PDA

View Full Version : CONVERSION: Sieg SX2 Plus build



russell
07-05-2012, 09:04 PM
My new mill arrived last week. TNT did a good job of getting it from Leicester to Montpellier (South of France) in 40 hours but it then took them 6 days to get it the last hundred miles. However, it arrived in good condition.
5956
Mill as received. I then stripped it down, washed everything in paraffin, and checked everything. The base was pretty good really but I blued up the surface plate and scraped it in:
5957
The Y slide fitted pretty well despite the visually rough machined surface so I just smoothed it off a little with a diamond lap leaving just enough texture (I hope) to hold some oil.
The only fault I've found so far is the gib strips:
5958
The photo shows a close up of one end of the X strip. The dimples are only partly formed and are such that any movement will give a wedging action and stiffen things up. I will deepen the dimples and put points on the adjusting screws (as recommended by Arc Euro) and see how they go. If they are still unsatisfactory I will then fit a dowel to prevent side to side movement.

The base is the same as the X1 rather than the X2 type. There is very little room to fit ball screws so I will make adjustable delrin nuts to fit the existing leadscrews - must try to find an M12 x 2 ACME tap.

I'm wondering what the best option is for the Z axis. There seem to be three options:
1. Lead screw at the side as with the CNC Fusion kit. Don't really like having the leadscrew so exposed.
2. Lead screw above the machine as in the Hoss conversion. However there is very little space between the motor and the column.
3. Lead screw behind column. This seems to be the neatest solution but I'm a little concerned that it may encourage the head to tend to tip and give jerky motion.
Any views on the best solution please.

All for now.

Russell.

rnr107
07-05-2012, 10:23 PM
I'm wondering what the best option is for the Z axis. There seem to be three options:
1. Lead screw at the side as with the CNC Fusion kit. Don't really like having the leadscrew so exposed.
2. Lead screw above the machine as in the Hoss conversion. However there is very little space between the motor and the column.
3. Lead screw behind column. This seems to be the neatest solution but I'm a little concerned that it may encourage the head to tend to tip and give jerky motion.
Any views on the best solution please.

Russell.

Hi Russell,
There is one more option you haven't listed which is better:
Remove the rack (cremaillere in french) and place the lead-screw in the slot between the head and the column. ;)

Laurent

russell
08-05-2012, 12:45 PM
Thanks for the suggestion Laurent, however the slot between the head and the column is just under 12 mm and even if the recess on the head is enlarged there isn't enough room for the leadscrew and nut. I'm leaning towards a modification of option 2 with the leadscrew (ballscrew) fixed to the top of the head between the motor and column and a rotating ballnut mounted high up so that it clears the motor in the up position of the head.

Russell.

rnr107
08-05-2012, 02:10 PM
It should works, just not as elegant as that:
http://www.thompsonmachine.net/mm2kit.htm

;)

russell
08-05-2012, 08:44 PM
Thanks for that link. I hadn't seen it. However that conversion is for an X2 mill. The SX2 Plus is somewhat different and has less room behind the motor.

Russell.

craigrobbo
09-05-2012, 03:15 PM
Good job, i wanna watch this as i wanted to convert mine.

russell
25-05-2012, 04:38 PM
O.K. It's two weeks since I started this so it is time for an update.

I've finished scraping in the X and Y slides and improving the gib strips. Both slides work nice and smoothly now without side to side movement.

I've fitted thrust bearings to the leadscrew supports and modified the ends of the leadscrews as shown:
6049

6050
As the leadscrew has an undercut next to the start of the thread I machined the tread away for 1 mm so that there is a good location for the thrust bearing. The thread on the end of the leadscrew was cut off and the remaining end reduced to 1/4 in to suit the flexible couplers. The end of the 8 mm part was threaded 5/16 x 32 for a nut to adjust the preload on the bearings. I'll fit a grub screw to lock the adjustment. Just waiting for an ACME tap to be delivered so I can make the acetal anti-backlash nuts. I had to reduce the width of the bosses on the mounts to 26 mm to get everything to fit.

I've also been drawing up a Z-axis arrangement and find that I can just fit the ballscrew behind the motor if I make a cut out in the plastic motor mount.

Russell.

rnr107
26-05-2012, 08:13 AM
6050

Nice machining....! How did you do that? Have you got an other mill? Lathe?



I've also been drawing up a Z-axis arrangement and find that I can just fit the ballscrew behind the motor if I make a cut out in the plastic motor mount.

Good! I knew it must be possible some how. This should looks good!

RNR

russell
26-05-2012, 07:54 PM
Nice machining....! How did you do that? Have you got an other mill? Lathe?
That was done on a sixty year old Atlas lathe with a homemade collet adaptor to ensure concentricity.



Good! I knew it must be possible some how. This should looks good!
I've just had another idea for mounting the Z ballscrew. Thread the end of the ballscrew M10 and drill and tap the top of the head casting to take it with a grub screw at the rear to lock it in place. It should all fit without any other mods. The ballnut has to be fixed above the motor in it's highest position so the assembly will be quite tall.

Russell.

RNR[/QUOTE]

russell
19-06-2012, 10:28 PM
Three weeks since my last update. I don't know where the time goes (well, yes I do - gardening, making new wheels for sliding patio doors, other projects).

Still, the M12 x 2 trapezoidal tap arrived from China last week so I was able to make a couple of anti-backlash acetal nuts for the leadscrews based loosely on the Brian Pitts design:
6170
Y leadscrew anti backlash nut.
6171
X leadscrew nut.
and finally (for now):
6172
Base assembly (almost) complete. Just have to fit some mounting brackets for the home/limit switches.

I can now adjust it to give me 0.5 thou backlash on the Y axis without it getting unduly stiff. However, the X axis gets stiff when the backlash is reduced below 2 thou so I have some investigation to do there.

Now to start on the Z axis. I've done some layout drawings and find it is just possible to get a 16 mm ballscrew to fit behind the motor on top of the head so that's the way I'll go. The ballnut has to be mounted rather high up though.

So - back to the workshop.

Russell.

rnr107
19-06-2012, 10:53 PM
Now to start on the Z axis. I've done some layout drawings and find it is just possible to get a 16 mm ballscrew to fit behind the motor on top of the head so that's the way I'll go. The ballnut has to be mounted rather high up though.


Hi Russell,
Can you post a drawing to show what you are trying to do? I don't understand why the Ballnut would have to be placed high up ?
Are you attaching the stepper on the ballscrew or are you fixing the screw and spinning the nut ?

Laurent

russell
20-06-2012, 11:14 AM
Can you post a drawing to show what you are trying to do? I don't understand why the Ballnut would have to be placed high up ?
Are you attaching the stepper on the ballscrew or are you fixing the screw and spinning the nut ?
Hi Laurent,
I am attaching the ballscrew to the head and spinning the nut, much as seen on the Hossmachine site here (http://hossmachine.info/cnc_conversion.html)However the SX2+ has very little space behind the motor and the ballnut has to be mounted above the motor height at the top of the Z travel. I will try to post a drawing later.

Russell.

rnr107
20-06-2012, 12:22 PM
Ha.... OK, all clear now! :-) I like it.... very nicely done...

russell
20-06-2012, 10:54 PM
O.K. some dxf drawing files follow:

Leadscrew_nuts.dxf (http://dl.dropbox.com/u/3521093/Leadscrew_nuts.dxf)
leadscrew support modified.dxf (http://dl.dropbox.com/u/3521093/leadscrew%20support%20modified.dxf)

and finally my proposed Z axis layout (provisional):
z-axis plan.dxf (http://dl.dropbox.com/u/3521093/z-axis%20plan.dxf)

To view these download them by right clicking and selecting save link as. Then open them in your favourite CAD program.

Russell.

russell
21-06-2012, 10:17 PM
Fixed the X axis backlash problem. Because of the restricted space for the X axis nut the flange on the acetal insert has to be trimmed rather close to the 5/8 in diameter. It was flexing a bit under load so I've backed it up with another plate the same as the one that gives the compression at the other end. I now have lass than 1 thou backlash on both axes with a torque requirement of about 40 oz.in.

I'll do the same to the Y axis nut to prevent problems under cutting loads.

Russell.

John S
21-06-2012, 11:11 PM
My SX2 isn't to hand at the moment, packed away whilst a workshop move is going on.

However looking at the design of the Z and Hoss's machine is there any reason that the Z screw cannot be mounted to the side in the conventional manner ?

rnr107
22-06-2012, 08:35 AM
However looking at the design of the Z and Hoss's machine is there any reason that the Z screw cannot be mounted to the side in the conventional manner ?

The only down side of having it on 1 side rather than in the center is that the screw will tend to pull on one side and therefore twist something...
and may not pull straight. You could fix that by having a linear bearing on a round bar on the other side?

Laurent

John S
22-06-2012, 09:55 AM
Laurent,
I can see your point but in the case of the SX2 where the nut would be very close to the gibs I don't think this will apply, there have been literally thousands of CNC mills built or converted of the Bridgeport type where the nut block is secured to the front of the quill which would have the same effect but in a front to back application instaed of a side to side.

In fact my big Beaver mill has exactly the same layout on the head and the Y movement is even worse as the ball screw hangs off one side, only the X has a central screw.6186

rnr107
22-06-2012, 10:23 AM
Hi John,
OK, I beleive you... ;-)
I like your lubricating system! Nice!

Laurent

russell
22-06-2012, 03:29 PM
However looking at the design of the Z and Hoss's machine is there any reason that the Z screw cannot be mounted to the side in the conventional manner ?
I guess you mean something like this (http://cncfusion.com/images/X2mounts/Z-axis/X2photos.html).

Obviously it can be done as many people have done it. However I have the same concerns as Laurent. I have read a number of threads where people have had to add a counterweight to avoid jerky Z movement and I don't want to get caught like that. It seems to be almost universal with the X1 where the leadscrew is behind the column and I guess it is due to the head nodding on the column. It strikes me that the closer the lift is to being in line with the C of G the less the problem will occur. Ideally I would like to mount the ballscrew further forward still but that would involve moving the motor out to the side.

You are the expert on these machines so I would appreciate your opinion. Perhaps it is not a problem if the gibs are properly adjusted.

A secondary consideration is the difficulty of keeping the ballscrew clean from swarf at the side. On top it is well clear of any flying chips.

Russell.

russell
01-07-2012, 09:55 PM
Still making some progress if a bit slow. I've now fettled the column dovetails (they weren't too bad really) and made the same mod to the gib screws as for the X and Y. It all works smoothly now.

Checked the alignment of the spindle to the column:
6218
.. found the spindle is parallel to thee column within 2 thou over 8 in. However the head is skewed giving 20 thou variation in 8 in when measured as above. I'll have to separate the two head castings and re-align.

Russell.

John S
01-07-2012, 10:29 PM
Russell,
That last picture, any chance of a shot of the column from the top looking down inside ?
Mine has the original wet noodle tilting column, only a couple made this way before we designed the fixed column. Unfortunately my fixed column is in customs somewhere.

Just wondering if the rack groove could be deepened ?

russell
02-07-2012, 11:47 AM
John,
Picture from top of column:
6219
Yes the groove could be deepened by up to 10mm bringing the thickness to the same as the rest of the walls. This would enable the ballscrew to move back but the motor still won't clear the pulley on top unless a smaller one is used (I currently have 40T and 20T on the motor).

BTW the column section on my file "z-axis plan.dxf", see my post about a week ago, is to scale.

Hmm... it could just fit with a 30T pulley. Would there be any danger of the column distorting if that much is removed? I don't know how well the column has been stress relieved.

Russell.

ian823478
02-07-2012, 10:29 PM
Checked the alignment of the spindle to the column:
. found the spindle is parallel to thee column within 2 thou over 8 in. However the head is skewed giving 20 thou variation in 8 in when measured as above. I'll have to separate the two head castings and re-align.

Russell.

Hi Russell
Thanks for sharing your build, I'm enjoying following it!
Have you had a chance to address the misalignment yet? I have the same problem but am living with it at the moment. Any experience you can pass on will be gratefully received!
For powering the z axis I'm a bit embarrassed to admit that so far I've just bunged the motor on instead of the handwheel, there's at least 1.5mm of backlash through the rack and pinion, but the repeatability is pretty good, I just have the clearance plane in cambam set to 3mm.
Looking forward to the next installment!
Ian

John S
02-07-2012, 10:36 PM
At least i will be able to see what I'm faced with, ARC reckon my column has landed at Leicester, need to pop down there anyway.
However my build will take a while as I have commercial work in the way but I often get a few minutes to work things out.

russell
03-07-2012, 10:47 AM
Have you had a chance to address the misalignment yet? I have the same problem but am living with it at the moment. Any experience you can pass on will be gratefully received!

Yes, not a real problem. I took the head off the column, loosened the four bolts that hold the two parts together but kept them tight enough so that the friction was enough for them not to slip in handling, then put it back on the column. Then tap the head with a hide hammer until the test bar showed no (or little) dial movement when moving the head along the column. It took half a dozen taps and tests to get it right. Took the head off, tightened the bolts and tested again. The misalignment is now just under 2 thou along 8 inches of Z movement.

Russell.

russell
03-07-2012, 10:53 AM
For powering the z axis I'm a bit embarrassed to admit that so far I've just bunged the motor on instead of the handwheel, there's at least 1.5mm of backlash through the rack and pinion, but the repeatability is pretty good, I just have the clearance plane in cambam set to 3mm.

Yes, I've seen that done. It depends what you're using the machine for but I would be worried about cutting forces moving the head up and down. How well do you find it works? What material are you cutting?

Russell.

ian823478
03-07-2012, 06:22 PM
Yes, I've seen that done. It depends what you're using the machine for but I would be worried about cutting forces moving the head up and down. How well do you find it works? What material are you cutting?

Russell.
Hi Russell. It seems to be accurate to at least 0.1mm as far as I can tell, probably nearer 0.05mm. I have a counterweight on the head and the gib reasonably tight.
I mainly cut aluminium but had a play with some 304SS over the weekend. Here's a couple of pics (if I can work out how to upload them.......)
62286229
Cutting the stainless was painfully slow (10mm/min 1mm DOC with a 2.5mm slot drill) but that was to prevent heat buildup (no coolant and just HSS bits). I've definitely got a lot to learn and loads of tools etc to save up for!
I'm still on the original leadscrews/nuts and have the DIYCNC system 4 with 3.1Nm steppers driving direct. (I won't get the max from these motors cos the system4 only gives 3A while the motors can take 4A.) I'll probably change this to a 2:1 reduction even if I don't eventually move to ballscrews.
I'm trying to convince myself I can make a bit of pocket money from it before spending much more!
Cheers
Ian

russell
04-07-2012, 10:12 PM
Why didn't I think of it before?

If I replace the 70T spindle drive belt with a 60T one the motor moves forwar by 25.5 mm giving enough room behind the motor for a much neater arrangement for driving the Z axis. I have to scrap the plastic belt cover and make a new motor mounting plate supported on pillars but that's no problem.

Revised Z layout CAD file here (http://dl.dropbox.com/u/3521093/z-axis%20plan2.dxf) ..still provisional of course. I've also changed the stepper motor drive belt from 65 to 75T to enable the motor to drop down behind the column.

Russell.

russell
19-07-2012, 09:14 PM
Time for another update:

I've drilled and tapped the column for the ballnut mounting and also at the back to lower the electronics enclosure to make room for the stepper motor.

Bolted the column in place and fitted the head to check the alignment of the column:
6395
The side to side alignment was good at less than 1thou over 8 in. However the front to back alignment was out by 5 thou over the width of the table. I corrected this by fitting shims ( a few layers of kitchen foil) at the front edge of the column mount. OK., I know I should have filed and scraped the mounting pads but fitting shims was easier.

The end of the ballscrew was turned down and threaded M10 and a mounting plate made:
6396
I had originally intended to tap the head to take the ballscrew but decided it would be easier to make slight positional adjustments with the mounting plate. The plate is tapped to take the screw and a brass grubscrew was fitted to lock the screw to prevent rotation.

I had originally intended to use an angular contact ball bearing in the ballnut assembly but couldn't find a suitable one so I used a normal bearing backed up with a needle roller thrust bearing like this:
ballnut_assy.dfx (http://dl.dropbox.com/u/3521093/ballnut_assy.dxf) The spacer is to ensure that the thrust from the thrust bearing is taken on the outer race of the ball bearing.
Note also that I have threaded the end of the bush and made an adjusting collar (again with a grubscrew to lock it) to allow the preload on the thrust bearings to be adjusted. This isn't very well detailed on the drawing but can be seen in the picture of the parts:
6397
Finally the parts assembled:
6398
Just have to make the support now.

All for now
Russell.

rnr107
19-07-2012, 09:34 PM
Nice work my friend! :-) You are getting there!

RNR

russell
06-08-2012, 09:10 PM
Getting there at last.

Made the support for the ballnut and motor for the z axis:
6519
Sorry, that's not a very clear picture. The side cheeks are rebated so that I could sit the mounting plate on top and slide it fore and aft to get the best alignment with the ballscrew (with the head in it's highest position). The assembly was then clamped and the position of the fixing screws was spotted through onto the sides of the mounting plates for accurate drilling and tapping. Side to side alignment is achieved by slotting the screw holes in the ballscrew bottom mounting plate slightly to allow adjustment.

The stepper motor is mounted on a 5 mm ali plate which can be slid back and forth on two arms at the back of the mounting plate to adjust the belt tension before being clamped in position.

I checked the backlash of the z axis and could not measure it on my dial indicator. It is certainly less than half a thou. However, leaning heavily on the head with it at the top of it's travel (with the gib tightened) gave me about 3 thou movement which must be due to the column flexing. If I'm going to want high accuracy I will have to keep the cuts light but then it is a light machine.

A replacement spindle motor mount was made to move the motor forward sufficiently to clear the ballnut and pulley assembly:
6524
The observant amongst you may notice the skinny drive belt. This wasn't part of the design. I ordered a 15 mm belt from the French RS Components and they sent a 9 mm wide one instead due to an error in their on-line catalogue. Rather than sending it back I am using it and doubt that I will overload it in use. The spindle drive electronics have been mounted lower down on the column to clear the z stepper motor and connected up. The spindle still works :smile:

Just the x and Y limit switches to fit and look for a cheap second hand computer. Then the learning will really start!

Russell

rnr107
07-08-2012, 07:44 AM
The braket looks good, nice and solid!
The big pulley on the nut, why is there no edge ? It would stop the belt from sliping. But should OK I guess.... :-)

Now, when are you making the first swart ? :D

russell
07-08-2012, 10:58 AM
The braket looks good, nice and solid!
Yes, I used some 12.25 mm dural plate I found cheap on Ebay. Wanted to make sure nothing would flex.

The big pulley on the nut, why is there no edge ? It would stop the belt from sliping. But should OK I guess.... :-)
Some of the pulleys I've seen with flanges have steel ones pressed on which would have made the machining difficult. I machined the pulley with a recess each side to locate the ballnut and the bearing and keep everything lined up. The pulley on the motor is one I rescued from a dead belt sander and has flanges both sides so that should keep the belt in place, if not I can always make a flange for the big pulley.

Now, when are you making the first swart ? :D
I hope I don't make any SWART :-) but I've given up predicting when things will be finished!

rgds.
Russell

rnr107
07-08-2012, 02:33 PM
I hope I don't make any SWART :-) but I've given up predicting when things will be finished!


Hummm.... yes, I meant swarf... :-)

russell
24-08-2012, 08:43 PM
Another two weeks gone - where does the time go?

It's been too hot here to do much in the workshop (at about 35C) but at last the conversion is now finished - well as much as anything is finished:

6655

The box under the monitor houses the drivers. I bought a 4 axis TB6560 board from Fleabay. Yes, I know, but they are cheap and seem to work OK if you don't want high speed. I've built a b****y great fan into the box and the board doesn't get hot at all. One fault I've noticed is that one of the motors gets hot when the axis is stationary while the others don't. The board includes a circuit to reduce the motor current when there is no drive and I guess it's not working on one driver. I'll investigate when I've time and can borrow an oscilloscope but in the meantime I can live with it.

I must build a pull out shelf under the bench for the keyboard rather than balancing it on my knees.

Yes, I have at last made swarf but only operating the machine from the Mach3 console. Not done any programming yet. I've installed Mach3 and EMC2 and need to spend some time to understand them. Any recommendations for other software I might need?

Off now to start reading up on Gcode!

Russell.

Sooty
18-02-2013, 09:44 AM
How are you getting on with the TB6560?

russell
18-02-2013, 04:17 PM
How are you getting on with the TB6560?

One channel makes the stepper motor a bit noisy but I'm not using that one at present. Otherwise fine - no failure yet. I,ll sort out the noisy channel when I get round to making a rotary table.

Russell.

jmarkou
18-03-2013, 05:24 PM
Hi there! that's some brilliant work there thanks for sharing!
Can you re host the dxf for the leadscrew nuts as the link seems broken, unless im missing some thing!

thanks again!

russell
18-03-2013, 06:42 PM
Sorry, I don't know what happened but for some reason a load of files have gone from my dropbox. Here they are again:
z-axis%20plan.dxf (http://dl.dropbox.com/u/3521093/z-axis%20plan.dxf)
Leadscrew_nuts.dxf (http://dl.dropbox.com/u/3521093/Leadscrew_nuts.dxf)
leadscrew%20support%20modified.dxf (http://dl.dropbox.com/u/3521093/leadscrew%20support%20modified.dxf)

Hope that works. Let me know if you have any problems.

Russell.

casperro
10-06-2015, 10:11 AM
Many Thanks for this Russell; I have one of these as a gift - long story. This is exactly what I need, but two questions: Roughly, what isthe Max height dimensions with the Z conversion fitted; and what stepper are you using to drive it.

Rob C

russell
22-06-2015, 03:58 PM
Total height with head in highest position is 1.07 m. I can't remember the model number of the stepper motors but they are NMEA23 size and about 80 mm long.

Russell.

casperro
23-06-2015, 09:45 AM
Thanks,

I have enough headroom and I have Nema 24 motors 425 oz-in; I have done the screws and fitted the thrust bearings. Also have the acetal and the the Acme tap is on it way from china (20 quid). Will post when I have completed the nuts and mounted the X/Y motors and tested.

RC

Dburt520
14-09-2015, 05:08 PM
Hi Russel,
I am getting ready to convert my Z and really like your design. I would like to make mine the same way.

Is there any chance that you can email me or upload the ball nut assembly dxf again? I clicked the link but it appears it is no longer valid.

Thank you, and again nice work setuo!
Dave

russell
19-10-2015, 07:59 PM
Hi Dave,
New link for ball nut assembly drawing: https://dl.dropboxusercontent.com/u/3521093/ballnut_assy.dxf
Not sure what happened with the old one.
Russell.