[louieatienza]

Hi Ian and others

For what it is worth, I will spell out what I am currently thinking would work.

I agree with Ian that a friction drive would likely work, up to a certain loading. My problem is that I do not know what that loading might be, and the possibility of a slip during a machining run worries me. On the other hand, a belt drive with the right belts would have very little backlash - small enough that a servo LOOP approach would be fine.

* I would use a mechanical box arrangement similar to what Ian and Katran have described, as far as the bearings go. Steel or aluminium is probably not critical at this stage. Line boring the holes for the bearings would be critical however.

* I am not sure whether I would bother with a tubular shaft: I have seen relatively few cases where a thru-chuck holding arrangement is really needed. That decision affects the bearing diameter of course.

* For front end bearings I would go for either twin opposing deep-groove bearings or twin opposing angular bearings, with a 3rd bearing at the rear to take the drive load.

* I would use a 25 mm wide 5 mm pitch GT3 belt running on steel/iron pulleys for the first reduction of about 5:1. These use a fibreglass tensile member in a Neoprene matrix, and wrap around the pulley in a non-chordal manner. Very little stretch! That means the rotation does not have a ripple superimposed on it like with a chain drive. Yes, a jockey wheel for tensioning would be needed.

* I would use a 15 (or 20) mm wide 5 mm pitch GT3 belt for the second reduction of again about 5:1. Yes, I suspect that I would need a two-stage reduction to match the motor holding torque to the table torque requirements.

* I would be happy to use a DC servo motor for the drive. They can have more grunt. (Happens I have a stack of Baldors and better on the shelf. That helps. But a large stepper is viable too.)

* I would use feedback from an encoder on the table shaft. This avoids all the hassles about slippage and backlash. Two possibilities exist:

# ~5:1 reduction using a 2 mm pitch GT3 toothed belt to a HEDs-style 500 or 1000 line encoder

# direct coupling from the main shaft to a 10,000 line optical or magnetic encoder

* Standard feedback servo drive in the Mach3 style.

I have been vague about the reduction ratios for a very good reason: the exact values do not matter! On the drive side, a DC motor won't even be aware of the ratio. On the output side the lines/rev in case 2 does not matter either because I would calibrate the whole system after it is built. Mach itself contains software for this calibration, but I could do it myself very easily.

The bearings, belts and pulleys are not all that expensive. A 500 line HEDs encoder is not expensive. A 'genuine Western machine-quality' 10k line encoder would cost under $400 - and would probably be worth it. After all, the quality of the sensor sets the quality of the result in a feedback system. But it is optional.

Well, there you have it. Fire away.

Cheers

I think I agree with most everything here, as it is a sensible plan with known qualities and performance for all the parts selected.

I might forego the twin AC bearing design in favor of a surplus crossed-roller bearing, which would easier to implement (no need for adjusting preload) and more compact.

If encoder feedback is to be used I'd lean more toward using a servo instead of a stepper.

I think a hollow shaft would be very handy for things like milling keyways in leadscrews, making multiple parts...

[RCaffin]

Hi Louie

I might forego the twin AC bearing design in favor of a surplus crossed-roller bearing, which would easier to implement (no need for adjusting preload) and more compact.

To be honest, I had not looked at them seriously. I will do so.

If encoder feedback is to be used I'd lean more toward using a servo instead of a stepper.

More grunt!

I think a hollow shaft would be very handy for things like milling keyways in leadscrews, making multiple parts...

Hum ... true. Good point.

Not this year: I am in the middle of a large production run.

Cheers
Roger

[handlewanker]

Hi, the bearings for the 4th axis arrived today, 6007RS.....35mm bore, 62mm OD, 14mm thick and sealed both sides, so now I need some steel to build the box.

The box build should be straight forward with complete plasma welding etc.

Originally the construction was going to have hot rolled steel, with the base 20mm, sides and top from 10mm and the two end plates from 15mm, but that would mean a varied cutting list and a cost for each cut, so rationalising to make it all from 15mm thick X 100mm wide bar stock would mean a single piece of 530mm long material 100mm wide and 15mm thick to be cut to size at home.

Cutting 15mm with a power hacksaw will take a bit of time so I might just get all the bits cut to size with the original cut list and be done with.

With the jury still out for the drive, whatever method is used, I don't think the full capability of the drive will be exceeded by my immediate job plans as and when they occur, but in any case going for the friction drive method might well involve so much effort that the end result would not be worth recording.

The main exercise is to produce a 4th axis that gives zero backlash, and the work involved getting there might be like climbing Everest just to see the view.

I tend to get extremely focused on a point of view and need to step back and see where it's going to take me.......that's the cold hard light of day scenario when the effort could be far more than the results achieved.

I'll draw up a cutting list and pester the local steel suppliers to see what the damage comes to, then it'll be weld, weld, weld until the cows come home and afterwards a lot of machining.

BTW, Roger, I did like your summing up for a logical approach to the 4th axis build plan....it makes me wonder if the friction drive will be worth it.......I just hope I haven't ignored the potential for the belt drive by being overly pedantic.

I did a test last night using my metal rolls to see how much friction can be generated with just a piece of 50mm wide 6mm thick steel plate and the grip of the steel rolls on it in pinch mode........without a figure to quote, I can only state that the grip is powerful and even with oil on the surface the thrust from the steel piece on exit from the rolls was impressive.....even banging the steel plate on the end with a block of wood did not cause it to slide back in the rollers.

The rollers are 50mm diam and 300mm long, arranged one above the other with a clamping screw on top to each side giving the downward pressure for the nip.

Normally for bending a third roller is raised underneath the plate to deflect it up and around for cylinder work.

So as I stated previously, it requires a definite pinch mode application to get the drive to work, and for the record the rolls are made from mild steel and so is the plate test piece, and no indentation of the surfaces took place.
Ian.

[Eldon_Joh]

Try cutting steel like that with harbor freight 4.5 inch cutoff disks 3/64's thick. 8$ for a pack of 10.
honestly I think they are cheaper than metal cutting bandsawblades.

[handlewanker]

Try cutting steel like that with harbor freight 4.5 inch cutoff disks 3/64's thick. 8$ for a pack of 10.
honestly I think they are cheaper than metal cutting bandsawblades.

Hi, I think that would be a suicide pact with 15mm thick steel plate 100m wide....LOL.

It could be done, and I have a power hacksaw, just not worth the effort if the steel supplier can do it for me with their industrial strength bandsaws.
Ian.

[RCaffin]

Hi Ian

the work involved getting there might be like climbing Everest just to see the view.

Well, you need some justification for it! And the view is very fine IF you can see it (and get back down safely).

Seems to me that you will have to face off the bottom AFTER it has been welded, so it will sit flat. Then you have to line bore the bearing holes so the axis is parallel to the bed. This can be hard on a simple mill. Got any plans?

it makes me wonder if the friction drive will be worth it

Well, good question. Only one way to find out: do a full design and see. :-)

Mind you, 15 mm plate/strip is very solid - any idea what it will weigh at the end?

Cheers

[RCaffin]

I might forego the twin AC bearing design in favor of a surplus crossed-roller bearing, which would easier to implement (no need for adjusting preload) and more compact.

Well, yes, but as far as I can see, the double angular contact (AC) bearings would have a far higher load rating than a crossed roller (CR) one. For instance, with a 50 mm bore:
Timken AC 7210WN: static rating 28.4 kN
THK CR RA5008: static rating 7.19 kN

Part of the reason is that the balls and the race on the AC bearings are MUCH more solid than on the CR bearing. I guess that means the CR bearing needs to be supported a lot better, but even so, the load rating is a lot less (as far as I can see). Where the CR bearing win easily is the profile: it is tiny compared to 2 off AC bearings. That's no advantage here though.

Ah - except that the OD for the AC bearings is huge in comparison (of course). 7210WH OD is 90 mm, which makes the 100 mm wide strip seem very small. RA5008 OD is 66 mm: much more reasonable.

Yes, I know that Ian is talking about a 35 mm bore (eg 7207WH, OD 72 mm), not a 50 mm bore. I couldn't find a CR bearing with 35 mm bore for comparison!

-------
However ... consider this. 28.4 kN is roughly 3000 kgF. 7.2 kN is roughlt 750 kgF. Are we at risk of going way too far overboard here with load ratings? I mean, seriously, 3000 kgF is 3 metric tons! On a rotary table with a 100 mm chuck and a <1kW spindle drive. Um ...

Fun.

Cheers
Roger

[RCaffin]

OK, here is an idea which may be so far out of court you can all fall about laughing.
Ian has been talking about welding up a box. Lots'a'work.

Has anyone ever considered taking some 20 mm ground plate and just sticking solid two pillow block bearings on it - with taperlock fittings for the front one perhaps? I am assuming good quality bearings of course.
If not, why not?

Cheers

[handlewanker]

Hi Roger, having been down the welded box path on a number of occasions, in this build the box will be fully welded and then machined all over.

My Ajax mill, which is like a King Rich 2000, but has auto feeds and quick traverse to all axis with a table size of 54", would mince that steel box without even slowing down.

I'm going to weld it with my plasma welder and then clean the outside by running all over it with an angle grinder to level the welds, then It gets machined on all faces to square it up.

Boring is a simple job, first it'll get rough bored to 3mm undersize, both ends separately, after which it gets cut along the centre of the bores to make two halves that after machining the faces of the split get bolted together with 10mm bolts in the end plates.

I'll put .25mm thick shims between the faces before bolting down, then use my boring head in the mill to bore both holes to dead bearing diam size.

When the shims are removed the bearings will have a nip when the bolts are tightened to keep them in position.

The design I am going for relies on the deep row radial bearings having enough stiffness in their manufacture to not have any axial or radial float once they are mounted on the spindle and clamped in the housing.

This design does away with the need for the conventional spindle layout of angular contact bearings etc, and will be mechanically located not only in the box but on the spindle too.

Having sealed ball races simplifies the need to keep dirt and coolant out of the bearings and lubricant in.

I'm going to have the spindle preassembled by mounting the bearings on it with a long spacer between them and anything else that needs to be fitted on the drive end, like the drive pulley, or friction disc assembly and the retainer nut that locks the whole lot together.

Then the complete spindle assembly can be inserted into the box and the top bolted down to clamp it in place and make it a complete assembly.....no bearing cover plates or retainers etc to complicate the design.......and no need to have a press fit for the bearings.

I also intend to directly bolt the chuck to a back plate that is part of the spindle, attaching it with SHC screws before fitting the bearings and inserting into the box.

This will allow the chuck to fit very close to the box with minimum clearance and over hang.......I hate chucks screwed onto the spindle.

After the box assembly, the drive will be decided as to how much reduction and whether to go for a compound 1:20 set-up or just a single 1:5 stage reduction.

The one component that will make or break the design will be the driven friction wheel, as it has to be machined, hardened and ground to very close tolerances in both concentricity and surface finish of the inner and outer diams....... actual diams are not important, and I think that would be a good time to source a cylindrical grinding service to do the job properly.

I may yet just turn it and then have it hardened and do a pseudo cylindrical grinding operation myself in my tool and cutter grinder, but that is further down the track......first the steel has to be bought already cut in pieces ready to weld.

Once the box has been machined all over I will do some pocketing in the faces of the sides and top surfaces by milling the middle of the face out to half the thickness of the 15mm thickness, finishing the corners of the pockets with a ball end mill.

The bottom 20mm thick plate can be cut right through to the inside as the centre is not needed to bolt down on....this will remove a lot of excess weight and give it a bit of shape too......that will need to be done before final boring of the bearing bores in case of any thermal distortion being released.
Ian.

[mactec54]

Nice... I wonder how much DelCam 5-axis costs. More than all my entire machines I suppose.

The 5 axes Powermill software is around $10,000 to $18,000 US depending on what you want with it

The Hermle C50U 5 axes is a cool 1 million US

[RCaffin]

Hi Ian

Boring is a simple job, first it'll get rough bored to 3mm undersize, both ends separately, after which it gets cut along the centre of the bores to make two halves that after machining the faces of the split get bolted together with 10mm bolts in the end plates.

The bit you didn't explain is the bit which interests me right now. How do you make sure the axis of the rotary table is dead parallel to the bed of the mill? It will need a bit more than just bolts through the end face to guarrantee that. Bolts through the base plate maybe? Need to mill the bottom surface and the bore in one go maybe?

The design I am going for relies on the deep row radial bearings having enough stiffness in their manufacture to not have any axial or radial float once they are mounted on the spindle and clamped in the housing.

If they have a micron end float ... tolerable. :-)

Then the complete spindle assembly can be inserted into the box and the top bolted down to clamp it in place

Yeah, I like that.

that will need to be done before final boring of the bearing bores in case of any thermal distortion being released.

An hour or two in the kitchen oven at 100 C?

Cheers
Roger

[mactec54]

An hour or two in the kitchen oven at 100 C?

Why would you have someone wast there time to put the part, in the kitchen oven, at 100 C for an hour

When you stress relieve Mild steel, after fabricating, you need to heat between 500C & 650C, soak for 1 hour per 25mm of thickness to relieve any stress

100C would warm up the part & do nothing to the part at all, apart from it being hot

[RCaffin]

When you stress relieve Mild steel, after fabricating, you need to heat between 500C & 650C, soak for 1 hour per 25mm of thickness to relieve any stress

Fair enough. I don't do much steel work you see. I am happy to be told.

Cheers
Roger

[louieatienza]

The 5 axes Powermill software is around $10,000 to $18,000 US depending on what you want with it

The Hermle C50U 5 axes is a cool 1 million US

I believe MasterCAM is in the same ballpark. VisualMill is around $4000. I think SprutCAM is cheaper.

I think there is a point here lost in this whole exercise... Aside from wrapping text around a cylinder or some indexed milling operations, it would be very difficult to utilize the full capability of a 4th axis without the aid of CAM. And simultaneous 4 axis CAM is not exactly cheap. I can purchase it as a module for my CAM and it's still not a trivial amount of money. There are some inexpensive options out there that I have not tested yet but hope to do so.

[handlewanker]

Hi Roger, the method for boring and base alignment is simple really as I'm a fitter and turner it comes as second nature.

The box in the first stage after welding gets cleaned up all round with an angle grinder and checked for low areas along the welded seams, then they are rewelded to fill in and re-ground again.

I intend to mill pockets in the top and side faces so it gets done now.

The next stage is to face off the bottom to clean it up.

Then it gets mounted in the lathe 4 jaw chuck to rough bore the end plates, it can also be done in the mill but needs some heavy drilling.....having marked off each end for the centres of the bores......each end is bored individually to give a minus 3mm undersize to allow for splitting the box and re-machining the bores to size true to the base.

Once rough bored, the box is held in the mill vice and drilled in the four corners of the top half down past the split line with a tapping size drill for the Socket Head Cap Screws and the top half redrilled for clearance for the screws and counter bored for the screw heads.

Next the box is cut in two halves with a slitting saw along the centre line all round to give top and bottom halves.

Next the two halves are machined on the faces to correct any spring that develops from the welding when the box is split.

The two halves are now bolted back together with .25 shims placed between the halves.

I prefer to bore the box in the mill with a boring head to get both bores the same diam and in line as a lathe invariably turns taper over a length of 100mm or so and the finished bores need to be the same diam and true to the base.

With boring in the mill you can bore the front bore and know the other end will be the same diam without having to struggle to get a long telescopic gauge or inside mike on a stick down to the bottom.....but I always do that anyway just to be sure.

Setting the box vertical in the vice or against an angle plate and using a dial indicator on the bottom face ensures the bores will be parallel to the base.

In actual fact when I do a job like this I often either grind the base on a surface grinder parallel to the bore centres, or put it back in the vice and set the two bores parallel to the mill table and give the base a lick with the fly cutter.......a hundredth or two here is not a problem, but we strive for perfection.

You can measure the thickness of the material between the bottom of the bore and the base at each end to check if they are the same and parallel, using a roller or ball anvil mike in the bore.

When the shims are removed the bores will clamp the bearings without too much force, as in a press fit, and make assembly very easy.
Ian.

[RCaffin]

Once rough bored, the box is held in the mill vice and drilled in the four corners of the top half down past the split line with a tapping size drill for the Socket Head Cap Screws and the top half redrilled for clearance for the screws and counter bored for the screw heads. Next the box is cut in two halves with a slitting saw along the centre line all round to give top and bottom halves.

I guess the idea here is that the bearing races themselves will provide enough alignment? No dowel pins needed?

I prefer to bore the box in the mill with a boring head to get both bores the same diam and in line as a lathe invariably turns taper over a length of 100mm or so and the finished bores need to be the same diam and true to the base.

Ah yes, here we get to the good stuff.
My Adept is a combined machine, but not that 'mill hanging over a lathe' design. You would need to see one to fully understand the design, but the mill table serves as the lathe saddle. The lathe chuck is very solidly mounted. It has its uses.
I adapted my boring head to go in the lathe chuck, and mounted 'bits' on the mill table (=saddle) to bore them out. Subsequent testing showed that it worked quite well, although boring from both the front and the back helped to avoid any taper.

Setting the box vertical in the vice or against an angle plate and using a dial indicator on the bottom face ensures the bores will be parallel to the base.

This is the bit which has me nervous. How good is the angle plate? OK, OK, I am maybe fretting too much. As you said, we strive for perfection!

Cheers
Roger

[louieatienza]

This is a great excerpt from NC Cams, from an older thread about steppers versus servos, in regards to a Bridgeport mill, with servos driving ballscrews via timing belts...

In our situation (2:1 ratio belt driven preloaded ball screw, gibb based Bridgeport Eztrak mill with with high $$ true ball screw support bearings) we found the "error" contributed by the OEM "gilmer" style square tooted timing belt to be inconsequential.

Granted, we spent a small fortune on the ball screw bearings and weeks tuning and adjusting the machine to remove/eliminate slop. During the tuning process, we also spent a ton of time playing with belt tension and the like.

A belt in good condition had virtually NO effect on net accuracy once we got the slop out of the rest of the system. The BIGGESt contribution came when we had a pro tune the servos so as to get the follower error of the servo's "tuned out".

We tested via cutting perfectly round circles via a canned circle milling program. We then checked the form for roundness on our camshaft measuring machine (measures to 0.0000025" resolution).

When the machine was properly tuned, the circles were round within 0.0003" or better and the "flats" at the 4 direction changes were 0.0001" or better - the "flats' are nearly IMPOSSIBLE to eliminate due to gibb slop needed to allow the table to move without binding. We ultimately learned that the above "errors" were due to other hysterisis issues that simply weren't practical to resolve on a gibbs based "bridgeport mill".

Does the average guy need such accuracy? No.

But we CONCLUSIVELY showed that "belt induced errors" are much less in reality than what one might "expect" IF you have a well adjusted/tuned system made of parts with decent integrity.

Worn/tired crap isn't worth spending the time on. Get good sprockets and premium belts, properly tension them and make sure they "wrap properly" and spend your time and money on other stuff that provides a better cost/benefit ratio - like a good servo/stepper tuning regime and the best damn ball screws ans ball screw bearings that you can hardly afford.

[handlewanker]

Hi Roger, when the machinery is on the light side, you have to go where the Devil drives.....if you can mount the box on the lathe saddle then a boring head or a boring bar between centres is the best way.....in other words line boring, where the line bar is just a big diam piece of mild steel bar and a hole through it for a cutter, making sure the saddle can traverse left and right enough to cut both bores with the same tool bit.......2 tool bits spaced apart will not work as adjusting the tool bits for the diam is the tricky bit, also measuring the diam of the holes.....you need to remove the line bar to measure the holes .....ACCURATELY.......that is why the bar is between centres.....it can be replaced and be exactly in line as before after adjusting the tool bit.

With the bar out of the lathe, the tool bit can be adjusted with a dial indicator or a micrometer if you have one big enough.

If the tool bit re-setting is a problem, then after the bore sizes are measured the bar can be put back in the lathe between centres and a dial indicator used to move the tool bit, remembering that the tool will cut both sides of the hole so any adjustment will give twice the amount to come out.

This is the most accurate way to bore a hole when the holes are spaced apart and must be in line and the same diam.

You can also bore the box held in the 4 jaw chuck using a long sturdy boring bar in the tool post, but if any taper is present in the bores you will need to re-cut the second hole a few thou to make both the same size......measuring the hole diam closest to the chuck face is tricky, but a telescopic gauge is ideal here.

And yes, the two bearings will align the top and bottom halves automatically, no dowels need to be used.

When the job is bored and if you find there is a "run" on the base, it's an easy job to just mount the box in a vice and skim the bottom face to make it parallel with the two bores......"corrective surgery".....LOL, bearing in mind that the sides may not be parallel so gripping in the vice can be tricky.

It's easy with a biggish mill as you have the height to get at the parts with boring head, and a boring head will give the same results as a line bar.

A solid bar held in the 3 or 4 jaw chuck, that is not between centres, with an adjustable tool bit, can be used in lieu of a boring head if you don't like line boring, but small cuts need to be taken or you'll get some vibration.
Ian.

[RCaffin]

Hi Louie

Thaks for the quote. You are of course preaching to the choir here.
But would you have the URL for that quote?

Cheers
Roger

[handlewanker]

Hi Louie, that's interesting.....the problem with vee slides etc is one you can't overcome as the friction is the enemy and adjusting for smooth running the other enemy.

I would have to strongly think that if/when you are going to convert a manual machine to CNC one of the first head scratching decisions to be made would be "can I remove the dovetails enough to fit linear ways" without which the re-build is a compromise.

I see a lot of retrofits that only go to the ballscrew/stepper stage, and in my mind, even if you have to get the parts machined outside, it will be 100% worth it.

It was also an interesting observation re the toothed belt analysis, but the point Mike Everman made from experience in the early posts is....the tolerance in the belt manufacture is something you have no control over and will give mixed and varied results where used for determining resolution.

If you are looking for a solution to a problem, bending with the wind is not the answer.
Ian.