Howdy,

I just finished my first enclosure over the last few weeks, and now it's time to do the machine conversion. I searched for a thread on how to install this kit and have decided to just dive in.
Enclosure thread:
http://www.cnczone.com/forums/showthread.php?t=42921

I found a thread that had a link to pdf guide for the preparation of the machine that helped with the tear down.

I'll be replacing the bearing blocks and lead screws, so I took a few short cuts on the X&Y axis tear down.

X-Axis: I locked down the X-axis, then removed the cap screws from the bearing block on the end of the table with the hand wheel. I then turned the hand wheel counter-clockwise, this pulled the bearing block loose from the table and pulled the other end of the lead screw from the other bearing block. I then slid the table off, and removed the other bearing block.

Be careful and wear gloves, the table has allot of sharp edges (everything has sharp edges) I guess they don't have files in china.lol

Y-Axis: I removed the hand wheel, bearing block, and removed the y-axis saddle. Flip saddle over and remove the bolts that attach the x-axis brass nut block. I then pulled the machine forward till about 4" of the machine hung over the front edge of bench. This allows access to the bolts that attach the y-axis brass nut block.

Ready to start assembly of the x&y axis, I still scared of the Z-axis tear down and assembly, so I'll start that tomorrow.

The ball nut blocks don't have taper pin holes, so here's what I did to align the blocks.
I attached (bolts snug) the x-axis ball screw assembly to the y-saddle, slid x-table and gib into the y-saddle, adjusted gib, bolted the x bearing block to the end of table and finished tightening the ball nut block to the y-saddle. To keep the weight down I removed the table again by unbolting the bearing block and slid the table off.

I slid the y-saddle onto the machine base and slid it back against the z column. To get the y-axis ball screw/nut assembly in place I needed to lift the saddle up just a hair for the ball nut flange to pass over the nut block pad and snug up the bolts. I Slid the saddle back to the bearing block, installed and adjusted gib, bolted the bearing block the the saddle and tighten the ball nut block.

It's time to remount the x-table, gib, and tighten the bearing block to the end of the table.

Now it's time to tackle the ZZZZZZZZ

Cheers,
Jack

Howdy,

I started the Z today?
I cranked the the z towards the top of the machine, cut a piece of 2 by 4
to put under the head and cranked it down till the head rested on it. I removed the lower back panel and disconnected the green ground wires from the base of machine. Removed four bolts holding the back cabinet and then moved the cabinet to the side of the machine without unhooking the motor wires.

Next I removed both side plates that connect the head to the lead screw. I didn't remove the locking screw (z's), the side plate on that side will not come off without removing the locking screw but, it will allow the plate to come far enough to come off the pins that hold it together.
I then unbolted the top and bottom bearing blocks and pulled the whole assembly off the machine.

Time to dis-assemble the z-axis lead screw. I started unscrewing the lead screw from the brass nut, the top bearing block will come down and hit the brass nut, keep unscrewing and the top block will press off. Keep going till the brass nut comes off, remove the four bolts that holds the brass nut to the cast iron support piece. I took a punch and tried to hammer the taper pin out. Did I say, I tried to remove the taper pin, It would not even move a little bit. Yes, I made sure that I was going the correct direction, I also noticed that the collar had a crack in it thru the tapered hole.

I got tired of messing with it and went over to my benchtop bandsaw and cut just on the other side of the pin. I got my pulley puller out ( brand new in package) and it had one undersized bolt in it keeping me from using the bolts that came with it (?is someone/something trying to keep me from completing my task?), so I rigged it and removed the bottom bearing block. HA!

Dinner is ready,

This is great info. I am getting my CNC fusion kit tomorrow from ups for my x3. Can't wait to dive in.

I'm hoping this thread will help someone who is getting ready to do the same conversion, such as yourself. I'm also hoping that if I start to do something wrong (stupid) that someone will tell me.

I thought I'd finish up tonight,... did I say I thought?

I started cleaning and noticed something about the cast piece that holds the ballnut and alum adapter plate. The surface that the alum adapter will mount to, had been machined with a round spotface tool and the alum piece was rectanglar and wasn't going to mount up flat. I got out a file and started filing with the file flat in a crosscut motion making sure to leave wittness marks from the spotface. I unbolted the alum adapter from the ballnut and mounted it to the cast piece and then re-mounted the ballnut.

I cleaned the top bearing block and checked it for fit on the top of the ballscrew. It's a loose slip fit, which is good because, I was woundering how I was going to press it on and get it in the right place for mounting.

I cleaned the bottom bearing block and thrust bearings and started asking myself. I said, self, where do the two small washers that came with the ballscrew (z) assembly GO? Both on top of bearing block, bottom, or one on each side.... Oh well!

I started looking at the timing pulley that mounts to the ballscrew. It doesn't have any set screws holes in it, now I'm wondering if it just got missed somehow or if I'm just missing something (like brains).

I sent an e-mail asking about the washers and set screw holes.

I guess I'll go ahead and mount the motor mount plate for the x-axis and mount up the x & y motors (425oz/in).

Jack

P.S. I was holding the z-axis ballnut and raised it up in a vertical position and as soon as I did the weight of the screw sent it spiraling down thru the nut and could have easily spun right out of the nut.

So my next project is to do a counter-weighted system for the z-axis. If the belt was to break, the weight of the head would send it crashing to the table. (Others may just want to add the gas strut)

Oh yea,

If others out there who have already done this conversion would like to add their 2 or 3 cents, PLEASE DO.

TIA,
Jack

I got an e-mail back already concerning the pulley and washers (quick reply)

They switched to a different pulley manufacturer (solid steel instead of nylon) and they (manufacturer) had missed a few. That has since been fixed.

The two washers should go, one on top, and one on bottom of the thrust
bearings, then the locking collar and pulley.

Suggestion from CNCFusion,
To align your X-Y axis,
Run the Y axis toward the stepper motor as far as possible, then readjust the motor mount. Run it back and forth looking and listening for anything odd. When you are satisfied fully tighten the Y mount.

For the X axis, run it back and forth with the motor mount and bearing mount
only snug. Adjust the motor mount when the table is all the way left, and
vice versa for the bearing mount.

To adjust your gibs, run your table back and forth about 15-20 IPM faster
than you plan to run your mill. Then tighten the set screws slowly until
your table starts locking up at max rapid travel. Now you can back off the
max IPM 15-20% to normal running speeds and you should not have any lockups.
Your stepper motors have more power the slower you run them.

Jack

P.S. I may need to re-adjust the way I did it. I ran the table back and forth and using my fingers to turn the ballscrew thinking, if I could turn it with my fingers it must be close.

Ohhhh no

Just having too much fun watching you!!

LOL

Ahh, come on!

It is kinda funny, I was scared of the electronics, but everything worked out ok.

I was putting off the Z-axis because I was scared, lol, and had a few ?'s that I figured everybody else already knew and were having to much fun watching me fumble with things.

And I was right! lol
Jack

Hi Jack,
I finished installing the deluxe kit a while back and noticed that you can drive the Y axis off the nut. There is no stop to prevent the screw from unwinding outwards. I emailed Michael at the time but never got a response.

I am still waiting on motors and electronics to be delivered so I'm not in a hurry but I will be doing something about it before I run the machine.

Can you let me know if you have the same issue with your machine.

Apart from this I am very happy with the kit. Just to let you know with the motor attached to Z the head does not slide down the column but you are correct that if a belt broke the result would not be good.

Howdy Rod,

I got my kit about 5 weeks ago, and Yes, mine is like yours. I too have been thinking of a way to attach a collar or a clip (like the old y leadscrew). I wounder, what will happen if I rapid into the collar or clip? Tear the nut block off the machine?

I've already started wiring my control box for limit switches.
I'll also setup Mach3 with soft limits.

I need to make a chain and sprocket or cable and pulley type counterweight system, I can"t make up my mind.

Cheers,
Jack

Thank you for the progress updates. They say those deluxe kit screws are very nice....

You inspired me to look at CNCfusion, and it looks like they are way back-ordered on the deluxe kits.

-Yojinbo

Hi Jack,
I was thinking along the same line about a clip/collar on the screw but would have liked a response from Michael on this. The other option I was looking at was to fix a block on the slide and another on the base to act as a stop.

I have built four CNC gantry routers and have had heaps of trouble with false triggering of limit switches. It becomes such a pain that I end up disconnecting them. You might have better luck than me as you have different electronics so I do not want to discourage you from having a go. Given my history on this I would much rather have a mechanical stop so I don't end up searching on the floor for weeks trying to find ball bearings. :)

You have me thinking about counterweights now and of course there is the enclosure and flood coolant to add to that list.

Good luck with finishing off your machine and I'll keep watching to see how you get on.

Thanks Jack for taking the time to document the CNC installation process.:)

I've had my X-3 for about 9 mos now, using it manually. I have a homebuilt CNC router and hope to convert the X-3 to CNC some day. If you don't mind, could you post the approx. amount of money you'll have in your conversion? I'm trying to figure how many soda cans I'll need to pick up to bankroll the project.;)

Also, does this kit come with drawings to give one the option to machine some of the alum. parts like brackets & such to save $$?

Nope just the parts. I just got mine in at 11:30 this morning.

My problem is that I ordered the Xylotex 4 axis kit with 4 425 oz nema23 motors and the Z axis on my CNC fusion kit is a nema34 mount. not a major problem but will delay me a bit. I also will need to make a new bushing for the 1/2" pulley that was sent with it. no major problem though.

Howdy Rod,

(Hi Jack) don't say that in a air port. lol

I'll have to look into the mechanical stop idea. The motor would just stall and maybe not tear anything up.

Later on,
Jack

Howdy DickyBird (that kinda fun to say),

Thanks, I was hoping it might help someone along the way, and it's a way of giving back. I've picked up allot here on the zone.

I not sure I want to add up what I've spent so far. I not sure I want to know.

I worked in manufacturing for 21 years and had never noticed any benchtop machines, didn't even know anybody made them. After a MX bike crash (forgot my parachute), I started looking for a hobby I could do while sitting down.

One night I saw a guy who was making his own RC cars using benchtop machines. I got kinda' excited and started researching and saw a cnc converted benchtop mill and I was hooked.

To start with I bought a benchtop lathe (lathemaster 8x14) and mill (seig x3). Then I started buying tooling needed to build the parts needed for the conversions.

Things happened (life) and I had to put the hobby on hold for awhile, which allowed me to do more research. It also gave me time to decide against going with the least expensive stuff, as others had written that they wish they had got better stuff to begin with.

I guess what I'm trying to say is, I could have done it for less BUT.....
(I have champagne taste on a beer income) lol

3-Gecko G203v with all the newbie safty features..............$147.00 each
CNC4PC C11G breakout board (bob).................................$109.00
A4 power supply 5v and 12v for bob................$ 28.50
A1 parallel cable..........................................$ 10.69
A7 limit switches..........................................$ 2.00
A9 latch twist-release e-stop.........................$ 22.50
C3 index pulse card.......................................$ 22.88
Kelinginc KL-6515 power supply........................................$179.00
KL34H280-45-8A (640 oz/in)..........................$ 89.00
2-KL23H286-20-8A (425 oz/in)..........................$ 54.00 each
CNCFusion Deluxe Kit......................................................$899.00
Action Electronics 4-pin female cable mic connectors............$ 2.50 ea.
4-pin male cable mic connector........$ 2.50 ea.
5-pin female cable mic connector......$ 2.50 ea.
5-pin male panel mount connector.....$ 2.50 ea.
6-pin female cable mic connectors.....$ 2.50 ea.
6-pin male panel mount connectors....$ 2.50 ea.
4-wire sheilded cable 18 awg 100 ft...$ .56 ft.
6-wire sheilded cable 22awg 100 ft....$ .50 ft.
Mouser electronics (resistors, capacitors, fuses, etc..............

I didn't add it up cause, I don't want to know.... lol

Cheers,
Jack

P.S. I've still got to get supplies for a counter-weight system and coolant enclosure. IT'S NEVER ENDING!

G'day Jack

I'll remember to adjust my greeting for Jack in future. :)

Congrats CJ,

I don't remember where I read it but, IIRC, CNCFusion had posted that they could make a z-axis motor mount for nema 23 motors. You might e-mail them, it couldn't hurt, who knows they may have some sitting around ready to send ya.

Good luck with your build,
Jack

Ouch, oh man....sorry to hear about the crash Jack. Sounds like you have the right attitude, dude.:) I had my share of tumbles on the ol' Elsinore back in the day but fortunately the only permanent body damage is a footpeg shaped scoop of meat missing from my upper ankle. Stoopid a$$ me testing a new bore/piston job at 2:00 AM after a few brewskies.

Thanks for taking the time to type out your equipment list. Yup, gonna have to win the lottery before doing the CNC thing!:(

My X-3 didn't come with the column gas strut and I quickly grew weary of cranking the head up. I cobbled up a compound pulley arrangement from some old pipe, some pulleys and a piece of Home Depot plastic coated wire cable. The weight was made from a couple of Jag front brake discs and some other steel scrap bolted together. I can spin the crank wheel with one finger now.

The ol'Elsinore was the start of my mx dreams back in the day.

My x3 didn't come with a gas strut either. I like your pimped out (jag brakes) counter-weight system. I was thinking of getting a couple of motorcycle counter shaft sprockets and a chain. Mounting the sprockets to a couple pieces of angle iron with sprockets mounted between them. Then mounting that assembly to the top of the machine. Or, cable and pulleys mounted the same way. Now I have a third option. Thanks!

Good day,
Jack

Howdy,

Got an e-mail yesterday (Friday) morning, it was to let me know I had a package on it's way (pulley).

I thought I'd go ahead and assembly everything onto the z-axis screw (bottom and top bearing blocks). I mounted it to the machine and made sure it turned smoothly. It felt really nice, and then I noticed that there may not be enough room to mount the pulley with the z-screw already mounted. Yep, sure enough, I will have to remove the assembly to mount the pulley. I'll know next time..........lol

I hadn't attached the x-motor mount plate yet, so now is a good time.

Let me say that everything has mounted up great and has fit without a hassle.
All of the parts come with their own mounting bolts. The motor mount bolt heads were the same diameter as the slotted mounting holes. I checked the bolts that came off the machine and their head diameter were .012" smaller. I thought I'd use them but, they were too long. I got the file out and filed the diameter until the bolt head knurl marks were gone. That should do it, so I started tightening them down and about 1 turn from tight they started getting harder to turn. I then remembered the bolts that came off the hand wheel bearing block. Which were not being used and were a little shorter then the ones that came with the kit and the diameter was right. I used them and it worked great.

Well, I guess while I'm waiting on the pulley, I'll program a plate to cover the hole where the z-axis crank/hand wheel was located. Should be a simple
little program (3 holes and a diameter profile cut).

Later on,
Jack

G'day Jack
You can bring your cables for the X and Y steppers through that plate so might be worth thinking about.

Thanks Rod,

I had not thought of that, sounds good!

later,
Jack

i just last week or so went through all of this myself.. i too am concerned about the freewheeling aspect of the z screw, so i'm considering either a counterweight or a gas spring, havent decided yet.. a couple things.. i emailed cnc fusion before ordering my kit, and just asked for a nema23 z motor bracket, and got it.. perfecto.. the only real problem i had was getting my old z handwheel shaft out of the base.. i couldnt get my gearpullier in there to get that little bevel gear off, and i couldnt get the key out.. so i ended up with a long drift punch and driving it part way out from the rear, and then going around front and yanking on the handwheel to get it out... not my most gracefull attemt at something, but i got the job done..

right now all i need is a computer to put out in the shop and a powersupply.. i'm still looking around, but i have decided that something in the 36-48v range @ 8A will do nicely... i'm using all 3 500oz-in nema 23 4 wire tripple stack bipolar motors and 203v's.. they have a rating of 3A, and according to gecko i only need 2/3 of the rating per motor, so thats 3motors *2A=6A, and that leaves me a little room for a 4th axis later on..

i did have this question... how tight should the belt on the z be? i got mine to where i can flex it some but theres really not any "slack" to it...

Howdy Project5k,

I had to yank on mine (z-handwheel) pretty hard also but, it came out.

What's the voltage on your motors?

I'll be finishing my z-axis tonight and will see how tight my belt feels. I'm thinking your good to go.

Jack

i managed to get the X and Y ballscrews on mine and found a couple of things that made installation a bit more time consuming.

On the y axis, there were some brass pins that were used to line up the original brass block. These holes were not in the bearing block that CNcfusion sent.

Also, mine has spanner nuts on the ends of the leadscrews and I have no way to tighten them up. Did they send a spanner wrench for your kits?

I have made an adapter plate to the z axis mount that came for a nema34 motor. If someone has this same trouble, I have the plans for the adapter and can post it here if need be. It does require some lathe access and some Aluminum bar about 3 1/2" diameter (1/2" of it anyway.) It fits perfectly snug in the original plate using the original nema34 bore.


Now one more question,

On the z axis, what prevents the head from crashing down with the power off on the controller. My motor has no real resistance when it is powered off at the controller so am I supposed to tighten the gib lock on Z before I shut it down?

They don't send a spanner and the advice is to grip the nut with a set of pliers or multi-grips. I wasn't happy about this but it works OK.

The locating pins you mention are for the original nut and not used with this kit.

My Z axis does not drop but I am using a Nema34 640 oz motor. I will be fitting a gas strut or couterweight after reading these threads - just to make sure.

I like your idea of making an adapter ring for the plate to convert to a Nema23.

IIRC, by leaving the pin holes out, the kits fit more of the different batches of machines. I'm guessing the quality control or the method that the manufacturer used to put in the taper pins were in question. That's the reason I assembled x-axis to the y-saddle on the bench and tightened up the nut block... etc...(post#1)

No, they don't send a spanner with the kit. My kit came assembled and I never took it apart. I did check to make sure there wasn't any slack in the fit. I thinking, if you get the nut too tight, it'll put more fiction in the bearing causing heat and drag.

An adapter will work.
I think it was Project5K who posted that at the time he ordered his kit he e-mailed them, and they sent him a nema23 z-motor mount with his kit.

When I put the z-screw in the other nite I noticed it didn't try to take off. I hadn't attached the head yet but, there was the extra weight of the cast iron nut block. The added drag of the thrust bearings I think were keeping it from taking off. I think it was Rod who said with the belt on it would not take off on it's own. I thinking I'll lock her down just in case. I'll have to test it and see, I'll lock it down, turn everything off, release the lock for about a second and see if it tries to take off.

I better go outside and finish up,
Jack

tauntdesigns- yea it was me.. i emailed back and forth several times with Michael before deciding that his kit was the way to go... and one of the things that he and I talked about was that i was going to be using all nema 23's.. one other thing that we talked about was that all my motor shafts are 3/8" and not 1/4" like the pullie that he normally ships.. so part of the deal was that he hooked me up with an all nema23 kit, and a pullie that fit my motor(he reamed it out for me)

All of my parts just dropped right into place, i didnt have to modify anything to make it work, i was so impressed... Michael really did his homework on this one...

i ordered my powersupply parts, breakout board, and limit switches last night, so they should be here late this week i hope.. I did hook up one of my geckos to the x axis with my modified computer powersupplies this afternoon just to see it move.. man.. i could actually get it to rapid at 270IPM with 24V!!!! took out my dial indicator, and what mach thinks is an inch, measures out at an inch...

i also made a couple preliminary measurements for backlash and pushing hard enough to slide the mill on the mounting surface(bolts arent in yet) i could only manage about .00075 backlash on my x... thats WAY better than i need any of my parts to be... most everything i wanna make can be within .002 and be fine... i think i have more machine that i really need, but thats fine, cause as i learn it, i can grow into better projects.

now i gotta scrounge up a case or something to put the powersupply, geckos, breakout board, and connectors in and on... hopefully whatever i end up with will shield the magnetic flux from the transformer enough so that i can have it near my monitor without making it go all wiggley...

the only real concern that i have right now is that there isnt any kind of mechanical stop on the y axis to keep you from accidentily running the screw out of the ballnut.. i'm actually thinking of 2 ways to fix this.. the first means pulling the y screw out.. and turning it down so that i can put a locking colar on it.. the other is to put a stop strap on the saddle to the column bolts so that it will stop..

anyone else see this? how did you fix it?

Hey guys,

My taper pins that align the column stick up a little and interfere with the motor mount bracket. Did anyone else have this problem? I've got a 4 1/2" angle grinder, I was thinking of just grinding the tops off the pins.........

I've looked at the column/electronics cover and am thinking I'll need to cut some metal away to allow for the motor mount bracket.

What did you'll do?
Jack

They don't send a spanner and the advice is to grip the nut with a set of pliers or multi-grips. I wasn't happy about this but it works OK.




I wish that would work for me. I have nearly 2 inches to tighten on my x axis. And probably at least an inch on the y axis screw. I guess I will have to make something. I guess I will take a socket and fit a 2 pronged insert into it for a tool to do this with. I was hoping I missed the tool in the box.

Maybe I need to make some tools like this for sale. :rainfro: Then sell them to Micheal. lol

Hey guys,

My taper pins that align the column stick up a little and interfere with the motor mount bracket. Did anyone else have this problem? I've got a 4 1/2" angle grinder, I was thinking of just grinding the tops off the pins.........

I've looked at the column/electronics cover and am thinking I'll need to cut some metal away to allow for the motor mount bracket.

What did you'll do?
Jack


Exactly what I had to do on grinding down those pins in the z column. And the back plate on mine had to be cut out also. I took the screws out of the backside, then plasma'ed the hole for the motor. I did take off the board and stuff though before I did it. I also had to open up the angle frame on either side a bit for bracket clearance. I think they have changed that back plate a bit on the sx3. At least since the pictures that Micheal had put up. his appeared to be a separate screwed on panel from the rest of the electronics.

Rod,

In a earlier post, I believe you said you were having noise problems with your limit switches. I just read a post that quoted Marriss from Gecko as suggesting putting a .1uf cap between the limit switch's leads. Here's a copy of the partial quote from Marriss:

"One way to use non-gold contact switches reliably at low currents is
to place a capacitor across the contacts. The cap will deliver enough
pulse current to clean the contacts on every switch clousure. This
energy is stored in the cap while the switch is open. A .1uF cap at
5VDC will store 1.25 uJ of energy, more than enough to clean any
contact. The noise filtering will probably be welcome as well.

Mariss"

CJ,
Thanks for the reply/conformation of my question. I kinda' knew it needed to be done, but thought I would ask anyway.

Really, I'm going out to finish now,
Jack

Jack
Thanks for the tip on using capacitors. I have used resistors but these didn't work. I'm not strong on electronics so I appreciate any advice.

On the dowel pins I was able to knock mine down. Worth a try before taking to them with a grinder.

. I was hoping I missed the tool in the box.

Maybe I need to make some tools like this for sale. :rainfro: Then sell them to Micheal. lol


cncfusion should REALLY consider supplying this tool. How much could it possibly add to the price of the kit? The nuts don't need to be very tight, so the cheapest c-spanner available would do the job. I used a vise grip but it does seem a little caveman-like.

my column pins werent sticking up at all, so i didnt have that problem, but i woulda probably looked into taping them downward, or maby dremmeling them down if it were an issue..

as for the spanner nuts, i too went the route of the vice grips.. but i'm sure that in time i'll use my mill to make the wrenches...

on the electronics sheetmetal, i was worried about the bold head clearance, but once i had the motor mount in, i slid the sheetmetal in place and it fit perfectly.. so i didnt modify anything on that.. if i had had a clearance issue, i woulda just dremel'd it, or maby used my metal nippers.. guess i got lucky on this one..

Howdy,

Monday night I took the z-axis off the machine, attached the pulley and re-assembled the z again.

OK, I had to see what would happen if I unlocked the z without the motor and belt on. I loosened the brake enough to turn the screw by hand rising it up off the 2 x 4 prop about a 1/4" or so. I then loosened the brake all the way, oops, good thing I left the prop in place. So, yes, without the belt and motor on it, bad things could happen.

After assembling the motor and belt, I repeated the same test, No crash.... WooHoo!

Tonight I got everything electronic outside and hooked up. Did a few quick tests. I put a block on the table, got out my cheap 2" dial indicator, set zero and started making .100" jog step moves.

X: each click of the mouse got me .100 right on the mark in both directions. Backlash checked at .0005"

Y: same thing, the .100" moves were right on the mark (both directions). Backlash checked at .0008"

Z: Hit the mouse button and boom, right on the money, changed direction and got .015" backlash, @#%$&. Then I remembered the brake was still on (dummy me). loosened the brake, now backlash was at .004", I thinking that's pretty good for a shaft collar type set-up.

I'm Happy,
Jack

G'day Jack
I still get a thrill when I first run a machine so I hope that CNC grin stays on your face for a few days. :) Pleased to hear your results on the backlash.

I got the last of my electronics from Hong Kong and started the enclosure tonight so hopefully I won't be too far behind you.

Must be time for you to post a couple of pictures of you work.

Have you decided on counterweight or gas strut or nothing yet?

Howdy Rod,

A friend of mine has a guitar amp he modding and wants me to make some cut outs for some more switches and gadgets. After that comes the counter weight and then I'll start a conversion for my small benchtop lathe.

I need to do some testing on the z-axis and see how it acts on plunge cuts. I'm wondering if the .015" (backlash) will show up again. I wish They had used a locking nut on z instead of the shaft collar. A locking nut would allow adjustment to the amount of backlash. Oh well!

Cheers,
Jack

Gidday Jack,
Been thinking (dangerous) about moding the collar to have a threaded section that can be pushed against the bearing and the collar once the main body of the collar is clamped. In other words expand the collar. A pin and capstan style could be used to tighten it. I hope I have used the right terms here. Do you think it would just make the collar slip away? Any other thoughts?

Howdy Rod,

It may not slip, the pulley may help kept collar in place. How much more room will the threaded piece take up? I ask because if you add a piece to the assembly the pulley alignment may become a problem.

I'd like to turn down the shaft to a smaller diameter up to the bottom of the shaft collar and then thread where the collar was for a nylon inserted locking nut. I'd have to buy a new pulley with a smaller bore or bush the original pulley.

Just another idea, any more out there (ideas).
Cheers,
Jack

Hey all

Sorry to jump in late here- been busy this week.

I finished an SX3/CNC_Fusion conversion (ballscrews with low backlash nuts) a couple of weeks back. On the electronics side, running 70V w/ 3 203V Geckos, driven by Mach3. So, a couple of observations/cautions-
- The thing can move like a router. Ill have to check exactly, but we are talking 200ipm sorts of speed on X/Y. So, there is definately the potential for some inertial damage if you allow the speeds for rapids.
- The screws (at least, the X) have some bend to them. Apparently, this is 'within tolerance,' but in combination with tight ballnuts, there are some unhappy sounds coming from the X at high speeds. Y is happy.
- Be very careful about your use of a gas spring on Z. Because most of the examples I have seen are single strut, you get asymmetric loading on the Z/head, and you would be horrified to see the amount of deflection this allows in one direction on X. Presumably, this would be less a factor if you used dual struts, one per side. But, I suspect just a simple counterweight of 40-50lbs is probably a better option- with reasonably tight gibs the head is not going to crash with a belt break.
- Apparently, most of these SX3 units have some 'sag' in the head which requires something like .005" shim in the base of the head->Z bearing interface. With that, Im showing less than .001" variation in tramming the spindle/table, which is a lot better than the .007/8 I was seeing before.
- I havent done much testing since some of the tuneups, but it does not like rapid plunges- there is too much slop in the Z axis bearing, and a lot of twist in the column. If you just stick a dial indicator on the table touching the spindle, and push on the side of the SX3 head, you will see .004" deflection without muh effort (and probably triple that with a gas strut in place). I added a lot more torque to the Z axis column bolts, and that helped a bunch.

Anyway, I just ordered the spindle speed upgrade kit from Syil, since 1800rpm is not going to do it for smaller cutters and aluminum. So, hopefully that will help a little with feed rates.

I threw together a fogbuster-style unit for cooling and chip clearing, and that works wonders. Definately a lot more effective for me than was air and a can of WD40, and a lot less effort.

Any questions I can help with- let me know
Rob

Oh, one other thing-

Maybe this is just an SX3 thing, but be really cautious of chips getting thrown into the Z column cover (electronics housing). Because you need to cut into the box for the motor/mount, you might be surprised to see how much junk finds its way into that section of the mechanism. So, aside from having nice shards of metal on your Z stepper (and perhaps on the belt...), you are one airgun shot away from blowing metal shavings all over your PC board.

Unfortunately, there does not seem to be much in the way of affordable bellows- as least not that I can find. Easy enough to rig up some plastic curtains or whatever, but they always seem to fold exactly in the wrong place- allowing chips or coolant right on in :(

R

Welcome Rob,

Better late, then never!
I been reading about rapid feedrates of 100, 200, and even 300 ipm. I'm new to the electronics side of things and not fully comfortable with everything yet. I set mine to 50 ipm the first time I ran my box and motors and still haven't changed anything since then. (50 ipm will rapid end to end 3+ times in minute)

I going the counter-weight way. I heard of people using 70, 90, and even 110lbs for their counter-weights. I may need to tighten up my z-axis gibs.

I thinking of a flood coolant and machine enclosure (I don't have an air compressor yet).

I been thinking about taking all the electronics from the back of the machine and put them in a new box mounted on the front of the machine bench (outside the machine enclosure) .

Cheers,
Jack

Yea, 250 IPM (rapids) is easy, and I suspect 300 is possible also- havent tried that. The problem is, the construction on these mills is just horrible. For example, the X axis is loose at full travel -X, and stuck at 80% +X travel. Im sure there are people here who can fix that in 10 minutes, but I dont have a clue as to where to start- maybe some blue and a scraper? Probably cheaper to just buy a new table (I think they are $120 or so).

I probably have too much backlash in the system, even though I have the preloaded nuts (Nook I think). I see about .004" on Y and Z, and .007" on X. I can compensate for it in Mach, but that sort of seems to slow things down a lot for 3d surfacing. I could live with .002, but .007 is pretty bad IMO considering this is a new setup, with matched screws/preloaded nuts. I dont think it is in the mounts or whatnot though, since I have double-checked all that, and since the error seems consistent at .004 or more for all 3 axis.

On the Z- my opinion is that the designers intended the head to use its weight as part of the bearing system. That is, it seems to have a TON of slop if you were to hang it from its center of gravity. So, unless you are pretty handy with bearings, it is my opinion that removing 40-50lbs weight with a counter-mass might reduce quality of cut, and probably will significantly increase slop on plunges. Then again, maybe my bearings are just really bottom of the barrel and others have a better setup.

R

On the Z- my opinion is that the designers intended the head to use its weight as part of the bearing system. That is, it seems to have a TON of slop if you were to hang it from its center of gravity. So, unless you are pretty handy with bearings, it is my opinion that removing 40-50lbs weight with a counter-mass might reduce quality of cut, and probably will significantly increase slop on plunges. Then again, maybe my bearings are just really bottom of the barrel and others have a better setup.
extrapilot,
no no no.. something is wrong with your spindle bearings if you need to somehow use the weight of the head for accurate cuts. I would check them out ASAP.

I run a fully counterweighted head on my x3. This should have no effect at all on cuts, plunges or drilling.

As for the Y axis, yep, they could have designed the base better for more dovetail area. It gets sloppy near the extremes + travel. But you should still be able to get a solid and firm held 6" of travel and it should not bind anywhere.

Are your nuts preloaded with just larger balls or are you using 2 spring loaded nuts? IMHO 2 spring loaded nuts which is what I use should have no more than .001" backlash.
IMHO I would slow down those rapids. You'll just wear things out at a fast rate.
Steve

Howdy Steve,

I agree with you on the counter-weight, most, if not all of the machines that I've ran in the industry has some sort of weighted system on them.

Jack

Steve

I seriously doubt that this issue is spindle-bearing related. I think what I was trying to suggest is that the bending moment on the Z-axis bearing seems to imply the designers built-in the weight of the head as part of a coupled system. Taking that weight out changes the load profile on the bearing, and, at least in my case, significantly increases slop in the head. As I mentioned, there is a lot of deflection on the head independent of the spindle- if you just put a dial indicator on the table measuring the side of the head (and push the head in the +/-X with your hand). Even if the gib is locked, you still see a bunch of deflection. The same happens in +/- Z, though it is a little harder to demonstrate. If someone has a single gas strut mounted to the side of the head, just try this X axis bending test with and without the strut (make sure the strut is at least partially compressed). I would be interested to hear if others have better luck.

Again, since all these are made by slave labor in China, Im sure no two are remotely the same. So, maybe my Z bearing is just the worst of the worst. My note was just cautionary- as I dont think people are necessarily running these tests- based on my experience with a retrofit on a very recent gen SX3. Frankly, I think I probably need to mill a new Z axis gib, as mine was obviously profiled with a hand grinder and not cleaned up in any way.

Regards,
R

R,
My Bad.
When you said Z bearing, my brain went and associated that with spindle bearings. I now understand you meant the z axis ways/tapered gib.
Well testing fully counterweighted vs no counterweight was a snap for me since I already have a setup in place to hold the counterweight in place while I remove the connection at the head.

So I did tests at around mid height for the z axis testing the deflection in each plane
Standing directly infront of the head using only my highly calibrated index finger for force I can deflect the head by .002" in either drection side to side with as much force as I can muster with 1 finger. Locking the gib has no effect. I uncoupled my counterweight system with a bar link at the same height. There is no change whatsoever in deflection amounts.
This is not a fault of the dovetails IMHO. This seems to be pure column deflection and not ways /gibslop. Slop would tend to allow the head to move. The head is not moving but is deflecting. It springs back to exact zero in both directions.
So yes, you have found a weak area for sure. A finish cut will be in order for all my parts after observing this flex! So thanks for pointing this out.
I should have tested this long ago.
Also tested up/down motion and again the same results. The counterweight does not effect the readings at all.

Before everybody freaks out about this, I can get even worse deflection on my 9x2O lathe and still turn parts within .0005" with little effort.
Just take a light finishing cut and all should be fine.
Just 2 pics to show I'm not blowing smoke and did actually just do these tests.
http://i109.photobucket.com/albums/n48/S_J_H/x3zaxissloptest001.jpg
http://i109.photobucket.com/albums/n48/S_J_H/x3zaxissloptest002.jpg
Steve

Yea, sorry to not be precise. Im an aerodynamicist, not a machinist- so the vocabulary is coming along slowly with the new machine :(

It is interesting that you are not seeing variation with/without the counterweight. In my case, it was a gas strut on the left side, which imparted a non-symmetric (rotational) load on the head/Z bearing. I suspect that my Z gib is not making correct contact with the ramp in the head, and is thus forming a sort of pivot point. Id guess that this problem would be less obvious if the strut were mounted on the right.

One thing to note here though- the deflection I saw was using more force (call it 30 lbs or so), and most of the flex was traced to the column. But, and here is a variance from your test, I also measured the deflection at the spindle- where it was higher. This implies that the head is twisting (call it a 4th axis move...) as well as deflecting (in the X axis). Of course, nothing in the world is perfectly rigid, but any X3/SX3 people might want to measure both the side of the head and also the spindle to see if you are getting twist; in theory you could have one without the other depending on where you place your gauge. Since Im running an SX3, which has a rotating head, I would expect that there is more potential for flex in that axis.

I have been told by people who know these machines very well that, with a finish pass, this deflection is trivial. Now if I could just get rid of the backlash (which is not trivial), things would be happier!

Regards,
Rob

Gidday Jack,
This time I am really going to "HiJack" your thread. :)

I am working on the driver board and thought you might be interested to see the progress. I'm not sure of the drivers as a friend got them from a supplier in China but they look similar to Kelling and the numbers match so they are probably generic. They were a little cheaper than Kelling but without support or likelhood of warranty so I wouldn't rush at them again.

There are three power supplies - 48 volt for the Nema34 and 36 volt for the two Nema23 and the smaller one is 5 volt for the breakout board and 12volt for the fans.

I cut the front and rear panels with one of my gantry routers out of scrap 3mm aluminum sheet and the case is a stripped out computer box. The yellow tags are cable ties to route the wires. Maybe this weekend will see some movement.

Howdy Rod,

HiJack away, Looks like you're pretty close to having it up and running. A few wires here, a few wires there.........

I like your front and rear panels, nicely done..... I wish I had a gantry style router... Maybe after I get my little lathe done.

Good luck,
Jack

Gidday Jack

Be careful this CNC can be addictive. :)

Gantry routers are fun but they are noisy - much better to work on a mill.

Howdy Rod,

Does the router spindle make most of the noise or axis movement? I've watched a few videos of routers cutting wood on youtube and they're pretty load.

Here's a link to a pic of a machine I ran 10 years ago. It's a 2-gantry (bridge) type 5-axis, 6 spindle (3 per gantry) , 1 operator per gantry (ride along). The whole machine was 120 feet long by 30 feet wide. It was a beast and I always put someone else on it if I could.

http://www.profab.com/images/snk1.jpg

Later On,
Jack

Gidday Jack

Yeah it is the router that is noisy. It would be nice to get a quiet spindle but the cost is just too high over here.

Now that is what I call a CNC machine. :) What was it used for?

Howdy,

The company (ProFab) did contract work for the aerospace industry. The 5-axis machines mostly just cut alum parts. Most of the 5-axis A/B head machines were single spindle. The large machines, the gantry moved and the table stayed still. The medium and small sized 5-axis machines the gantry stayed still and the table moved. I always tried to run the 60 inch SNK 5-axis. I was 3rd shift 5-axis zone leader/on the job trainer so, I could chose what machine I wanted to run. They also had a few 5-axis horizontals that cut stainless and titanium parts.

That place had everything you could want in machinery (waterjet, laser, 5-axis A/B head verticals, 5-axis horizontals, my favorite 4-axis horizontals, 3-axis verticals, and cnc lathes)

The guy who owned the place started in his garage.

Cheers,
Jack

P.S. My favorite 4-axis horizontal, cut with a 2" insert cutter at 20,000rpm, 200ipm, .5" deep, and full width of cut. 2000" ipm rapids.

Jack,
I never imagined CNC machines as big as those.
Thanks for sharing.

Hey all-

Sorry to bring back from the dead this thread on the X3 (SX3). But I wanted to relay some additional info from recent tests/mods I made.

OK, so with the CNC Fusion kit on the SX3, I was seeing about .004-.007 backlash on all axis (plural?). For X and Y, this turned out to be the Lovejoy flex couplers. I replaced them with the Helicals CNCFusion now offers, and the lash is well below .001 on X and Y. One note though- on my machine, the mount for the X axis bearing block was offset (or the holes were drilled a bit off on the kit's block replacement), and this was causing the X screw to bend. The nut did not like this at all, and was causing some really nasty motion. So, if anyone is seeing some weirdness on X (sort of behaves like whiping, but even at lowe speeds), just check to see if when the X axis motor is removed, the screw is centered up on the motor axis. If not, you cannot simply slide the mount in the Y direction (since the screw is not parallel with the table, the offset will change depending on the X position of the table...). I just overdrilled the X mount holes in the SX3 base and that provided enough offset to align the screw.

Z is just a problem. It seems that .004" is about as good as it will get unless you counterweight the head at the head's center of gravity. I have tried gas struts at the Z bearing, and this does not help the lash at all. I think this is more a simple physics problem of stiction and the Z axis bearing being about 8 inches offset from the head's CG (causing sag). When the head is counterbalanced at its CG, the lash is gone (well, below .001 anyway).

So, with this kit and (possibly) some minor tweaks to the mill, it is possible to get it below .001 in all 3 axis, and not have to use software compensation (which really does not work well in 3d surfacing operations, anyway).

Good luck-
R

Sorry to bring this thread back from the dead a second time, but I'm doing the same cnc retrofit to my X3 and was wondering what kind of maximum travel distances are you getting on the X, Y, & Z axis?

Thanks

Howdy All,

I really should know that....
OK, I'm guilty, I've been running the machine without any limit switches.
I've been setting up my projects in the center of the table and they've been small parts.

MRM, I'll get you some travel distances this weekend.... Maybe some other kit owners will chime in with some of their distance readings.

extrapilot, I'll be looking at my x-axis this weekend. I have a low clicking sound in the first couple of inches of travel.

Cheers everybody,
Jack

MRM

The irony is that, at least for my SX3, the theoretical limit and the practical limit are different. At least for X, the table tends to bind as it moves to its far left. So, the option for me is to loosen the gib and then have some slop, or live with a couple of lost inches in movement.

Z is probably not relevant, simply because there so so many variables (if you retain the cutter shield, if you have a vise, length of cutter, etc). And frankly, at the upper range of travel, you get so much twist on the column that it is not going to be useful anyway.

Y is the problem with these mills. Im going to redesign the motor mount and ask CNC Fusion to build for me a longer screw. If the thing could do 7", or even 8", it would be alot more competitive with the 'bigger' mills. As it stands, with approx 5" travel, you are severely limited for single-fixture setups.

R

My Z travel is now 360mm. You say it's a bit twisty way up there, but it sure is easier than cutting norfmal drills down to stubs.
A few more mods and the head will hit the table lying on it's side. Gooid excuse to remove guard hing junk.
New counterballancing arrangement. X,Y and Z better than 0.001".
Put the head at the bottom.
Rotate 90 degrees.
3" 3 jaw chuck. (stolen from4th axis, with R8 adapter)
Instant CNC lathe! ( with weird tool holders)

MRM,

My X travel is 16 inches, but I'm only using 14 inches because I only run my table in the positive direction till the edge of the table reaches the edge of the Y saddle.

My Y travel is 6 inches and the table appears to be centered in the travel of the axis.

I haven't checked my Z axis yet.

I just ordered my kit Friday. Could any of you take some pics of the Y Axis set up. I have an idea to get ~1in of extra travel on each end. With limit switches, should be able to reach each end of the table.

This will be fun.

Kelvin

I just ordered my kit Friday. Could any of you take some pics of the Y Axis set up. I have an idea to get ~1in of extra travel on each end. With limit switches, should be able to reach each end of the table.

This will be fun.

Kelvin

Congrats,

The very first post of this thread has 3 pics that show the y-axis saddle, y-axis dovetail on the base, and y-axis saddle mounted to the x-axis table.
I didn't take any pocs of the y-axis screw assembly,but there's pics on CNCFusion's site. They have changed the ballnut mounting block since their pics were taken. They did this in order to center the axis travel to the table. Here's a link to their pics:
http://www.cncfusion.com/images/X3mounts/DeluxeX3photos.html

When I measured the y-axis travel for you, I turned the screw by hand until the bearing block/motor mount assembly hit ballnut and then turned the screw by hand in the other direction until there was about 3/4 to 1"" of thread left showing before the balls would start coming out.

Cheers,
Jack