The Meticulous Z-Axis


This is a companion discussion topic for the original entry at http://maslowcommunitygarden.org/The-Meticulous-Z-Axis.html
6 Likes

image

9 Likes

That’s a thing of beauty.

3 Likes

really nice. thanks for posting and I’d like to give it a try. there appears to be an additional sheet of material (?laminate) on the bottom of the sled – could you comment.

the weight of this is sufficient that you’ve left off additional bricks/weights?

Just going to add a link to the original discussion. I will be posting build logs to this thread when I get to it, to try to keep the information in one place.

The bottom layer is 1/4" HDPE. It’s a low-friction plastic that I’ve secured to the bottom face of the sled. Without it, the plywood would catch and drag a little as it moved across the workpiece. It’s not entirely necessary, but it helps. I haven’t really detailed that part too well so I can see the confusion.

I haven’t weighed out the parts yet, so I’m not entirely sure what the final weight will be. I use these 1 1/2" x 1/2" steel bars to weigh down my sled, and plan on swapping them over to the new sled once it’s built.

1 Like

got it. thanks for the reply

Hey, one comment. I have purchased the materials for this. The two pulleys from McMaster Carr are press fit. They do not have a set screw or key. You will have to drill and tap holes in order to get them to work. I would recommend something else.

3 Likes

I’ve drilled and tapped mine the way that you had suggested.

If you want pulleys that already have set screws, I think these should work. You’ll need to order 1 with 8mm bore for the lead screw and if I’m remembering correctly a 6.35mm (1/4") bore for the motor.

I don’t think the T5 belt I originally spec’d won’t work with those pulleys, though. The pitch is off by a little bit. Might work alright, but then again you could loose some accuracy. The pulleys from aliexpress need an XL timing belt.

1 Like

I just looked at the updates to the project. I have a question about the updates to the pulley specification. Reading the BOM.md folder, it looks like there are several pulley sizes. However when I read it, it appears there is a range of pulley sizes that go on the Lead Screw; with increasing pulley size resulting in higher peak z-axis speed. I think something has to be backwards because increasing pulley size on the lead screw should decrease the maximum z-axis speed, correct?

Thanks,

2 Likes

You’re entirely right and I got that backwards. Sorry about that, I will make the corrections.

EDIT: It’s fixed now. The pulleys I linked in the BOM are 1/4" bore, which I think is the motor output shaft diameter, but my motor is in a different State than me right now. Can anyone check that to make sure we’re ordering the right sized pulleys? I’d hate for someone to order pulleys that don’t work for them.

2 Likes

Hey @MeticulousMaynard, thanks a bunch. I did already purchase the linear rail kit. I am realizing now that they ship from China, and it is going to be a month before they arrive. I am wondering if we could find a way to get some of these kits local, so that it doesn’t take a month.

My first thought is we could work with @bar to help create a community z-axis design, and have the required hardware available within the Maslow store. Could you (whoever is reading this) provide feedback on whether or not you think this is worthwhile? I was very impressed with how well the Frame Design went that I have an optimistic view of how the Maslow community can come together to create a design.

I have already said the Rigid router z-axis did not pass my gut-check, which is why I am seeking an alternative. In addition, whatever we come up with could be used with a large number of router options, opening up options like used routers from craigslist, etc.

2 Likes

I think this is a GREAT idea. I’m really really interested in ways to work on making more community designs available to the community. It’s important to be able to buy the parts for something (which is why all the projects on the community garden come with a “buy” tab). I would really like to find a way to do it without us (Hannah and I) doing more logistics. We really aren’t passionate about dealing with customs and shipping things. Maslow doesn’t need to make any money from it, we just want to create a space where collaborative designs can live.

4 Likes

I’ll have to take a look to see if there is an option to easily source these within the US. With a lot of these CNC parts, it’s much easier and cheaper to source them from China. Searching eBay and Amazon may produce some results that are much closer to home, though. If I find something that works well that doesn’t take so long to ship I’ll update the BOM.

3 Likes

Here is an idea. Combining both @bar 's and @MeticulousMaynard 's idea. There are already individuals on ebay selling simple kits and stuff for specific projects like this. Could we contact some people, who are already selling similar things on ebay? We could define our kit, the Maslow Z-Axis, and all the contained components and ask them to package and provide for sale. I am sure they would appreciate the business. That way we wouldn’t have to assemble the kits ourselves and @bar doesn’t have to spend a bunch of time dealing with shipping nuances.

If we define the kit and there is reasonable volume, it would seem like an enterprising e-bay seller would jump at the opportunity.

3 Likes

Another comment. It makes sense that this be the basis for a community z-axis. I would like to propose one change to the design. Could we eliminate the need for linear rails by building them out of plywood? I have seen linear guides built out of plywood elsewhere on the internet. UHMW tape can be used to reduce friction between moving parts. I think a Maslow would be capable of cutting these parts out accurately enough.

What are your thoughts?

3 Likes

I’ve added a $31 dollar version of the linear guide kit from Amazon that is the actual length needed to the BOM. It looks like shipping time is about a week as opposed to a month via Bangood.

This could probably work. I would be concerned that the plywood rails would warp, and that could hurt accuracy. Which ultimately forces me to ask if it is worth it. The linear guides are certainly going to be more reliable over time. An 8 mm lead screw is ~$12, whereas the kit that I linked is $31. Is it worth saving the $20?

But then again, the rest of the system is built from plywood so that could be a moot point.

I’ve been thinking about whether or not I want to provide a kit myself, much like David or Logan’s linkage kits. Maybe have different options, whether it’s hardware only, hardware + 3D printed parts, or with all the plywood parts/plastic parts/hardware included. I remember seeing a little while back that people were curious if anyone would pre-make them a sled so they didn’t have to go the temporary sled route themselves. Is this something that people would be interested in? I could probably buy the Z-Axis components in bulk and ship them out to people in the US. Not sure if anyone would be willing to spend the extra $$$ for international shipping. There is a post office in my front yard, so it would not be hard to ship things out.

I’m also not opposed to having an eBay seller ship the hardware kits. That would certainly save me the effort of shipping things out.

2 Likes

My preferred solution would be to have a z-axis kit set up so that I can use the temporary sled to cut out the wooden parts, and assemble it myself. I am also not opposed to spending $10 if it is a solution I can be confident in. Everyone is probably somewhere on the spectrum of willing to spend money, willing to contribute time, and willing to debug.

I would like to have a validated community-developed z-axis. I see this as a requirement, prior to offering kits for sale. The more people who contribute and the more who use it, the more confident I will be, and the better the solution will be. I am willing to put in work to develop and validate it. However, I will just be a consumer.

I would like to discuss more specifics, and get more people’s opinions.

  1. Can others comment on their thoughts on plywood vs steel linear rails.
    a) Plywood linear rails save about ~$20 in materials.
    b) There are examples of plywood being used for linear rails elsewhere on the internet.
    c) There is the risk that the plywood could swell, shrink, or warp.
    1. I don’t think this is a risk because of the short dimensions of the parts, on the order of a few inches.
  2. I have questions about the dust-collection features.
    a) I am not looking for anything fancy here. As far as materials go, plywood would be sufficient for me.
    b) It is required that a 2.5" hose be able to plug in, and not interfere with the travel of the axis, over its entire range.
    c)
  3. I would like the CAD model to be parametric, such that it can be easily customized to routers of various diameters. I know how to do this in FreeCAD, but I am less confident on how this works in other CAD packages.
    a) A “diameter” parameter in the CAD model should be referenced by every dependent part in the model. When it is updated, the assembly should update. It should be very straightforward to generate the updated g-code from the updated dimensions.
  4. Can we all settle on a default max z-axis speed. If we can decide upon this, the pulley options would be reduced. I am looking for feedback on this.
  5. Can everyone agree upon a travel distance? For me, 6" sounds about right. Any thoughts?
    a) This could also be parameterized within the CAD model.
1 Like

I feel the same way, but I have seen people who would like to have a sled ready to go so that they don’t need any temporary parts. If I go the route of providing kits, I’d like to have the option available for those that feel that way.

Of course. I hesitated to even talk about kits before I’ve made one myself to confirm this system works. I’m willing to foot the time on proofing the design. In fact, part of me even regrets sharing the design until I had a working prototype. I feel like I’m letting users like you down by not having proofed this myself.

I originally tried designing a dust hood that was Maslow-able. It’s not to say that it’s not possible, but I wasn’t able to come up with a design I liked. The round section near the spindle isn’t the issue, in fact, you can still see the CNC influence in the current design. It was the hose adapters that I had trouble with and ultimately decided that it was easier for me to simply print them. Printers lend themselves to more complicated geometry with undercuts and the like. Admittedly, even deciding to print the parts put size limitations on me. That required it to be broken into several parts so that they would fit on my print bed.

One of the parts in the STL folder is a hose adapter for 2.5" hoses. It will need to be tested to ensure the fit is right, but it wouldn’t be hard to tweak to get right.

I realize that I haven’t put a link in for the Fusion model in this thread (I think I’ve put it elsewhere).

Fusion Model

I almost always design with user variables, that way the model can be changed easily. After working with architects for awhile, its really has paid off to be able to adapt the model on the fly like that. This is the variable list for this model:

The variables you had asked for would be SpindleDia and ZHeight. I plan on eventually upgrading my own spindle, so that’s part of the reason I put that variable in. The ZHeight I had to tweak quite a bit to get something that wasn’t too top-heavy and still gave a good amount of travel distance.

I can’t promise that Fusion will be able to resolve the geometry if you made a huge change to these variables, however.

I think that would be that the fastest we can make the axis move without stalling the motor (with a little margin). I’ve ordered options from 10 teeth to 30 teeth to do trials myself, but I will need time to test it. I will not be terribly upset if someone beats me to this. Ultimately, I don’t think the correct solution is to use the pulleys to gear up the motor, but to simply order a motor which turns at the correct rpm for our needs. When I dial in this speed, I will spec the correct motor to use with this rather than a series of different pulleys. If I go the route of offering a hardware kit, the new motor will be provided with it.

Keep in mind, that, even though the lead screw is 300mm, there is only about 125mm (~5in) between the bottom of collet and the bottom of the sled with the spindle fully retracted. I think that might be overkill, but I’m not sure.

In case this isn’t carrying through in my responses, I’m very glad that you’re asking these questions. This is helping solidify the design further and I think it’s really helping improve it. It’s probably also helping the lurkers who want to build this Z-axis as well who have similar questions.

2 Likes

Let me ask another question. Based on the picture, a 2.5" hose is necked down to fit through a smaller square hole. Is that correct? I was thinking the 2.5" hose was a minimum diameter. It is surprising how much airflow is needed to mitigate dust.

1 Like

Let me preface this by saying that I am not an airflow engineer and I have not done simulations to see how effective this would be. The design is entirely to contain and provide a physical connector for common shop vac and dust collector hoses.

Essentially, yes. It’s actually a round hole that the adapter interfaces with, but the schematics aren’t that important (as far as I know). I would have liked to made the chute wider to improve airflow, but I had to fit it between the linear rails and I only had so much room with the 45 degree linkage kit.

CNC routers do make a ton of dust, so I do mostly agree with you. However, I’ve been using a smaller diameter hose attached to my shop vac so far and it’s done alright with the dust. From what I’ve seen here on the forums, most people are using a shop vac. I wanted to make sure that the option was available without forcing people to make their own adapter, putting a larger hose on their shop vac, or upgrading their dust collection system.

The pictures I’ve posted so far are with a 1 1/2" adapter. This is what it looks like with the 2 1/2" adapter:

1 Like