[WB-1]

Hi everyone, I have been reading through the forum and haven't been able to find anything (but I'm sure there is) about my z-axis design. I have seen the design on youtube and some pictures elsewhere but I would like to know what people think.

Rather than mounting the rails on the X-Axis plate, why not mount the bearings on the X-Axis plate. This way as the spindle move up on the Z-Axis the rigidity should improve (spindle will be closer to bearings).

I want to be able to machine a maximum of aluminium, whilst also softer materials. So if I were to cut aluminium I could increase the thickness of the sacrifice board so the Z-Axis will the cutting higher. (Does this make sense)

I'm trying to keep the Z-Axis with a machining area of 220mm.

This makes sense to me but I'm wondering why others haven't done it.
Cheers

[DonFrambach]

Actually, lots of people do it this way.

Here's a link to a photo showing it on my current build with linear acturators:
http://www.cnczone.com/forums/diy-cn...ml#post1051005

Here's a link to a photo showing an earlier version with parts from CNCRouterParts.com:
http://www.cnczone.com/forums/diy-cn...tml#post843806

[HorridHenry]

I'm no expert but I'm thinking thats adding more weight to an already heavy axis thats supporting the spindle and axis itself if made out of thick aluminium,I mean it can't be healthy for the stepper motor?

[WB-1]

Thanks for the replies its nice to know I'm not breaking away from the crowd!

HorridHenry you say it will add more weight. I should say as well I'm having a fixed gantry with a moving table (Y-Axis). So the extra weight would only be for the Z-Axis motor. I may need to gear down this axis.

The details of the design need to be finalised. Like how to fit the ballscrew. The rails and bearings are only 30mm deep which isn't enough room for the ballscrew which needs a minimum of 45mm.

[crane550]

I have never once heard someone say "I wish I had built my table lighter." Only stronger.

Weight in the right places is good. Weight in the wrong places doesn't make much difference. The heavier you make your components the less it will flex. I say beef it up- the cost is not much different and you will be much happier with the results.

[HorridHenry]

The heavier you make your components the less it will flex.

Yeah but....the floorboards will.

[WB-1]

crane550 do you think using 10mm steel plate would give it that extra weight minimising vibration. I do tend to forget that the overall mass of the machine is very important.

I have a question about linear rails. How interchangeable are the linear bearings. Is it possible to put thk bearings on to a nsk rail? Also if the linear bearings don't have ball bearing retainers how do you change them over??? Specifically are nsk sh20 linear ball bearing retained?

Cheers.

[louieatienza]

I also have my rails mounted on the carriage plate, exactly for the reasons stated. There's always some debate over this, but I agree. When you have the carriage all the way up, the collet/bering block distance is at its minimum; this is the best case scenario. When the carriage is all the way down, the collet/bearing block distance is at its maximum; this is the worst case scenario. Mounting the rails to the carriage plate does add the mass and ridgidity to that part; sizing your motor and screw right will ensure it moves right.

If you mount the rails to the saddle (or bearings to the carriage), the bearing blocks must be arranged on the carriage so that it is the maximum distance from the collet to allow for the same amount of travel; in other words, the collet/bearing block relationship will always be at a worst-case scenario.

The exception to this is if your Z axis travel is minimal, and you can space the bearing blocks farther that the distance between their ends is a majority of the total length of the rail.

I feel the moving rails design gives more flexibility; more travel; and simplifies the carriage design. On the flip side, it is not as easy to compact all the linear motion components.

The moving bearing design is easier to implement, since mounting rails to the saddle doesn't present the screw clearance and installation issues that mounting the bearings do. But it may not be the best choice if you need more Z travel.

So in conclusion there are pros and cons to both ways, and you need to base your design on your needs, components used, and ability.

As far as I know, you cannot interchange blocks and rails from two different brands. In fact you cannot interchange them with different series from the same company. For example, a THK SHS block will not fit a HSR rail. You could, however, use a HSR-C or HSR-A or HSR-LC, etc. block with an HSR rail. Also if I recall correctly you should be able to remove any THK block past 15 series from the rail, since the bearings are retained in the block. It is a good idea however to put the blocks on a stub rail or holding jig, since dropping the block could loosen the bearings, and you'd protect the bearings from contamination.

[WB-1]

Hi louieatienza, thanks for your detailed response, much appreciated. Have you any thoughts about where the best place is to mount the ballscrew and motor?

The ballscrew bearing mounts could be mounted on either the carriage plate or the x-axis plate. I prefer the idea of the motor mounted on the x-axis plate so its stationary in the vertical direction. Mounting the motor on the carriage would mean its moving up and down vertically....not so keen on this.

What do you guys think?

[louieatienza]

Hi louieatienza, thanks for your detailed response, much appreciated. Have you any thoughts about where the best place is to mount the ballscrew and motor?

The ballscrew bearing mounts could be mounted on either the carriage plate or the x-axis plate. I prefer the idea of the motor mounted on the x-axis plate so its stationary in the vertical direction. Mounting the motor on the carriage would mean its moving up and down vertically....not so keen on this.

What do you guys think?

On my current machine, the leadscrew and motor are both mounted to the gantry saddle. It's actually a modified Solsylva plans design. On my new machine, I will add a plate perpendicular to the saddle and mount the motor upside down, and run the ballscrew via timing belt and pulleys.

[Falcon69]

WB-1

Take a look at my build. The z-axis is louieatienza approved! LOL

Anyway, the holes on mine were to just lighten it abit, but after talking with the folks who might be cutting it out on their waterjet advised me not to, as the cost is per inch the machine moves. So it adds up quick.

You may want to take that into consideration when you build your machine. As others has stated, heavy and ridgid is better. I'm new at this as well. So only advise I can give is take your time on designing and do it right. Draw it up 3, 4, or even 50 times if you have to, and only cut it once. Steel is expensive.

Watch ebay, craigslist, anything you have access to. Pay close attention to auctions and such, you may just find yourself getting a nice deal on some parts, like Profile Rails and Blocks, Motors, etc. There has been a few members on here that have gotten lucky and gotten some really nice deals on used, and new, equipment. Take your time and keep a look out.

I made a mistake and bought rails and pillow blocks for my z-axis, as well as a break-out-board before getting the much needed advice. Now I'm stuck with parts I won't be using. Just take your time, ask lots of questions, and don't buy stuff more than once. Keep your costs down.

As a newbie/beginner in CNC, much like yourself, this is the best advice I can give you. G'Luck on your build!!!

[WB-1]

Hi Falcon69, I appreciate your comments. Would your machine be capable of cutting aluminium? I see the y-axis is belt driven. Belt driven axis are a lot cheaper (from first impressions) but I would have thought the belts wouldn't be capable of cutting aluminium???
What do others think about this??? Any examples of aluminium cutting/belt driven machines.

Your z-axis design seems very complicated and intricate for what real gain???

I would say the stationary bearing/moving rail design is the way to go for the z-axis.

[Falcon69]

I believe it will cut aluminum without a problem.

It's not really a belt drive. It uses the belt to do a 2:1 ratio step down. But I'm going to be using a rack and pinion design. The belt is only for the Drive assembly. The Belt will be under tention and there will be a heavy duty spring keeping the drive assembly meshed together with the Rack. This is a great way of making sure that the pinion has contact with the rack and all times, plus, as an added bonus, you don't really have to have the rack perfectly parallel with the pinion because of the spring keeping the tight tolerance between the rack and pinion. I got the drive assembly parts through CNCRouterParts.com. Ahren is nice guy and happy to answer your questions.

The z-axis is the most complicated design, yes. But, I believe the way I have it design is probably the most ridgid. The ridgidity of it should be strong enough for cutting aluminum. I'll be using a 3-1/4 HP Porter-Cable router.

One thing you have to keep in mind also, and what was I told, is that once you get your pieces welded up, anywhere you have a bearing block or rail mounted, you need to get it machined flat. That's what I will be doing to all my axis' once I get it welded up. Just like where I live, I'm sure there's many shops around you that can do the job. If you want an accurate machine, And you plan on putting some money into your machine, the extra amount it costs to get the parts machined flat will go along way in how your machine operates.

[louieatienza]

I believe it will cut aluminum without a problem.

It's not really a belt drive. It uses the belt to do a 2:1 ratio step down. But I'm going to be using a rack and pinion design. The belt is only for the Drive assembly. The Belt will be under tention and there will be a heavy duty spring keeping the drive assembly meshed together with the Rack. This is a great way of making sure that the pinion has contact with the rack and all times, plus, as an added bonus, you don't really have to have the rack perfectly parallel with the pinion because of the spring keeping the tight tolerance between the rack and pinion. I got the drive assembly parts through CNCRouterParts.com. Ahren is nice guy and happy to answer your questions.

The z-axis is the most complicated design, yes. But, I believe the way I have it design is probably the most ridgid. The ridgidity of it should be strong enough for cutting aluminum. I'll be using a 3-1/4 HP Porter-Cable router.

One thing you have to keep in mind also, and what was I told, is that once you get your pieces welded up, anywhere you have a bearing block or rail mounted, you need to get it machined flat. That's what I will be doing to all my axis' once I get it welded up. Just like where I live, I'm sure there's many shops around you that can do the job. If you want an accurate machine, And you plan on putting some money into your machine, the extra amount it costs to get the parts machined flat will go along way in how your machine operates.

It would be worthwhile to register at THK's website. There are numerous resources there on how their (or others') linear rails are to be mounted for maximum accuracy. It would be a good idea to have the machine shop leave a ledge on the surfaces for the bearing blocks; that will give a good reference point for mounting the bearing blocks. Same with the rails.

[WB-1]

Just came across some cnc plans known as the Momus Design. Its driven by belts on the X and Y with some very impressive rapid movements. I think the cutting area is only 16" by 16", not very big. Is there any reason why this couldn't be enlarged to approximately 40" by 40". Its a large increase by what do you guys think? There are videos showing it cutting aluminium and others who have built it seem very happy.



PLANS to build CNC 3 axis router table, milling machine, engraver. PDF download. | eBay

[PaulRowntree]

I would be worried about belt stretch and bounce. With a wider, heavier carriage the belt is going to start acting like a rubber band at some point.

[louieatienza]

Just came across some cnc plans known as the Momus Design. Its driven by belts on the X and Y with some very impressive rapid movements. I think the cutting area is only 16" by 16", not very big. Is there any reason why this couldn't be enlarged to approximately 40" by 40". Its a large increase by what do you guys think? There are videos showing it cutting aluminium and others who have built it seem very happy.



PLANS to build CNC 3 axis router table, milling machine, engraver. PDF download. | eBay

The designer of the Momus will tell you that it is easy to make this machine deeper, but widthwise would be a problem with it's single-side gantry drive and 'cantilevered bearing' design. You would have to mod the plans to have dual drive for the gantry.

I don't think the belts would be the main concern as far as stretch. There are many Bridgeport mill CNC conversions, and almost all of them use belts (and that knee ain't light!) For light occasional work it would be fine, as the videos show.

[PaulRowntree]

I don't think the belts would be the main concern as far as stretch. There are many Bridgeport mill CNC conversions, and almost all of them use belts (and that knee ain't light!)

How fast are these converted Bridgeports moving? Are the belts they have running from steppers to the leadscrews such that the belts are short and seriously geared down? I understood the suggestion to be running the gantry directly from the belt motion.
Cheers!

[WB-1]

I was suggesting running the belts only, no screw. Not sure if louieatienza meant belts and screws for the Bridgeports. Any chance you could post a link for those Bridgeport examples please?

louieatienza I didn't mean to copy Momus design. I was only interested in seeing a belt driven machine cutting aluminium. I think I'm going for a fixed gantry design with a moving table.

If I was concerned about the belt stretching why couldn't you use dual belts driving the y-axis (table) and x-axis.

I have seen a couple different methods of belt drives. Some have the motor fixed at one end of the belt, whilst others have the belt moving along the belt (like a R&P style). Does anyone have any thoughts about these.

[louieatienza]

I was suggesting running the belts only, no screw. Not sure if louieatienza meant belts and screws for the Bridgeports. Any chance you could post a link for those Bridgeport examples please?

louieatienza I didn't mean to copy Momus design. I was only interested in seeing a belt driven machine cutting aluminium. I think I'm going for a fixed gantry design with a moving table.

If I was concerned about the belt stretching why couldn't you use dual belts driving the y-axis (table) and x-axis.

I have seen a couple different methods of belt drives. Some have the motor fixed at one end of the belt, whilst others have the belt moving along the belt (like a R&P style). Does anyone have any thoughts about these.

I realize that. I didn't mean to imply that a Bridgport conversion run on only belts. But the steppers used on them are pretty powerful, and there's really not much stretch.

They do make timing belts in different grades with different reinforcements, and different widths. You can get them with steel or even Kevlar reinforcement. I don't think belts would be ideal for aluminum cutting, but like I mention, for light occasional work it can be done.