[louieatienza]

here is the stepper:
eBay Österreich: Stepmotor Schrittmotor 4,2A 3Nm CNC Fräse Fräsmaschine (Artikel 270712289529 endet 08.03.11 13:41:39 MEZ)
It is in german at all, but the graph speaks for its own.

Yeah the lower extrosion is a really welcome place for cable chain.
But that is not very independent at the moment. - I just want to optimize its performance.

Can someone tell me how I can calculate the cutting force?

That explains it! These aren't your typical steppers; they can run at high rpm...

I gave up on the translation, but found a company here that sells similar motors, and ordered some! As I plan to use high lead ballscrews on my machine with gear reduction to gain back accuracy, the higher rotation steppers should give me back some lost speed...

[louieatienza]

That explains it! These aren't your typical steppers; they can run at high rpm...

I gave up on the translation, but found a company here that sells similar motors, and ordered some! As I plan to use high lead ballscrews on my machine with gear reduction to gain back accuracy, the higher rotation steppers should give me back some lost speed...

Found the steppers, they're made by Sanyo Denki; have many frame sizes as well...

[eaglezsoar]

But apart from that I am very unsure that I should bet on ballscrews.
Many of the builders here in the zone are really promoting a R&P system.
But I dont know where I can buy them from and how expensive they would be.
Mainly I think that they will wear of much more faster than a ballscrew. When just a small piece of dirt gets on the rack, will result an false aligenment.
Will they be the same accurate as a ball screw would be for the same price? That does only matter for me. Why should I make it more expensive when it is useless(?). So please help me.

Link to the R&P CNCRouterParts

[MaxTheModder]

Thank you for the link. I did already stumbled across them a while ago- but I think their R&P systems are some kind redicoullus.
Their real resolution is 0.13mm per full step - and I never trust that microstepping ratings. even a normal step has a tolerance of +-5% - what is microstepping going to be - +-50%?
Anyway I dont think that I will cut at 1m/s speed. And at low speed the R&P´s resolution is just really bad. Maybe I would have to reduce the speed more with some extra belts and pulleys but $$ and of course : bore belts - more flex
On the other hand it is able to handle only a little cutting force. for example:
the CNCRouterParts R&P system is geard down by 3:1 that means that a stepper with 3Nm force will only be able to hold the rack with an 25mm Dia on the end pulley - only till about 720N. thats too low for me and I am sure that its going to backdrive too often for me. Am I wrong with the 3:1 reduction (I thin I read that in the specs)
Simply mulitply the steppers torque (in Nm!) by 240 and you will get the holding force (only for the CNCRouterParts R&P system)

What do you think? So now I come to the conclusion that R&P is really good for rapids, but they really fail at holding force and accuray and if you want the same accuracy as an ballscrew you would have to gear it down pretty much which results in an high flex on the belts.

[Pplug]

Thank you for the link. I did already stumbled across them a while ago- but I think their R&P systems are some kind redicoullus.
Their real resolution is 0.13mm per full step - and I never trust that microstepping ratings. even a normal step has a tolerance of +-5% - what is microstepping going to be - +-50%?
Anyway I dont think that I will cut at 1m/s speed. And at low speed the R&P´s resolution is just really bad. Maybe I would have to reduce the speed more with some extra belts and pulleys but $$ and of course : bore belts - more flex
On the other hand it is able to handle only a little cutting force. for example:
the CNCRouterParts R&P system is geard down by 3:1 that means that a stepper with 3Nm force will only be able to hold the rack with an 25mm Dia on the end pulley - only till about 720N. thats too low for me and I am sure that its going to backdrive too often for me. Am I wrong with the 3:1 reduction (I thin I read that in the specs)
Simply mulitply the steppers torque (in Nm!) by 240 and you will get the holding force (only for the CNCRouterParts R&P system)

What do you think? So now I come to the conclusion that R&P is really good for rapids, but they really fail at holding force and accuray and if you want the same accuracy as an ballscrew you would have to gear it down pretty much which results in an high flex on the belts.

One alternative could be a spinning ball nut. The ballscrew would be stationary and therefore would have no whip. The motor would be attached to the gantry and a belt would connect the gears to the ball nut. My techno lc4896 4x8 ft uses this design and is very efficient & sturdy. The techno also uses 50oz/in servos instead of stepper motors. Seems pretty weak but it does work well.

[louieatienza]

Thank you for the link. I did already stumbled across them a while ago- but I think their R&P systems are some kind redicoullus.
Their real resolution is 0.13mm per full step - and I never trust that microstepping ratings. even a normal step has a tolerance of +-5% - what is microstepping going to be - +-50%?
Anyway I dont think that I will cut at 1m/s speed. And at low speed the R&P´s resolution is just really bad. Maybe I would have to reduce the speed more with some extra belts and pulleys but $$ and of course : bore belts - more flex
On the other hand it is able to handle only a little cutting force. for example:
the CNCRouterParts R&P system is geard down by 3:1 that means that a stepper with 3Nm force will only be able to hold the rack with an 25mm Dia on the end pulley - only till about 720N. thats too low for me and I am sure that its going to backdrive too often for me. Am I wrong with the 3:1 reduction (I thin I read that in the specs)
Simply mulitply the steppers torque (in Nm!) by 240 and you will get the holding force (only for the CNCRouterParts R&P system)

What do you think? So now I come to the conclusion that R&P is really good for rapids, but they really fail at holding force and accuray and if you want the same accuracy as an ballscrew you would have to gear it down pretty much which results in an high flex on the belts.

There are many large commercial CNC routers that use r&p, that are very accurate...

[MaxTheModder]

what do you think - is it better to put the ballscrew on the outside?
In my mind it will make it more easy to protect it from dust , but are there any other advantages?

At the moment I am designing the protector for the bullscrews but it seems a bit too complicated for such a simple reason...

tell me where you would plase the ballscrew on the X- axis

[louieatienza]

It might be simpler to leavev it as-is, and just put removable polycarbonate "sides" to keep dust away...

I have my x rails above the table, with my leadscrews below my LM rails, with dust collection, and I STILL get dust on my leadscrews!

[Pplug]

On my techno they are totally enclosed on the x&z axis. They are supported by steel beams that protrude from the linear bearings and a rubber or vinyl strip covers the slot they ride in. I'll take some Picts later. My shop is a giant dust ball and the bearings still look as good as the day they were made!

[ger21]

What do you think? So now I come to the conclusion that R&P is really good for rapids, but they really fail at holding force and accuray and if you want the same accuracy as an ballscrew you would have to gear it down pretty much which results in an high flex on the belts.

The difference you're seeing between R&P and ballscrews is an issue of gearing more than one being better than the other.

Gear them so they both move the same distance per stepper revolution, and the specs will be similar.

The real issue here comes from using steppers. Because steppers lose torque as rpm's increase, they have a rather low maximum rpm, so speed vs resolution is always a tradeoff.

Most rack and pinion users go that route to get more speed, and give up resolution to get that speed. If you gear a rack and pinion system 10:1 instead of 3:1, it'll compare much more favorable to the ballscrew.

Along the same lines, you can use a ballscrew with 25mm or even 50mm pitch, and it'll perform much like a rack and pinion.

You decide on the performance goals you want to achieve, and then figure out which system will meet your goals. If very high speeds AND high resolution are required, than servo's may need to be considered.

And fwiw, the $100,000 routers I've used all use a belt reduction to drive their ballscrews, moving over 1000lbs at over 1000ipm. Someone already mentioned it, but it's just a matter of using the right belts for the application.

[louieatienza]

The difference you're seeing between R&P and ballscrews is an issue of gearing more than one being better than the other.

Gear them so they both move the same distance per stepper revolution, and the specs will be similar.

The real issue here comes from using steppers. Because steppers lose torque as rpm's increase, they have a rather low maximum rpm, so speed vs resolution is always a tradeoff.

Most rack and pinion users go that route to get more speed, and give up resolution to get that speed. If you gear a rack and pinion system 10:1 instead of 3:1, it'll compare much more favorable to the ballscrew.

Along the same lines, you can use a ballscrew with 25mm or even 50mm pitch, and it'll perform much like a rack and pinion.

You decide on the performance goals you want to achieve, and then figure out which system will meet your goals. If very high speeds AND high resolution are required, than servo's may need to be considered.

And fwiw, the $100,000 routers I've used all use a belt reduction to drive their ballscrews, moving over 1000lbs at over 1000ipm. Someone already mentioned it, but it's just a matter of using the right belts for the application.

Since you're using high-speed steppers, you could gear down quite a bit and still get good speed and resolution from r&p. It might even be easier to shield an r&p drive as opposed to ballscrews.

[ahren]

Interesting thread. First off, I don't want this post to sound defensive, but did want to point out some strengths of our R&P system, which I don't consider "redicoullus"

I'd agree that R&P is probably not the right drive system if sub-0.001" accuracy is your primary goal. Ballscrews or linear motors shine in these applications when set up properly. However, most router builders don't need this, and for larger format machines, the cost of ballscrews becomes pretty prohibitive. It's a trade-off, as others have mentioned, but R&P offers a nice combination of decent accuracy, good cutting forces (720N is quite a bit of force for most applications...), excellent rapids, and the ability to affordably build machines that can handle 4' x 8' sheets of plywood or larger. We have cabinet builders holding 0.002" on their cuts (0.050mm), so there is something to be said for micro-stepping.

Dust with the R&P is really not a big deal -- they're remarkably resilient, and it's pretty easy to mount them with the racks pointed down or sideways. And ger is right -- our system is set up to work primarily with stepper motors, which are affordable and easy to set up for the DIYer. Independent of the drive system chosen, collecting dust at the source is always the best idea, as most mechanical systems don't tolerate it well.

Maybe R&P isn't right for this application, but we've sold over 1200 of them since they came out in late 2009, so it does appeal to a large segment of builders. Ballscrews also have their place, and we're investigating setting up some ballscrew drives for z axes. If accuracy is your paramount goal, you might consider a moving table setup like those used on Vertical Machining Centers for added stiffness. In many cases, the frame design has far more of a real world impact than the drive system used.

Enough out of me though. Cool renderings, a nice looking gantry design, and good attention to engineering detail. I'll be interested to see how this machine turns out!


Best regards,

Ahren
CNCRouterParts

[louieatienza]

These guys use r&p on all 3 axes; no one would argue the accuracy of their machines...

Home

[MaxTheModder]

so first I have to excuse me for my strange word choice -
I meant "redicoulus" with the meaning of the resolution compared to my needs.
Anyway I didnt meant that in an condescending way and I never said that the R&P systems from CNCRouterParts are bad or something like that. I just meant that their R&P system will not match my needs far away from that what I want.

So at the moment I think I will keep the ballscrew design.
I let the ballscrews over the table, because as louieatienza already said - It will be just ALOT of work to change the position of the screws compared to the usability created then.

For this reason the design keeps a bit more compact and the bed is delimited by the new chipkeepers is created. I will post the new renderings ofdesign version 1.1 in 3-5 hours.

[ger21]

These guys use r&p on all 3 axes; no one would argue the accuracy of their machines...

Home

I downloaded one brochure, and it said it used ballscrews.

[MaxTheModder]

yeah thiere machines are looking PRETTY amazing!
But I also think that it uses a R&P on the X and Y axis - maybe a ballscrew on the other axis in that head.
These machines are about 6mx4m and with the rapids in the vid -I think no normal ballscrew is going to do that

[louieatienza]

I downloaded one brochure, and it said it used ballscrews.

You downloaded the wrong one HAHAHAHA....!

Their top of the line FZ100 uses helical rack and pinion on all three axes (see the brochure, I think it's page 8). Then again, I guess it's a helical GEAR, since it's too big to be a pinion!

[ger21]

I was in a hurry, and looked at the FZ50. Went back and saw that the bigger one was R&P.

[louieatienza]

yeah thiere machines are looking PRETTY amazing!
But I also think that it uses a R&P on the X and Y axis - maybe a ballscrew on the other axis in that head.
These machines are about 6mx4m and with the rapids in the vid -I think no normal ballscrew is going to do that

Onsrud Extreme Duty series rapids at 6000ipm, with ballscrews!

[MaxTheModder]

time was rare...

but now it is here!
The new updated optimized gantry design.

-The cutting forces in X-direction will now be a lot better asorbed.
-added 33x30x33 U-Steel profile for more stiffness in each extrusions weak direction

The cutting bed is now 20mm smaller to allow place for the protection for the ballcrews.

I also quick-designed a dust shoe - at the moment it is not very well minded, but i have added everything i want later to it:
The top will be made out of 10mm plexiglass
the tool will be illuminated with 4x1W high power leds
And a little webcam is also planned to record the process.

Here are the new renderings:







But then I came to a little problem:
That CNC is going to produce LOTS of dust - and dust bags are too expensive for me to be changed evry 10 minutes.
So i decided to plan a cyclone dust seperator.
I have searched around and i finally stumbked across Bill´s cyclone design
(Bill's Cyclone & Dust Collection Research - Cyclone Plans)

But I wondered how effective that design is going to be.
And finally I simply re-modelled his design and passed it through an flow simulation. The simulation is based 1:1 on his original plan and the inlet air flow is also 1000CFM as he suggests on his website.
The results are pretty good:
Air speed at the walls is about 15-20m/s and on the very bottom the air really stops flowing. (really good to see in the last pic)







But I think that his design can still be optimized.
Fact is that I am going to mill out my cyclone´s body out of two large foam blocks and face the inside with glassfibre or only epoxy.

At the moment I cant sayy that my design is final now - so I can still make changes.

YOUR SUGGESTIONS ARE WELCOME