Diameter of shaft for linear-guide

[Phantom]

Hi,
what diameter should a shaft for a linear guide have? 10mm? 15mm? 20mm? Or even more? The router where this is planned for will probably be made of wood.
Greetings
Martin

[mxtras]

I think we are going to need a LOT more info to give a recommendation.

How large will this router be? What style - gantry? The design plays a big part in selection also - will the guides be supported? What type of guides - round?

Dig around on this site for a few days and you will likely find one that suits your expectations, then ask the fellow that built it specifics - this will likely get you where you want to go much quicker.

Scott

[Phantom]

Hi,
I am thinking about a router made of wood, like perhaps 600*500*100, not supported guides. The y-axis on a bridge which does not move, the x-axis table below should move.

[jeffs555]

With the moving table designs, the x-rails have to be at least twice the length of the table, so for a 600mm table, you would need at least 1200mm guides. That is very long for unsupported guides. A 20mm steel rod, 1200mm long would deflect about 1.1mm for every 50 newtons of force applied. Going to 30mm rod would change the deflection to .216mm for every 50 newtons of force.

[jans123]

Originally Posted by jeffs555 With the moving table designs, the x-rails have to be at least twice the length of the table, so for a 600mm table, you would need at least 1200mm guides. That is very long for unsupported guides. A 20mm steel rod, 1200mm long would deflect about 1.1mm for every 50 newtons of force applied. Going to 30mm rod would change the deflection to .216mm for every 50 newtons of force.

It is possible to cheat a little. Make the table 650, keep the rod 1200 and put a support in the middle under the rod. By letting the table have one bearing on either side of the support you will increas the stiffnes dramatically. Anyway, the thicker the better.
Another cheat is to only use a thick bar on one side, letting it guide the direction and on the other side of the frame just use a flat iron standing on the edge. A simple ball bearing carriage (ball bearings above and under the flat iron) is all it takes on that side to keep the direction and you get around the problem with keeping two circular guides exactly parallell.

/jan
(Sorry about all bad typing in English. It isn't my native lingo, I just do the best I can.)

[jeffs555]

Martin,
Those are very good suggestions that Jan made. Adding just one support in the middle with bearings on either side, will increase the stability greatly. The deflection would be nearly 15 times less than an unsupported rod. I also like his suggestion about using a larger rod on one side with full bearings, and only vertical bearings on the other side. That is the way I have been building my machine which is an over-head rail design. This should hopefully let me get by with a leadscrew only on one side.

Jeff

PS. Your English is excellent Jan, and yours also Martin. I could not tell from your writing that English was not your native language.

[Phantom]

Hi,
first of all, thank you for your help!
The hint with the support in the middle is really excellent. If I will have more questions concerning something like this, I will perhaps ask again. In the attachement you can see, where about I've thought.

[jans123]

Originally Posted by jeffs555 Martin, Those are very good suggestions that Jan made. Adding just one support in the middle with bearings on either side, will increase the stability greatly. The deflection would be nearly 15 times less than an unsupported rod. I also like his suggestion about using a larger rod on one side with full bearings, and only vertical bearings on the other side. That is the way I have been building my machine which is an over-head rail design. This should hopefully let me get by with a leadscrew only on one side. Jeff PS. Your English is excellent Jan, and yours also Martin. I could not tell from your writing that English was not your native language.

Thanks Jeff, I'm really working hard with the language.
I had no idea the single support increased the stiffness so much. I figured it would do some 4...8 times, but you are probably right. I just never done any calculations on it.
Regarding the rail system I mentioned and where you agree with me, I just wanted to upload a simple picture to visualise it for those who didn't understand what we where talking about.
On the sketch the "A" bar and the "B" bar share the vertical load (Z). The small boxes are just symbols for bearings (Solid or roling).
The "A" bar takes care of the forces in the "Y" direction as well as the torques around the "Z" axis and the "Y" axis.
The "B" bar takes care of the torque around the "X" axis.
I guess that 80% of the mass in the "A" bar isn't needed exempt for damping vibrations with it's mass, otherwise it could just as well be a tube.
This design solves the great problem with aligning the bars perfectly parallell- a problem that otherwise would cause extra loads and wear on the bearings.
Now, it's "simply" to join all the bearings to one table and adjust The "B" bar a little so it dosn't bent the table too much.

/jan