[h3ndrix]

Hi ppl,

I know I ve been asking questions and jumping from subject to subject, but I am in my research phase for the CNC that I am building. After I found out that I was not going to be able to use the small bridgeport for my project, I decided to make my own CNC, from ground up (well...almost). Here is some progress:

This is the spindle design, with automatic tool changing capability, that will fit a Mini Mill head from Harbor freight, Grizzly etc.:




It will ride on two angular contact bearing OD: 62 mm ID: 30 mm.
All parts are stell except the orange part and dark green part on the back, which might be made from silicon bronze.

-It features a NMBT 30 collet with standart 45 degree retention knob.
-It has a holding system which is very similar to a quick connect hose fittings that are on compressor hoses. The bright green are the steel balls that will fall into the groove on the spindle when the bar is compressed down.
-It has a die spring which has a rated force of 500 N. It will be under compression when the collet is in the spindle.
-The spindle nose is a standart nose (measurements taken from Machinery's Handbook) , but it is missing 4 screw holes and the two ears that I will add later.
-The system will be activated by a linear actuator.
-The hole running thru the middle is for coolant to the collet.

Please give me some feedback on the design guyz. Oh, and if someone wants to make it before I ever get to it, feel free to do so, but let me know of the progress.

Thx,

H3ndriX

[Ken_Shea]

I like it !,
Good thinking.

Thinking on the the use of the silicon bronze, would not these pieces be carrying the entire load and be subject to quick wear?

This material makes a great bushing but not much for strength.


Keep us up to date on your project.

Ken

[h3ndrix]

Thx Ken,

This silicon bronze I am talking about is unbelievably hard, they usually use it on casting sculptures that are big in size . But, then again, you might be right, I am still researching the materials I will use on the spindle and silicon bronze, like you said, might not be a good idea.

Thx

H3ndriX

[HuFlungDung]

A few observations: I don't see the bearing surface for the large angular contact bearing. It should be right behind the main flange, but you are showing an angled surface there.

It is not a good idea to use only two bearings. The heat buildup needs to be compensated for, or the bearing preload will increase as the spindle warms up and bearing failure will ensue because they will run really hot.

So, use pairs of angular contact bearings, with the larger main bearings being a preloaded set that mounts back to back in the lower end of the housing. This pair is securely locked in place to position the spindle accurately (axially) in the housing.

The upper pair of bearings should be fastened securely to a shoulder on the upper end of the shaft. They are also a preloaded set that are locked back to back on the shaft. However, the outer races on this pair should be a close slide fit in the housing. This is where the expansion and contraction is compensated.

I am not sure about the die spring. Belleville washers have advantages because they might give better balance. They are also very stiff with a very high spring rate. But, one thing I have not figured out about the "real cnc's" with this retention knob system, is what is applying the pressure to release the tool, and what is bearing the load while this takes place. I'd like to see something that is not pressing the hell out of the spindle bearings and Z axis ballscrew supports. Ideally, this would be something that locks onto the spindle shaft itself, and then bears against that lock and the spring to do the tool release.

[wms]

Here is shot of a Famous name machine tool draw bar.

Hope it helps. You are very close to it's design.

[h3ndrix]

ThX Hu,

Right after the angled surface on the spindle nose, there is a flange, which is for the inner race of the first bearing, the second bearing, which will be on top, does not have a flange, this is because the flange is on the casting side and the inner race will be secured by the space and nut that will fit on the spindle outside (which is not on the drawings).

The reason why the spindle was made this way was to be able to put it into a mini mill head, which can only take two bearings. Ideally 4 bearings would be the way to go, but the casting does not have any machinability on it for extra bearing surface. BTW, the reason I am using a mini mill head is because I don't have enough money to get a cast iron head cast and dont know anyone with the facilities.

I am still researching the belville washer deal, but I do not know how they opperate to allow for constant force, which is what i need.

I will post a drawing of the head when I get home, so it will be easier to understand.

Thx,

H3ndriX

[h3ndrix]

THx Wms!

It is close to impossible to find a image like that, I gave up looking and decided to design my own, then something flashed in my mind as I was getting ready to spray paint my table with clear coat when I plugged the hose into the gun and bam!

Quick connect!

Then I realized how similar the quick connect male connector is to the retention knob design, then I went from there.

Thank you for posting that, that just makes me feel alot more better

H3ndriX

[WOLOG]

There is a quick release kit that you can buy for a R-8 spindle. The R-8 collets have a built in retention knob. I think it uses a die spring also. Maybe you can get some ideas from that. I will look for the bookmark on my computer when I get home and will send it to you.

[h3ndrix]

Wolog, thx for the heads up, I have seen what you are talking about, but it was really hard to understand anything from it, and they dont have any mechanical drawings of it. Costs around $1000 too. Thx though

Here is a pic of the assembled pieces with the mini mill head:



I hope it is clearer why I have to use 2 bearings instead of 4 and why there is no upper flange. Hu, If you have any ideas on how I could fit 4 angular bearings instead of 2, I would be glad to hear it

Thx

H3ndriX

[HuFlungDung]

H3ndrix, those are real pretty pics

You can count on the spindle becoming hotter than the housing. The length between those bearings simply increases the total differential of expansion. Your bearing preload will be lost with the bearings put in that way.

If you turn them around, they will get too tight.
Edit: they won't get tight if they are not restrained in the housings, but instead, the bearings will move away from the shoulders in the housing. This will make the Z position slightly unstable.

Here is something to consider: instead of mounting the bearings directly into the casting, you could machine some bearing housing blocks that would fit in like extensions, either outwards, or inwards, as you might find room.

[HuFlungDung]

I had to rush away, but I'm back. I wanted you to consider the overhang of your current design. A reasonable boring bar overhang/diameter ratio (in lathework) is between 4:1 to 6:1 max. This affects the vibration of the bar.

So, you would check your estimated tool overhang from your frontmost bearing, divided by the 30mm diameter of your spindle at that point. Already, your tool is working on the extreme limits, and vibration will be a problem.

So, if you create an extension "pillow block" that slides (with a pressfit) into your existing housing bore, then, you can use larger ID bearings on the main end, and move them further down towards the end of your spindle, where you now have that empty tapered zone.

[Stevie]

What you need is called Amco Bronze; we use it for slide guides in Injection molds; very hard to machine and tap; the wear rate is very low
I'm sure it's an alloy of Alumi and Bronze

and for sure the bearing needs to be extended into the tapered zone; or simply add another

[HuFlungDung]

Yes, Stevie has made a good recommendation: Ampco 18 aluminum bronze is one of the popular types. There are generic copies similar to this alloy made by other companies, too. Just phone them and let them do the "type crossover" for you.

[WOLOG]

It seems that the spindle needs to have a cup/cone type bearing at the spindle end and another bearing(s) that can slide as the nut on top is tightened up to remove any play. I rebuilt a r-8 spindle the other day. That was the arrangement of the bearings. My Haas service tech is coming next week to install my VF-5/50. I will ask him about the arrangement on the Haas VMC spindle. Maybe he can shed some confus... I mean light on the subject.

By the way, what do you use to produce those drawings? They look great.

[WOLOG]

Maybe I should look at the picture better. You are doing basically what I said. Its been a long week.

[John S.]

h3ndrix
This is very similar to one I'm working on at the moment for a Taig mill.
Because the Taig is a lot smaller than the one you are doing I'm using my own taper which is smaller than a 30.
Some holders will be made for ER collets and some will be Weldon type end mill holders .

Because I'm stuck for room in the original tail head I will have to machine a new head up from an alloy block.
I propose to use two angular contact bearings at the botton for rigidity and accuracy and one standard ball race at the top for support.

May I point some area's out where you could make improvements?
You are lacking a spacer inbetween your bearings in the drawings to set preload on the bearings. That bronze bush with the balls in to hold the pull stud can be steel as it revolves with the rest and doesn't need a bearing surface. If you made the recess a bit shallower inside when you operate the drawbar the bottom of the recess will eject the taper.
As someone else has mentioned a stack of belville washers will work better.

You also have far too much overhang at the end of the spindle, move your bearings out and if nessesary go bigger and try to get a bearing as close as possible to the larger diameter of the spindle.
You state you have to put the hols in and the ears, I presume you mean drive dogs ?

If you aim to use a tool changer you have to take into account lining up to engage the dogs. If you make your spindle fatter at the business end and fit a roller clutch needle bearing you can use this to drive the chuck / holder on a parallel section of the same. This way you don't have to line up if using a tool changer and this does away with the drive dogs.

John S.

[ballendo]

Hu,

That is an OUTSTANDING idea! Great post.

Ballendo

P.S. I would also point out that the recess at the retention knob end is too long. And if you're using through-hole(spindle) coolant, you may want an o-ring up there so whent the tool is in place the coolant goes where it is supposed to...

Originally posted by HuFlungDung
<snip>Here is something to consider: instead of mounting the bearings directly into the casting, you could machine some bearing housing blocks that would fit in like extensions, either outwards, or inwards, as you might find room.

[ballendo]

Originally posted by HuFlungDung
I had to rush away, but I'm back. I wanted you to consider the overhang<snip>
So, if you create an extension "pillow block" that slides (with a pressfit) into your existing housing bore, then, you can use larger ID bearings on the main end, and move them further down towards the end of your spindle, where you now have that empty tapered zone.

Another GREAT post!

Hu, you asked how a commercial machine pushes against the tool release? It IS just a large diameter short throw single acting air cylinder(a spring return is fitted and the shaft which "carries" this spring also has an actuating disc for the down/up limit switches),

The piston has a central "bolt" which acts against the end of the drawbar.

So the pressure IS being directed onto the spindle bearings with every toolchange... (But I wouldn't worry about it; the force required to release the tool is FAR less than the working loads imposed on the spindle.

One additional point: The lower pair of bearings (in the 4 bearing spindle setup) should be spaced apart for a "fixed" style mounting. This adds considerable rigidity to the spindle--for side loads especially.

Hope this helps,

Ballendo

P.S. I'd be concerned about using anything less than tool steel-hardened- for the sliding part in the spindle. The retention knob is hardened, as are the bearings gripping it. these bearings will deform the holes they work in in short order, which will compromise the tool-holding, perhaps with seriously bad results.

I think anyone working in a shop long enough has seen, heard and maybe felt the fear when a toolholder releases "accidentally".

3/4 pound of metal with a sharp end , spinning at some thousands of RPM's is NOT a "good thing"...

[HuFlungDung]

I whipped up a quick 2d of the bottom end of the spindle and housing. Let's see if the image will fit.

Nothing is to scale whatsoever. The green bearings represent a pair of flush ground angular contact bearings. The orange nut clamps the inner races together against the shoulder behind the spindle flange.

The outer races of the bearings are clamped inside the blue outer housing extension which has a back end designed to fit into your existing machine's bearing bore. Again, another nut is used to clamp the two outer races together. Because the bearings are ordered as a flush ground set, the preload is automatic, and you can tighten the nuts up right tight, with no regard for overdoing it.

I've also shown in light blue, what I call a grease labyrinth. This will allow the bearings to expel the excess grease, yet provides a none contact baffle between the pristine cleanliness inside the bearing housing, and the dirty outside world. In my experience, contacting rubber seals create way too much heat in a spindle. The labyrinth is simply a couple of meshing plates that the grease has to work its way through to get out, and vice versa for the dirt coming the other way. Each half of the labyrinth needs to be firmly attached to its mounting surface so that they cannot actually contact one another.

Perhaps someone else has some insight into the seal problem. Maybe there are suitable seals out there. They definitely need protection though, from sharp chips.

[HuFlungDung]

Thanks for that info, Ballendo. It must take a fair bit of pressure to overcome that retention spring, would it not? So is there actually enough tension in the retention knob system that the tapers want to lock a little bit, like a regular NMTB taper with a drawbolt?

Seems like it would or could be quite a push to release, so a person would have to be mindful of creating the Z axis quillbox with adequate strength for this purpose.