[JoanTheSpark]

What is the reason, that I don't seem to find a lot (actually none) designs/builds that try to scale down those large portal mills to shop format?
IMHO the design should be sturdier than a gantry design (moving weight can go into rigidity of essential parts) and the rails/drives are higher up and away from chips.
What am I missing?

[mactec54]

What is the reason, that I don't seem to find a lot (actually none) designs/builds that try to scale down those large portal mills to shop format?
IMHO the design should be sturdier than a gantry design (moving weight can go into rigidity of essential parts) and the rails/drives are higher up and away from chips.
What am I missing?

The dust will be much the same, If the moving gantry rails are under the table they are less prone to dust

There are lots that build them like this, the only problem you have is less access to the table, and loading material is not as easy

A well designed moving Gantry does not have any rigidity problems, and will work just as well, and you will have full table access

[handlewanker]

Hi.....by posting the sketch of "that" design, what are you trying to say?.......it's a monstrosity.....my opinion in case Mac wants to fire a broadside.

BTW.....what is a portal mill????.....the sketch you posted is of a gantry router.....there is no comparison between a router and a mill..
Ian.

[JoanTheSpark]

I just got that from one of the threads here, where someone posted an image of a big one and they call it a portal milling machine - I mean I can find them via that with google.
I assumed that would be the difference between a gantry (legs/columns connected with the beam, rolling on the floor) and this design with the high walls (beam moving alone, no legs).. probably wrong then?



I checked wikipedia and came across this (https://en.wikipedia.org/wiki/Gantry_crane):
"They are also called portal cranes, the "portal" being the empty space straddled by the gantry. The terms gantry crane and overhead crane (or bridge crane) are often used interchangeably, as both types of crane straddle their workload. The usual distinction drawn between the two is that with gantry cranes, the entire structure (including gantry) is usually wheeled (often on rails). By contrast, the supporting structure of an overhead crane is fixed in location, often in the form of the walls or ceiling of a building, to which is attached a movable hoist running overhead along a rail or beam (which may itself move)."
Not the kind of distinction I'm after really, but it conveys the thoughts I had about gantry/portal mills before I read the article about cranes.
I don't think one would want to call this design an overhead mill/router or even a bridge mill/router - the last one already being taken by 'bridge mills' from what I have learned so far?

As for 'my monstrosity', what's wrong with it?
From my experience the flat pieces for the walls and bed can be welded without skew (SHS) and should come out pretty flat, just being bolted together to get the box. I've done stuff like this before, just not that 'fat' or for such an application.
Don't you guys usually suggest that bigger, heavier and stiffer is better when people rock up with those router designs that sport thin uprights for the gantry / thin cross sections /etc.?
And I kinda want to have the dirt contained from the get go if possible. Some sheet metal on the walls from the 'inside' and a door at the front and it's closed for the worst of it, not messing up the shop.


@mactec45
Loading/unloading shouldn't be worse than for one of those vertical machining centers then I guess? I somehow expected a more 'sinister' reason that I didn't recognize.. cool. Thanks.

[mactec54]

I just got that from one of the threads here, where someone posted an image of a big one and they call it a portal milling machine,]

Don't bother with what Handlewanker says, he is very miss informed, on machine building, that's no router Handlewanker it's a 5 axes Milling machine

There is nothing wrong with your design, others have made machines just like this, and there is no reason not to build them like this, the only thing like I said is loading large pieces, there are some large commercial routers also manufactured just like your design

Loading/unloading shouldn't be worse than for one of those vertical machining centers then I guess? I somehow expected a more 'sinister' reason that I didn't recognize.. cool. Thanks.

You can't compare your router, to a machining center, it would be a very large machining center, that has a 1.6m Y axes, front loading machines like the Tormach are very easy to load as you park the table at the front, with your design the table does not move, the big portal milling machines like your photo you can walk inside if you have to

[JoanTheSpark]

Well, I was expecting that once pieces get heavier than 15-20 kgs I need to use a mobile crane anyway.
But anything smaller I should be able to put in the front there, doesn't matter for the mill where it is located on the table.
And for larger lighter stuff it should be easy to put it in by hand.
If I have to clamp down somewhere I don't reach I can always climb into the machine (if it wouldn't support me without budging I did something wrong anyway) - walking into would be nicer naturally - but I don't have that kind of money ;-)

Anyhow, thanks for the clarifications.
Just one more question that remains, as I wouldn't have started this thread if I'd found any examples for something like this in this size.
Do you have any links to builds like this - large vertical/z travel as I'm especially interested how stiffness/rigidity has been solved before?
I'm sure I can learn a lot by looking at others peoples work, at least that's how it usually works for me. Still enough that goes wrong even with that kind of help :-)

I'm currently doodeling around in my cad tool with the x/z node, trying to get it stiff and manufacturable (by the means that are available to me locally).
I can get parts lasercut up to 20mm mild steel and machining of some faces of smaller parts is also cared for.
So yeah, nothing carved in stone. I'm still open to input and options and will have more questions as time goes by.

[mactec54]

Well, I was expecting that once pieces get heavier than 15-20 kgs I need to use a mobile crane anyway.
But anything smaller I should be able to put in the front there, doesn't matter for the mill where it is located on the table.
And for larger lighter stuff it should be easy to put it in by hand.
If I have to clamp down somewhere I don't reach I can always climb into the machine (if it wouldn't support me without budging I did something wrong anyway) - walking into would be nicer naturally - but I don't have that kind of money ;-)

Anyhow, thanks for the clarifications.
Just one more question that remains, as I wouldn't have started this thread if I'd found any examples for something like this in this size.
Do you have any links to builds like this - large vertical/z travel as I'm especially interested how stiffness/rigidity has been solved before?
I'm sure I can learn a lot by looking at others peoples work, at least that's how it usually works for me. Still enough that goes wrong even with that kind of help :-)

I'm currently doodeling around in my cad tool with the y/z node, trying to get it stiff and manufacturable (by the means that are available to me locally).
I can get parts lasercut up to 20mm mild steel and machining of some faces of smaller parts is also cared for.
So yeah, nothing carved in stone. I'm still open to input and options and will have more questions as time goes by.

That's a good configuration for that axes support, some will say you don't need 3 linear rails, but I say more the better, something for you to look at, these do not all look like yours, but have the same concept

This is one of many of similar concept, https://www.finelineautomation.com/p...-4?taxon_id=54

This is made of wood and metal, a lot of these have been made, same concept, not what you have in mind, but the same idea is there http://www.cnczone.com/forums/momus-...176-forum.html

There is another very good build that is on the Zone somewhere and was built for a 5 axes build

[handlewanker]

I just got that from one of the threads here, where someone posted an image of a big one and they call it a portal milling machine - I mean I can find them via that with google.
I assumed that would be the difference between a gantry (legs/columns connected with the beam, rolling on the floor) and this design with the high walls (beam moving alone, no legs).. probably wrong then?



I checked wikipedia and came across this (https://en.wikipedia.org/wiki/Gantry_crane):
"They are also called portal cranes, the "portal" being the empty space straddled by the gantry. The terms gantry crane and overhead crane (or bridge crane) are often used interchangeably, as both types of crane straddle their workload. The usual distinction drawn between the two is that with gantry cranes, the entire structure (including gantry) is usually wheeled (often on rails). By contrast, the supporting structure of an overhead crane is fixed in location, often in the form of the walls or ceiling of a building, to which is attached a movable hoist running overhead along a rail or beam (which may itself move)."
Not the kind of distinction I'm after really, but it conveys the thoughts I had about gantry/portal mills before I read the article about cranes.
I don't think one would want to call this design an overhead mill/router or even a bridge mill/router - the last one already being taken by 'bridge mills' from what I have learned so far?

As for 'my monstrosity', what's wrong with it?
From my experience the flat pieces for the walls and bed can be welded without skew (SHS) and should come out pretty flat, just being bolted together to get the box. I've done stuff like this before, just not that 'fat' or for such an application.
Don't you guys usually suggest that bigger, heavier and stiffer is better when people rock up with those router designs that sport thin uprights for the gantry / thin cross sections /etc.?
And I kinda want to have the dirt contained from the get go if possible. Some sheet metal on the walls from the 'inside' and a door at the front and it's closed for the worst of it, not messing up the shop.


@mactec45
Loading/unloading shouldn't be worse than for one of those vertical machining centers then I guess? I somehow expected a more 'sinister' reason that I didn't recognize.. cool. Thanks.

As I said....it's just my opinion.....you either like it or not.

One thing you come to realise when you work with machinery all your working life........the hands on experience which Mac doesn't have......you cannot scale nature and attempting to scale a full size machine down is also not practical.......something Mac doesn't understand as he's not a skilled engineering person, just an armchair engineer who likes blowing off now and again

So, we're talking about a moving gantry router not a mill....portal or otherwise......that limits your range of materials.

Attached is a pic of my favourite "portal"??? mill design made in Hindustan......which in actual fact is commonly known as a moving table CNC gantry router.....call it what you like......a rose by any other name is still a rose.

If I were to build a serious CNC router that is what I would base my design on.

At the end of the day you have to work with what you are capable of achieving.

What was your working envelope or table size for a......machine......to satisfy your needs, and what range of materials are you going to use on it?
Ian..

[JoanTheSpark]

Attached is a pic of my favourite "portal"??? mill design made in Hindustan.
{...}
If I were to build a serious CNC router that is what I would base my design on.

I think I've run across your current machine and it is of that design, a moving table fixed gantry mill.

Space wise you need longer y-rails and ball screws to move the part past your milling head (rail length = 2x table length).
Might be OK for small designs, but the y-dimension of your gantry envelope vs. the y-dimension of the work piece length (or table) make this not economical for longer tables (strength of ballscrew vs diameter, flex, etc.).
Also the y-drive then needs to work for your heaviest work piece, while for the moving gantry design the moving weight is relatively small/constant.
The only pro I can see is that you get a single drive for the y-axis which means your gantry doesn't skew and that you can build the gantry very heavy to counter flex.

Also, per common wisdom (I'm lazy and will just follow the herd here ;-) ):
https://www.reddit.com/r/CNC/comment...outer_vs_mill/
"I believe the most succinct differentiation is that a mill makes cuts using torque, while a router makes cuts using rotational speed.
A mill will have a heavy structure relative to the working area and will use a spindle with finely grained speed control.
A router will be relatively lighter for the same working area and will mount a high speed spindle, trim router, or full-size router or similar tool."

TL;DR: mill/router doesn't differentiate between the positional arrangement of the axis drives, but only concerns with the cutting force/speed/materials involved.

And portal vs. gantry for cranes doesn't exist, it's the same, thus same should be true for mills/routers, with the caveat that there are no overhead mills/routers which would distinguish the moving column from the fixed column versions (both moving gantry). This might be a reason some manufacturers refer to the ones with fixed columns/moving gantry as portal instead of gantry.

What was your working envelope or table size for a......machine......to satisfy your needs, and what range of materials are you going to use on it?

wood, plastic, etc - 1200x800x300
aluminium, mild steel, etc - 1000x300x300

I'm still torn between building a single machine or making two - space available vs money vs reality.
I suppose there is no spindle for hobby kind of money that can cover both, otherwise I'd have come across one by now on this forum I guess.
A BT30 milling spindle with motor in the 1.5-2kW size as a single item/unit seems not to exist either.

This is the example I'm mostly hanging onto for the size/power/capabilities of the machine:
http://www.cnczone.com/forums/uncate...aluminium.html

[handlewanker]

I think I've run across your current machine and it is of that design, a moving table fixed gantry mill.

Space wise you need longer y-rails and ball screws to move the part past your milling head (rail length = 2x table length).
Might be OK for small designs, but the y-dimension of your gantry envelope vs. the y-dimension of the work piece length (or table) make this not economical for longer tables (strength of ballscrew vs diameter, flex, etc.).
Also the y-drive then needs to work for your heaviest work piece, while for the moving gantry design the moving weight is relatively small/constant.
The only pro I can see is that you get a single drive for the y-axis which means your gantry doesn't skew and that you can build the gantry very heavy to counter flex.

Also, per common wisdom (I'm lazy and will just follow the herd here ;-) ):
https://www.reddit.com/r/CNC/comment...outer_vs_mill/
"I believe the most succinct differentiation is that a mill makes cuts using torque, while a router makes cuts using rotational speed.
A mill will have a heavy structure relative to the working area and will use a spindle with finely grained speed control.
A router will be relatively lighter for the same working area and will mount a high speed spindle, trim router, or full-size router or similar tool."

TL;DR: mill/router doesn't differentiate between the positional arrangement of the axis drives, but only concerns with the cutting force/speed/materials involved.

And portal vs. gantry for cranes doesn't exist, it's the same, thus same should be true for mills/routers, with the caveat that there are no overhead mills/routers which would distinguish the moving column from the fixed column versions (both moving gantry). This might be a reason some manufacturers refer to the ones with fixed columns/moving gantry as portal instead of gantry.


wood, plastic, etc - 1200x800x300
aluminium, mild steel, etc - 1000x300x300

I'm still torn between building a single machine or making two - space available vs money vs reality.
I suppose there is no spindle for hobby kind of money that can cover both, otherwise I'd have come across one by now on this forum I guess.
A BT30 milling spindle with motor in the 1.5-2kW size as a single item/unit seems not to exist either.

This is the example I'm mostly hanging onto for the size/power/capabilities of the machine:
http://www.cnczone.com/forums/uncate...aluminium.html

Hi.....steel?......you would need to know your stuff to design for that beastie and have a good workshop inventory to do it.

CNC routers, no matter how you are gifted in the design or manufacturing department, just aren't up to it......wood, plastics, maybe aluminium, but steel......forget it.

You "could" over engineer anything to make it work, but a practical design to make it work so far has not eventuated in any of the builds, that had steel as a desirable material to cut, on any I've looked at on the forum.

One reason gantry type routers are not easy to build is the mass required to absorb cutting forces and deflective loads.........a .02mm deflection at the cutter due to a flimsy build means the cutter is on a self destruct course from vibration.

The big problem for all machines that go as benchtop or floor standing is having enough grunt in the spindle to make cutting a reality instead of a joke.

Mills tend to have a better configuration as they can have a separate spindle and more powerful belt driven motor to get the low down torque that you must have for steel hacking while at the same time stepped pulleys or a high speed spindle can be attached to do whatever you need in the higher rpm area.

It's all a matter of personal preference.....we could debate the attributes of many designs for ever, but in the end it can turn out to be just another common or garden CNC router of one description or other.....and the Earth will never move beneath your feet.......everyone plays that game......no two designs are exactly the same, but they all do the same thing more or less.
Ian.

[NIC 77]

I'm still torn between building a single machine or making two - space available vs money vs reality.

Make one. You will undoubtedly want to change something (fix mistakes and make improvements) in the way you do things after the 1st one, so the 2nd one is better.

I suppose there is no spindle for hobby kind of money that can cover both, otherwise I'd have come across one by now on this forum I guess.

I don't think you've grouped those together correctly. A spindle that is good at cutting aluminum should be good for wood too. For steel, it's high torque at lower RPM's you want obviously. Most (99%) of the machines we make wouldn't be rigid enough to cut steel regardless of what spindle was used. Builds that can do it are rare. Perhaps that is also a reason you haven't come across it.

This is the example I'm mostly hanging onto for the size/power/capabilities of the machine:
http://www.cnczone.com/forums/uncate...aluminium.html

That's a fantastic build, one of the best ever. I'm super jealous of it. I've read some of his comments on his YouTube channel.

This one I find very informative

"EG is under table, steel frame is filled with EG, and bridge is also filled with EG. Machine handles steel, I even made video of milling stainless 304 steel, which I will post in short time.
Mild steel can be machined without any problems , but I didnt use tool more than 12mm in diameter. For aluminium, I tried 50mm face mill, and it works without any problem.
Additional spindle is BT30. I use it for steel. However this spindle will be removed, actually both of them, because I already ordered ATC BT30 spindle water cooled for up to 12k rpm, and I will use only ATC, so no more 2 spindles.
About aluminium gantry.. It wont handle steel if its not filled with EG. Even when milling aluminium, machine was much more sensitive to parameters. Without EG, it happned that sometimes I hit parameters (mainly tool rpm) which came in resonance with machine structure, and it made ugly high pitch noises.
Aluminium structure, steel structure.. Those are all very very bad materials to build machine bed out out of. Grey cast or EG, or UHCP are way to go.

Sure there were some lessons learned.. Like use ballscrew with high pitch - 10mm or more. It is best if ballscrew rotates slow - less vibration, quiet and smooth operation.
None of commercial made machines use 5mm pitch, like us hobbyst do.This is why I will change both X and Y ballscrews to 10mm pitch, and make higher ratio with pulleys.

Then there is so important Z axis.. Aluminium or steel plate is again very bad solution. Any even half serius machine should have gray cast square struture for Z axis. Or somekind Of EG filled Z stage. Vibration dampening of Z axis is super important."

Here's another one that has great performance:

https://www.usinages.com/threads/cnc....63531/page-54

You can see what kind of spindle he has. Many design iterations, so look at the end and check out his youtube channel to see videos of cutting steel.

For my spindle, I'm going to be using a spindle that came out of a Biesse Rover. Not sure yet if it was a good idea or a mistake. It weighs near 100 lbs. Also, I still don't know what some of the pins are for and can't find any documentation. As a bonus it has an ISO 30 ATC (which I plan to use as a quick change, not auto change) and lots of power.

First off, let me say, that I am a fan of your design and thought process. But, there are things about your current design iteration that I really do not like.

1. The base is overly complex. Also, fewer, but much larger, square tubes would give IMO, better performance, and be less prone to weld distortion.

2. Instead of adding weight with prefab concrete blocks, I think it would be better to fill the frame with something, epoxy granite perhaps.

3. I don't believe that gantry will give you the best stiffness to weight ratio. I believe the piece where the rails are bolted may distort like crazy when welded in place. And there's no way you could decide to fill it with any vibration dampening material later on (because it's open).

4. It all depends on what kind of a spindle you have as to whether the spindle body helps with the rigidity of your Z axis. For a long Z axis like you have, I think a square tube concept is better.

What I do like

1. You're looking at designs that work and have a great design iteration process and methodology.

2. Ribs in the gantry.

Of course, these are just my opinions, and I don't claim to be anyone special. I think at the end of the day you may come up with something great, perhaps better than any of us could, I just think you should explore some different ideas to have something really awesome.

Handlewanker,

Nice to see you here mate. Good to see you can dish it out as well as you can take it!

[JoanTheSpark]

The base is overly complex. Also, fewer, but much larger, square tubes would give IMO, better performance, and be less prone to weld distortion.

I haven't seriously stress tested this yet, but here is a simple one - 50kg centrally at the top tube with the bottom face the fixed bearing. Gives 0.12 mm lateral displacement. Now imagine two of those and the fixed area being at least half up vertically and the force being more realistic - it's good enough for what I will be doing - especially if other people manage with aluminium.



I also don't really want to go any bigger. The largest tubes I have cut & welded were 75x75x4 mm ones for a welding table. That stuff already looks & feels sturdy as hell.

And on complexity I wholeheartedly disagree. Anything I've seen made for steel out of tubing is way more complex and prone to welding distortions - I will stay within the 2D plane(s), as are the welding stresses. These grates will come out fine.

Instead of adding weight with prefab concrete blocks, I think it would be better to fill the frame with something, epoxy granite perhaps.

1) I would need to cut holes into each and every one of them or redesign with compromises to get the EG in there
2) a block cost me $3, for 10kg, another $5 for the PU glue
If I ever need to dismantle it - the blocks are knocked out easily and then the thing cut up.

From my POV it doesn't matter if you attach the dampening mass on the outside with some rubber/glue or from the inside.

I don't believe that gantry will give you the best stiffness to weight ratio. I believe the piece where the rails are bolted may distort like crazy when welded in place. And there's no way you could decide to fill it with any vibration dampening material later on (because it's open).

Not finished. The image was just to show were I'm headed vs. were I came from.
It will be enclosed and possible to put heavy mass in/on there and welding wont distort it, at least not worse than a certiflat welding table, who's principle I will adapt and make use of..

It all depends on what kind of a spindle you have as to whether the spindle body helps with the rigidity of your Z axis. For a long Z axis like you have, I think a square tube concept is better.

The 200x200 SHS 'tube' for a gantry has a couple of problems:
1) twisting moments are better taken care of by a round tube, the ribs will transfer those to it and then to the rails on the outer walls (imagine the z-ram fully down and apply a force in y-direction)
2) I need more vertical distance of the rails for the sled and can afford some overhang at the rear - shape becomes triangular
I would love to put the full triangle on the y-rails, but I'm stubbornly set on 1,600mm long y-rails and a 1,200mm y-envelope, so that's that.

The spindle in the image up there is a 3kW water cooled ER20 spindle with up to 20k rpm or thereabouts.
As I wrote earlier, I didn't come across a ATC 'ready' one, that has got a BT30/NT30/etc. adapter and is able to mill steel in that size class and in a single unit.
There is all sorts of spindle powerheads, which are then driven by a low rpm/high torque system on a parallel axle and the ATC is bolted on (all very big and clumsy and not in one single axis). Also I don't like the pneumatic ATC, I'd rather have something electric.

I can imagine a swappable spindle assembly. One for high rpm with ER20 collet for routing and another one for high torque with a BT30 cone and ATC for milling.

This video on youtube was pretty interesting (imaging an outrunner BLCD on a BT30 power head), but doesn't look like he made any progress as he got himself an ATC spindle with high rpm since then :-(

Make one. You will undoubtedly want to change something (fix mistakes and make improvements) in the way you do things after the 1st one, so the 2nd one is better.

I don't think I want to build a 2nd one in short succession.
Till I'm ready for one 3D metal printing will nearly be economical for home users ;-)

" ..Additional spindle is BT30. I use it for steel. However this spindle will be removed, actually both of them, because I already ordered ATC BT30 spindle water cooled for up to 12k rpm, and I will use only ATC, so no more 2 spindles..."

The question for me then is - what is the new ATC spindle with BT30 running at for steel milling if he wants to use the 50mm face mill?
What spindle is that?

" ..Sure there were some lessons learned.. Like use ballscrew with high pitch - 10mm or more. It is best if ballscrew rotates slow - less vibration, quiet and smooth operation.
None of commercial made machines use 5mm pitch, like us hobbyst do.This is why I will change both X and Y ballscrews to 10mm pitch, and make higher ratio with pulleys..."

Very good catch. Will take that to heart.

" ..Then there is so important Z axis.. Aluminium or steel plate is again very bad solution. Any even half serius machine should have gray cast square struture for Z axis. Or somekind Of EG filled Z stage. Vibration dampening of Z axis is super important..."

Not there yet. Will keep it in mind though.

Here's another one that has great performance:
https://www.usinages.com/threads/cnc....63531/page-54
You can see what kind of spindle he has. Many design iterations, so look at the end and check out his youtube channel to see videos of cutting steel.

Ha, came across that, but after 6-7 pages it was still a complex tubular mess, so didn't bother to go deeper in that thread.
His final build has a lot of welds in all directions = lot's of distortion, I try my best to avoid that.

For my spindle, I'm going to be using a spindle that came out of a Biesse Rover. Not sure yet if it was a good idea or a mistake. It weighs near 100 lbs. Also, I still don't know what some of the pins are for and can't find any documentation. As a bonus it has an ISO 30 ATC (which I plan to use as a quick change, not auto change) and lots of power.

50kgs for the power head alone, that's hefty. Can you drive it?
I won't make it past 3kW where I am, unless I somehow manage to get a solar feed battery system with a custom inverter going, then only the sky is the limit (or my pockets) ;-)

[Goemon]

What makes you think they aren't popular?

I found loads of portal mills when I was doing research for my build. They aren't always referred to as "portal mills" specifically though. I have seen them called "dual column mills", "gantry mills", "fixed gantry mills", "bridge mills" etc... but... it's all the same thing - I.e. A fixed gantry / moving table machine.


You can find both routers and mills in any of the commonly used designs. I.e. Both mills and routers can be either fixed column, dual column or moving gantry designs. Either can be too light and flimsy and either can be well made, stiff and weigh many thousands of LB's. If there is any distinction at the premium end of the market, I think it is only what they are designed to cut.

[iwaskevin]

Big 5 axis cnc first run - YouTube

[JoanTheSpark]

@iwaskevin
Cool, thanks for posting. The y-direction bending/torsional stability of that gantry/portal would be worrisome to me in that design though.
I'd expect some struts and wire-strapping could make it stiffer bending wise.
Same as they use around here for those agricultural watering sprinkler cranes:


Torsional-ly not much that can be done, besides beefing up the gantry.
Worst case is full z-depth and in the middle of the gantry and a force in x-direction.

[dharmic]

So, Iain, Mactec, 7 hours on is there any clear picture of who has the biggest e-wang yet?

One of the first videos (and I'm annoyed I can't find it anymore, gave up 20 pages into the search) I saw of a 5 axis machine was essentially a box with the front and top removed. Y rails on the top of the left and right sides, gantry for X sitting directly on the rails with no moving pillars to flex or add torque load to the linear bearings, Z on a carriage same as with a typical gantry machine and an A+C trunnion table fixed to the table beneath.

Aside from the access and ability to overhang large workpieces off the edge of the machine issues which, already as identified, is just as much a PITA with any enclosed machine anyway, I can't see any disadvantages to compete with the advantages of getting the Y rails out of the swarf (not dust, I don't care about dust) path and eliminating the flex of those side walls which can be framed or whatever as heavy as you like.

But I've always been too scared to ask, so thanks JoanTheSpark

Also: Ian, I think it might be a bit silly to automatically write off any gantry machine as "just a router". As a simple, straight off the top of the head example I look at the Datron machines... not at all what I'd think of as a router.

[JoanTheSpark]

..I've always been too scared to ask, so thanks..

I would have asked earlier, but never had time to sit down and draw something up, as I couldn't find examples ;-)

some will say you don't need 3 linear rails, but I say more the better, something for you to look at, these do not all look like yours, but have the same concept

Yeah, as short torsional lengths as possible, while the carriages as far apart as possible.
The biggest worry is the parallelism of the opposite mounting surfaces for the rails there.
And being able to get everything together, piece by piece, aligned and tight, while still being able to reach the bolts.. hehe.
The weight of the z-ram and the shorter x-directional spacing of the bottom runners on the gantry begged for an additional rail on the top.
A nightmare making those planes parallel though.

[mactec54]

The biggest worry is the parallelism of the opposite mounting surfaces for the rails there.
A nightmare making those planes parallel though.

Not to bad on a large Mill, when I do them like that I pocket all the rails in one operation, the other way, if you have your beam Blanchard ground on both faces, you can mill the rail pockets 1 face at a time

[JoanTheSpark]

Not to bad on a large Mill, when I do them like that I pocket all the rails in one operation, the other way, if you have your beam Blanchard ground on both faces, you can mill the rail pockets 1 face at a time

Thanks for using the word pocket - made me think twice ;-)

A question in regards to this - the mounting reference for the rails - does it have to be one solid piece, or can this reference also be a bolted-on piece of metal (base only needs to be ground/milled flat, not also pocketed)?
Also how important is this - as most DIY builds seem to omit this as far as I can see?



Just leaving this interesting essay on guide rails here - Picking the Right Linear Rails for Your Application

[louieatienza]

Thanks for using the word pocket - made me think twice ;-)

A question in regards to this - the mounting reference for the rails - does it have to be one solid piece, or can this reference also be a bolted-on piece of metal (base only needs to be ground/milled flat, not also pocketed)?
Also how important is this - as most DIY builds seem to omit this as far as I can see?



Just leaving this interesting essay on guide rails here - Picking the Right Linear Rails for Your Application

You only have to do that to the master rail and master blocks - provided you bought your LM bearings as a matched set.