Single THK HSR45 truck enough?

[nicad]

I have been thinking about building a 2kw Dunham 5c spindle based cnc lathe design. It will be for turning 1" aluminum and I would like to hold better than 0.0005".
My question is, would a single THK HSR45 "long" (quad bearing races) bearing block on one rail be ridged enough support for the Z axis? The tool tip would be 5" above the rail, so any rotational movement (Moment "Mc" in the datasheet here: http://www.thk.com/archive_file/tech...n/A262-287.pdf) of the block on the rail would result in part diameter variances. Example: if the block rotates 0.1 degrees, the part diameter changes by 0.0042". This sounds like a tall order, but I have to keep telling myself that the HSR45 rails are 1.77" wide and the blocks are aprox 5" x 5", which is quite "large".

I guess I am concerned that a single block will have no pre-loading and there will be some rotational "slop" existing. I do not have a rail and block here to test.. Anyone have any idea?

[will gilmore]

is there a reason not to use two smaller (cheaper) rails?

[nicad]

Yes, but that is not my question.

[The Blight]

It will most likely work, but I too would go with a dual rail setup.

[Ajax-oz]

If your after accuracy use two rails with 2 bearing blocks on each rail.

Remember the bearing type you wish to use has 3 rotational axis that produce an error to be concerned about.

By adding another bearing block to a rail at a reasonable spacing will drastically reduce 2 rotational error. by adding another rail will reduce the 3 rotational error.

The further apart the bearing blocks the greater the reduction in axis rotational error occurs.

AJ

[fatal-exception]

I think it depends on how heavy your cuts will be. If you are skimming a thou at a time a single rail would probably be fine. 100 thou, you might have some problems...

[Jason3]

I'd have thought it should work fine for light cuts. Could you plan your machine so that you could swap to a 2 rail set if it's not rigid enough?

I'm very keen to hear how that goes for you actually - I'm planning a lathe using the Dunham spindle too. Mine will use a single THK HRW35 (the wide one) block & rail for the Z, and a single HRW50 for the other axis. I've used these sets in small desktop mill style machines & been very happy with them. Perhaps those blocks might suit you too?

Keep us posted on your progress!

Jason

[LeeWay]

I have no experience with the larger blocks except to have seen some. They are beasts for sure. They can take quite a load.
That said, I would think that a single block would wear much quicker than at least a dual block design. 4 blocks are better. You loads will be spread out far better on these if installed correctly. I think this is the reason that most equipment designed with these blocks and rails, come with 4 trucks and two rails.

For economics and footprint, I have found that the THK HSR 25's and the HRW 27's work very well.
These are what I used on my mill and I am using the same ones on my little lathe project.

I bought a lot of 10 trucks brand new on Ebay for like $200. I bought the rail from THK. The HRW's were bought used with trucks from Ebay. I did use three trucks on each rail on the mill table. It really is a sturdy little setup that did not break the bank.


You have to consider where your cutting forces are going to be as well. A single block will narrow your rigid zone down.

[Ajax-oz]

Here is an ex sample
The attached image shows the three rotational errors that occur in a bearing block and rail setup

MA, MB, & MC

lets say the spec of a bearing is + or - 1deg of rotational error in MA, MB, & MC

so with one rail and bearing block you will have a error of + or - 1deg
Now we add a second bearing rail and block next to it 2" apart with a support plate bolting both bearings together

Now the rotational error has reduced to + or - 0.1deg (I have not done the maths, but it can be worked out.)

But if you increased the distance between the rails the rotational error will be reduced even more.

With two rails and one bearing block on each rail MC has been reduced But this will also reduce MB

But MA will stay the same.

If you have second bearing block on the same rail spaced 2" apart then the rotational error of MA will be reduced + or - 0.1deg. It also reduces MB.

Increasing the spacing of a bearing blocks on the same rail will also reduce the rotational error of MA & MB.

It does come down to what type of rotational error you wish to remove.

Hope this clears thing up.

AJ

Please Note: I have not done any maths to confirm the figures I have stated and I pulled a 1deg bearing spec out of thin air.
Please refer to bearing specs for these tolerances.
This is just to demonstrate the affect of adding extra bearing blocks or rails to a configuration.

[nicad]

Thanks for all of the replies. I fully understand the advantage of a multi-block setup.
I already have my main design with dual TKH HSR20 rails and quad HSR20 blocks for both the Z and X axis. Based on a granite plate ofcourse.

However, I have always thought that it would be kinda fun and sooo much simpler to use one of those massive 45+ sized blocks for each axis, which has about the same rail width and height as my entire Taig lathe bed (Also thought about building a Taig replacement with a HRS45 rail AS the lathe bed, and the one truck as the slide)... so I wondered if anyone had any experience with doing something like that.
It might work, except that they are also quite tall, so by the time you stack two of them on top of each other to form the Z and X axis, and mount a tool plate to the top, you are getting pretty far up from the base, which will have a lot of leverage on the Z axis rail. I have pretty much given up on the idea.