CNC Project X Axis Y Axis Wiring Control electronics

Z Axis

The Z axis, is by far the most complicated item of the machine and the most challenging too, it requires the most machining and a major part was fashioned out of the rack and pinion of a Clark bench drill press.

The body of the assembly being machined out of aluminium block and the bearing surfaces are also of aluminium, no additional bearing surface was required as the amount of travel compared to the rails was very slight, but I will as and when time allows rebuild the assembly with a more sturdy design.

The arbor being machined from silver steel ( drill rod) inclusive of collet chuck, this was turned between centers to ensure accuracy and to minimize shaft run-out, the chuck is to take ESX Collets.

the first section of the spindle was the internal collet taper which was turned using a taper turning attachment set to 7° this then formed the center taper at one end of the arbor before centering the far end, this was then fitted between centers, the drive to the shaft is via the original spline coupling now a slightly tighter fit in order to reduce any backlash and chatter, the Z axis has a maximum stroke of 96mm which is almost one full rotation the pinion, The Z drive is provided by a third stepper motor coupled via a 32:1 worm gear, this means that a 200 step stepper motor giving one rotation of the pinion will almost give 0.5mm (0.48mm) per step or 0.24mm per half step of Z axis movement.

The arrangement of the Z axis can be seen in the photos below, the worm gear drive to the rack and pinion shown on this side of the assembly, also shown is the collet chuck, the nut was obtained from a local friendly machine shop as damaged and the shaft was machined to fit.

The collet chuck is designed to fit the complete range of collets to cover all eventuality's, this will enable the work head to use both commercially available wood router bits and hand made 3mm flat bits, which can be made from 3mm carbide shafts ( broken carbide bits, you will undoubtedly collect a number of these) that is first ground to rough dimensions then honed prior to use.

The cutting rotational drive was to be provided by a standard off the shelf wood router motor, directly coupled into the Z axis arbor, A 500 watt hand held router gave a reliable no load speed of 28,000 rpm and ran reasonably smooth, the mounting of the router is simplicity in its self as a hole is bored into a aluminium block a remainder from an earlier casting and a slot machined from the top to allow the spindle stop button to be accessed which also prevents the router from turning in the mount, no other form of securing was considered.

i have since reduced the drive speed to 13000 RPM by off setting the router and spindle and fitting a drive belt this with the correctly selected pulleys gives a 2:1 reduction in speed

The mounting of the router, and adjustment to the drive belt is provided by two slots in the mounting plate, which provide sufficient movement to tension the drive belt, simplicity is best and this method seems to work well, As a belt drive is now being used a second top bearing was required and was built into the motor mounting plate by the addition of a removable bush that holds the bearing and gives access to replace the belt as and when it becomes necessary as well as providing greater mechanical rigidity

As promised earlier this is the re-modeled Z Axis design

I have decided to remodel the Z Axis because the current design has too many faults, which were partly designed in,and partly because of the material used to build it, for it has too much play which is causing rough edges and rattling, it is not sturdy enough, nor has it any where near enough Z axis travel, an early design flaw which is rectified in this revision.

These revaluations have only come to light during the use of the machine generally a good method of testing and shaking out faults

The major problem is that the quill was not long enough, because as the cutting point is extended the top of the quill is going into the main sliding bearing, this means that there is less shaft available to keep it steady, as these bearings were machined out of aluminium each time the quill slides through a black metallic film is being deposited on the top bearing this means that on each stroke the bearing is getting bigger, also due to the increased hole clearance and subsequent knocking on the side of the bearing the hole is starting to pull, resulting in that every time the machine is used the worse things get.

The point i do like with the current Z axis is the sliding quill, this will be included in this iteration with an alternative drive mechanism and use taper roller bearings so that the shaft can be preloaded against the bearings to remove play and runout.

With the new design the sliding shaft will always be fully held in the bearing block, this results in a more stable bearing, and providing greater support plus the increased mass of the bearing will make for a more solid spindle design, i am this time going to use direct drive and move away from belt drive, to this end the motor and spindle shaft will all move up and down as one unit so i will not be using spline's to drive the cutting head, which is the current design.

The Z axis lead screw is driven by a 90° gear attached to a stepper motor which will give around 3:1 reduction these gears are then fitted into an aluminium housing attached to the side of the main bearing of the Z axis, see sequence of photos below

It would seem that my Z Axis designs are never simple!!

The Z Axis feed drive is built on a right angle drive gearbox based on a helically cut bevel gear, as taken out of an old disk cutter the last photo in the sequence is what is left of the angle grinder i bought at a secondhand sale for a few pence, it looked good but smelt burnt and after hanging around under the bench for a while, i found a use for it.

This gives a reduction of about 3.5:1 therefore 200steps*3.5 gives 700 steps per revolution of the lead screw so 5/700 0.007mm per step not a nice round number!! one advantage in having such a small step is that the holding torque is multiplied by the reduction ratio of the gearbox the down side of course is speed of travel, but as i have said before it is only going a short distance.

The main deciding point of using this type of gear box was because the motor and feed mechanism has to be below the motor mount so the motor is pulled down and pushed up, if the lead screw motor is at the top by the main spindle motor then the lead screw will move down along with the cutter and would poke out from under the Z Axis support which would be extremely hazardous.

So the basics of the new Z head.

I am going to use a 280mm length of 50mm linear shaft which is pre hardened and ground to a fine finish this is first drilled out to take rotating shaft and then the ends are bored out to fit the bearing race, a taper roller bearing type 30203J fitted in each end, the rotating steel shaft will be fitted from the work end and a left hand retaining nut will provide loading on the bearings and simultaneously retain the shaft in place.

The main sliding bearing is a 100mm long split bronze bearing, held in a aluminium housing , the two being cut at the same time, the reason here is to provide adjustment to the bearing simply by tightening the two bolts that run across the split, as has been sucessfuly done in many machines before.

The housing will hold two further linear rails which are to stop the shaft from rotating in the opposite direction to the motor and are for guidance, the gearbox is secured to the main aluminium bearing block with two M6 hex socket head bolts, the top of the gearbox is to be flush with the top of the main bearing so that the lead screw runs alongside and parallel to the main sliding shaft.

Photos 1~ 3 main bearing block Photos 4 ~ 5 sliding Z Axis shaft

The start point of the Axis is the main Bronze Bearing, this was turned to be a good sliding fit to the 50mm shaft, next the aluminium block is machined, i use a brush with light oil as a cutting lubricant but because of the act of machining especially when boring, it becomes literally too hot to handel, and dimensioning must take this into account.

I bored the hole to rough dimensions then finely trimmed the hole to take the bronze sleeve, an easy sliding fit is all that is needed as long as the main aluminium block is hot, finnish off the machining and slip the bronze bearing sleeve into the aluminium block run a little adhesive over the bronze sleeve before inserting into the aluminium, and stand on a cold surface table, leave over night to cool, the action of the heat will help set the adhesive, when it has cooled it is also held in place by the contraction of the aluminium body.

Next slip the sliding shaft through the block mount onto two "V" blocks and clamp both the shaft and aluminium body, mill off the edge for the two M10 nuts to sit, also drill through and counter bore the holes on the head side, next unclamp and remove the sliding shaft and saw the adjustment slot, here i mounted a tungsten carbide saw blade in a milling arbor and with liberal amounts of cutting fluid made the cut in two passes, remove the block from the mill and fit the two M10 clamping bolts, by using these bolts to clamp the bearing block firmly onto the sliding shaft, i then mounted the shaft by one end into a dividing head and very carefully adjust the shaft using a"V" block to support the other end adjust the shaft for paralell and square by using a dial gauge, to as close a tolerance as can be achieved.

(you may have to tighten up the grib strips on the mill table to take out that any play!!)

Having trued the shaft, next is to mill two slots these slots are to hold the two vertical fixed slides in my case i attempted to machine two semi-circles using a bull nosed cutter, ( Bad idea) STICK TO A SQUARE SLOT AND MAKE SURE IT EXACTLY FITS THE BARS USED, it would be best to use an undersized cutter and make two or more passes two get the slot center and true.

drill the side bars counter bore the holes, drill & tap the block and fit the side bars, check for square, this is very important as all must run paralell.

Next machine the sliding bar,this bar is hardened linear rail and providing you do not want to make the hole too big it is relatively easy material to work, in my case i had to drill use a long series drill but be careful do not force the drill or the hole will run off center, on finishing and fitting the bearing shells into the shaft, here i noticed that the ends to take the bearing cages were slightly enlarged this must corrected, i found the best method is by lapping each end with a diamond flat sharpening steel and thin aerosol oil whilst in the lathe, this is very hard case material and will slowly cut, but be careful do not take too much off or there will be play at the end when the shaft is retracted.

Next is the retaining collar, this is turned out of mild steel drilled on four square and shrunk fitted onto the shaft, it may be an advantage to add a grub lock screw to ensure that it is locked into place, four M6 studs are secured into the collar with anaerobic adhesive.

Having fixed the collar into place spend a little time adjusting the bronze bearing for a very good sliding fit, it should not be difficult to adjust the bearing for 0.00" play and maintain a free sliding action, try to bed it in a little by holding the block in a vice and sliding the shaft back and forth adjusting the clamping bolts to a fine fit, check and double check for any lateral play and readjust if necessary.

The next task is to machine the rotating shaft and collet, I would suggest that the first end to do is the collet taper and thread for the collet nut, here i used a 1" piece of silver steel (drill rod) held in a 5c collet, this is because a much better and repeatable center location can be achieved than with a conventional chuck, but one should always check with a dial gauge, the collet end is then used as a center at the tail stock end, turn the work around and center the other end, turn the work around again and remount in the collet and center, then proceed to machine the bearing journals, do not make these too tight but a tight sliding fit, as you will have difficulty in removing the shaft at a later date, the best method i have found to accurately hone the bearing journals is to run the lathe at high speed and use the diamond lapp the soft material will cut with ease and with plenty of aerosol lubricant give a fine finnish that looks like it were ground, when all is done and the bearings fit,the bearing closest to the collet chuck is held in place with anaerobic adhesive and is permanently secured in place.

there you go 4 bolts and it is off the compleate z axis

oh by the way, i Just thought i would show the grizzely old git!

"Just to prove the system does work, and the fact i could not have cut it out by hand the motor actuator plate, which i cut on the CNC Machine as is, close inspection of cut edges will show slop and play in existing Z Head"

The next very difficult part is to fit the sliding bearings onto the aluminium plate as shown in photos 10~12, here i used two oilite sintered bronze sleeved bearings, this did not work well, because unless all of the sliding shafts were sprayed with oil they would dry and size this overloads the stepper motor stalling it, i have since opted for plastic or PEEK bearings simply drilled through with a milling cutter and the hole sawed through, this gave a degree of adjustment these bearings will run dry providing they are not too tight and will bed in quickly.

there is a M5 bolt that i run through the slot of the PEEK bearing to give some adjustment

the last two photos in the series shows the assembled Z Axis, less the spindle motor these will be fitted as soon as the drive coupling becomes available.

Z Axis

Design Modification

I have found a design flaw where the distance of the center of the lifting screw is too far away from the center of the main Z Axis sliding shaft, this resulted in the assembly jamming and would then require a thump to free it only to jam again the next time the sliding shaft returned to the same point, i have therefore been revising the design and i have come up with an " an improved" design which is more sturdy and free form jamming.

An undesired problem was that the overall weight of the motor, sliding shaft, spindle motor, and motor mount, totaled a massive 15Kg, this proved too much I think for the Nema 23 frame stepper motor to handle even with the 4:1 reduction gearbox as detailed above. and i have unfortunately had to resort in a bit more gearing than what i wanted, how ever as this mechanism seems not to suffer from jamming and as the movement is much smoother, with a bit more thought i may well be able to re fit the ousted gearbox, the downside of the extra gearing is that the time to travel the full extent is well far too long, however it works and will do until i get the itch for more tinkering, or get so bored wating for "Z" axis movment.

so for now on this section that is it, FINISHED,.........well just a bit of wiring then No MORE "Z" !!!!!

CNC Project X Axis Y Axis Wiring Control electronics