O.K., I stopped talking and placed an order with Harbor Freight for an X2. This will be my second Asian mill. I have an old Enco that will be very handy converting the X2 to CNC. In fact, the X2 base will fit nicely on the Enco's table. This should be fun!

Bill

My x2 arrived today via Yellow freight. The driver set it down on my driveway with his lift gate. The box is very stout. I was surprized to see that inside the box was a massive block of rigid styrofoam that completely enclosed the machine. Inside the foam box, the machine was totally bagged in heavy plastic. No red grease! The ground surfaces are all heavily oiled. Fantastic packing. Everything looks exceptionally nice. What a cute little machine! I'm impressed so far.

Bill

I dont know if you need this but I found this website lastnight and will be getting this dvd. 10 bucks doesnt seem like its that much for some knowledge. http://smartflix.com/store/video/912/Teardown-Tuning-Tramming-MiniMachines-101-volume-5 I am also order the rest of the series and I let you know how it goes.

I am working on an idea to eliminate the inherent weakness of the X2 mill: the pivoting column feature. Although this might be considered of value by some, I think it diminishes the little mill’s utility. Both the X2 and the Taig have this useless feature.

Several people have created good fixes that strengthen and stabilize the pivot. I like the use of a bracket or angle plate affixed to the rear of the column providing better support and rigidity. Turnbuckles would also do a good job.

However, I am considering something different. I am thinking of building a wooden work station/table with a tall “back splash” that is built with great strength and rigidity. I would then add a sheet metal flange to the rear of the column, and bolt the column to the table’s back splash. This should make it column very stable, make tramming a cinch, and resist twisting and deflection.

Has anyone tried this?

Bill

See the sketch of my idea:

Here is another alternative http://warhammer.mcc.virginia.edu/ty/7x10/vault/Mills/G8689-MiniMill/Projects-Mods/BaseBrace/

Using a wooden structure to stiffen a machine tool is a bad idea. Wood by it's nature is flexible and is not dimensionally stable over time. A better approach would be to weld up a massive steel fabrication.

Phil

I am working on an idea to eliminate the inherent weakness of the X2 mill: the pivoting column feature. Although this might be considered of value by some, I think it diminishes the little mill’s utility. Both the X2 and the Taig have this useless feature.

Several people have created good fixes that strengthen and stabilize the pivot. I like the use of a bracket or angle plate affixed to the rear of the column providing better support and rigidity. Turnbuckles would also do a good job.

However, I am considering something different. I am thinking of building a wooden work station/table with a tall “back splash” that is built with great strength and rigidity. I would then add a sheet metal flange to the rear of the column, and bolt the column to the table’s back splash. This should make it column very stable, make tramming a cinch, and resist twisting and deflection.

Has anyone tried this?

Bill

The GrizHFMiniMill Yahoo (http://groups.yahoo.com/group/GrizHFMinimill/messages) forum has a few discussions going on right now about column flex and stiffness.
The Photos section also has several solutions people have tried.
Very informative.
This pic shows how I beefed mine up.
Hoss

In terms of using the wood table to stabilize the machine, if done properly, wood can be extremely stiff. Remember, one of the highest performance fighters of WWII, the Mosquito, was made of plywood! Also, there are whole sections of this site dedicated to making CNC machines from wood. The wooden structure would have to be a "stressed skin" design in order to perform up to expectation, but I still think the concept has promise.

Bill

I say go for it. Plywood is cheap and not hard to come by. Take measurements of before and after to compare the numbers and see if the project is worth it. I know I cannot do it down here(florida) the humidity is killer and all the tools live out in the garage. Something else to consider is making a form out of plywood and pouring concrete. If done properly you could absorb alot of vibration and make it pretty stiff. Use the 5000psi or better. Also it sticks to the not very expensive/easy to find materials. Goodluck with everything

I wonder if anyone can provide the necessary "footprint" for a CNC Fusion converted X2. I am ordering the kit for my little mill, but it would be nice to be able to get started immediately on appropriate bench. I suppose that I could just add 6-8" of additional room on each side of the X?

I wonder if cement board might be good?

http://www.nationalgypsum.com/literature/gypsumconstructionguide/PermaBase.pdf

Ya know Regnar posted a perfect example, that first pic with the angle plate mounted to that pivot bolt is perfect. Besides what do you plan to mill that the machine needs such rigidness that you seek? I think a lot of newbies(including myself at one time) expect those mini mills to do some serious metal hogging. The KISS method is the way to go(Keep It Simple Stupid). Besides you,ll spend so much time trying to make the machine rigid and precise that you'll forget what the hell you were going to mill!!!

Ya know Regnar posted a perfect example, that first pic with the angle plate mounted to that pivot bolt is perfect. Besides what do you plan to mill that the machine needs such rigidness that you seek? I think a lot of newbies(including myself at one time) expect those mini mills to do some serious metal hogging. The KISS method is the way to go(Keep It Simple Stupid). Besides you,ll spend so much time trying to make the machine rigid and precise that you'll forget what the hell you were going to mill!!!

I will be using the new mini mill for miniature CNC projects in aluminum, plastic and wax. Big stuff goes on my 700 lb. 2hp Enco mill-drill.

I'm seeking as rigid an Z axis as possible given the size and weight of the X2. The pivot point at the base of the X2 spoils an otherwise fine design. The long lever arm of the column, secured only at the lower end, makes for a very flexible machine.

I have examined the angle plate fix, it is very well done. Although I'm sure it is a huge improvement, it doesn't fix the underlying problem. The column ideally needs to be completely immobilized along it's entire length. I do not need or want to ever pivot the column.

Since I must build a work bench for this machine anyway, I want to see if I can make the bench work with the machine and improve the it's functionality.

I know cement board flexes, I just redid my bathroom with this stuff. I was thinking more along the lines of a concrete counter top. Make it a few inches thick and reinforce with rebar and chicken wire. You could even use epoxy cement although that starts to get pricey. I am still working on the ideal for an enclosure for mine but I have a feeling I will be make it out of concrete,80/20 and plexi glass. I will be ordering the fusion kit next month so I would be able to give you the numbers then but you could just email fusion Im willing to bet he has it memorized.

I think you may be cofusing strength with stiffness. The design criteria for an aircraft is not the same as for a machine tool. Aircraft are required to be strong and light, for which wood and aluminium provide a reasonable compromise, machine tools are required to be stiff and massive, for which cast iron and steel provide a good compromise.

Can you build a bridge out of chocolate, yes. Would you build a bridge out of chocolate.....

Phil:)

Remember, one of the highest performance fighters of WWII, the Mosquito, was made of plywood!
Bill

The structure that I am describing would use a stressed skin panel, or torsion box. These are extremely strong, stiff designs that resist torque. They are variations on the monocoque designs used in aircraft design like the plywood airframe of the Mosquito. Additionally there was a successful sports /racing car, the Marcos, that used a plywood chassis. Cast iron is best, but I can't build a bench from cast iron.

Here is some pictures. This guy built his with the mini mill and a skill saw. I wish I could give you the 23 pages of thread he wrote up about this and other little things he did to the mill but its part of a pay for site.

Here is some pictures. This guy built his with the mini mill and a skill saw. I wish I could give you the 23 pages of thread he wrote up about this and other little things he did to the mill but its part of a pay for site.

Wow, now that is a nice piece of work! An elegant solution to the problem.

Here is some pictures. This guy built his with the mini mill and a skill saw. I wish I could give you the 23 pages of thread he wrote up about this and other little things he did to the mill but its part of a pay for site.

Regnar: what site would that be? and is paying to see worth it?

Thanks,
Java77man

Its a site that is dedicated to machining your own firearms. For me 30 dollars a year is well worth the wealth of info the site gives me. If you have no intrest in firearms I am willing to bet you would find the site not beneficial. Here is the link if you are interested. http://www.homegunsmith.com/cgi-bin/ib3/ikonboard.cgi?

You will notice that you are able to look at some of the forum but once you buy your password the main page doubles in size.

I will ask the author if he wouldnt mind posting this over here.

I am preparing to order all of the componemts for my CNC conversion. Here are my selectons:
1) The CNC Fusion 3 axis ballscrew kit with preloaded ballscrews.
2) The complete Keling 3 axis package for the X2 using 425 oz. in. stepping motors, drivers, breakout board, power source, etc.
3) The Keling MPG2 pendant and the C22 adaptor card.
4) Mach 3 software

Does this look correct to the experts in this group? Any advice before I start spending?

Thanks,

Bill

I think you can get by with the 282 oz motors seeing some are using 269ozers But for only another 30 dollars to bump up to 425 it is probably worth it.

I am pretty sure I have seen a pendant for less. I will have to go threw my favorites and find the link for you. I just hope I saved it. Another thing is that you could make your own pendant and use it with key grabber. Its nothing that you need to start off with. But it does look like a nice system just need a second printer port. Cant do that on my laptop so I need to go the other route.

Does the "Pendant Interface Card" do the job of a second printer port? I'm a commited Mac user and i've never owned a PC, so I'm a little dumb on these issues. BTW, what is the best/ most popular PC option for running a CNC set up? I'm going to have to buy one soon.

Bill

Here is the interface card for the pendant:

You might want to go here and read this description. http://www.cnc4pc.com/Store/osc/product_info.php?cPath=40&products_id=160 It also show the c22 setup and it uses a 2nd parellel port. I am not for sure on this but I think the c22 just makes it an easy setup feature so that you dont have a bunch of wires going into the pendant. Sorta a plug and play. It might even opto-isolate

If you limited by only being able to have one parellel port you can use a usb or keyboard socket and make your own or you might even be able to use thier pendant. This item has has a choice between 28 or 56 inputs. Plenty to cover limits all the buttons you need, a joystick, let your imagination be the judge. Here is a link the the Ipac http://www.ultimarc.com/ipac1.html

Here is a link to what could be done http://www.flickr.com/photos/11132086@N02/1051114611/sizes/l/ http://www.flickr.com/photos/11132086@N02/1209027756/in/photostream/ http://www.flickr.com/photos/11132086@N02/1314455704/in/photostream/

Here we go.

The mill base is up on the table of my old Enco getting prepped for the attachment of a rear bracket to support the column. The back of the base was not flat and square, so I milled the back off using the dovetails as the registration surface. Now the base is square to the the movable axis’s.

I also included a couple of pictures of how the X2 was packed. Pretty impressive job. It arrived in perfect shape.

In the course of breaking it down for conversion, it was notable that none of the fasteners were properly tightened. Every single bolt as loose. Great for easy disassembly, but not great for proper machine adjustment and operation.

A bit more progress.

The first photo show the mini mill base up on the table of the Enco in preparation for drilling holes that will be used to attach the support plate for the column. The old Enco has a lot of Z. Sometimes bigger is better! The support plate will be a piece of 6” wide aluminum channel. The channel has 2” “legs” and is about .420 thick. It should provide the stability I am seeking.

The second shot is a close up of the collet chuck holding a pilot drill. I will be using an identical R8 collet chuck in the mini mill too. Next I will drill to the proper size for the tap.

Here is the finished product with six 5/16 X 18 holes tapped.

The machine work continues. In these pictures, I am doing the modification suggested by Hoss: increasing the Y travel. The opening that the Y axis ballscrew works in is now lengthened on both ends. BTW, the gray cast iron used on these mini mills machines very nicely.

Bill

I received by CNC fusion kit yesterday. Because I am going to try and get the maximum possible travel on the Y axis, I decided that I had better take a precaution to prevent running off the end of the ball screw and spilling the ball bearings. So I drilled and tapped the end of the ball screw and installed an aluminum disk as a "keeper". My plan is to extend the stock X2 "Y" travel from the OEM 3.7" to 4.5". Here are a couple of photos. These CNC Fusion ball screws are really good looking pieces.

At long last, my column brace assembly is completed.

It consists on a piece of aluminum channel,( 6” wide, 2” deep, 12” tall and .437” thick), a 3/8” thick spacer (the piece with the curve), and 12- 5/16x18 cap head bolts. The channel bolts to the bottom edge of the machine base with the spacer inserted between the channel and the machine base.

The cast iron machine column was drilled and tapped then is bolted to the channel with 6 more bolts. The holes in the channel for the column bolts are drilled oversize to allow the column to still be able to pivot a couple of degrees to each side allowing adjustability during tramming. Once trammed, the bolts are all torqued down and it should really stay accurate unless the machine is moved to a new location. I think this is a pretty good solution.

Your thoughts?

Bill

A piece of steel channel of the same dimensions would have been 3 times stiffer. A piece of box section would have been several orders of magnitude stiffer again. Fixing the bottom of the channel to a common, thick base plate would also have help significantly. It's nicely implemented but sorry, technically it's unsound.

Phil


At long last, my column brace assembly is completed.

It consists on a piece of aluminum channel,( 6” wide, 2” deep, 12” tall and .437” thick), a 3/8” thick spacer (the piece with the curve), and 12- 5/16x18 cap head bolts. The channel bolts to the bottom edge of the machine base with the spacer inserted between the channel and the machine base.

The cast iron machine column was drilled and tapped then is bolted to the channel with 6 more bolts. The holes in the channel for the column bolts are drilled oversize to allow the column to still be able to pivot a couple of degrees to each side allowing adjustability during tramming. Once trammed, the bolts are all torqued down and it should really stay accurate unless the machine is moved to a new location. I think this is a pretty good solution.

Your thoughts?

Bill

This is structural 6061-T6 aluminum ASTM B308,Federal Spec QQ-A-200/8/16. It is worth noting that lots of aluminum is used in tabletop milling machines. Check out the Sherline and Taig.

Certainly using steel would yield greater stiffness for any comparable sectional profile, but this is a massive hunk of aluminum and is very stiff. It provides excellent support for the rear of the large pivot bolt, a major weakness of the original X2 design. Also, it means that the pivot is not the only attachment for the column.

Could I have designed an even stiffer support? Yes, but I believe that this is a huge improvement over no column support :)



A piece of steel channel of the same dimensions would have been 3 times stiffer. A piece of box section would have been several orders of magnitude stiffer again. Fixing the bottom of the channel to a common, thick base plate would also have help significantly. It's nicely implemented but sorry, technically it's unsound.

Phil

All aluminium alloys have effectively the same modulus of elasticity. In fact 6061 is slightly worse than most. So the fact that it is structural 6061-T6 aluminium ASTM B308, Federal Spec QQ-A-200/8/16 is irrelevant. You are confusing strength with stiffness. Massive it might be but the mass is in the wrong place. More than 2/3 rds of the mass is as close to the neutral axis as you could possibly get it. For maximum stiffness it should be as far from the neutral axis as is possible. Also, as somebody pointed out to me on another forum, open section beams are not particularly stiff in torsion.

This is why square column mills are made with, well ... square columns.

Will it add stiffness to the column, yes, is it as good as it could have been, no.

Sorry Bill but you did ask.:)

Phil

This is structural 6061-T6 aluminum ASTM B308,Federal Spec QQ-A-200/8/16.

Yep, I did ask. No need to be sorry. BTW philbur, are you an engineer?

My employer think so.

Phil

BTW philbur, are you an engineer?

mechanical?

I did a test on the machine to evaluate deflection of the column with and without the brace. I placed a indicator 6" up the face of the column with the magnetic base on the saddle way. I used a digital fishing scale to apply 20 pounds of force straight forward along the Y axis at the top of the column. Without the brace, the indicator showed .001 deflection. With the brace, the deflection was .0005. Not too bad.

I haven't as yet devised a test for torsion of the column. Probably the most significant benefit of this modification will be the ability the easily and accurately adjust the X axis when tramming.

In order to achieve 4.5“ of travel on the Y, it was necessary to mill off part of the front edge of the CNC Fusion motor/bearing mount. These pictures show the milling in progress, the finished mounting block, and the block mounted on the X2. Notice how the saddle can move forward much farther, overhanging the front edge of the base. Next step, limit switches.

.001? I put my Z as high as it would go and put a dial indicator from the spindle referenced against an angle plate on the bed and using a modest amount of effort pulling on the head, I can get up to .005 deflection. I can see deflection of .0005 just adjusting the Z height with the handwheel. I have replaced the mounting bolts with hardened bolts and have torqued them all equally - I guess what I am saying is that I'm suprised you are only getting .001 without the brace - of course maybe it is because you only tested 6" up? Do you have any idea what you torqued the mounting bolts to?

I did a test on the machine to evaluate deflection of the column with and without the brace. I placed a indicator 6" up the face of the column with the magnetic base on the saddle way. I used a digital fishing scale to apply 20 pounds of force straight forward along the Y axis at the top of the column. Without the brace, the indicator showed .001 deflection. With the brace, the deflection was .0005. Not too bad.

I haven't as yet devised a test for torsion of the column. Probably the most significant benefit of this modification will be the ability the easily and accurately adjust the X axis when tramming.

I am sure that the deflection at the top of the column would be .005 or more, but I haven't figured out a way to mount my indicator higher up.

I also don't have allen sockets that will work on my torque wrench, I leaned hard on the allen key, but the handle is very short; maybe 30-40 lb ft?

.001? I put my Z as high as it would go and put a dial indicator from the spindle referenced against an angle plate on the bed and using a modest amount of effort pulling on the head, I can get up to .005 deflection. I can see deflection of .0005 just adjusting the Z height with the handwheel. I have replaced the mounting bolts with hardened bolts and have torqued them all equally - I guess what I am saying is that I'm suprised you are only getting .001 without the brace - of course maybe it is because you only tested 6" up? Do you have any idea what you torqued the mounting bolts to?

I redid the test. First, I moved the indicator up to 17.5” above the table. I clamped the mill base to the table with a huge C clamp.

Without the brace, placing the palm of my hand against the top of the column and pushing HARD, the column deflected .008. In fact, the column would deflect .002 with thumb pressure.

With the brace, pushing equally hard, the column deflected .0015.

Certainly not a scientific test, but highly repeatable. Conclusion, A brace, even of flawed design and inappropriate material selection, is highly effective in reducing Y axis flex.



.001? I put my Z as high as it would go and put a dial indicator from the spindle referenced against an angle plate on the bed and using a modest amount of effort pulling on the head, I can get up to .005 deflection. I can see deflection of .0005 just adjusting the Z height with the handwheel. I have replaced the mounting bolts with hardened bolts and have torqued them all equally - I guess what I am saying is that I'm suprised you are only getting .001 without the brace - of course maybe it is because you only tested 6" up? Do you have any idea what you torqued the mounting bolts to?

Nothing wrong with your design or material.
You fixed the main problem with column flex on the X2,
no support between the rear of the column and the base.
Good job.
I had a similar amount of deflection when I tested mine.
Made a similar fix with a plate bolting the column and the base together.
Makes a big difference from a simple solution.

As long as you are happy Bill that's the main thing. Certainly seems to make a marked improvement. I think the main improvement probably comes from support for the cantilever at the swivel/Base. A thick steel plate with a curve slot would allow you to still swivel the column and retain the siffening at the same time, or alternatively you could rely on only the swivel bolt at the bottom end.

Phil

I redid the test. First, I moved the indicator up to 17.5” above the table. I clamped the mill base to the table with a huge C clamp.

Without the brace, placing the palm of my hand against the top of the column and pushing HARD, the column deflected .008. In fact, the column would deflect .002 with thumb pressure.

With the brace, pushing equally hard, the column deflected .0015.

Certainly not a scientific test, but highly repeatable. Conclusion, A brace, even of flawed design and inappropriate material selection, is highly effective in reducing Y axis flex.

Here is my installation of the Y axis limit switches. I attached the micro switches to an aluminum angle with two 4-40 button head screws. The screws were easily tapped into the pre existing holes in the switch body.

The angle is screwed to the underside of the base. All attachments on the angle are slotted to allow adjustablity. The two switches will be wired in series in the closed circuit position. When the the striker bar trips either switch, the circuit will open.

I intend to have all the limit switches mounted in “internal” positions where they cannot be bumped, damaged, or contaminated with swarf.

Bill

Nice I like the Ideal. Can you post a few pictures of the underside. Just wondering how you attached the angle. Thanks.

Here are a few additional photos.


Nice I like the Ideal. Can you post a few pictures of the underside. Just wondering how you attached the angle. Thanks.

Like the looks of the Y so far. I am curious as to your z stiffener solution. You or someone else posted a picture of the same mill with a large triangular plate with a bottom plate that went under the mill and the whole thing bolted to the z axis. That looks SUPER rigid, is there a reason you did not use that method? You obviously have a mill ready to do the heavy machining. Not bagging on your method but the large triangle looked to be over the top.... I really like the smaller x series of mills, they have a quality look to them but there is a lot of this type of flex in the machines and they are basically not all that heavy to begin with. If I had one and intended to go full bore cnc with it I would probably have bolted it to a freakin' I beam or something to stiffen her up. The Hoss website features a X machine that is just about as modded as I have ever seen a milling machine and much of it was to stiffen and strengthen the basic design. The toolchanger is a really cool idea as well and if I ever get my machine up and running I may actually try to reproduce that thing. Anyways, That is a very nice mill and I look forward to reading further about your adventures with it. peace man....

Pete

Thanks Pete,
I had several goals when I created my Z brace: 1) it had to be effective in addressing the well know flex of the X2 mill. 2) it had to be a simple design that didn’t interfere with the practical operation of the mill, and 3) it had to look neat and professional.

I sketched out a number of Z brace concepts that met objectives one and two, but would have been hard to build and/or looked ungainly and cobbled together. The rejected designs involved angled braces of welded tubing, turnbuckles, etc. These designs were overly complex and would have looked excessively “Rube Goldberg”. The final Z brace is effective, simple, adjustable, easily built, and relatively unobtrusive in appearance.



Like the looks of the Y so far. I am curious as to your z stiffener solution. You or someone else posted a picture of the same mill with a large triangular plate with a bottom plate that went under the mill and the whole thing bolted to the z axis. That looks SUPER rigid, is there a reason you did not use that method? You obviously have a mill ready to do the heavy machining. Not bagging on your method but the large triangle looked to be over the top.... I really like the smaller x series of mills, they have a quality look to them but there is a lot of this type of flex in the machines and they are basically not all that heavy to begin with. If I had one and intended to go full bore cnc with it I would probably have bolted it to a freakin' I beam or something to stiffen her up. The Hoss website features a X machine that is just about as modded as I have ever seen a milling machine and much of it was to stiffen and strengthen the basic design. The toolchanger is a really cool idea as well and if I ever get my machine up and running I may actually try to reproduce that thing. Anyways, That is a very nice mill and I look forward to reading further about your adventures with it. peace man....

Pete

I thought this might interest you guys. I have made a way cover bellows to use on the X2. It is made from PVC vinyl sold by Home Depot in the outdoor area as an pond liner. I made a template from aluminum and cut out identical pieces. The are glued together using cyanoacrylate glue (super glue).

A half inch thick piece, (first picture) expands to over six inches (second picture) and snaps back closed when released.

This is my first attempt and there are a few flaws, but it really seems to work great. Also, any size, shape and length is possible. The cost? less than $10 for materials.

Bill

Have you applied a repetitive torture test yet? :p

I like that, might have to try it out. PVC Pond Liner you say...

Here are some detail photos of the trigger for the Y axis limit switches.

Photo 1 shows the milled grooves in the underside of the saddle that allow clearance for the switches and the recess for the trigger. The small hole is tapped 4-40.

Photo 2 shows the trigger screwed into place in the recess.

Photo 3 shows the trigger sitting on the machine base in the position that it will take when it is ready to trigger the limit switch. The saddle is flipped up on it’s edge showing the attachment recess.

Photo 4 is a zoom in on the trigger and the switch.

As long as everything is torn apart, I decided to investigate the spindle bearings. I was able to get them out using a puller. The verdict: completely unmarked except for the bearing numbers. If you are unwilling to put your name on the product, it's a pretty good bet that they aren't top quality. So, I'm going to replace them with good Japanese or equivalent bearings. Has anyone installed new bearings? I am thinking freezing/heating shrink method, but your suggestions/advice would be much appreciated.

The picture: The stripped head, old bearings and spindle. In the foreground is an Erickson DA 180 collet chuck that fits the R8 spindle. I have three of these already in use, on my 9X20 lathe, on my mill/drill, and in my my rotary table. They work great holding endmills, and holding stock in the lathe.

The new SKF bearings are installed at last. These are ABEC 3 bearings, not super quality, but much superior to the OEM no-name Chinese units.

The results are very good. The new bearings are rated for 10,000 rpm and have significantly less runout. The spindle runout was .0005 with the original bearings and is now less than .0001(smaller than I cam measure). All for an afternoon of work and about $40.

I have a belt drive conversion that I will be installing. In preparation, I removed all of the plastic gears, counter shaft assembly, etc., from inside the head. There is now only an aluminum spacer where all the drive parts were.

I wonder if anyone has tried to increase the speed with the belt conversion beyond the 4500 RPMs that the kit provides? It would be pretty easy to make a larger pulley for the motor that would provide the potential for 10,000 RPM. Has anyone done this? Does the motor have sufficient power to handle such a strain?

I wonder if anyone has tried to increase the speed with the belt conversion beyond the 4500 RPMs that the kit provides? It would be pretty easy to make a larger pulley for the motor that would provide the potential for 10,000 RPM. Has anyone done this? Does the motor have sufficient power to handle such a strain?

I have mine spinning at 6100 RPM with the stock bearings.
Only noticed the head getting warm to the touch after a long run.
I drew up an even larger pulley for the motor that would top out at 10,500 RPM.
Haven't made it yet though. Maybe someday.
Hoss
here's a video of the upgrade.
Belt Conversion Overdrive

I learned the trick for perfect running, non-binding ballscrews: exact measurement. In these photos I am measuring the “droop” of the Y axis lead screw shaft, front to rear. Using a depth micrometer, the screw must be precisely aligned with the top of the base and the dovetails. In my case it was necessary to insert an .008 shim at the top of the bearing carrier to raise the unsupported end of the screw. Shims can be placed at the top, bottom, and sides of the bearing carrier allow you to achieve exact alignment.

How do you know when everything is perfect? If the ballscrew binds when the ball screw carrier screw is tightened, the shaft needs improved alignment.

Bill

Hi Bill,

That's a useful way to check vertical alignment but I think you need to measure off the base of the ways not the top of the base, they might not be in the same plane. Also horizontal alignment is equally important and that's a little more tricky to measure as the "vertical" ways are of course angled. A couple of lengths of ground bar clamped in "Vee" on each side would get you started. The rest depends on what measuring equipment you have available.

Just some thoughts.:)
Phil


I learned the trick for perfect running, non-binding ballscrews: exact measurement. In these photos I am measuring the “droop” of the Y axis lead screw shaft, front to rear. Using a depth micrometer, the screw must be precisely aligned with the top of the base and the dovetails. In my case it was necessary to insert an .008 shim at the top of the bearing carrier to raise the unsupported end of the screw. Shims can be placed at the top, bottom, and sides of the bearing carrier allow you to achieve exact alignment.

How do you know when everything is perfect? If the ballscrew binds when the ball screw carrier screw is tightened, the shaft needs improved alignment.

Bill

Here is the first test of my improved motor drive for the spindle. New bearings, machined pulleys for the motor and spindle, and mounting plates. The motor is a 4,000 RPM, 700 watt Electro-Craft brushed servo motor ($20 on Ebay). The speed controller is a Leeson Speedmaster ($18 on Ebay). With the combination of pulleys, I should achieve approximately 10,000 RPM on the spindle. The spindle runs extremely quiet and vibration free. I need a tach to determine the actual top speed.

Bill

bilinghm, did you modify your servo motor to drive your milling machine's spindle? I have a small servo I would like to use to drive a small lathe...can you help me? I don't know a lot about servos. here are the pics

I am no expert on this matter; I'm sure others can offer better advice, but based upon my experience, it may be possible. In my case, no modifications were necessary to the motor.

To evaluate this conversion, first you need to find the specifications of your servo motor to determine if it will be appropriate for your intended use. The motor I used is a 90 volt, 7.7 amp motor, (693 watts, approximately 3/4 HP) and 4,000 rpm maximum speed. This was about double the power of the OEM motor. Does your motor have the power and speed you need? If so, you must next see if can you obtain a speed controller that will drive the motor. I was able to use an off the self Leeson DC drive, However, your servo seems to have a tachometer, so you may need a more elaborate drive with a tach feedback feature. It certainly may work, you just need to do some research.

Hopefully others will share their experiences.

Bill

Thanks Bill, it's more help then I have gotten...

I received my Ebay laser tach today and did some testing. Yikes, not even close to 10,000! My max speed is currently only 6,500 RPM. However, the motor is only turning 2950 RPM rather than the specified max RPM of 4,000. But it is obvious that I goofed up the pulley ratio (6,500 divided by 2,950=2.2), I won't be seeing 10 grand. Even if I achieve 4,000 RPM on the motor, my max speed will be only 8,800.

Making pulleys is trickier than I had supposed. It is really difficult to know the actual pitch you are creating until you test. In my case, calculating diameters didn't deliver the planned ratio of 2.5. I will have to make a bigger motor pulley, or a smaller spindle pulley.

Next step is to start adjusting the trim pots on the Leeson drive and see if I can get the motor to speed up to 4K.

A summation of where my CNC conversion stands.

After many months of work, I continue to be an enthusiastic fan of the X2 package. It provides just what I was seeking in a benchtop milling machine:

• Good work envelope for the small parts I plan to make
• High quality iron castings for all major assemblies
• Good basic design with accurate dovetails on all three axis of movement
• Readily converted to CNC configuration
• Low initial price

However, as well as I like the machine, there are three major aspects of the X2 package that I found, in my opinion, to be in need of attention:

1) The attachment of the column to the machine base and subsequent deflection of the column.

2) The spindle motor, speed, noise and vibration issues.

3) The counterbalancing of the mill head.

Up to this point, I have corrected the column and the spindle issues to my personal satisfaction in a reasonably effective and low cost fashion. I now have turned my attention to the problem of counterbalancing the head.

I believe that the two counterbalancing solutions that have been offered by the factory, springs and gas struts, are very effective on a manually operated machine, However, springs and struts are only marginally effective on a CNC machine. Both methods support the weight of the head, but provide only upward force lifting on the head. Unfortunately, this design impedes the downward movement of the head.

I tip my hat to Hoss for his elegant, simply designed counterweight system for his highly modified X2. I believe that he has the right idea. My counterweight is a variation on the Hoss solution.

As you can see, the counterweight I have built uses two idlers and a 1/4” wire rope. The two pulley arrangement allows the deadweight to be cut by 50%. Therefore, it takes only 16 lbs. to offset the 32 lb. weight of the head. The weight is suspended from the end of the cable below the table top. It is enclosed in a section of 4” PVC tubing to prevent swinging when the head is moving.

How does it work? Like a charm! The movement is smooth and effortless. The head moves with finger pressure an holds any position when released without movement.

Pretty dang spiffy I must say!
:cheers:

Advice needed:

I have assumed that with CNC, perfect balance of the head weight is the ideal way to adjust the head counterbalance. Is this true? Perhaps perfect balance isn't best? It is easy to change the counterbalance weight to make the head biased towards easier movement upward or downward. Any experienced voices out there?

I should also mention that with the head perfectly balanced, it still takes some force to get the head moving in either direction. The counterweight makes this vastly easier, but the head still weighs 32 pounds and the counterweight adds 16 pounds more.

Bill

Advice needed:

I have assumed that with CNC, perfect balance of the head weight is the ideal way to adjust the head counterbalance. Is this true? Perhaps perfect balance isn't best? It is easy to change the counterbalance weight to make the head biased towards easier movement upward or downward. Any experienced voices out there?

I should also mention that with the head perfectly balanced, it still takes some force to get the head moving in either direction. The counterweight makes this vastly easier, but the head still weighs 32 pounds and the counterweight adds 16 pounds more.

Bill

In theory perfect balance is ideal.
However when machining most of the time the cutter will produce upward trust that is best meet with your head gearing under slight tension.
So adding the weight of your tool at a worst case scenario (heavy tool) you will want to still have a slight upward force on the head.
This will ensure your gears are under tension when positioning the head down.

By the way your counter balance will not add to the weight of the head.
It is equivalent to a 32 pound head moving horizontaly.
Good Luck

Thanks, very helpful advice. Also, I was unaware of how the machine would "see" the weight of the counterbalance. Very interesting that it is essentially invisible.

A couple of additional photos, the complete 32 lb. head installed and a shot showing the 16 lb. counterweight below the table

In theory perfect balance is ideal.
However when machining most of the time the cutter will produce upward trust that is best meet with your head gearing under slight tension.
So adding the weight of your tool at a worst case scenario (heavy tool) you will want to still have a slight upward force on the head.
This will ensure your gears are under tension when positioning the head down.

By the way your counter balance will not add to the weight of the head.
It is equivalent to a 32 pound head moving horizontaly.
Good Luck

I love your Speed Controller and Motor. Wonder if it would work on a Taig mill. It's so clean!


-Jason

I love your Speed Controller and Motor. Wonder if it would work on a Taig mill. It's so clean!


-Jason
The Electro-Craft servo motors show up pretty frequently on Ebay. 3/4 HP and only 5 pounds. It might fit well on a Taig.

At long last, I am doing some work on the X axis, specifically the limit switches. One my objectives is to have the switches be completely internal with no chance of them being contaminated with swarf or knocked out of adjustment. This presents a challenge, as there is very little space under the table.

My strategy is to machine a space in the saddle for the switches. The first photo shows the saddle on the table of the mill drill.

Next view shows the lengthwise groove that will hold the switches and their mounting bracket.

The end-on shot clearly shows the groove (white arrow).

The final three photos show the saddle with the machining completed, the aluminum angle bracket set into place, and finally the bracket with the switches attached and attached to the saddle, The switches will be triggered by machine screws threaded into the underside of the table.

Bill

That is a great idea. Good job.-Adam

bilinghm what is the make of the mill/drill doing the work on the saddle?

It's a 24 year old Enco, updated with a 2HP VFD controlled motor. I bought it new. (i've been doing this for a while :-)

Bill

bilinghm what is the make of the mill/drill doing the work on the saddle?

Why not use one limit switch and 2 cams.

Phil

At long last, I am doing some work on the X axis, specifically the limit switches. One my objectives is to have the switches be completely internal with no chance of them being contaminated with swarf or knocked out of adjustment. This presents a challenge, as there is very little space under the table.

My strategy is to machine a space in the saddle for the switches. The first photo shows the saddle on the table of the mill drill.

Next view shows the lengthwise groove that will hold the switches and their mounting bracket.

The end-on shot clearly shows the groove (white arrow).

The final three photos show the saddle with the machining completed, the aluminum angle bracket set into place, and finally the bracket with the switches attached and attached to the saddle, The switches will be triggered by machine screws threaded into the underside if the table.

Bill

Why not use one limit switch and 2 cams.

Phil

Phil,
I considered that. However, using two switches, back to back, allows me to have more than two inches of adjustment with stationary triggers. A single switch would have required movable/adjustable triggers to fine tune X limits. This arrangement seemed simpler.
Bill

As I worked on the X axis, it became apparent that I would not be able to achieve full travel without providing some clearance for the Lovejoy coupler on the end of the ballscrew. The coupler would rub on the saddle when the table moved fully to the right.

The solution was to cut a small recess in the saddle. The pictures show the cutting, the recess (white arrow) and the ballscrew in place. This was a simple fix, and the X now has full movement.

Bill

Here is a better view that shows the clearance created for the Lovejoy coupler.

Bilinghm, Its been awhile since I last read this post but with all the work and effort you are putting into this your going to be really unhappy with the lovejoy couplers. I would say after a couple of weeks of using the mill with them on it I was starting to see about a .004 - .006 backlash on all axis. I contacted CNC Fusion and told them what I had thinking maybe the screws were not loaded right or something well they pointed me to this problem area. You can make your own spider out of a harder material was one of their suggestions or you can buy the Heli-Cal couplings that they sell for the X3. I went with the Heli-Cals and I cannot measure the backlash at this time. I have a .001 dial indicator. I just wanted to give you a heads up because of all the work your putting into it. Great mods by the way. I wish I had another mill laying around to do some of the work!

Regnar, I know you are right on the Heli-Cals, I am looking for a bargain set:)

I went with the Heli-Cals and I cannot measure the backlash at this time. I have a .001 dial indicator. I just wanted to give you a heads up because of all the work your putting into it. Great mods by the way. I wish I had another mill laying around to do some of the work![/QUOTE]

You're doing such quality work, I'm sure you could use solid couplings.
Super cheap to make and never any backlash.
Nice work.

I'm going to copy that counterweight design. Beats the heck out of what I was planning. Think 'gymnasium equipment' :)

The pulley ratio is 2:1 ? Formal education was sooo many years ago.

Cyclestart,

A single pulley doesn't provide a mechanical advantage, it only changes the direction of the applied force. Two pulleys, regardless of their sizes, give a 2:1 advantage. However, the trade off is how far you must pull: half the effort, twice the distance pulled. So to raise a 32 pound head 9", a 16 pound weight moves 18". That's why I put the counterweight below the table where it has plenty of room to move.

Bill


I'm going to copy that counterweight design. Beats the heck out of what I was planning. Think 'gymnasium equipment' :)

The pulley ratio is 2:1 ? Formal education was sooo many years ago.

Borrowing a great idea from skmetal7, I decided to try to create a LED spindle light. I used an underwater ring light used in ponds and pools, it fits perfectly around the spindle.

However, there is a big problem. The LEDs are super bright (24 of them) and they cast a very dark shadow, exactly where you need the light! This would be less of a problem without the extra length of the 180DA collet chuck, but I use 180DA on everything in my shop, so it cannot be avoided. Rats!

Bill

Bill:

I think you need to angle your LEDs from a wider circle toward the
center, in order to reduce the shadow effect. In my work I have had
a lot of white led applications that had hot spot, shadow, and angle problems.

Wish you the best.

Jesse
(Java77man)

Bill:

Also, if you are getting hot spots you might want to look for a diffuser
material to soften the narrow focus of most leds.

Jesse
(Java77man)

Thanks Jesse,

A diffuser! Maybe I can scatter the light enough to reduce the shadow? I do vacuum forming, so I could experiment with designs.

Bill

Bill:

Also, if you are getting hot spots you might want to look for a diffuser
material to soften the narrow focus of most leds.

Jesse
(Java77man)

Now that I have all of the necessary items installed on the head, including the counterweight system, its time to check where things stand. Specifically, how square is the column to the base? To test the squareness, I clamped a machinist's square to the column, supporting it between the bottom of the head and the sliding stop. I then zeroed a dial indicator against the blade of the square with the table moved all the way back. The table was then moved forward and the difference was observed. My first measurement showed that the angle was greater than 90 degrees (the column was leaning back).

I did a trial and error process with shims placed between the column and the aluminum channel brace. With a .024 shim, I was able to get a zero reading front to back.

Now that I have established that the column is square, I can test the Y axis tramming. If the spindle needs adjustment. it will require shimming between the two halves of the head casting.

Bill

My headstock is completely disassembled at the moment. I've been wondering about alignment and tramming.

Since the headstock is in two loosely-fitting pieces, you could get the spindle trammed in on vertical, but the column could be leaning slightly to one side, causing the whole headstock to move slightly in the X plane as it moves up and down.

I guess the first step would be to run the headstock up and down to make sure the column is vertical. The next step would be to tram the front half of the head left to right... but the two halves are held together by Allen screws you can't get to when it's mounted on the column.

What am I missing here? Or, in the general scheme of things, is a small X error acceptable over the full range of Z travel?

X2cnc

Get a flat piece of material to place on the table. Then make (or buy) an arm to mount a dial indicator to the spindle. Turning the spindle by hand will sweep the flat surface. This will measure vertical alignment in all directions. Despite the bad press the X2's tilting column gets here, many "real" mills have a tilt feature. Of course those machines generally only tilt the head :)

EDIT/ Hope this link works. In the shop we used a ground ring as the flat surface and a much better arm for the dial.
http://images.google.ca/imgres?imgurl=http://www.cnccookbook.com/img/OthersProjects/tramming_bar.jpg&imgrefurl=http://www.cnccookbook.com/CCBlogJune2006.htm&h=596&w=589&sz=34&hl=en&start=4&tbnid=MJXRhS_Mh9FIDM:&tbnh=135&tbnw=133&prev=/images%3Fq%3Dbridgeport%2Btramming%26gbv%3D2%26hl%3Den%26sa%3DG

edit-2/ You'll have to scroll down a bit unfortunately.

The tramming bar picture about halfway down?

That would let me get the spindle vertical, but since the headstock is in two pieces, the part with the dovetails isn't necessarily in alignment with the spindle. Once trammed, the spindle would always be square to the table, but it wouldn't necessarily stay in the same X position.

As an extreme case, imagine that the front and back halves of the headstock are 30 degrees out of alignment with each other. You tram the spindle, and the spindle stays 90 degrees to the table, but the whole headstock slides left to right as it goes up and down.

How do you guarantee the spindle is parallel to the dovetails?

How do you guarantee the spindle is parallel to the dovetails?

To be clear, the shop equipment I referred to was a kneemill with a head that tilted both side-side and for aft. There are different views on the best way as seen in the discussion here
http://www.practicalmachinist.com/vb/showthread.php?p=843600

However the technique I mentioned can be used on the X2. The fact that the head assembly has 2 pieces is a red herring afaict. The parts don't move in relation to each other unless the bolts joining them are loose. Now let's say you get the spindle perpendicular to the table in all directions with the head 4" (or whatever) above the table. That's trammed in as far as normal adjustment goes. This might require some shimming. Remember It's a cheap mill. Now if the spindle reads non-perpendicular at different height :eek: Are the gibs loose allowing rock ? It's either set a compromise setting or rework the column. What else ? I'm going to check this out on my X2. Could be an unpleasant surprise.

Anyway, this is bilinghm's thread, so enough OT from me unless he wants to pursue it.

EDIT/ Oh, I see where you're going now! The spindle could keep perpendicularity to the table but change position in the x or y plane as the head is raised. Sorry, I have no clue how that would be corrected. Not by any normal adjustment.

Thanks to cyclestart for sharing some excellent information about tramming. Indeed , that is exactly what I did last evening. My hope was that with the column set at a right angle on the Y axis (I’ll worry with the X later), that the tramming of the spindle would be perfect without the need to put shims between the two halves of the headstock. Happily, the spindle trammed within .00025 of perfection over a 4.5” swing. In my book, that’s pretty darn good.

I guess my main point here is that given the design of the X2, there is the potential of alignment errors with both the column and the head. Either error will present as the spindle not tramming within specifications. However, if the column isn’t completely square to the table under real operating conditions carrying full weight, no amount of head shimming will correct the error.

Thankfully, the inexpensive little X2 seems to have an accurately ground column and headstock. Therefore, when you square the column, the head (assuming properly adjusted gibs) will tram out nicely. However, squaring the column without some form of a column brace that allows adjustability would be pretty challenging, if not impossible. In my opinion, the unsupported end of the big pivot bolt must be locked in position to keep the column square.

Bill

One issue with the CNC Fusion kit is that the Z axis ballscrew mount blocks the gib adjustment screws. I fixed this problem by placing a .75” spacer behind the top mounting plate to move the entire ballscrew assembly forward.

Picture 1 shows the spacer made from a big chunk of dark gray PVC plastic (white arrow). I originally intended for the plastic spacer to only be a prototype, replaced by an aluminum part, but it worked so well that replacement isn’t necessary.

Picture 2 shows how far forward the ballscrew mount is moved, easily clearing the gib screws.

Picture 3 & 4 show the new mounting holes that are drilled in the block. The screws are 1/4" X 20 tapped into holes in the side of the head. All four screws will be used. This allows the block to be mounted and adjusted from the outside of the head.

Picture 5 shows how the back edge of the ballscrew mount is trimmed to give full access to the adjustment screws.

As a project, this rates as very easy to do, and very effective in solving the problem with the CNC Fusion kit.

Bill

I built a small aluminum plate to support the Z stepper motor. I just didn't like the way the CNC Fusion kit only uses two bolts to secure the stepper and mounts it diagonally. It looks wrong to me.

This mount is very solid, uses all four motor mounting holes, and is "aesthetically" pleasing to my eye.

Hi everybody,

I’ve been pretty busy lately, but I am back at work on the machine.

One lingering problem was a way to protect the Z axis ballscrew from contamination with chips and trash. I absolutely want the guards to be fully integrated into the design, not afterthoughts. The X and Y are no problem, the rubber bellows the I previously posted will work great. But the Z has had me stumped.

I think the solution is a spring cover made from a telescopic wound flat spring. I will use a spring above and below the ball nut. The springs will extend and retract to completely protect the ball screw.

The photo shows the two springs. the expanded one is held partially closed with a piece of wire. When fully extended, the length is 250mm (about 10 “).

They are not particularly expensive, about $50 each. You can find them at Centryco.

That is how the X axis is done from the factory on my Bridgeport BTC-1 and it is very efective. I am glad to see that they are available from an outside source because they will look good on a few of the projects that I have going.
Mike

Great thread. Lots of info. Thanks.

Here are the springs installed. The upper mounting sleeves are pieces of black plastic that hold the big ends of the spring on the outside diameter. The small ends are supported with sections of aluminum tube the fit the inside diameter of the spring. Neither end holds the spring tightly, the spring can move freely at both ends.

I must say that this way of protecting the ball screw has completely exceeded my expectations; it just works great. The ball screw and nut will stay perfectly clean.

Bill

Your mill is shaping up nicely. I agree, thaose ball screw covers are going to make sure the screws are nice and clean. Very nice instalation too.

I just read your thread right through. Yours is the best planned X2 conversion I have ever seen. Thanks for documenting it here.

In terms of using the wood table to stabilize the machine, if done properly, wood can be extremely stiff. Remember, one of the highest performance fighters of WWII, the Mosquito, was made of plywood! Also, there are whole sections of this site dedicated to making CNC machines from wood. The wooden structure would have to be a "stressed skin" design in order to perform up to expectation, but I still think the concept has promise.

Bill

For the love of engineering and machining do not use wood.

bilinghm,

Excellent conversion and writeup!

What are your X, Y and Z travels after all of these modifications? Were you able to use the ballscrews that came with the standard CNC fusion kit? Where did you get your limit switches?

Thanks

Sorry for the slow reply, The X and Z are unchanged, the Y is increased to 4.5", a valuable gain. The limit switches are from Surplus Center.

bilinghm,

Excellent conversion and writeup!

What are your X, Y and Z travels after all of these modifications? Were you able to use the ballscrews that came with the standard CNC fusion kit? Where did you get your limit switches?

Thanks