Wishbone stabiliser for round post drill/mill

[irving2008]

I had a little dig around and found some SKF cast pillowblocks today....and some 50x20 rectangular tube....Also some spare HR1530 linear bearings and 650mm long rails.......I may be a defector

Hi Kip, I'm interested in what you have in mind... I like what Bill's doing but it would take me forever to create something that impressive (and the roller nuts are way beyond my current machining skills)!

[BillTodd]

Hi Kip, I'm interested in what you have in mind... I like what Bill's doing but it would take me forever to create something that impressive (and the roller nuts are way beyond my current machining skills)!

Er, it's only a matter of facing, drilling and tapping four lengths of steel rod,

The joints are 18mm lengths of 1 1/2" x 5/16 ally extrusion (bent in a vise where necessary)

The rose joints are cheap pressed steel ones. They had a bit of play, but this is easily removed by (using a pair of old bearings as anvils) squeezing them in a large vise.

Making the post clamp is going to be tricky. Simply holding the piece, while boring a 115mm hole will be a challenge. But, none of it is difficult. The only difficulty with the roller nuts is fitting the ****** things to the mill.

If you have a lathe and a mill of some description you can do it too

[Kipper]

Hi Kip, I'm interested in what you have in mind... I like what Bill's doing but it would take me forever to create something that impressive (and the roller nuts are way beyond my current machining skills)!

I was interested too...until I realised a fundamental "thickie twin" flaw and abandoned all hope.

Bill, try pipeline center <American company and have a look at 115mm pipe clamps, or a 115mm "U Bolt"

[BillTodd]

Bill, try pipeline center <American company and have a look at 115mm pipe clamps, or a 115mm "U Bolt"

Now you tell me

It was tricky; I had to make a 'face-plate' to hold the 160mm x 15mm piece of scrap iron whole I turned part way through each side then jig-sawed out the hole, prior to boring and finishing the outside (pity those threaded holes were right in the way )

I'll weld on a couple lugs for the rose joints drill the clamp holes then cut it in half tomorrow (Oh no!, it Christmas tomorrow. - If his birthday caused this much disruption while he was alive, it's no wonder they nailed him to a cross )

[BillTodd]

I managed to find time to press on a bit today...


I cut the clamp in two, drill and taped the clamp bolts then fitted the wishbone (WB) joints on small eccentric height adjusters; as long at the two WB axes are at 90' to the post (i.e. equal height) and parallel the head should track a true vertical.

I had to re-think the WB joint; The upper WB has to be wider than the bottom in order to clear the edge of the head but, because the rose joints have only a limited angular movement in the 'side' plane (IYSWIM) I've had to space the rods apart at the WB joint end.

Tomorrow, I'll make up some blocks and a couple more eccentrics to mount the upper WB to the head

[sharpshooter90]

Bill,
My compliments on your workmanship, but I can't help thinking that there will be flex in that joint assembly. Being as far away from the head as it is, the flex will multiply itself at the spindle center. Also, the length of the arms will have to be equal at the full down position to prevent a bind- this may require 2 of the arms to have right and left hand rod ends for adjustment.

[BillTodd]

Bill,
I can't help thinking that there will be flex in that joint assembly.

Yes of course. The problem is that the two wishbones have to fold flat at the bottom of the post so one (or both) have to have an offset where they join. The offset inevitabley leads to a twisting moment and therefore flex :Shrug:

I looked at the possibility of extending the axes above the bottom of the head and below the bottom of the post, but there are too many things in the way.

Measuring the 'finished' version shows about +/- 10 thou" flex as the head is wound up and down but is does spring back towards the centre as the twist force is removed.

Being as far away from the head as it is, the flex will multiply itself at the spindle center.

Actually if you think about it, the opposite is true. if the joint were at the centre of the post any give would be multiplied hughly, as the joint is moved further away the less effect the play in the joint will have.

Also, the length of the arms will have to be equal at the full down position to prevent a bind- this may require 2 of the arms to have right and left hand rod ends for adjustment.

Yes, noted. All the arms are adjustable; they're fixed axially at one end so it's possible to turn the rod, screwing it onto or off of the rose joint.

[BillTodd]

I made the head mounting blocks, drilled the head and fitted it...

Well, it works! It's not quite right yet;

The upper wishbone ball joints (either side of the head) are binding against their fixings because the angle to the WB joint to too sharp (it needs a wider joint block )

The motor fan case just collides with the upper wishbone just before the bottom of the head travel - I can see no easy way to fix this, other than to take the case off and 'modify' it slightly (Engineers, cover your eyes while Bill wields his hammer)

Apart from the above it works OK. I trammed up a vertical straight edge using the quill a a reference, then shifted the head up and down to see what happens:

There's about +/- 10 to 15 thou" 'flex' as the head lifter is turned (there is considerable twisting force as the handle is several inches off the side of the head) but it generally springs back to a few thou" (say, +/- 3 to 4) of correct. (The 'zeros' in the pictures are genuine, but not really 'typical' )

[Kipper]

Looks like a winner Bill!

From concept to completion very fast!

[BillTodd]

I was originally going to fit an electric screw jack onto the upper wishbone to lift the head but found the jack was just too big to fit.

While searching through a junk draw in the shed I found a geared motor and wondered if it would be strong enough the lift the head.

So, I made a prototype:

The Doga motor I found is a 24vdc ~125w geared to do 80 rpm (no load). It's rated at 6Nm. A quick test with a torque wrench showed I needed ~11Nm (8ftlb) to turn the lifter handle (after removing the lift mechanism I found the worm screw had been rubbing against the rack - once fixed, the torque dropped to ~6ftlb).

The only suitable gears I had gave me a 3:1 reduction (I guess they're from a photocopier). Fortunately, they were easy to adapt to fit the motor and handle shaft. I simply bored the small 20 tooth one to 12mm and cut a key way with a slot drill. The 60 tooth gear had to be sleeved down to 17mm for the handle shaft.

Luckily, I had four 1/2" x 60mm stand-offs left over from another job, I drilled and threaded one end to to take a 1/4" x 20 to fit the mill and left the other M6.

I knocked up a motor mounting plate from a small piece of 4mm sheet aluminium (Now that I know it works I'll make a proper one out of something more substantial) - I managed to get the holes in almost the correct place first time

And, well it works fine: the bench PSU will have to be replaced with something a little smaller but 30v @ 2.5A lifts the head, bottom to top, in about 30 seconds (maybe less).


I thought that using an electric lift would remove the twisting moment so stopping the head moving at all during the lift. However, for some reason (maybe the fact that the rack teeth are angled?) there is still a jump to one side on lift, and surprisingly, a jump to the other side on lower. At the moment, with still some play in my rose joints, it amounts to about +/- 5 thou" error

[sharpshooter90]

So what was the final verdict on this design?

[BillTodd]

The wishbone works well - I changed the upper wb joint slightly, to clear the motor (now I need to set it up again).

[handlewanker]

Hi Bill, been following your work on the problem of the rotation on lowering and lifting the head on these type mill drills with the round columns.

I almost bought one in 1980 before I emigrated to OZ, but the round column put me off and I settled for a Beradi Italian made Jig borer, about the same size but without the round column.

Long story short, during the fifties when I was serving my apprenticeship we had a small mill drill type machine, make unknown, and it also had a belt driven head that rose and fell on a round column, but the column had a Vee shaped keyway in it, which ensured the head was always (more or less due to age) in position on raising or lowering.

The best bit is that the column was attached to the base by a substantial flange mount that upon slackening the flange clamp allowed the COLUMN to rotate, and so the raising and lowering and rotating of the head were with two different mechanisms.

I've seen this method on a tool and cutter grinder made in USA called a Cutter Master, which also has a seperate rotate and raising mechanism.

I realise this is major design "adjustment", but in the interest of efficiency anything is allowable.

In my opinion, the column of the mill drill should be attached to the base casting with a flange that would clamp it and allow it to rotate when needed, whilst the head should be made to LOCATE to the column without allowing it to rotate about the column.

A design I have in mind would be a base bolted flange that had a bore/length ratio of 3:1 and a clamping split at the back to hold the column tightly.

It's not very often that the head needs to be moved from it's central position, but it's VERY often that the head need to be raised and lowered and position maintained when drilling and reaming to allow tool removal etc.

The Beradi mill I bought back then has a large head casting that is attached to the column by dovetails, which means no column rotation for off table deep hole drilling etc, but this is not a problem as I decided years ago to change the spindle from a 2 morse taper to 30 int taper, and as that project was started in 1984 it might just get finished now that I'm retired, LOL.
Ian.

[BillTodd]

Hi Ian,

but the column had a Vee shaped keyway in it,

I came across the idea of a key-lock column here in Tony's Lathe archive.

" The startling performance of this key in maintaining accuracy has to be seen to be believed. Its very simplicity is deceptive. When raised or lowered on the jacking screw and re-clamped, the outer end of the bed can be relied upon not to deviate by more than .001" to .002" in radial alignment. "


As used on the multi-purpose LaborMill. It looks ideal for a round post.

[handlewanker]

Hi Bill, the big problem is to make a keyway in the column and bore of the head accurately enough with whatever means you have at hand, and that means some pretty accurate machining with large machinery.

The other problem is you only get the key to locate the head in one position, and if you want to swing the head to either side you will lose register if the key is removed or disengaged somehow when raising or lowering the head.

Having the head and column located seperately allows the head to be raised or lowered without losing register even if the head is swung to one side or the other.
Ian.

[BillTodd]

video of various mods:

http://www.youtube.com/watch?v=7y7pPm5vPys]RF30 CVT and other modifications - YouTube

[handlewanker]

Hi Bill, seeing is believing, or, one picture is worth a 1,000 words, a video 10,000....LOL.

I absolutely fell off the chair when you wound up the drive motor to adjust the speed......it's so simple in concept.

I would have thought that applying a variable pulley drive would have meant seriously modifying the head driven pulley, but having the side jockey pully adjustable makes it so simple.....my hat's off to you for ingenuity.

The wishbone device is quite nifty......how accurate is it, as in do you get a movement in the radial position at all when raising or lowering?

This is a vital need as if'n there is any variation it makes the problem ....oooooHHHHH heck!

I wonder if'n the original design had been made with a square column, same proportions diam/square wise, and the rotating base, if'n this would have solved the problem from day one?

This I suppose would have added cost to a relatively inexpensive machine, but the dividends would be 100 fold.

It's easier to bore a round hole than a square one, which would need a slotting action or very large broach, but other designs have gone to the square column/dovetail design which is better.

Just a wild thought at the advantages of a square column, I wonder if a square could be "introduced" into the head casting where the column hole is bored....not necessarily a complete square, but just enough to get a square column into the round hole with corners only locating and a rotating base.......it doesn't have to have a full bearing quality fit, as when it's in position the head is always locked to the column, and movement is slow when raising or lowering.

But.....the square column would be much smaller than the full round one as it is across the corner dimensionally larger but overall in the cross sectional area less, and this would make it like a lollipop on a stick with the vibration from flimsiness added, but if'n the top of the head had a triangular brace fixed back to a wall it would probably damp out any head movement without adding to the design complexity.

It would be a grand plan if'n a head casting could be re-engineered at the pattern stage to get more body in the column area so that a square hole could be made instead of the round hole......the square could be larger in area dimensionally and this would stiffen up the vertical plane a lot, and the square would only need 30% of the face at each corner to make a secure location......a truly positive locaton without any loss of position.

I think if'n anyone wanted to fabricate a vertical mill, a square column made form large section channel iron welded together to form a square and machined true at the corners would be a starting point, and incorporating a rotating base flange too.

BTW, I think the variable speed design you videoed could be made into a kit to modify similar mill drills without too much trouble, and maybe you could retire a millionaire...LOL.....and if'n you motorised the design to allow push button adjustment from the front...then billionaire status is a possiblity...LOL.

(Bill says he's already retired?)
Ian.

[BillTodd]

how accurate is it, as in do you get a movement in the radial position at all when raising or lowering?

Depends on how well it's set-up - You notice when the base clamp is rotated it tends to lift, that throws the alignment off somewhat.

If one carefully aligns the top and bottom ball joints (as I did when i first fitted it) then better than +/- 10thou" is possible - certainly good enough for drilling on a POS drill/mill

Just a wild thought at the advantages of a square column, I wonder if...

You're over-thinking it! If you want "accurate" you need a better mill


Bill

[handlewanker]

Hi Bill, one of my problems....overthinking and finding solutions to problems that really shouldn't exist.

I've always succumbed to the notion that things can be improved by a bit of re-engineering, and a lot of DIY improvements go down this path, purely from economic reasons in the first instance.

I noticed that whan the wishbones are extended there is practically no rigidity present....do you still get some allignment when the head is at the top of it's travel?......010" is pretty good if'n it's consistent.

You can get some variation in the vertical allignment when raising and lowering the knee on a Bridgeport when it's been in service for a while....this is a kind way of saying there is wear present in the dovetails of the Knee/column......but the solution in a strip down and rescrape is daunting.

Most times the variation can be decreased by adjusting the gibs carefully and always raising the knee slightly after you drop it down and lock it.

The wear is a constant factor in the travel, so if'n you always raise the knee and lock it up you should get back to vertical alignment every time, provided the quill is clocked square to the table and left set.

In the case of the Bridgeport, the quill does most of the work for X axis, and you only need to move up and down to get the extra height.....so saying unless the quill is dead blind square to the table, you will get some misalignment when the quill is racked in and out.....and in both X and Y planes too.

I've seen people square up the head by setting to the circular degree markings only, no clocking of the spindle to the table etc, and then wonder where the misallignment came from when the quill was racked out.
Ian.

[D.L]

I may make a variation on this theme using 4 chevy rods with plate stiffeners, pivoting on steel bars.
I'll put some cad image later after i've finish building my friends router table.