[4]

X2 CNC Conversion

[3]
[10] Like what you see?
Click here to donate to this forum and upgrade your account!

Kennyd

Active User
Registered
Joined
May 1, 2011
Messages
478
Mike, I have been using that same pendent control on my BP for 6 months or so with good results.
 

Hawkeye

Active User
H-M Supporter - Gold Member ($25)
Joined
Jun 17, 2011
Messages
1,885
I'm an electrician and the guys at one of our wholesalers are really good about giving me the lowest price if it's for my personal use. The box cost me about $13 including tax. I never have to ask for a discount. I found the same box on their web site for $35.
 

7HC

Active User
Registered
Joined
Jul 7, 2012
Messages
640
I think I'm okay as far as a lathe is concerned. But then, that's what I thought about CNC in general. :thinking:

I've gotten this far without pictures, but I think the mods are getting ready to revoke my parking pass. I got the control box done today. The setup has been running the adjusting lever for about an hour and the fan on the power supply hasn't come on yet. The big fan I put in the control box is doing it's job.


A 12 x 12 junction box is just the right size for the power supply and control board, with a bit of room for small extras. The white terminal strip at the bottom will handle the connections for the homing switches.
View attachment 43533

Nicely done Mike!


M
 

Hawkeye

Active User
H-M Supporter - Gold Member ($25)
Joined
Jun 17, 2011
Messages
1,885
Catch-up time. I decided to extend the Y-axis as part of the conversion instead of a later upgrade. Since I didn't have any steel in a suitable size (I wanted a piece 2 1/2 x 2 1/2 x 6), I picked up that piece of mystery-cast I had used to make the base for the surface gauge (http://www.hobby-machinist.com/showthread.php/5435-Surface-Gauge). It had eaten a 10' bandsaw blade when I cut off the first piece, so I have been hesitant to use it for anything.

I figured one of the previous problems could have been grit and such in the surface from the mold. I used my radial arm angle grinder to cut about 1/4" into the piece all the way around.
PC160672a.jpgPC160673a.jpg

Then, out to the big bandsaw, running at its slowest speed. And yes, I did use a bit of coolant. No problems.
PC160674a.jpg

Some careful facing with an indexable end mill ...
PC160679a.jpg

And on to cutting the dovetails. The portable bandsaw would hardly touch this metal, but drills and shaper work fine. It even dulls centre punches. I drilled rows of holes and broke out the waste outside of the dovetails, then smoothed it off with the indexable. I took out all the waste I could below the slopes with a 1/8" woodruff cutter.
PC220680a.jpg

I spent most of the day today using a brand new 3/4" dovetail cutter to very carefully cut the dovetails, 0.010" at a time.
PC220681a.jpg

I carefully measured the original dovetail with a pair of 1/4" carbide shafts and I was quite proud of myself when I finally finished the extension to the same measurement. I took it out of the vise and held it up to the base of the mill. For all my life, I never expected that the mill would have a different angle than 60[SUP]o[/SUP]! And it's not 45[SUP]o[/SUP] either. It's only a few degrees off, but it totally screws up all the work I did today.
PC220682a.jpgPC220684a.jpg

I think what I'll have to do is mount the block on the base, then mill them both to 60[SUP]o[/SUP]. But first, I'll make an indexable dovetail cutter. With the change in angles, I'll need a new gib anyway, so I'll make it thick enough to counter the loss of metal from the correction.

PC160672a.jpg PC160673a.jpg PC160674a.jpg PC160679a.jpg PC220680a.jpg PC220681a.jpg PC220682a.jpg PC220684a.jpg
 

Hawkeye

Active User
H-M Supporter - Gold Member ($25)
Joined
Jun 17, 2011
Messages
1,885
Time for another update. I finished shaping the y-axis way extension and mounted it on the base. Then I cut out enough of the front wall on the original ways to allow the ballnut to pass.
PC260700a.jpgPC260704a.jpg

I got the dovetail cutter made and reworked the y-axis ways. The metal the extension is made from is a lot harder than the original. I'm pretty sure it's white cast iron - made by cooling cast iron quickly. Really tough stuff.
PC260705a.jpg
PC260706a.jpgPC260707a.jpg

I wanted to make sure the X and Y axes were at 90[SUP]o[/SUP] to each other, so I mounted them together and dialed them in, then milled the saddle dovetail that mates with the base dovetails.
PC260708a.jpgPC260710a.jpg

The ballnuts are a lot bigger than the originals, so a bit of cleaning up is required to let them fit. A 1" ball end mill comes in really handy.
PC280712a.jpgPC290713a.jpg

Next came the ballscrews. I turned the end down to 12 mm and threaded part for the bearing nut. A 3/8" shaft finished off the end, with two flats milled at right angles to take the coupling setscrews.
PC300714a.jpg

The couplings are made from aluminum and are one-piece. One end fits the 14 mm motor shaft and the other takes the 3/8" ballscrew end. I made a mandrel to hold the couplings while I cut the spring slots on the big bandsaw.
PC310716a.jpgPC310717a.jpgPC310719a.jpgPC310720a.jpg

The couplings will flex slightly to handle any slight misalignment in the motor mounts, while having no slip in the rotational movement.
PC310723a.jpg

PC260700a.jpg PC260706a.jpg PC260707a.jpg PC260704a.jpg PC260705a.jpg PC260708a.jpg PC260710a.jpg PC280712a.jpg PC290713a.jpg PC300714a.jpg PC310716a.jpg PC310717a.jpg PC310719a.jpg PC310720a.jpg PC310723a.jpg
 

Rick Leslie

Active User
Registered
Joined
Jul 30, 2012
Messages
480
I see what you mean by "projects within a project". Very nice work so far. I caught the bug a while back and set a ridiculously low limit for my expenditures, thinking this would save me from myself. Ha! Fat chance. I'm in the process of converting my tiny Sherline. So, I'm following your progress anxiously. I love the couplers you made. Is there a specific number/size or rotation of the spring cuts? Also, I saw a warning at the beginning of Mach3 demo that said you must reboot before running the program. That may have caused the issues you experienced at first.
As a comparison, I went with the cheapest hardware I could find. We'll see how it fairs. So far, the seller has been very responsive and supportive on questions I have had.
Good luck and Happy New Year!
 

Hawkeye

Active User
H-M Supporter - Gold Member ($25)
Joined
Jun 17, 2011
Messages
1,885
The cuts on the couplings are arranged in pairs cut from opposite sides. One pair is at 90[SUP]o[/SUP] to the other pair. The cuts stop about 1/4"short of coming out the other side.

My couplers are a fair bit larger than the commercially available ones, due to the 14 mm shafts on the oversized steppers I'm using. They are 1 1/2" long by about 1 1/4" diameter.
 

Hawkeye

Active User
H-M Supporter - Gold Member ($25)
Joined
Jun 17, 2011
Messages
1,885
I've been locked out of the site for a week due to a technical glitch - mostly fixed now, thanks to Nelson and Tony. Time to add to the build record.

I got the ballscrews done. Each one required a different mount for the ballnut. An interesting exercise if you don't want to buy the pre-made kit.
PC310721a.jpg

The X and Y motor mounts are finished. I decided to buid them up out of steel because I didn't have large enough blocks of aluminum and didn't want to risk casting in the winter weather.
P1010728a.jpg
P1020729a.jpg

I cut and spliced the two way covers to make one long enough to do the back section. I'll have to make up something for the front part later. The light spring works to pull the cover up as the table moves to the back of the ways. Without it, the cover becomes pinched behind the table and blocks full movement.
P1020730a.jpgP1020732a.jpg

I added a pair of microswitches to use as homing limits so the computer can set the table to a consistent 0,0 position. Sorry about the poor focus.
P1060736a.jpgP1060738a.jpg

Since I extended the Y-axis about 2", I wanted to have the centre of travel close to the centre of the table. This required extending the spindle about 13/16" out from the column. I milled some aluminum to thickness on the Victoria and drilled it to take some 3/8" bolts in place of the original 8mm ones. I retrospect, I should have just milled the counterbores deeper and used the original bolts. It would have saved at least a day of shop time.
P1130746a.jpg
P1150747a.jpg

While the head was off, I used the old plastic gear to mount the magnets for the MachTach tachometer sensor and mounted the Hall Effect sensor inside the head.
P1120744a.jpgP1120745a.jpg

Today, I finished the bracket to mount the Z-axis ballnut. It is mounted where the quill handle used to be, using a custom-fitted expanding shaft to take the majority of the force and a couple of screws into existing holes to hold position. The ballnut will bolt into the large hole in the end of the bracket back beside the column.
P1190749a.jpgP1200753a.jpg

So now you're up to date. More to come as it happens.

PC310721a.jpg P1010728a.jpg P1020729a.jpg P1020730a.jpg P1020732a.jpg P1060736a.jpg P1060738a.jpg P1150747a.jpg P1130746a.jpg P1120744a.jpg P1120745a.jpg P1190749a.jpg P1200753a.jpg
 

Hawkeye

Active User
H-M Supporter - Gold Member ($25)
Joined
Jun 17, 2011
Messages
1,885
Re: X2 CNC Conversion - It's Alive!

I got the Z-axis installation done today, as well as replacing the old spring support with a gas strut I've had for some time. I've run the three axes using the pendant. Smooth and quiet. I still need to build a tray for the keyboard and mouse, and move the 3-ton arbor press somewhere else.
P1310772a.jpg

The MachTach is ready to go as soon as I get an enclosure built. I really should cut the display window on a CNC machine ... hmmm.

P1310772a.jpg
 

Hawkeye

Active User
H-M Supporter - Gold Member ($25)
Joined
Jun 17, 2011
Messages
1,885
I managed to get the homing function working last night. This morning I drew up a simple shape and ran it without any tooling, just to see how the mill moves. I had to retune the motors to get them to run smoothly, now that there's some resistance on the motors. I took a couple of videos, but this software doesn't support AVI files, that I can see.

I'm having a problem that will keep me from using the mill to cut anything. Rapid returns don't work. The table moves through all it's paces (I haven't measured yet to see if it's accurate.), but when it comes time to move to the start point, or to retract the bit on the Z-axis, it gives a quick twitch, but doesn't move very far. Then it goes on to the next operation. In this morning's test, the head ended up a few inches lower at the end than at the start. Guaranteed table crash.

Any ideas on what to tweak? I noticed this when the motors were sitting on the floor in my living room, but figured it could wait until the machine was finished. I guess I could try editing the G-code for normal (slower) feed rates on returns, but I'd rather have full functions.
 

jumps4

Global Moderator
H-M Platinum Supporter ($50)
Joined
Apr 30, 2012
Messages
2,236
it sounds like you have the motors tuned wrong if you send me a copy of your "mach3mill.xml' file in the mach3 directory i will check your settings
i also need to know what you set the drivers to for microstepping, amps, motor sizes and power supply voltage
it is easier than going step by step in here typing
steve
 

Hawkeye

Active User
H-M Supporter - Gold Member ($25)
Joined
Jun 17, 2011
Messages
1,885
Thanks, Steve. I think it'll have to wait until at least tomorrow. I picked up some kind of bug yesterday in a very old dust-filled attic full of dead mice and wood-chip insulation. Feel like I got pulled through a hedge backwards.
 

jumps4

Global Moderator
H-M Platinum Supporter ($50)
Joined
Apr 30, 2012
Messages
2,236
hi hawkeye
I sent you an xml file named hawkeyemill.xml to use to run your mill and its pulse per inch should be really close.
before calibrating with the autoset function in setup screen you need to set backlash compensation.
there are a few different methods but here is how i do it....
first you need to adjust the gibs so they are snug but not too tight
back off the gib untill it is loose and in step mode set to .001 per step, move the axis 1 step at a time untill you see a movement on a dial indicator against that axis. when it moves on the indicator zero the indicator. now move the axis the other way one pulse at a time untill you see movement and count the numbers of steps you made untill it moved and write it down -1 the last step made it move and we are looking for the steps without movement, or the backlash.
now slowly adjust the gib tighter until you see the steps required to see movement go up by one , this shows that the gib is just starting to bind the axis. back off just enough untill it returns to the original amount and lock the gib.
repeat this for every axis. on the z-axis make your measurments moving in the up direction it wont work moving down because of the free weight of the head takes out backlash.
now take you measurements you wrote down and enter them into each axis box in backlash compensation and remember to enable the compensation and leave the compensation feed rate at 20 percent.
dont forget to save your settings ( config pull down bottom of list " save settings )
now you can use the auto set in mach3 setup and backlash will be included in the settings.
steve
 

jumps4

Global Moderator
H-M Platinum Supporter ($50)
Joined
Apr 30, 2012
Messages
2,236
also in the xml file i set the max axis movement speed to 20 inches per minute.
your machine will be able to move much faster than that but 20 is a good starting point for beginning to learn cnc and g-code because it will give you more time to hit estop or space bar to stop if something is wrong like forgeting to zero the axis before running the code.
seeing the axis take off at 100 inches per minute in the wrong direction or straight down towards your table is exciting to say the least.
and because of stored energy in the driver and mach3 having to process the command the stop is not instant using a breakout board pin for estop or pause.
steve
 

Hawkeye

Active User
H-M Supporter - Gold Member ($25)
Joined
Jun 17, 2011
Messages
1,885
Thanks, Steve. Once I got the file in and sorted the odd thing out, it started moving accurately - at least from the point of view of the scales. I'll do a DTI test soon.

Too many dry runs before first cut. I forgot to turn on the spindle first. :banghead: I'll have to look into adding spindle control to the system. Anyway, aside from a ragged entry hole, it was impressive to watch the foam fly.
P2060785a.jpg

I didn't set it up with tabs. That kind of detail will come soon enough. This isn't any particular part. Just an example I drew up for test purposes. If I'm bored, I might cast it anyway, just to try out lost-foam casting.

P2060785a.jpg
 

jumps4

Global Moderator
H-M Platinum Supporter ($50)
Joined
Apr 30, 2012
Messages
2,236
looks pretty good
are you going to try placing that in the sand or cover it with thinned drywall mud and remove the foam with lacquer thinner after its dry?
i made a few castings with placing the foam directly in the sand but i havent tried coating it yet
i tried the green foam they stick plastic flowers in and it didnt melt, it insulated so well the molten aluminum sat on top of it lol
if you need any more help just ask
it looks like your on your way
steve
 

Hawkeye

Active User
H-M Supporter - Gold Member ($25)
Joined
Jun 17, 2011
Messages
1,885
Steve, I'll go half-way between the two. I'll coat the form with drywall mud and set it into dry sand. My casting bucket has an air agitator already, so sinking the mold should be fairly easy. I'll let the foam burn out with the pour. Outdoors. Out of sight of the strata council. Standing upwind.

Getting ready to go down to the shop and let the mill cut the front panel for its tach. I actually thought of casting an aluminum box for the tach, but welding up an ABS one is so much faster. Especially considering the time you need to let the drywall mud dry completely.
 

jumps4

Global Moderator
H-M Platinum Supporter ($50)
Joined
Apr 30, 2012
Messages
2,236
I have tried pouring into foam uncovered with mixed results. i havent tried the coating method yet but it has to work better as long as it's good and dry.
i'm looking for a kiln to try for drying the molds and melting before i start casting again. i missed a nice one two weeks ago for $125
steve
 
[5] [7]
Top