PDA

View Full Version : CONVERSION: Buget Warco Major 5 axis conversion



Tweaky
14-02-2010, 09:50 PM
Greetings to all, and thanks to many.
I have been researching for a while now, and found many helpful and informed comments on MYCNC........ Many thanks for a great site and comunity.

I have for many years wanted to have a cnc mill to make prototype parts, and possibly manufacture small runs to sell.

After many failed purchace attempts on Ebay, I finnally bought an old Warco Major, made in Korea, and fairly poor condition, but at £125 I could not refuse.
Found a man with a van (and tail lift) to deliver it for £40.
Built a stand from 4x2.
I stripped it for inspection ...... ooooo! ..... it looks like i was used to cut stainless mainly on the Y axis, no covers for the ways and lots of swarf embeded in them.
There was a dip in the middle of travel, of approx 7 thou!, so I spent many, many hours with stones and then lapping paste and it is down to 1/2 thou dip. (just read of an enineering company in Bedford, who will recut the ways for £30 ish) ....we live and learn.

The X axis was OK.
Quill OK
The spindle bearing are fairly worn and get hot at 2500 rpm, will replace those later.
Motor (2HP single phase), was replaced with a 3 phase 2HP for two reasons:-
1/ the single phase motor was the source of much vibration and fettling did not improve it (un-even air gaps)
2/ I already own a VFD.

For the conversion, I gleaned much of the spec. from here, Robin Hewitts post, but combined with info from Gary at Zapp and others.

Bits:-
Ballscrews where purchaced from China (sorry to the English Companys) for £92 inc post and end machining. That is 2 x 1605 screws 400 and 600mm long, 2 x ball nuts + machining. They are single nuts and C7 rolled screws.
Direct drive size 34 steppers (4.5 nm) from Arc Euro for X & Y, & A, size 23 (3 nm) for Z & C. (forgot to say, it is going to be 5 axis).
Z screw from an aircraft breaker, (a flap screw I think) anyway it has a 1605 screw and rotating nut.
Driver was one of the cheap 3.5A 36V, sorry 1.5A 24V boards from Ebay, but more on that later.
Mach3.
Old 2.8Ghz computer and a 15" ELO touchscreen from a skip(thanks to my partner) which I repaired.

The work:-
Made blocks for the nuts, used the original bearing blocks and used a pair of angular contact bearings on each ballscrew, with the opposite end floating.
1723

The motor mounts
1725 1724

Motors on
1726

Having mounted the motors, I was keen to try it out, so I connected the motors to the driver 36v 11a PSU at min voltage of 32v and hey presto!
Chuffed with my progress, I came indoors and did some more reseach.
I read page after page saying that the driver board I was using was pants! Most blew-up at 30v or more (I had run mine at 35v with no problems {35v is max input for the 5v reg.}).
More research, looked at modding the driver, then decided this was a waste of time and money. So i bit the bullet, phoned ArcEuro and ordered 2 x 9a and 1 x 4.2a drivers (for X, Y & Z). Bought an opto board from ebay, also found some IP67 micro switches for the limits (£25 for 5)
Using the 36v PSU (can go to 80v on the 9a driver) connected up and spent many hours, initially without any movment from the motors, then to one direction only (9a drives have switch for step/dir or CW/CCW orientation wrongly described) and finally working well.(Required 6uS setting in Mach3 for both step and dir.)
Since this is never going to be hyper acurate, I have used 2000s/rev = 400 steps per mm.
It runs at the maximum Mach3 can output in 35Khz mode which is 5250mm/min. so I am happy with that.
1727
Made sone covers for the ways, and fitted X & Y limit switches (one each).

Found another bargain on Ebay, my fourth axis, apparently from a Swiss gear cutting machine.
1728

Boy is it heavy, must weigh 40-50 kilos!, bit worried about the balance of the table, but with the trunniun table and end support it should be OK, I hope!?

OK enough already, I will add more a I go.

Tweaky
14-02-2010, 10:01 PM
Should have said that this is my first post, ever, so I hope all is correct?.

John S
14-02-2010, 10:02 PM
Tweaky, Drop the speed down a load, it's not a race and reliability is worth more than speed over such a short bed length,
If you know it does 5250mm min, halving it will give you a 100% safety factor.
Set the velocity in Mach to 2500.

Tweaky
14-02-2010, 10:29 PM
Hi John, thanks for your coments.
Should I increase the resolution, which will also decrease the top speed?

Tweaky
14-02-2010, 10:35 PM
I should add that whilst setting up, I set Mach3 to 45Khz, turned up the PSU to 40v and it ran hapily at 6000mm/min, so I thought I had some margin already.

John S
14-02-2010, 10:53 PM
If you are happy at that then fine but 6 metres a minute over a bed length of this size doesn't give you much time to hit the E-stop.
Another point is that Mach is more happy at lower kernel speeds less chance of errors and the base 25K is plenty for mills etc.

The high speeds were brought in for routers and such with servo motors that require 2,000 to 4,000 steps per rev against steppers at 200 steps [ in full steps ] these were running out of steam.

You don't really achieve much on stepper driven machines as the stepper drivers are the choke point and get egg bound way before the set maximum.

Tweaky
14-02-2010, 11:05 PM
You are quite right, it leaves no time to hit the Estop, in fact whilst playing I found that the limit switch had to be set to be nearly 25mm inside the EOT to enable it to stop before *!?
Cheers

Tweaky
14-02-2010, 11:12 PM
Here is my box for the electonics (ex alarm box from a skip). It is not tidy yet because I am still sorting it out.
1729
and here is the Y axis limit switch.
1730

Robin Hewitt
15-02-2010, 11:30 AM
Hi Tweaky

Looking good but no pics of the Z, have you done that one yet? A honking plate? :naughty:

On mine the pin that stops the quill turning had raised a burr down one edge of the groove it runs in. You can't see it, you can't even feel it, but when you do the Z you may lock the quill orientation and move the burr out of the groove it has cut for itself. If you have the quill out, easy job, look for a groove on the inside of the housing by the guide pin. If there, assume a burr and stone it off.

Also, are you sure the spindle bearings are shot? l had a lot of clatter on the splined shaft and tried to fix it by liberal application of heavy grease, maybe a previous owner did the same to yours, it is an obvious fix :whistling:

The grease eventually found it's way down in to the spindle bearings causing them to run hot at top revs, same as you've got.

Might be worth running oil in at the top, with the spindle running, until it comes out oil coloured at the bottom. Changing that God awful original motor fixed most of the clatter for me, flushing the spindle bearings stopped the bearings running hot.

best regards

Robin

Tweaky
15-02-2010, 09:58 PM
Hi Robin,

Z is next, have a smaller honking plate than yours, only 25mm thick. It is a disk 150mm dia., with the quill bore offset.
I have seen thinner ones, but when I saw your plate, I suddenly felt inadequate!. Do you think 25mm will flex?

Thanks for your help with the quill (does this mean I owe you a Hobgoblin?), I did find burrs on the quill groove and the pin when I stripped it. The pin is quite worn.
The bearings are marked on the rollers and the race, although once cleaned, it did tighten nicely ( no soft area), however, like you I would like more, 5-7000 rpm area. I have found some angular contact bearing that will fit, and they are rated at 14k rpm, dinamic load of 24Kn, do you think they would do the trick?

Last question, for the mo' , as you can see my finish on ally is poor, I have not yet aligned the column, but have read much which suggests that finish is related to vibration. I still get some vibration at almost all speeds, feels to me like column flex. Does your do it?, have you filled the column with Epoxy Granite or balanced you pulleys (mine has 2 inters .....pain)?

Thanks again.

Regards
Mark

Robin Hewitt
15-02-2010, 11:47 PM
I have seen thinner ones, but when I saw your plate, I suddenly felt inadequate!. Do you think 25mm will flex?

Hi Mark

I don't know if mine is thick enough, won't find out until I run it in to an immovable object. (3Nm*pi)/0.005m pitch = 1885N driving force :whistling:

I'll do my best to answer your questions but I've only been doing this for about 20 years so I'm still learning.


does this mean I owe you a Hobgoblin?

Indubitably, if we ever meet I will expect beer :beer:


I have found some angular contact bearing that will fit, and they are rated at 14k rpm, dinamic load of 24Kn, do you think they would do the trick?

Probably be fine, you're going to find out for us. Probably when a fat tool gets lost in a lump of iron and starts trying to cut triangular shaped holes with a sort of hammer action. :naughty:


Last question, for the mo' , as you can see my finish on ally is poor, I have not yet aligned the column, but have read much which suggests that finish is related to vibration. I still get some vibration at alost all speeds, feels to me like column flex. Does your do it?, have you filled the column with Epoxy Granite or balanced you pulleys (mine has 2 inters .....pain)?


What's 2 inters mean?

Obviously you want to avoid vibration but cut quality depends on a lot of variables. I still get a crap finish at times so I have yet to sort it. My biggest problem is aluminium chattering when I change direction into a radius.

If I was asked to list the variables that result in a good finish in descending order of importance, I'd guess...

A machine that is much too big for the job in hand.
Minimise overhangs.
The right tool for the job.
The right lube
The right feed rate and rpm
Cutting downhill

My mill is somewhat different to yours. I have a cast iron box section plinth at the back below the round column, often as not the head is wound down until it rests on the plinth. If I lean on the spindle nose I can bend it about a thou, maybe two if it isn't hard down. Perhaps they changed the design because it was too flexible?

Bestest

Robin

Tweaky
16-02-2010, 02:13 AM
Hi Robin,
I'll try the bearings and let you know.

By "inters" I meant two intermediate pulleys, so four in all. Every other Warco Major I have seen, had one intermediate pulley.
From my perspective, it is just another component to add errors to the setup. I guess I could dump one pulley, and get longer belts.

Thanks for the short list, by overhangs, do you mean the machine itself, or mounting of the workpiece?


The right tool for the job.
The right lube
The right feed rate and rpmThese are things I know almost nothing about yet. My experience with machine tools is confined to 1 year spent at a broad base engineering school, as part of an electronic apprenticeship, and that was 32 years ago. I have read much over the last few years, but there is no knowledge like experience, so I Know very little at the mo'.

I did not notice the difference in your mill until I read your message. It does seem then that the column is the problem, that you can remove it from the equation by lowering the head to the support, it's case solved then.
I have had a few ideas for improving things here, one was to mill square a large cast-iron wieght, and then mount the column support onto it, with the column through it.
I have enquired about a solid steel column, which could be cut with a groove to locate it radially, a la' Arborga mills.
A square column is the ideal I guess, but not a good use of time for me (silk purse and sow's ear etc. ) better to buy another machine.
So maybe i will have to be satified with small cuts.....?

A Fursty Ferret is calling me, so time to go.

Many thanks

Mark

Robin Hewitt
16-02-2010, 11:53 AM
Hi Mark

Overhangs are tool tip extension from the chuck, how far the chuck projects below the quill, distance you wind the quill down to make the cut, height of the workpiece above the bed.

Small cuts are not a problem for CNC once you have got past the teething troubles and trust your setup and G Code not to go hideously wrong. A large job may take hours but that's okay for a hobby machine which doesn't have production deadlines and can simply be left to get on with it. Trust is the key to that, when it stabs you in the back find out exactly what went wrong and fix it once.

You will "hog it out" as our American chums would say. Rapid removal of the excess material followed by fine finishing cuts. Only problem with that is chatter can leave vertical striations which echo in to the finishing cut. A speed change can help.

You have to experiment and find out what you can get away with on your machine, unfortunately there doesn't seem to be much rhyme or reason to it. You settle on a fine finishing cut, then one day it goes horribly wrong and ploughs in to the body of the work. An enormous cut that comes up with a fine finish and you wonder why you aren't cutting it all that way.

HSS tooling takes a sharper edge than carbide but doesn't last as long. If you want a fine finish don't fall in to the carbide trap. A centre cutting tool is a bad idea if you are trying to get a fine finish on a horizontal surface but is essential if you want to plunge.

Lubes. A bottle of Rocol RTD is a good start. A suds pump is better than a brush, especially if you want to run it unattended, but that probably precludes the milling vice because it will dump the run off on the floor. Incidentally, very tempting to buy an enormous milling vice but that's a big overhang. If using suds a wet and dry vacuum cleaner makes chip removal and clean down a doddle. I run the suds back in to the tank through a filter bag to protect the pump. Cast iron should be cut dry but I have recently had luck with paraffin as a lube, expect to find hard spots. Avoid cutting stainless, it is horrible.

Cutting downhill is a good idea unless you have backlash issues. If you cut an external counter-clockwise the tool will tend to bend in to the job, if you go clockwise it will shy away from it. The tool tip will always leave a line regardless, some people hone a tiny radius on the end to prevent that, I've never tried it, prefering to go full depth on the final pass. It all depends on what your machine will let you get away with.

I have probably rambled enough, but let's have one more... avoid tool changes unless you have a really sneaky method to reset the tool height with incredible precision.

Have fun

Robin

Edit: One more, cockpit checks. When you get in an aeroplane you go through a list, hatches and latches, tight and secure, brakes on, off, pressure exhausted etc. Before you set the milling job in motion have a check list. Particularly, are the chuck, drawbar and clamps on that round column stupidly tight. Do you have enough movement on the table to do the job? Will a G0 run the tool through a hold down fixture? The spanner Warco supply to clamp on to the round column isn't up to the job, suggest a socket or a box spanner, or extend the nuts so you can get a ring spanner on them. The Warco drawbar head is held on with a pin and is best replaced in toto.

Tweaky
17-02-2010, 03:00 AM
Hi Robin,
Wow, that’s probably more useful information than I have gleaned from memory and reading. Thanks

The overhang scenario is obvious, now you have pointed it out. I had assumed that if the extensions were solidly mounted and flex free, they would become the mount rather than a pivot.
You will probably laugh, when I tell you, that all my parts have been made using a humungous 6” tilt & swivel vice, bought on the premise that mass and solidity were king! …. Clamps from now on.
This leaves me with two problems now, my trunion table with a rotary table on top 170mm min., and the column, as the head will be that much higher.
Incidentally, the head clamping is by 3 large nuts (& bolts) which I tighten with a large ring spanner, so much so in fact that the column is free of grease (copperslip)
and won’t fall or crank, until I wind it up!

I have a fairly extensive range and quantity of HSS cutters (all imp.) so I am happy they will be useful, although I guess I will need a Tool cutter/sharpener in the future.
Are indexable cutters a good idea, I was thinking of buying a face cutter?

I had intended to use Castrol Carecut as my lube, because the heinous stuff we used in the 70s gave me dermatitis, but I have no idea how good it is as a cutting fluid, yet!

Well, lots to be getting on with, Z axis next to make, and I’m looking forward to applying this new found knowledge.
I do not have an angle plate large enough to mount the Z axis bearing block to bore them, so I am going to use the vertical rotary table (truniun mount to be) and find out how solid it is as a mount.

Many thanks for your time and for sharing this knowledge.

Regards
Mark

Robin Hewitt
17-02-2010, 11:10 AM
When designing the Z remember that you will want to hammer tooling out of the spindle taper.

You don't want to hammer against the Z screw so some way to release it is a good idea.

The quill lock is probably not enough because you will want to draw them in tight. I use a sledge hammer with a broken haft because it fetches them with one blow.

A release also lets you use the machine as a drill without writing G code, which is totally handy.

Robin

Edit: I don't use indexable tooling and I have never thought about resharpening cutters.
Face cutting accuracy and finish depends on how square the spindle is to the bed. Inaccuracies becoming more pronounced as the radius of the cutter increases.

irving2008
17-02-2010, 11:24 AM
So how do you do it Robin, I looked at the pics of your current Z and your previous Z I now have and i dont see how it is released?

Robin Hewitt
17-02-2010, 12:04 PM
So how do you do it Robin, I looked at the pics of your current Z and your previous Z I now have and i dont see how it is released?

The enclosure around the Z screw nut extends down to the honking plate and bolts on from below. To release it I remove the bolts.

The snag is the screw wants to be as close in to the quill as possible but the tin housing at the top of the machine overhangs, limiting the vertical space. I didn't want the screw to extend below the honking plate.

On the Mk 1 I had one 8mm bolt in the hole where the depth stop used to go. On the Mk 2, I used two 8mm bolts so I could extend the screw down to the bottom of the honking plate, passing between them.

It is a weak point because on the Mk 2 I cut threads in aluminium rather than steel which will wear and, eventually, strip. Hopefully I have cut them deep enough to see me out, one of the advantages of getting old :naughty:

irving2008
17-02-2010, 12:18 PM
So you remove the bolts for every tool change? i can see why you'd want to minimise tool changes then!

Robin Hewitt
17-02-2010, 01:41 PM
So you remove the bolts for every tool change? i can see why you'd want to minimise tool changes then!

I remove the bolts if I want to change chuck, not tool. Swapping a collet doesn't involve hammering. Changing from collet to drill chuck does, but I want to free it anyway if I'm going to drill.

Setting up a CNC mill is not what you could call 'a quick job' by any stretch of the imagination, bunging in a couple of bolts is not a big factor.

I have my own method, it goes something like this...


Glue a laser printed paper template to the workpiece using Pritt.
Saw off as much excess material as possible.
Drill mounting holes, using the template and the .dxf file that created the G code to get the X,Y right.
Remove all clamping and clean the machine down.
Change drill chuck for a collet chuck and fit a centre drill.
Drop the milling head down and lock it.
Bolt stand-offs to the table locating them with the centre drill.
Bolt the job to the precisely located stand-offs thus aligning it with where the mill thinks it is.
Drill a pip somewhere to get an alignment point I can refer back to if everything goes tits up.
Change the centre drill for a milling cutter.
Bolt the nut to the honking plate.
Move to a location that will be cut away.
Drop the cutter until it cuts through the paper.
Zero the Z axis.
Withdraw the tool and move closer to the start point.
Blow the suds return channels clear with the airline. (Best done dry).
Make sure everything is tight.
Get the suds in the right place, set the flow somewhere between "No flow" and "Splashing everywhere"
Main motor on.
Move to the computer and click on "Cut" then quickly move the mouse over the "Pause button".
Stand there, finger poised, until confidence wells up inside.
After about 30 seconds the suds has destroyed the Pritt so I can flick the paper template out.
Make tea, find cigarettes.


Edit: Forgot the splashguards, I use sheets of acrylic in the front and rear T slots. Probably forgot lots of other stuff but hey :whistling:

irving2008
17-02-2010, 02:27 PM
I see.... well I am learning something here too...

I have a set of MT3 collets rather than a collet chuck. I can see a collet chuck would be easier to switch tools with and could help avoid moving the head up n down. Have you used the collet chuck for drilling, and if not, why not? My 16mm drill chuck is huge.. I am thinking of getting a 10mm one to be of a similar size (vertically) to the boring head so avoiding head movements when switching between them. But of course you wouldnt use a boring head on CNC, you'd use a small mill on a circular cutting path instead, although deep bores might be tricky.

What happens if the part has no through holes to bolt it down with, or are these in the excess? Presumably the reason for the standoffs is to allow you to machine around the outside and inside in one operation (and to space it off the bed? - or do you use a sacrificial?)

Robin Hewitt
17-02-2010, 04:00 PM
Hi Irving

Don't 'learn' too much, this is just my way, room for improvement :heehee:

I use an ER32 chuck, much bigger overhang than you've got, but you need to hammer to change a tool. Swings and roundabouts.

I use a drill chuck because it is so much easier to change the bit and I don't end up scrunching a 5mm collet down to a 4.2mm point contact. That's got to hurt :eek:

Holding the work down is the big problem. I have other methods but that's my favorite.

You can clamp alternate ends from above but it makes the G code complicated. You can leave mounting lugs and saw them off afterwards. You might mill through holes first (if it has them), then run a big bolt down before removing the other clamps and milling the external. You might not be milling full depth and have excess that you can clamp sideways but space is limited if you don't want to compromise the splash guards. Water mess that makes :heehee:

Changing fixture part way through usually means cleaning out the tee slots with the work in situ and moving stand-offs. Not easy without losing X,Y or applying a tilt.

Don't use stand-offs and you hit the bed.

Cutting a sacrificial spacer for every job would be expensive.

Using scraps doesn't hold it true IMHO.

I cut my bolt down standoffs all from the same 3/4" thick bar and I don't face the top. That way I can make more and they are all very close to the same height. They are sort of pear shaped. The fat end just allows them to pass by each other in adjacent tee slots. The thin end, with the screw thread, is offset so they reach just past half the tee slot separation when turned 90 degrees.

I put the centre drill in deep when I drill for the thread so I get an annulus for alignment.

3/4" thick is pleasantly chunky, allows me to countersink an 8mm cap head flush to hold it down while still having some meat left to relieve most of the top so it stands proud around the threaded hole. The smaller sizes do get a bit chewed on occasion.

If you can come up with a better idea I am all ears :beer:

Robin

Tweaky
18-02-2010, 01:44 AM
Thanks Gents,
It will be a while before I can comment on much of this, but….

I had not included a release in my Z axis, which is similar to Robins (but not as pretty … no surprise there!), so you have saved me from myself, and a time consuming re-make.:eek: (two Hobgoblins).
I am using a smaller motor for the Z, as it will be geared and therefore should be approx. the same torque ( although slower).

Irving (Hi ), said you would not use a boring tool on a CNC, why not?

I don’t have one and have been cutting large holes using a mill and my rotary table, but I would have thought that having a fine a constant feed down would be beneficial, giving a more accurate (round) bore, with a better finish, would you enlighten me?
I remembered watching a program on TV, where a company building Drag Racers had received a part with a bore (about a foot across) cut in it, which was not round.
The chap commented that it must have been cut using a CNC machine (and therefore circular interpolation).
I am concerned that my bores for the Z bearings will not be very accurate, so I was intending to buy a boring bar.

I have just received my MT3 collets today, so haven’t tried them yet.
I use an Osborn Titanic collet chuck, which seems quite sturdy, so I will compare the cuts with the collets and see how much difference there is.

It seems that we have the wrong spindle for CNC, are we stuck with it? The R8 is presumably not any different in operation.
I guess the proper CNC machines have much larger quills, allowing the larger tapers to be used, is the lower bearing the limiting factor?

Cheers for now.
Mark

irving2008
18-02-2010, 10:56 AM
Thanks Gents,
It will be a while before I can comment on much of this, but….

Irving (Hi ), said you would not use a boring tool on a CNC, why not?


Well I've not got my mill CNC'd yet, but done a lot of boring (badly - my problems are getting the centre accurate after drilling because I need to move the mill head and not rushing it so as to end up on the diameter I want and not just over! - I've learnt that you need to take roughly the same amount off each time to allow you to estimate/compensate for the movement of the bar). The point about a boring bar is that you have to adjust it and creep up on the final diameter, so the CNCability of it is limited to just driving the vertical movement (ignoring the intial XY placement). When I said 'not on a CNC machine' I meant 'not under full automated XYZ control'. Of course you could use the z-jog or a bit of g-code to get a constant Z feed to a fixed depth and that would ease things a bit. I dont know why a milled bore would not be round assuming the final cuts were done light enough not to bend the tool/quill..

Robin Hewitt
18-02-2010, 01:23 PM
It seems that we have the wrong spindle for CNC, are we stuck with it? The R8 is presumably not any different in operation. I guess the proper CNC machines have much larger quills, allowing the larger tapers to be used, is the lower bearing the limiting factor?

You can get a Warco Major with an R8, it's an option. Now't wrong with MT3 though, apart from needing draw bars in 3/8", 1/2" and M12.

Production CNC requires an auto-tool changer so they prefer shorter fatter tapers.

Tweaky
22-02-2010, 12:45 AM
Auto tool changer would be nice though, complete the system, …..next time.:wink:

Cut my first metal using the X Y controlled by Mach. I faced the blocks for the Z axis, and cut shoulders on each.
What a difference, simply having DRO on the screen is a huge improvement, but coupled with little or no backlash (as opposed to half a turn on the X and a quarter of a turn on the Y) and a “go to” facility and constant federate it is a million miles from how it was.:dance:

I used a newly acquired ripper cutter, 0.5mm side cut, 12mm deep, feed rate 700mm/min. I like this too, fast removal of metal and small chips of swarf, I have not finished these surfaces yet, so have not encountered the “print through” that Robin talked about.

This was the first time I had used G-code as well, and that is not a daunting as I first thought (only simple stuff so far) but this is fun.
Back down to earth though, when I measured the blocks.
Not square by a long way.
I am still using my vice, as I don’t have many clamping bits yet, so I put a parallel in the vice with the DTI on the spindle, blimey, I can file squarer than that!:eek:

Before I started on the Z axis, I should have trued the column, so I’ll start there.

Set the DTI on the spindle, with a bar, so that I could measure the compass points, so about 3 ½” from the spindle.
Starting with West set to zero, South was +3 thou., East = +7 thou. and North = +5thou. So that is an error of 0.001” per inch! on the East West axis.
Four hours later, it was zeroed.
Cranked the head up and extended the quill to max, then repeated the tests, OMG.
The quill lock moves the spindle to one side by 0.005” (on a 10” extension from the spindle)?
Removed the quill lock, it is not even cut to fit the column, just an angle each side.
Re-ground the contact surfaces as they were a bit rough.
Now 0.003”, not good, but this must be wear in the quill housing, bugger.
Whilst doing this, I noticed that the head is actually quite well balanced, and if it still had the single phase motor, which is much heavier, I think it would have been near perfect.

Next, I put the vice back on, parallel in the vice and ….. still not square.
Stripped the vice, yuk, made in Poland, but exactly like Chinese machinery, a reasonable collection of bits, but completely un-fettled.
Re-assembled and back on the table it is better to operate, but still not square.
So the next step is to turn the vice over and machine the base, then maybe I can make my Z axis.
Hopefully my finish will have improved with things a little truer.

I have a day off tomorrow, so will hopefully get some actual progress on the machine.

Fursty Ferret time.:beer:

Tweaky
22-02-2010, 12:52 AM
Robin,
I have replied on your post "heads up Ebay", but I guess you are not very likley to read that.
Have you had any joy modifying your touch probe to incorporate the Z?
Also, how do you mount it?
Could it be rotated 90 deg. on it's mount?

Regards

Mark

Robin Hewitt
22-02-2010, 12:37 PM
I have replied on your post "heads up Ebay", but I guess you are not very likley to read that.

Hi Mark

It sends me an email if you post on one of my rants.

You are having teething troubles. When it misbehaves, figure out what went wrong, fix it, try again, find out if you were right. Nemo repente sapit :beer:

Robin

Tweaky
22-02-2010, 04:36 PM
Hi Robin,
of course it does, sorry, I am new at this.

Tweaky
22-02-2010, 04:49 PM
Fettled, filed and stoned my vice, did not skim the base (it was late and I was tired) because it swivels.:redface:

Tried my Z axis bearing blocks again, lots more G-code, swarf and finally ….. square, plus a much better finish.:smile:

Now for some more complex code, as I want angles on the outside and three different bore diameters.

“Apparatus est amo a infantia, totus Ego operor est nutritor is quod tarsus is”:wink:

Robin Hewitt
24-02-2010, 10:10 AM
My Latin isn't very good, "Gizmo's are the child's friend, thereby all work is food for the hands" ?

Nemo repente sapit, "Persevere and win".

Tweaky
25-02-2010, 12:54 AM
Hi Robin,
I have to confess, my Latin is virtually non-existent!, although it is something I would like to know.
I had to look up your phrase, it is the motto of a school in Oxford and means “no one becomes knowledgeable over-night “.
My phrase (I think) means “ the machine is like a child, all I do is feed it and clean it” :wink:

Have not cut much metal, as I am trying to plan my attack (so that I don’t make too many mistakes and have to re-make it), but hope to get more progress this weekend.

Regards

Tweaky

Robin Hewitt
25-02-2010, 01:33 AM
I had to look up your phrase, it is the motto of a school in Oxford and means “no one becomes knowledgeable over-night “.

I confess, I went to that ghastly dump, long closed, as did my father. I was taught latin by Mr Atkinson, an intimidating Victorian git we all called Snoop.

The actual school motto was Labor vincit omnia but we had the translation of both in the school song :rolleyes:

Andre
27-02-2010, 08:55 AM
Hello Tweaky,
That looks a very nice project I will certainly follow the progress.
Kind regards ,
Andre

Tweaky
28-02-2010, 10:14 PM
Hi Andre,
thanks for your kind words.
I must confess that after reading your posts, I found it hard to get some enthusiasm back for my rather inferior machine.
Your workshop must be the envy of many here, me at the top of the list.
I used to own a Bridgeport which was to be the basis of my project, but that’ a long and boring story.
Best of luck with your project, I’m sure it will be something to behold when finished.
(I really like your adjustable supports for clamping)

Regards

Tweaky

Tweaky
22-03-2010, 11:56 PM
Before starting on the Z axis, I needed to address the issue of positioning the work-piece accurately on the table.
I have to make an upper and lower bearing block for the Z axis ballscrew and cannot mount them together to machine them.
I watched ebay for a touch probe, but with digitising in mind, I decided to go the webcam route (also a lot cheaper).:heehee:
I bought a 1.3Mpixel camera and mounted it in a defunct waterproof cctv camera housing. This has two advantages, the cctv cam. has a C mount zoom lens with an iris control and is of course impervious to swarf and coolant etc.
I made an adjustable mount so that it can be aligned accurately.
After two failed attempts at making a magnetically coupled quick release, I ended up mounting directly to a 3 morse sleeve which I ground a groove into, so that the grub screws which adjust the x and y, also pull the mount to the sleeve.
I am very happy with this, not very pretty (as always!) :rolleyes:, but very solid.
I have checked its accuracy many times now and it is always spot-on.
I will be adding a laser at 45deg. to give an accurate height later, an idea I stole from a professional CNC camera costing $650 !
This will enable the camera to be used to measure distances, holes and angles.
Also in the future I intend to use some software to scan an object and create a point cloud for digitising.

Tweaky
23-03-2010, 12:37 AM
Ok so now I can get my work-piece accurately positioned, I made the upper and lower bearing blocks, which house the needle roller bearings for the rotating ballnut.

http://www.mycncuk.com/forums/asset.php?fid=1739&uid=601&d=1269300662

Ah!, having milled the bearing pockets, I checked the bores and was happy (0.01mm undersize), BUT, I measured the bore at the top and when I tried to fit the bearings.... nogo.:cry:
As Robin alluded to previously, the milling cutter shies away from the work, so my bores were both tapered! ..... bugger.
Spent about two hours with a 1" milling cutter in hand, scrapping the bore to make it parallel... what a pain!
I had used a 1/2" cutter, and ran the code 3 times on the final cut!?
I guess a different method is required next time.

The upper block also houses the two thrust roller bearings and this is closed by a plate, which is also the motor mount.
The bearing blocks mount into a 19mm plate which will attach to the front face of the head. They mount “into” the plate so that the ballscrew is as close to the spindle as possible.

Next I cut the plate, and for the first time clamped the work to the bed.
What a difference! (thanks Robin).

http://www.mycncuk.com/forums/asset.php?fid=1737&uid=601&d=1269300841http://www.mycncuk.com/forums/asset.php?fid=1740&uid=601&d=1269300847
The cuts where sooo much better, cleaner, faster and so much quieter.:smile:
And I still have more overhang to remove as I am still using a collet chuck (no metric tools yet for the 3 morse collets).
I was really happy for a few seconds, them I realised that this has severe consequences for my planned 4th & 5th axis, something I guess a lot of you already knew. :redface:
Maybe it will work OK with small cutters running at high speed, I will have to wait to find out.:eek:

Amazingly the blocks fit snugly and the shaft is in alignment!:surprised:
http://www.mycncuk.com/forums/asset.php?fid=1742&uid=601&d=1269300898

Next job is to modify a timing pulley to fit over the 37mm barrel of the ballnut, and then cut the depth stop lug off the front of the machine, and fit the assembly to the machine.

Cheers for now

Robin Hewitt
23-03-2010, 01:04 PM
I had used a 1/2" cutter, and ran the code 3 times on the final cut!?

If you want accuracy it is best not to buy your cutters purely on price :whistling:

Once they start to rub, rather than cut, you can repeat until Doomsday without removing the excess.

If the taper is more pronounced in X it's probably column flex. To flex in Y means bending the column, to flex in X the column acts more like a torsion spring.

Put a dti on it and see how hard you have to lean on the chuck to get problematical movement.

If the taper is even it could be a sloppy quill, does the dti reveal more flexing as the quill extends downwards?

Lots of lube helps keep tooling sharp.

Robin

Tweaky
24-03-2010, 12:41 AM
Hi Robin,

the cutter was a brand new Clarkson HSS 4 flute.
The column flexes as I have said previously, by 2-3 thou with a light pull on the head and 12-13 thou with a more substantial pull, at a height of 12” or so above the table.:thumbdown:

Yes, the quill is worse the more it is extended, about 3 thou at max extension and un-locked.
I am not using flood coolant yet as I have not made a tray to collect it, so I am squirting the lube by hand.
I guess I should have expected these results.

I must add that I have not always been using the collet chuck correctly, as I have discovered that the cutter must be screwed into the collet a fair way before it grips the end of the cutter.:redface:
I have only learned this after the problem with the tapered bores.

Maybe I should treat this as a training exercise and buy a better quality machine after I have learned how to use this one.
It is difficult for me to have down-graded as I have.
At my previous house, I had a large workshop and had just bought a Bridgeport B2J CNC with the motors removed and was in the process of refitting it, when the house next door burnt down, leaving our rented house un-safe.:cry:
So we had to move and now my workshop is 6` x 4`!
Oh well, se la vie.

Thanks again for your help.
Cheers
Tweaky

Robin Hewitt
24-03-2010, 09:28 AM
I must add that I have not always been using the collet chuck correctly, as I have discovered that the cutter must be screwed into the collet a fair way before it grips the end of the cutter.

You don't want to screw it in too far or the screw end can crack on the smaller sizes when you tighten. Assemble the collet without the cutter, back off about one quarter turn, screw the cutter in tight then tighten the nose.

It is a lot more picky on the diameter than ER collets but has better grip in the Z axis.

I got caught out with my first ER chuck, didn't realise you had to fit the collet in to the nose, cost me a collet :whistling:

Getting the right grease on the quill could remove some slop. If you got 10 micron accuracy at the top of the cut that is freaking amazing given the set up. Well worth persevering :eek:

I got moved out of the garage by her indoors moaning about the swarf getting walked in to the house. Luckily I own half of a small company so I was able to rent space up on the industrial estate as a business expense. I got 25 square meters for £3k pa, put in an extra floor and turned it in to 50 sq.m :smile:

Tweaky
01-04-2010, 12:45 AM
Thanks again Robin,
to say I got 10 micron accuracy would be overstating things somewhat, .... I crept up on it over the last 3 cuts!..... but it was round.:wink:
The swarf is a problem, I like the idea of a proper unit to keep the mess in, could go for a bigger machine, a power saw, grinder........mmmm.
I will look into that.
Must finish this one first though, more progress this Easter I hope.
Cheers:beer:

Tweaky
03-04-2010, 10:39 PM
Made a bit more progress over the last couple of days, finished the drive part of the z axis, and fitted it to the head.
Removed the depth stop from the head and mounted the assembly using the four mounting points from the name plate and an extra point on what is left of the depth stop bracket. All drilled and tapped M8.
Need to get the correct length belt and then sort out the quill plate.
Still haven’t decided how I am going to “quick release” the drive from the plate.
I might have to actually buy some metal (for the first time), to make the plate, as I am not very happy with my first one, having seen Robins honking plate.
I like to recycle things, so I will see if I can find a suitable part to modify.

More soon ……

Robin Hewitt
04-04-2010, 07:42 AM
How does that work then? Are you turning the nut?

Robin

Tweaky
06-04-2010, 01:23 AM
Yep, it is the internals from a flap control. Basically a screw jack with a 1605 ball screw and I think a double nut, roller bearing at the top, then two thrust rollers and a roller bearing at the bottom.
It came from a scrap metal metal chap, who deals in ex-MOD stuff ....... I think it was a fiver!
I ground off the beautiful 90deg. helical gear that was on the shaft and fitted a bored timing pulley. Shame, but I didn't want to have to sort out the end float on the drive shaft.
Fourtunaley, the screw is about 1/2" longer than the quill travel!

I am going to buy a piece of steel for the quill plate.

Cheers

Mark

swarf-boy
06-10-2010, 10:38 PM
Hi Tweaky,

I have just read this thread with great interest.

I too am contemplating the modification of my old warco major to cnc.

I am just wondring if you could bring us up to date with developments.

I have not seen any reference as to what motor drivers you ended up using with those NEMA 34 motors from Arc Euro Trade. Please could you enlighten me!

GeorgeD
07-10-2010, 01:39 AM
I have the same motors in that mod and the guy assured me these drivers would suffice for the motors....
160-020-00101 4.2A Bi-Polar Stepper Motor Driver

swarf-boy
07-10-2010, 04:26 PM
Tweaky,

May I also just ask, have you now completed this project ?
If so, do you still consider it a worthwhile project with regards to time and capital investment versus end usability, or would you advise otherwise?

Regards

Swarf Boy