[JermNZ]

Hi Everyone,

I've been designing this router for a few weeks and though I'd share it on here since I've learnt so much from other build threads on the forum.

Working envelope: Approximately 860mm x 610 mm x 144 mm

Components:
Bosch Rexroth 90x45 and 90x90 extrusion
Linear Slides SSR15x760 mm, SSR20x1080 mm SR15x300 mm
Ballscrews RM1605 - 290 mm, 790 mm, 1060 mm (x2)
Nema 23 425 oz-in motors
1.5kw Water-cooled spindle
CNC machined 6061 aluminium plate, mostly 15 mm thick.

I previously built an Openbuilds Ox router, but found it lacking in rigidity. I'd like to be able to comfortable machine aluminium, so I have tried to stiffen the gantry as much as possible with four braces between the extrusions, the end two also double as ball-screw mounting points.

The gantry side plates are 15 mm thick, do you think this will be sufficient? If not i'd have to consider adding some bolt on fin braces to the sides if needed.

Check out the images below and let me know what you think, I am very open to suggestions and constructive criticism on the design!

Attachment 269618Attachment 269620Attachment 269622Attachment 269624

Regards,
Jeremy

[aarggh]

Nice job so far Jeremy. If you want to mainly do aluminium machining though, you'll need to stiffen up the base of the gantry, as on fast moves and under load the whole gantry will skew sideways at the base. The top is really well braced, but you need some support beams at right angles to the gantry sides to stop flexing.

Another thing that's really important, and can dramatically help, is don't be afraid to add heaps of mass to the frame. The more rigid and heavy the frame and gantry are, the better it will dampen vibration and noise, and reduce tool flexing.

looking forward to seeing how this build progresses!

cheers, Ian

[ger21]

Get rid of the cutouts in the gantry sides. All they're doing is making them weaker.

I'd also lower the gantry if you can, so the Z axis doesn't hang down as far. The lower the beam, the more rigid the Z axis will be.

[wizard]

I have some comments below. I'm trying to keep in mind that this is a smallish router.

Hi Everyone,

I've been designing this router for a few weeks and though I'd share it on here since I've learnt so much from other build threads on the forum.

Working envelope: Approximately 860mm x 610 mm x 144 mm

Components:
Bosch Rexroth 90x45 and 90x90 extrusion
Linear Slides SSR15x760 mm, SSR20x1080 mm SR15x300 mm
Ballscrews RM1605 - 290 mm, 790 mm, 1060 mm (x2)
Nema 23 425 oz-in motors
1.5kw Water-cooled spindle
CNC machined 6061 aluminium plate, mostly 15 mm thick.

I previously built an Openbuilds Ox router, but found it lacking in rigidity. I'd like to be able to comfortable machine aluminium,

Unfortunately that phrase means very little, it could mean thin electrical panels or large aluminum castings. Given that I find your design troubling.

so I have tried to stiffen the gantry as much as possible with four braces between the extrusions, the end two also double as ball-screw mounting points.

I really don't like this approach in a gantry, in fact your gantry design is why I find the whole machines design troubling. I generally don't like T-Slotted extrusions in this sort of application, that is a gantry beam but I especially find it troubling when the beams are bolted together in such a haphazard way. I just don't see those four gussets doing much for you in connecting the two beams rigidly together. On sampler machines it is possible to use a single T-Slot extrusion of large relative cross section to function as a beam. You might even be able to come up with a good design using two extrusions if they are tightly coupled together.

In the end though economics will likely lead you to using a large square tubular beam.

The gantry side plates are 15 mm thick, do you think this will be sufficient? If not i'd have to consider adding some bolt on fin braces to the sides if needed.

Not even close! Not so much the thickness, thought that is a problem but the cut outs do nothing for you positive. Beyond that we come back to the issue of stiffness and a box section can make a lot of sense here. In fact even extrusions make more sense than flat sheet stock.

Check out the images below and let me know what you think, I am very open to suggestions and constructive criticism on the design.

Unfortunately I can't respond to specific pictures at the moment. The problem I see here is that if your open builds machine isn't cutting the mustard you aren't taking a big step up with this design. If you want to stay with extrusions for the gantry beam they need to be rigidly coupled together. Since I don't know what you mean by machining aluminum most likely they should be larger beams also.

To put this another way, almost any router type machine can machine aluminum if you work within its capabilities. However in your case you have already said that one machine you have simply isn't good enough, not rigid enough. If this is the case why not put significant effort into the next build to make it far more robust to meet your machining needs? If the current machine doesn't live up to your expectations then make darn sure the second machine does! My first look at the gantry had me saying what the hell.

Regards,
Jeremy

If you read my posts ove time here you will find I'm not a big fan of extruded aluminum framing for use as gantry beams, especially beams that end up as fabrications out of several beams, plates or whatever. Don't take that to mean I'm against the use of aluminum extrusions on machinery, on the contrary I just ordered a bunch of stuff for work. The problem is two fold, the stuff isn't cheap especially when multiple beams are fabricated together. The second problem is the stuff doesn't hold up well to twisting and shifting seen in a construction like you have pictured.

This second issue is bothersome to me when I see how you intend to bound the two beams together with those gussets. Unless you add gussets to the gussets I really would expect them to loosen up over time. You don't have details of the fastening of those plates but I'm imagining a couple of screws run into the plates from the extrusion T-Slots. If I was to do this I'd seriously consider arraigning the beams so that they are tied together with a single plate running the entire length of the beam and bolted to all T-slots it sits over with bolts every 40-50 mm. You want to effectively turn the beam parts into a single piece and do so in a way that remains secure for the long haul. The problem is by the time you get this all done a square steel beam will likely be cheaper and just as rigid.

[JermNZ]

Thanks for your replies everyone.

Firstly I would like to clarify that I'd like to be able to profile/pocket and drill 6-20mm aluminum plate.

I did wonder about the cutouts as initially I was trying to reduce the weight of the gantry unit, but as Ian (aarrgghh) mentioned I do think having a heavy gantry may actually be a benefit.

As I mentioned I was thinking about adding extra bracing on the sides of the gantry plates to stiffen them up, similar to the "ripper CNC". MAXMALI.COM | ripper – build log / part 4

I think lowering the gantry is also a good idea, It currently has a about 140mm of clearance, with the intention of maybe using a small vise as a fixture on the bed I kept it relatively high. I may bring it down to 120 mm or even 100 mm. I will also take a closer look that gantry base design to reduce the amount of sideways flex of the gantry.

In regards to the gantry braces, I don't really see why they won't provide sufficient bracing between the two extrusions if fixed properly, if I do have an issue I will remove them and replace with a folded sheet steel brace at a later date. I think that's the beauty of these DIY machines, the fact that they can somewhat continually evolve and be upgraded where needed.

I have also seen several people warn about using t-slot extrusion for the construction of CNC machines, although I have also seen many machines online which use extrusion and it seems to perform very well. For example Neo7CNC's machine which you can see machining aluminum in the videos below. There are also the obvious benefits of working with t-slot extrusion, ease of fixture to linear rails, ease of cutting and machining etc.

https://www.youtube.com/watch?v=D4-p...qJqxPDEAIQR1cw

In light of all of your comments and suggestions I will revise my design and post some more images in a few days.

[ger21]

Firstly I would like to clarify that I'd like to be able to profile/pocket and drill 6-20mm aluminum plate.

If that's the main purpose of the machine, than build it as solid as possible.

although I have also seen many machines online which use extrusion and it seems to perform very well. For example Neo7CNC's machine which you can see machining aluminum in the videos below.

People don't use extrusions because they're the best way to build a machine. They use them because they're the easiest way.
If you're going to use extrusion, then consider one big, heavy wall extrusion for your beam. It'll be stiffer than the two smaller ones.

[wizard]

Thanks for your replies everyone.

Firstly I would like to clarify that I'd like to be able to profile/pocket and drill 6-20mm aluminum plate.

I did wonder about the cutouts as initially I was trying to reduce the weight of the gantry unit, but as Ian (aarrgghh) mentioned I do think having a heavy gantry may actually be a benefit.

There are trade offs, a heavy gantry means you need more powerful drivers to get acceptable accelerations. One thing to consider is that a moving table design should allow you to build a stiffer machine with possibly less effort and materials. The big problem is that you are right on the border of being to big machine wise for this to work well.

As I mentioned I was thinking about adding extra bracing on the sides of the gantry plates to stiffen them up, similar to the "ripper CNC". MAXMALI.COM | ripper – build log / part 4

I think lowering the gantry is also a good idea, It currently has a about 140mm of clearance, with the intention of maybe using a small vise as a fixture on the bed I kept it relatively high. I may bring it down to 120 mm or even 100 mm. I will also take a closer look that gantry base design to reduce the amount of sideways flex of the gantry.

Think about this long and hard. Too little clearance significantly limits what you can do with the machine. If you have the intention of machining parts that might best go into a vise then you need the clearance. However that does mean you need a stronger machine especially with respect to the gantry. Honestly 40 mm isn't really much.

In a nut shell if you think you might have materials or processes that would use that clearance then it would be wise to build for that capability. It is sort of like the potential for a fourth axis, if you suspect that you will be using one it is best to design in the accommodation from the beginning. This does mean a significantly stiff gantry though.

In regards to the gantry braces, I don't really see why they won't provide sufficient bracing between the two extrusions if fixed properly, if I do have an issue I will remove them and replace with a folded sheet steel brace at a later date. I think that's the beauty of these DIY machines, the fact that they can somewhat continually evolve and be upgraded where needed.

Well you can try. Without additional bracing I just don't see a secure way to built them in place. The thing that bothers me the most though is that this is a very fiddly design. A big square tube or even a really large extrusion would be cheaper, easier to build.

I have also seen several people warn about using t-slot extrusion for the construction of CNC machines, although I have also seen many machines online which use extrusion and it seems to perform very well.

You can certainly produce a decent if limited machine, completely out of extrusions. The problem as I see it is that you already have a machine that was built out of extrusions that didn't work out well for you. You didn't go into specifics but the lack of stiffness probably impacted finish badly. If your primary focus for this machine is working aluminum then it would pay to address these issues directly. I just don't see the bolting together of extrusions that are obviously too small as the way to go, especially if the joining method leaves a lot to be desired.

The gantry is a critical part of any router type machine, it has to be strong enough to handle the challenges placed on it by the machining operations.

For example Neo7CNC's machine which you can see machining aluminum in the videos below. There are also the obvious benefits of working with t-slot extrusion, ease of fixture to linear rails, ease of cutting and machining etc.

Certainly there are benefits but there are also significant limitations and what are some peoples positives are other people's negatives. For example mounting linear rails looks easy but there are a couple of problems with those T slots. For one there is a huge amount of play in the slots. Second; the mounting points are less than robust which can be seen in any cross section drawing of the extrusions.

As for machining, Aluminum certainly is easy to machine relatively. However you are building a machine that I would imagine you expect to keep around for some time. As such the little extra effort to do steel isn't that significant. Beyond that there is nothing to stop you from buying a big aluminum tube except for possibly the costs. I just don't see it as being reasonable to dismiss the alternatives based on the difficulty to drill a few holes.

There are larger extrusions to consider, for example: Phoenix Mecano » Industrial, Handheld, and EX Enclosures + Automation Solutions and Mitsumi has 100 x 200 extrusions. Mitsumi also offers machining options to flatten the T slot surfaces for linear rail mounting. Spend some time looking around and you can find others. The point here is why not avoid a built up beam if you can get a solution that will do the job off the shelf.

In light of all of your comments and suggestions I will revise my design and post some more images in a few days.

Just remember these are opinions. In the end you need to be happy with the design.

A side note on that video referenced above. We never did see which of his machines was used for that video nor did we see the resultant finish quality.

[louieatienza]

For the gantry, you can space the extrusion with aluminum channel, drilled and bolted to the extrusion.

Not all extrusion is made the same. Compare 80/20 stuff with Misumi GFS series with 6mm walls throughout. They even sell them face milled for attaching LM rails.

I've cut aluminum with a wood frame router, phenolic frame router, and t slot router. It's more how you do it than anything. Stiff frame helps greatly, but make it as stiff as you can with the components you have available. You can check my build threads and thread on bits and endmills to see examples and videos.

[JermNZ]

There are trade offs, a heavy gantry means you need more powerful drivers to get acceptable accelerations. One thing to consider is that a moving table design should allow you to build a stiffer machine with possibly less effort and materials. The big problem is that you are right on the border of being to big machine wise for this to work well.

Yep I had considered more of a "mill" or moving bed design. Although working as a product development engineer I spend a significant amount of my time designing and machining prototype injection mould tools, so I have access to 3 axis Haas machines and 5 axis Deckels. All of these machines are setup with vise fixtures so this machine I am building is going to be predominantly for machining sheet and plate.

Well you can try. Without additional bracing I just don't see a secure way to built them in place. The thing that bothers me the most though is that this is a very fiddly design. A big square tube or even a really large extrusion would be cheaper, easier to build.

Yeah being in New Zealand materials can be quite expensive, I am looking into what other options I can find for my gantry. I chose this extrusion as it was available easily through my work.

You can certainly produce a decent if limited machine, completely out of extrusions. The problem as I see it is that you already have a machine that was built out of extrusions that didn't work out well for you. You didn't go into specifics but the lack of stiffness probably impacted finish badly.

The main cause of the lack of rigidity was the roller system which the extrusions moved on, play was inherent in the design.

Thanks for your input, I'll work on some modifications this week.

[JermNZ]

Hi Guys,

Since I last posted I have further developed my router design. I have taken considered everyone's suggestions and tried to provide a stiffer structure by raising the rails and redesigning the gantry base.

I have also opted for a welded steel base, two 150x50x4 side rails with four 50x50x2.5 cross rails. One of the reasons that I changed to steel was that the quoted price for the 90x90 Bosch extrusions which I had previously planned to use came to $175/m !! Some of the types of materials are difficult to get hold of for a reasonable price in this part of the world.

I am sticking with the twin 90x45 extrusion gantry at the moment and I have some ideas about how to brace the two rails together in order to further stiffen the structure.

For electronics I am thinking about splashing out and getting one of these kits from CNC Router Parts. I may also get a Smoothstepper motion controller to avoid the archaic parallel port.

4-Axis DIY Nema 23 Electronics Kit | CNCRouterParts

Here's some pics of the new design, the dark coloured components will be powdercoated steel and the lighter stuff is mostly 6061 Aluminium.

Please excuse the half finished model.

Attachment 272154

Attachment 272156

[aarggh]

Some of the types of materials are difficult to get hold of for a reasonable price in this part of the world.

I thought you must from OZ with that remark! :-)

Nice design now, looks robust and should work a treat, the only thing I would do further is to allow for a bracing sheet of aluminium or steel, doesn't have to thick at all, to screw along the rear of the gantry from end to end, that also screws into the top and bottom extrusions of the gantry. This will greatly add to the rigidity, and help further reduce flex and noise of the gantry, especially at higher speeds and load.

cheers, Ian

[wendtmk]

Not sure if it's a huge thing, but it looks like, in your renderings, you have the weight of your gantry outboard of your rails. This will put a constant rotational stress on your rails and bearings, due to the weight/mass of your gantry. I don't know if that will be a real factor in the life of your bearings or rails, but if it was me, I'd put the weight/mass of the gantry directly over top of the rails. The manufacturers put out specs for max vertical and horizontal forces on the bearings and rails, but I'm not sure if they put out specs for rotational forces on the bearings and rails.

[ger21]

This will put a constant rotational stress on your rails and bearings, due to the weight/mass of your gantry.

Wouldn't some triangular braces on the inside of the uprights eliminate that?

[wizard]

Hi Guys,

Since I last posted I have further developed my router design. I have taken considered everyone's suggestions and tried to provide a stiffer structure by raising the rails and redesigning the gantry base.

I have also opted for a welded steel base, two 150x50x4 side rails with four 50x50x2.5 cross rails. One of the reasons that I changed to steel was that the quoted price for the 90x90 Bosch extrusions which I had previously planned to use came to $175/m !! Some of the types of materials are difficult to get hold of for a reasonable price in this part of the world.

Wow, high prices for extrusions. So why not go steel for the rest of the machine?

I am sticking with the twin 90x45 extrusion gantry at the moment and I have some ideas about how to brace the two rails together in order to further stiffen the structure.

Why extrusions if they are so expensive relative to steel in your country. Further a big square steel beam will be stiffer than two small cross section extrusions. All things considered I would reverse the materials usage and put a steel beam in the gantry and use aluminum for the frame.

For electronics I am thinking about splashing out and getting one of these kits from CNC Router Parts. I may also get a Smoothstepper motion controller to avoid the archaic parallel port.


Here's some pics of the new design, the dark coloured components will be powdercoated steel and the lighter stuff is mostly 6061 Aluminium.

Please excuse the half finished model.

I'd still consider stiffer side plates to support the gantry, again something made out of square or rectangular tubing. I'd also center the side plates above the linear bearings to simplify things a bit. The other option is to weld up something out of plate material.

You indicated access to milling machines, if this is so (a great advantage) I'd seriously consider more welded up structures. That is structures welded up out of steel designed to be bolted together! Most people don't have access to a well equipped machine shop so it is hard to recommend this all the time, however it can lead to a very high quality machine at relatively low costs. This given proper structural design of those components.

In the end the gantry is a very important part of a router design. Is your primary goal is the machining of aluminum you should strive to make the machine as stiff as possible and stay with in your budgetary goals. In the end a stiffer machine means better finish quality or more aggressive machining capabilities.

[JermNZ]

I thought you must from OZ with that remark! :-)

We're even further away in NZ!

Nice design now, looks robust and should work a treat, the only thing I would do further is to allow for a bracing sheet of aluminium or steel, doesn't have to thick at all, to screw along the rear of the gantry from end to end, that also screws into the top and bottom extrusions of the gantry. This will greatly add to the rigidity, and help further reduce flex and noise of the gantry, especially at higher speeds and load.

cheers, Ian

Yep, this is what I had in mind, just haven't gotten around to designing/modeling it.

Thanks for the reply.

[JermNZ]

Not sure if it's a huge thing, but it looks like, in your renderings, you have the weight of your gantry outboard of your rails. This will put a constant rotational stress on your rails and bearings, due to the weight/mass of your gantry. I don't know if that will be a real factor in the life of your bearings or rails, but if it was me, I'd put the weight/mass of the gantry directly over top of the rails. The manufacturers put out specs for max vertical and horizontal forces on the bearings and rails, but I'm not sure if they put out specs for rotational forces on the bearings and rails.

Thanks for the reply..

I also had this though initially, but after further analyzing the situation I have to disagree.

Assuming that the right angle join between the gantry side plate and gantry base plate is rigid, the linear slide bearing block will experience mostly only vertical and horizontal loads. Only if that right angle join is allowed to flex will the bearing block see a moment force, this is why I have gone to the trouble of reinforcing that join with another aluminum block which is fastened to both the side and base plates.

Am I missing something?

[JermNZ]

Wow, high prices for extrusions. So why not go steel for the rest of the machine?

Why extrusions if they are so expensive relative to steel in your country. Further a big square steel beam will be stiffer than two small cross section extrusions. All things considered I would reverse the materials usage and put a steel beam in the gantry and use aluminum for the frame.

I actually have a scrap length of this 90x45 which I intend to use. As I said before, I like the ease of attachment to the T-slot rail, limit switches, end stops, cable routing etc..

I'd still consider stiffer side plates to support the gantry, again something made out of square or rectangular tubing. I'd also center the side plates above the linear bearings to simplify things a bit. The other option is to weld up something out of plate material.

You indicated access to milling machines, if this is so (a great advantage) I'd seriously consider more welded up structures. That is structures welded up out of steel designed to be bolted together! Most people don't have access to a well equipped machine shop so it is hard to recommend this all the time, however it can lead to a very high quality machine at relatively low costs. This given proper structural design of those components.

In the end the gantry is a very important part of a router design. Is your primary goal is the machining of aluminum you should strive to make the machine as stiff as possible and stay with in your budgetary goals. In the end a stiffer machine means better finish quality or more aggressive machining capabilities.

I have increased the side plated to 18 mm thick and unlike my original design there is barely any part of the gantry side plate that isn't supported/reinforced, The top section is covered by the extrusion ends and the bottom section is reinforced by the joining block. I believe this will offer more than sufficient stiffness for my purpose.

To be honest, I do not have that much experience with welding, especially structures which require relatively good geometric tolerances. Which is why I tend towards the machining process.

I am actually looking forward to attempting to weld this base frame myself and putting much consideration into how I can achieve a nice square structure, free of heat induced warping. If it goes well and I find that the extrusion gantry isn't working for me I may redesign the gantry and use steel beams instead. I think this iterative development process is the beauty of these DIY projects!

[wendtmk]

Wouldn't some triangular braces on the inside of the uprights eliminate that?

Not really, with the mass/weight still being outboard of the bearings.

[ger21]

But as the OP said, if the joint doesn't flex, then the load is vertical, not twisting. (although technically, everything flexes a little.)
Even though the weight is outside of the blocks, the blocks still see a mostly vertical load on them.

[JermNZ]

But as the OP said, if the joint doesn't flex, then the load is vertical, not twisting. (although technically, everything flexes a little.)
Even though the weight is outside of the blocks, the blocks still see a mostly vertical load on them.

I agree, obviously with the assumption that the gantry is a symmetrical setup which prevents either of the gantry sides from rotation about the axis of the linear slide.