.com/iKuChuA.jpg
Attachment 366288

Below are the changes I made:

1. Most important of all, altered the geometry so that the rails and ballscrews are protected. This also simplified the Gantry side plate design a lot!
2. Gantry side plates now are 20 mm thick (instead of 15 mm).
3. Strengthened the X-axis (i.e., the horizontal "gantry axis") by using 80 x 160 mm aluminum profile instead of 40 x 160.
4. Replaced ballscrews with 20 size (instead of 16), and linear bearings + rails with HGH20 size (instead of HGH15).
5. Added a fixed bearing block (with AC bearings) for Z-axis ballscrew (previously it was fully floating beneath the shaft coupler).
6. Added two more linear bearings (HGH15) for Z-axis.

Feedback?

I hope this one is OK to start assembling!

I wouldn't extend the Z axis so far past the bottom of the gantry. A clearance of 120mm is not bad, but you don't want to further restrict the height of parts that can be placed on the bed by having fixed parts lower than that which will bang into them.

Quote:

---

Originally Posted by gamingan

Appreciate the very specific and well thought out feedback, Louie and Gerry.

I tried my best to merge some of the changes that you guys suggested as well as some of my own geometric refinements.

So, here is the 2nd version of the design...

Front view: http://i.imgur.com/Ik9s7uV.jpg
Attachment 366286

Back view: http://i.imgur.com/iKuChuA.jpg
Attachment 366288

Below are the changes I made:

1. Most important of all, altered the geometry so that the rails and ballscrews are protected. This also simplified the Gantry side plate design a lot!
2. Gantry side plates now are 20 mm thick (instead of 15 mm).
3. Strengthened the X-axis (i.e., the horizontal "gantry axis") by using 80 x 160 mm aluminum profile instead of 40 x 160.
4. Replaced ballscrews with 20 size (instead of 16), and linear bearings + rails with HGH20 size (instead of HGH15).
5. Added a fixed bearing block (with AC bearings) for Z-axis ballscrew (previously it was fully floating beneath the shaft coupler).
6. Added two more linear bearings (HGH15) for Z-axis.

Feedback?

I hope this one is OK to start assembling!

---

I like it. I do think the simpler the design, the easier to make, and the easier to make more rigid since there are no excess connectors or brackets. Having the ballscrews away from the cutting area is a plus.

Quote:

---

Originally Posted by gamingan

Appreciate the very specific and well thought out feedback, Louie and Gerry.

I tried my best to merge some of the changes that you guys suggested as well as some of my own geometric refinements.

So, here is the 2nd version of the design...

Front view: http://i.imgur.com/Ik9s7uV.jpg
Attachment 366286

---

Your Y axis who I would call the X axis it kinda a bit of a mystery as to how the extrusions are held together. I'm not completely comfortable with the approach you have taken. The extrusions should be well supported but it looks like you have thin cross section extrusions for the lateral "rungs". You would want a stiffer approach to connecting the extrusions the rails bolt to to the rest of the bed.

I might have missed it but I didn't see any info on the Z axis stroke length. You have indicated 120 mm below the gantry beam which is probably good from a stiffness standpoint but you will want to make sure you can clear your tools when the Z is fully retracted. so you need to consider the longest cutter you are likely to chuck and make sure you can lift that tool to clear the beam. I'm not even sure the nose of the spindle will clear from the picture.

Mounting linear rails directly on T-Slotted extrusions can be a challenge at times. Depending upon the combination of rails, extrusions, nuts and so forth getting the two rails parallel can be a real challenge. Some guys have resorted to bolting aluminum plates to the extrusions to make this easier (the rails get bolted to the intermediate plate).

Quote:

---

Back view: http://i.imgur.com/iKuChuA.jpg
Attachment 366288

Below are the changes I made:

1. Most important of all, altered the geometry so that the rails and ballscrews are protected. This also simplified the Gantry side plate design a lot!
2. Gantry side plates now are 20 mm thick (instead of 15 mm).
3. Strengthened the X-axis (i.e., the horizontal "gantry axis") by using 80 x 160 mm aluminum profile instead of 40 x 160.
4. Replaced ballscrews with 20 size (instead of 16), and linear bearings + rails with HGH20 size (instead of HGH15).
5. Added a fixed bearing block (with AC bearings) for Z-axis ballscrew (previously it was fully floating beneath the shaft coupler).
6. Added two more linear bearings (HGH15) for Z-axis.

• The layout is certainly better now. As mentioned above I do have concerns about how reliable and stiff the X axis connections are to the table bed. The gantry side plates are certainly better. I still prefer to see box sections, of some sort, for the uprights but 20 mm plates are getting to the point where it doesn't matter. Generally keeping it simple is the smart approach.
• This should do all right for a machine of this type.
• I like to see nomenclature such that the X axis is the lowest axis the Y being the gantry and the Z being the the axis that moves the spindle into or out of the workpiece. This mainly due to consistent communications. People seem to like to argue about this but from my standpoint they don't have a leg to stand on. Consistency makes the chore of communications far easier.
  
  All that being said going to a larger extrusion here is a smart move. In may home built machines the gantry is the weak link either due to a scrawny beam or uprights that flex way too much. It is kinda hard to go overboard here, at least in the context of a limited budget most of us have.
• Sounds good.
• The bearing supports for the leadscrews are very important. They have to handle the trust loads that motors can not handle.
• Just be careful of your ability to clear tools mounted in the spindle collet. This is another design element that sometimes gets overlooked. In an ideal world you want enough Z axis travel to be able to lift the tool to clear the gantry beam. The "tool" here being whatever cutter is the longest that you commonly expect to use.
  

Feedback?

---

For the most part it is looking really good.

My biggest concerns are the Z axis stroke and the way the X axis rail extrusions are coupled together with the rest of the table.

Quote:

---

I hope this one is OK to start assembling!

---

Lots of luck!

Quote:

---

Originally Posted by wizard

Your Y axis who I would call the X axis it kinda a bit of a mystery as to how the extrusions are held together. I'm not completely comfortable with the approach you have taken. The extrusions should be well supported but it looks like you have thin cross section extrusions for the lateral "rungs". You would want a stiffer approach to connecting the extrusions the rails bolt to to the rest of the bed.

---

Again with this crusade... Just call it the gantry. If you're setting the CAM up that the material gets loaded from the front, then the gantry is the X axis.

Quote:

---

Mounting linear rails directly on T-Slotted extrusions can be a challenge at times. Depending upon the combination of rails, extrusions, nuts and so forth getting the two rails parallel can be a real challenge. Some guys have resorted to bolting aluminum plates to the extrusions to make this easier (the rails get bolted to the intermediate plate).

---

Yet most every person that builds with t-slot extrusions somehow manages to mount the linear rails! Besides, using plate would be redundant - you'd just get thick plate and forego the extrusion. Unfortunately, with a set budget, purchasing plate and having it machined can be costly for two gantry risers, let alone two long table sides.

Quote:

---

My biggest concerns are the Z axis stroke and the way the X axis rail extrusions are coupled together with the rest of the table.

---

The Z axis stroke shouldn't be a problem. If it was it would simply be a matter of adjusting the position of the spindle up or down in the clamp as needs warrant. As to the long axis rails (which are labeled Y) I really don't see issue here; pretty standard for t-slot extrusion build.

Quote:

---

Originally Posted by louieatienza

Again with this crusade... Just call it the gantry. If you're setting the CAM up that the material gets loaded from the front, then the gantry is the X axis.

---

It isn't a crusade, it is a quest for better communications. Generally I agree with you but I this case I think you are way off the mark. More importantly you can't be renaming axis based upon where you are standing at any one moment in time.

Quote:

---

Yet most every person that builds with t-slot extrusions somehow manages to mount the linear rails! Besides, using plate would be redundant - you'd just get thick plate and forego the extrusion. Unfortunately, with a set budget, purchasing plate and having it machined can be costly for two gantry risers, let alone two long table sides.

---

And yet there are many build reports where people have struggled with this. Either due to using the wrong nuts in the T-Slots or ill fitting linear rails. How bad this is will be varies with the extrusion manufacture but if one wants to go this way they should consider extrusions that are ground and suitable for such use.

Quote:

---

The Z axis stroke shouldn't be a problem. If it was it would simply be a matter of adjusting the position of the spindle up or down in the clamp as needs warrant. As to the long axis rails (which are labeled Y) I really don't see issue here; pretty standard for t-slot extrusion build.

---

It isn't a question of seeing an issue, it is something the builder needs to be aware of. I don't know how he will be using his machine or even material dimensions, I'm only suggesting that he makes sure he has clearance for the cutters he intends to use. I only mentioned this due to it being an issue that is often overlooked. Generally it is a good idea to at some point make sure you have all the clearances you need, on all axis, to achieve your working area design goals.

As for the original poster I really think he has worked towards a nice design. This is likely due to the variety of comments he has received so far.

Quote:

---

Originally Posted by wizard

It isn't a crusade, it is a quest for better communications. Generally I agree with you but I this case I think you are way off the mark. More importantly you can't be renaming axis based upon where you are standing at any one moment in time.

---

You name the axes as you have them set up in CAM and the control software. That eliminates any confusion. The gantry can either be the X or Y depending on how the machine is used. The machine don't care - it takes g-code and spits out pulses all day. It's not hard - left to right is X, front ot back is Y, up and down is Z. Rotary axis parallel to X is A. Rotary parallel to Y is B. Rotary parallel to Z is C. Very simple.

Quote:

---

And yet there are many build reports where people have struggled with this. Either due to using the wrong nuts in the T-Slots or ill fitting linear rails. How bad this is will be varies with the extrusion manufacture but if one wants to go this way they should consider extrusions that are ground and suitable for such use.

---

If someone used the wrong t-nuts or purchased rails that won't work with the rest of the system that's another story. The main reasons for using the t-slot extrusions are availability, and no need for machining, drilling, and tapping long plates.

I think your design looks good, however, don't stop where you are. Place your limit switches, cable guides, vacuum system. I have found that it is always a PIA to build extra brackets, drill holes, and change the thing after the major parts are fabricated. Often a simple change can be made in the early design to accommodate these parts.

Also, what are your electronics going to look like? Could you use a single servo on the Y axis instead of 2 steppers?

I will definitely add the limit switches and miscellaneous components. Love doing a complete CAD before building things; helps me check EVERYTHING (probably at the cost of the learning you get from failures though).

Electronics: I am going with 2 steppers simply because that was the easiest approach I imagined. Time to explore servos now then! Any servos you can recommend as a substitute for both steppers together?

Thanks for all the feedback guys. I've made one more change -- moving the rails onto the top (instead of side) of the extrusions on the Y/long axis, to help with alignment.

Theoretical question about Z-axis (I should probably post this in another category of the forum?):

If you were given only two linear bearing blocks (for the profile rail type, like Hiwin HGH-15 type) for the Z-axis, would you put them vertically on a single rail, or would you split them into one for each of two rails? I know four is ideal, but just wondering about the mechanics in this imaginary scenario that might pop up in the future when I want to maximize travel.

wizard, The variety of comments has certainly helped, thanks to all including you and Louie for contributing your own takes.

I just finished a CNC mill conversion and I used Clearpath SD servo's by Teknic. They are a little bit more expensive than steppers, but super easy. The drivers are built into the servo so the control box can be smaller and is easier to wire.

For the Z axis, use 4 bearings. Anything less and you would probably be disappointed in the rigidity.

I built a machine just like this a couple of years ago. The only thing I really had issues with was the z axis rigidity. If I had to build it again I would have built the z axis over what I thought was necessary, also I would use a square shaped motor. After I used one I never went back to round.

Also if you need rails I have a nice set of thk 25mm rails that I never used (2 rails and 4 blocks)


Sent from my iPhone using Tapatalk