[ishi]

I am designing the enclosure as a separate component in Fusion 360 in order to make it easier to hide it when necessary. The aluminum profiles are anodized black and the cast iron brackets have a black oxide finish. My goal is to give the outside enclosure a totally black finish, without having to use any additional paint or covering panels. Also, I am trying to design it in such a way that the machine enclosure does not look like a simple assembly of standard aluminum profiles, even though that is exactly what it is. In other words, I am aiming for a high-end look-and-feel while using standard parts with fairly uncommon finishes. The outside will be totally black, while the inside will be totally white. The contrast should be quite striking.

Generally-speaking, I am shooting for a distinctive look, but without having to design many custom parts.

Easier said than done...

[ishi]

The enclosure's frame will use these corner fastening sets:

https://product.item24.de/en/product...0-black-41608/

If necessary, they will be complemented by angle brackets to increase stiffness. But these make the mounting of cover panels a lot more challenging...

[ishi]

Here is the lower part of the enclosure frame.

[ishi]

We've added an enclosure frame mounting bracket on the back of the frame as well, with matching notch on the base casting. This will add a bit more rigidity to the enclosure's frame. It's not as critical as the one we've added on the front of the enclosure, but we might as well do it, just in case.

[ishi]

Here is the full frame for the enclosure, at the exception of the some profiles at the top that will be added once we redesign our ATC conveyor.

The profile view looks pretty nice...

***

High-res pictures on public drive:

https://drive.google.com/drive/u/3/f...3dQrx?ogsrc=32

[ishi]

The side vertical profiles for the machine enclosure have been moved forward by 300mm in order to fully enclose the ATC conveyor, but not more. That way, we still have full access to the table for the top loading of parts when the table is moved to its forward-most position alongside the Y axis.

This will also reduce the width of the side panels, from half of the machine's depth down to about a third.

Now, I am trying to figure out how I can design the front panels that would move up and down.

This one will be tricky... In fact, I'm not even sure that it's feasible at all, for a host of different reasons...

[ishi]

Following a first day of experimentations, I am in the process of redesigning our enclosure frame using double-width profiles like this one:

https://product.item24.de/en/product...natural-64938/

But the resulting geometries end up being a bit too complex for pencil on paper. At the same time, I'm not quite ready yet to work on the CAD model. Therefore, I went for LEGO bricks, which are a good middle ground. Here is what the right side of the frame might look like.

[deadlykitten]

hi ishi, i believe you are doing a great job ... pls, when you finish your cnc, instruct a good bunch of girls & women, because i am tired of working with men

.... i guess i am not the only one feeling this way

[rowbare]

To me there is way too much distance between your X and Y rails. They should be as close together as possible for maximum rigidity.

bob

[ishi]

hi ishi, i believe you are doing a great job ... pls, when you finish your cnc, instruct a good bunch of girls & women, because i am tired of working with men

.... i guess i am not the only one feeling this way

deadlykitten,

I will do my best. My class at the Community College last year was actually quite diverse. I would say a third of the participants were women.

Antarctica? Really? I would *love* to deliver a machine there! (^_^)

And thanks for the kind words.

Best regards
Ismael

[ishi]

To me there is way too much distance between your X and Y rails. They should be as close together as possible for maximum rigidity.

bob

Bob,

Thanks for the feedback.

Opinions on this matter seem to vary quite a bit. Some people agree with you, and some think it does not matter.

I am not a mechanical engineer, but my understanding is that you want to minimize deflection of the carriage sitting on top of the rail, therefore minimizing spacing is usually a good idea. By the same token, more spacing will facilitate squaring. Therefore, the best answer is probably somewhere in the middle.

But there are much more critical things to consider: for Y and Z, we positively cannot reduce that spacing, because of what goes between the rails: the rotary table's torque motor for Y and the backside of the spindle head for Z. Therefore, the spacing between the rails for Y is what it is and cannot change, and the same applies for the spacing between the rail beams for Z. Now, for X and Z, we could swap the rail beams and have rail shoulders be mounted from the outside instead of from the inside as is done today. This would shave off about 3" of rail spacing for both X and Z. Finally, for X, we could further reduce that spacing, but that would reduce the space available for chips to fall into the underside chip conveyor. Not a major problem, but still something to keep in mind.

Another important thing to consider regarding X: the downward force applied by the tool will always be coplanar with the vertical plane containing the longitudinal axis for the back rail of the X axis. That is because our tool does not move alongside the Y axis (our part does). As a result, this force won't have any impact on deflection for the horizontal carriage. Therefore, any deflection we could experience will be the result of static forces. Some of these forces will originate from the weight of machine components and could be compensated for at the software level in a static fashion. Others will come from the weight of the piece being cut and could be compensated for dynamically if that weight of is known in advance.

Bottomline: we will probably swap the rail beams (I don't see any downside in doing this) and narrow our chip evacuation canal.

Last comment: many top-of-the-line machines I have studied extensively (mostly from DMG MORI) have proportions similar to our machine's, therefore I doubt that we are too far off on that front.

[ishi]

Our enclosure frame will be made of two sub-assemblies: the inner enclosure frame and the outer enclosure frame. Pictures in attachment show the former one.

Several reasons motivate this design:

1. It will facilitate the design of our lateral sliding panels.
2. It will facilitate the design of our frontal rising panel.
3. It will make it easier to mount cover panels (using the outer frame).
4. It will make it easier to add square brackets for increased rigidity (onto the inner frame).
5. It will simplify the selection of profiles.

The last point is interesting. To better understand it, look at these profiles:

https://product.item24.de/en/product...ms-1001167391/

If we focus on the 40 x 40 profiles, we can see that they come in a variety of shapes, in relation to the t-slots that are either open or closed. For example, the standard one has all four slots open, giving it an X shape. One has only three, giving it what I call an I shape. One has two on opposite sides, giving it an H shape, while another has them on contiguous sides, giving it an L shape. The one with a single slot open has a C shape, and the one with all slots closed has an O shape.

This is important, because you want open slots if you are going to use them for mounting things, but you want them closed otherwise in order to prevent the accumulation of chips or dirt and give the enclosure a better look and feel.

For most profiles, the inner shell will tend to use mostly-open profiles, while the outer shell will use mostly-closed profiles.

This design is a bit challenging though, because it requires some profiles to be assembled in rather non-conventional ways, sometimes bolting two profiles together face to face. This is not an ideal solution, because it is not the most rigid assembly from a mechanical standpoint, but the enclosure is not going to deal with a lot of stress, and there are ways to make things more rigid by using mounting plates or cross-profile connections:

https://product.item24.de/en/product...ns-1001012006/

I got to learn a lot of that stuff when designing and building this monstrosity:

https://www.flickr.com/photos/841129...7606405587450/

You can think of it as a giant iPod on wheels...

Hopefully, I have learned a few things from past mistakes...

[Goemon]

Goemon,

Thanks for sharing this. This is a great project indeed.

I am quite familiar with what you are describing, and this is one of the reasons why our rails for X and Z are mounted on a steel beam that will either be incorporated into the cast or bolted onto it, using threaded inserts incorporated into the cast.

In order to familiarize ourselves with the process and design requirements, we are using this outstanding presentation from RAMPF:

https://drive.google.com/file/d/1t5g...ew?usp=sharing

Best regards
Ismael

I didn't achieve the best result with threaded inserts. I use them with success on my carbon fiber parts but they aren't subject to the constant linear motion stresses. Using them on granite on granite is one thing (as it is super hard) but I would be worried about them coming lose and hole widening with epoxy + constant linear motion stress.

I thing your embedded steel plate option is a better idea.

[ishi]

I didn't achieve the best result with threaded inserts. I use them with success on my carbon fiber parts but they aren't subject to the constant linear motion stresses. Using them on granite on granite is one thing (as it is super hard) but I would be worried about them coming lose and hole widening with epoxy + constant linear motion stress.

I thing your embedded steel plate option is a better idea.

Goemon,

Indeed, for the rails, I would not consider using inserts directly.

[ishi]

Here are some of the mineral castings done by our supplier. The last picture shows an example for a casting with rail beams embedded into the casting.

And it's interesting to note that they mount the rails from the inside, not from the outside as we were planning to. This is further confirmation that we might want to swap our rail beams, as discussed in an earlier post this morning.

[ishi]

Here is a first version of the enclosure frame with inner and outer sub-assemblies. Proper mounting brackets between the two sub-assemblies still need to be added, but the overall geometry is there. The 80mm section on the front of the machine will contain the front rising panel, which will be made of two horizontal sections, using 6 series profiles.

For ergonomics reasons, the top of the outer frame's front section is at the same height as the top of the rotary table.

[ishi]

Here is a third (and hopefully last) iteration of the enclosure frame. This version will provide a much more rigid assembly between the inner and outer sub-assemblies. Now working on the brackets to mount the outer frame onto the inner one and the fasteners to attach side panels.

Obviously, the frame will need to be adjusted in order to provide clearance on the right side for the chip conveyor and on the back for the pallet jack, but the overall geometry seems good enough for now.

[ishi]

These fasteners will be used for mounting static panels on the frame:

https://product.item24.de/en/product...zn-1001012498/

They will allow us to mount the panels perfectly flush with the outside surface of the profiles.

[ishi]

When using this angle hinge bracket, we still have 4mm of clearance for mounting our static panels flush with the profiles. Perfect!

https://product.item24.de/en/product...acket-8-45776/

[ishi]

Here are the openings for the chip conveyor and the pallet jack.

These also serve another purpose: adding some additional profiles to increase rigidity and reduce the size of individual static panels.