[jerber]

I hardly finished my conversion of a proxxon mill when my dad called and said that he got four pieces of INA linear rails with trains for free (sh*t happens I guess ).

Now I've got 3 INA kwve 45H (4 row bearing) and one kwe 45H (2 row bearing).
I think I can build a better machine with them than the proxxon, wich is a round column mill with alluminium ways.
I also wanted to make a better anti-backlash mechanism for the leadscrews than I have in my current setup.

The rails are 280 mm long and the trains are 139 mm long so I will get around 140 x 140 x 140 mm of travel wich is about the same as the proxxon in x and y and almost double in the z.

I've decided to go with a moving gantry design. I know that's not common for a metal mill but since it's so small I think it will work.
I like the fact that the workpiece is stationary.
Another plus is that I can mount the steppers "inside" the structure and make a real compact machine.

I made a semi-rigid a/b nut design using two nuts, made from acetal (delrin) gears, wich are driven apart by a spring and then clamped between two alluminium plates. I installed one on the y-axis and it works realy well.
If there develops any play I just un-, and retighten 6 nuts to eliminate it.
The leadscrews themself are trapezodial 12x3 stainless.
Like any homebrew cnc builder does at a certain point in his life I made my own tap from a piece of rod and it worked better than I thought.

I will try to update weekly with pics if real life doesn't get in the way.
I welcome your thoughts/critism.

[balsaman]

Pretty cool. My experience with delrin nuts is that altho they seemingly exhibit very little backlash, they actually compress under cutting loads. The result is that they have variable backlash depending on load which can not be compensated for in the software. In other words under light loads no backlash, under heavy loads, let's say .010 backlash, and everything in between. Put a dial indicator on the axis and push on the axis back and forth and you will see what I mean.

Eric

[jerber]

I did put a dial indicator on it and measured .05 mm backlash but I pushed reasonably hard.

I plan to use my current spindle, 300w and no endmills bigger then 6mm so cutting loads will be small. I don't use high feedrates/deep cuts due to the noise ( at least I try to :devious: ), I use the machine in my appartment.

I think it performs better than the split brass nuts I use in my xy-table (0.1 mm backlash).
On my current z-axis I use a split delrin nut and I have no problems/backlash with it (see pic).

In the future I plan to make my own spindle from a er16 collet holder I have, maybe then it's a good idea to upgrade to ballscrews. There's room for them.
But for now this is the most cost effective solution.

[balsaman]

Cool. I am sure it helps a lot that the nuts are quite thick. When I tried it my delrin nuts were about 1 thread diameter thick. You might have 2 diameters there at least, plus there are two of them. Good work.

E

[jerber]

Slow progress ...


I made a bracket for the motor and spindle of the Y-axis.
It will hold a nema 23 stepper wich is connected to the spindle by a timing belt.

I also finished the bridge section.
All the alluminium pieces I use are scrap pieces so there are some holes and scratches that have no function.
But so far I'm pretty pleased with the result.

Next job will be the column that supports the Z-axis.
I'll use a housing from a festo air cylinder for it.
It's an alluminium extrusion, 109 mm x 109 mm and 360mm long with a 100mm hole trough it.
It will be mounted on the train of the Y-axis.

[jerber]

Ok now it starts to look like something.
The collumn for the z-axis is mounted on the y-axis train with four pieces of M10 threaded rod and it's also clamped at the bottomn.

The plate at the top will hold the z-axis spindle and a nema 34 size stepper.
I'll probably have to reinforce it in order to prevent it from bending under the weight of the z-axis.
As you (maybe) can see the workpiece will be in front of the machine instead of underneath it unlike most gantry-style machines.

I dropped the idea of using my proxxon spindle and I"ll make new one from a ER16 collet holder driven by a 2/3 HP DC motor.
I already got a lot of parts for it so it shouldn't be to much work.

[jerber]

In the last couple of weeks I "designed" and made a spindle. It was more work then expected because I had to install my lathe first, it was still packed since I moved.

The ER16 collet holder has a shaft of 20 mm. It's hold by two A/C bearings in the bearing blocks and one radial bearing in the spacer.
The spindle will be coupled to the motor with a bellow coupling.

The label on the motor says it has an output of 500 Watt @ 4000 rpm.
But since this a chinese motor I think it's a little optimistic.

Next weekend I'll make the parts for connecting spindle and motor to the z-axis train.

[Kipper]

Looks nice :rainfro:

[jerber]

I've mounted the spindle and the z-axis leadscrew.
It took some time to make everything square but it runs very smoothly.

I planned to use a gasspring as a counterweight but I can turn the z-axis up and down with two fingers so I'll try it without first.

The next job will be the table that holds the worpiece.
I'll use a solid piece of alluminium that will be mounted on top of the extrusions that also carry the y-axis rails.
The top of this alluminium block will milled flat by the machine itself so it should be flat and square with the machine.
On this surface I'll mount a piece of toolingplate (I believe it's mic-6) with a grid of tapped holes.
But before I can do this I have to make the last parts on the old machine and then "transplant" the steppers from the old machine to the new one.

I can't wait to see the new one move.

[Jason Marsha]

Good work Jeroen, you will soon be finished. Do not forget to post a video of the machine in action when the time comes.

Jason

[jerber]

It has been a while since the last update but I'm almost finished.

The last couple of weeks I have been busy with the table.
I used a piece of scrap mic-6 plate of 200 by 220 mm.
I bored 49 holes into it in a 30 by 30 mm grid.
The plate had already a couple of holes in it but fortunately not in the way.
Instead of tapping the holes I counterbored the backside.
The size of the countebore is just smaller then a M6 nut so I used a M6 bolt to pull the nuts into the hole.

I also moved the stepper motors from my old machine to their new home and they seem to like it.
After some playing with the motor settings in Mach3 I get rapids of 500 mm/min on the x and y and 400 mm/min on the z.
This is more then twice then the speed of my old setup.
I only use my steppers at 24v and the torque drops at 100 rpm.
I can double it because I can run my PicStep drivers at 48 Volt, switching to gecko's for a higher voltage will even show more improvement but for now this will work fine.

The speedcontrol box for the spindle is mounted on the spindle motor.
It's the control and motor that came with my CH-350 mini-lathe (don't ask how I run that now :devious: ).
FIW : they're not made in china as a mentioned earlier but made in taiwan so the 500 Watt output is maybe a bit more "real".
As you can see in the pictures there's a lot of glue residence on side of the control box.
I'll make a plate with my own logo for it so I didn't bother to clean it up to much.

Today I'll mount the table and cut some metal (and try to make a video).

[jerber]

The machine is up and running.
I've got some issue's with the z-axis stepper but I think I got it sorted now.
Funny that it runs better with the coils put in series then parallel (bipolar).

My first project is a spindle light.
I made ring around my spindle with 6 white smd leds,
The leds are the same kind you find in mobile phones.

I drawed the pcb in autocad and used sheetcam to produce the g-code.
After I cut it and soldered the components on it I covered the top (the copper side) with hotmelt.

It works beautifully, there's almost no shadow at the tip of the mill.

Now I have a question for you guys.
I want to convert my lathe too.
But I've been thinking to just place the headstock of my lathe next/on the table of my mill instead of converting the lathe itself.
Big plus is that I also can use it as a 4th axis and it's cheaper.
Downside is that I can't use both machines at the same time and it takes more time to set up.
What's your opinion ?

[digits]

That really is a very nice little machine. Have you cut much metal with it yet?

[RotarySMP]

Great idea on the light.

[jerber]

Thanks for the nice replies.

digits:
So far I have only cut aluminium. All scrap pieces so I don't know the exact alloy.
I can cut alu with a 10 mm carbide endmill at a d.o.c. of 3 mm and a feedrate of 200 mm/min but I didn't pushed her (yes I decided it's a "she") any futher yet. Too much noise.

[Switcher]

Cool machine, good job on that light!


.

[digits]

Thanks for the nice replies.

digits:
So far I have only cut aluminium. All scrap pieces so I don't know the exact alloy.
I can cut alu with a 10 mm carbide endmill at a d.o.c. of 3 mm and a feedrate of 200 mm/min but I didn't pushed her (yes I decided it's a "she") any futher yet. Too much noise.

Wow that really is impressive - my X-1 can barely manage a 6mm mill 0.5mm deep at 250mm/min - nice to see that aluminium extrusion and THK rails are woman enough to cut aluminium!

[jerber]

Most of her strength are in the linear rails, wich are INA and not THK.
The rails is 45 x 45 x 280 mm so it's very rigid by itself.
Moment ratings are 2485 Nm and they're heavy preloaded.
The extrusions are simply used to hold it together.
The x-axis (the "gantry") rails is directly bolted to the Y-axis trains.
Only the column that holds the z-axis is realy a suporting structure.

There are more parts in my spindle then the rest of my machine.

I thought a X1 was capable of more.
I'm surprised, my goal was to build something x-1 like.
Is it the rigidty of the frame or is motor power ?

[digits]

Most of her strength are in the linear rails, wich are INA and not THK.
The rails is 45 x 45 x 280 mm so it's very rigid by itself.
Moment ratings are 2485 Nm and they're heavy preloaded.
The extrusions are simply used to hold it together.
The x-axis (the "gantry") rails is directly bolted to the Y-axis trains.
Only the column that holds the z-axis is realy a suporting structure.

There are more parts in my spindle then the rest of my machine.

I thought a X1 was capable of more.
I'm surprised, my goal was to build something x-1 like.
Is it the rigidty of the frame or is motor power ?

I'm pretty sure it's the motor power - 150W isn't much

[CountZero]

Nice machine!

Have to agree with everone else, the lights are impressive!