CNC conversion BL250G/700 lathe

[MrCrankyface]

Random thread start but bought this lathe a couple of years ago, maybe 2019 sometime.
It's had a semi-rough life since then, I'm not too kind to my machines.
It's a pretty good machine, especially when you start getting everything tuned in and roughly scrape the surfaces(had a gib not lay flat and the saddle didn't have great contact).



HOWEVER ... A recurring thing for me seems to be breaking gearboxes in machines.
Don't remember exactly what happened but I think I was powerfeeding and really trying to push things.
A second of inattention and I hear the machine bogging down, gearbox is now under so much force that I can't disengage the powerfeed and a second after that I just hear a crack behind the apron handwheel.
A quick teardown reveals this poor guy:




Reseller didn't even answer my emails and only factory I could find wanted ~250 USD for a replacement, for what is basically a 10mm shaft with a gear on it.
I considered making something myself from shafting and premade gears but lowest gear count I could find was 12t and this is a 11t gear.
I don't have the tooling to make gears and didn't feel like doing that investment so I drilled out the center and loctited a long screw in there to hold things together.
This somewhat worked for quite a while but over time you could notice how the apron handwheel "slips" when you're moving the carriage, highly annoying.
Basically the shaft was acting as a clutch because of the broken shaft and loosening loctite.

So a lot like with my mill, I was at the crossroads of either investing a bunch of money to make my own gears or just CNC converting the lathe(which I already wanted to anyway since I more often run into repetitive lathe jobs).
Took some fairly rough measurements and ordered a 1204 and 1605 "ballscrew kit", basically all holders/bearings, the screw and the nuts.
No idea if I could fit it but impulse-decisions is my thing.

This was probably a year or two ago and I haven't had time to deal with it since it's mostly worked "good enough".
Recently however the lathe has been declining in how well it works due to my dodgy repair, so it has become urgent all of a sudden.
First of I started with moving the lathe, it's been sitting on a wooden bench, only bolted through it's narrow feet's through the plywood sheet underneath, with the studs offset quite a bit from the feet..
The metal tray does some work to spread the load but not much, the lathe was definitely rocking when the chuck was out of balance.


Living on that epic student paycheck I resorted to using scrap steel to improve the situation.
Starting with this massive piece of 12mm steel from some old press.


Lots of hours later I have it cut down to size and milled flat, now around 10mm.
I've also managed to tear off my back-cover on the mill because the plate got stuck on it when I didn't notice.
Fix 2 things, break 1 thing. Rinse and repeat.
Luckily it was a prototype cover design so no big deal, if it's that vulnerable I need to redesign it either way.


A whole bunch of drilled and threaded holes later.
The idea is to screw this into the studs below and the plywood to spread the load as much as possible.
The lathe with it's tray will then sit on top of this and get bolted down into the threaded holes in the plate.
Made two of these so each "foot" on the lathe gets it's own.


Lathe finally in it's new place, same bench but moved over to the much stronger section.
This let's me run longer stuff through the headstock and combined with the reinforcement plates makes it sit insanely solid now.
I can push it all I got with my 280 pounds with no effect on it, real happy so far.
Next we just got the small and quick project of completely redesigning everything so the CNC-things will fit.
SFU1605 screw along the Z-axis and SFU1204 along the X-axis.
Nema34 steppers for both, 8Nm for the SFU1605 and 4Nm(if I remember correctly) for the SFU1204.
Because the lead is quite high on the screws I will gear them down 2:1, optimally I'd want 3:1 but then the pulleys get so big and hard to place..


Really didn't feel like "trial&erroring" my way through this so spent quite some time in CAD to come up with this.


I've tried to keep destructive modifications to the lathe itself to a minimum but in 2 instances I had no choice.
1. The cross slide needed to be milled out to accept the taller/larger ballscrew nut, furthermore two holes were drilled so I can attach the ballnut from above.
2. Because the ballscrew nut is also wider than stock I had to mill out a few millimeters where the nut travels, while doing this I also lengthened the cutout a bit for more travel, guess I got greedy.


The larger motor(Z) sits on a plate which gets bolted into the original mounting holes for the gearbox which no longer has a use.
The smaller motor(X) was more difficult to place but gets hidden in behind a bunch of plates that attach to the original holes for the apron.


Easier to see the motor here, just enough clearance to fit it this way, much prefer this look to having it protruding out the front.


Regarding the nut that I had to mill out both the saddle and cross slide, that's after I've already butchered the nut as much as I dared.
Lower part is cut off, top part is ground down and the grease port is blocked off with a M6 nub that I loctited in place and also peened out the edges on.
A hole was then drilled with a masonry bit to instead get the grease port along the screw instead.
The smaller block(towards the "camera") has a small cutout for an o-ring which HOPEFULLY seals this up once everything is bolted together.
The 4 holes on top go up and bolt through the cross slide, but most likely I will only be using 2 of these.


The worst part of this was having to empty and refill the nut, so I could safely modify it.
The smaller hole infront of my finger is the new "grease port direction".
Finger, balls and m4 screw for scale so you get an idea of how much fun I had trying to get them back in the nut later.


Then it's just a matter of trying to fit everything onto the lathe and hope I didn't build in too many obstacles.
Starting with the X-stuff, it quickly became obvious that widening the slot in the saddle hasn't quite worked out.
I suspect the endmill has dulled because the slot gets narrower the further down you go, despite several spring passes.
I only designed in a clearance of 0.5mm per side so wasn't much room to start with.
Also might as well clarify, everything gray is 3D-printed placeholders before I spent a bunch of hours milling steel.


Besides the slot below being a bit too tight, the missing top cover plate and one of the nut screws not threading in, this first testfit was a great success.


The much simpler part is fitting the Z-screw stuff.
Basically no modifications needed to anything, just need to figure out measurements.
Motor and ballscrew bearing sits on a plate which bolts in where the old gearbox sat.
Could only find one issue here, that the motor can't tension the belt because the slots are not far down enough, easy fix though.
Also need to get proper length screws.


Ballscrew goes out like the usual screw does and has it's nut hidden behind the X-motor.


Because it's so long and heavy I want to use it's end-support.
Need to come up with a reliable way of measuring this in perfectly parallel with the ways.
I could probably put a dial indicator on the saddle or something and let it slide along the screw.
Once I have those measurements figured out I can make a plate to extend from the lathe to the bracket.


So that's where I'm at currently.
Hopefully it was an inspiring or funny read for someone.
If you see that I've done something stupid somewhere, let me know, I'm just a happy amateur.

 

[DavidR8]

Darn impressive in my books!

 

[markba633csi]

Good start on a never-ending project!
Your CAD pictures are very nice looking
I think aluminum plate would be plenty sturdy enough to replace the 3-D printed plastic stuff

 

[MrCrankyface]

Thanks! Trying to keep any machine alive for a longer amount of time seems like a never-ending project in itself...

Spent a while trying to figure out the screw alignment, only to get more confused.
The problem I faced is the ballscrew was hitting the sides of 3D-printed plastic but in CAD I have plenty of clearance.
Took me quite a while to realize I had placed the carriage too far up, basically acting like the saddle was sitting above the prism.


Having that issue semi-figured out I could continue making bracket for the floating bearing.
It can be shimmed along the length of the screw and out from the side of the lathe bed and then have slots to adjust a bit up and down.
Should be close enough and enough adjustment to nail this later.


Bonus pic of this great holesaw I used, or tried to.
I was using flood coolant, trying to cut ~3mm construction steel.
After having moved down 10mm and still not through the material I was quite confused.
Apparently the cutter was basically butter and just smooshed itself against the surface instead of cutting.
I need to stop buying no-name holesaws.

 

[MrCrankyface]

Well this took an amazing amount of time to get right.
Milled, bored, drilled and tapped a new ballscrew nut holder, needed some special stuff because space is so limited.


Better view of the second piece that redirects the grease port, sealed up with an o-ring.


Only the "big" block has threads so the screws clamp the entire package together.


Quick test-fit and seems to work nicely.
The screw locks up a bit when you tighten down the ballnut, but this makes sense since I made the nutholder a bit short so I can shim it in perfectly.

 

[MrCrankyface]

Little bit of progress, doesn't look like much but sure took it's time..
Mounted the encoder that will track the spindles rotations.
Connects with one of those flexible couplers to the layshaft next to spindle, hence 1:1 ratio between them.
Not a lot of space to mount things in there without making a very long bracket so took some time to bend and hammer out this sheet metal just right. Even had some good use of my experience from working on car bodies, luckily no one will ever see this bracket.


Encoder sticks out a bit so had to make a new hole in the side cover. Not optimal but doesn't bother me that much and is a lot less complicated than having to relocate the encoder somewhere else.
Even managed to get a really snug fit with some sanding drums.

 

[mattthemuppet2]

Super cool project. But I have to ask - what's happening with the Volvo?!?

 

[MrCrankyface]

Not much, for many different reasons.
1. Always something more urgent that needs to be fixed(ie lathe is broken and I need it for my company)
2. Garage currently semi-occupied storing family members stuff
3. Multiple uni-courses stealing most of my time
Etc etc etc.

I'm hoping to resume working on it in april once I have more space available, going to "pause" all other mini-projects on the car and figure out where I want my seats, steering column and shifter stuff so I don't build myself into a corner.

 

[mattthemuppet2]

Makes sense, life and shop gets in the way sometimes Looking forward to the next update and good luck with your studies!

 

[MrCrankyface]

As usual whenever I try to do something, it ends up like Hal fixing a light bulb.

I figured I should scrape things in now rather than later, because if I do it later all my measurements will be off and the ballscrews won't be straight etc.
But I can't fix the dovetails if I don't have an angled master, so out with my old camelback and try to scrape it flat. It was planed on a machine when I bought it like 2-3 years ago but haven't touched it since.
Some kind of horrible red ink that's impossible to get off your skin and maybe 20 passes in.
I've gone from two tiny lines, one on each outside diagonally across(twisted surface).
Took me hours just to get to this stage and it's way too easy to scar the surface by accidentally misaligning the scraper blade.
Doesn't help that it's some kind of super-stiff scraper I have, really hard to get a good cut.


I'm obviously don't have "biax-money" to spend so let's take a look at my trusty "sawzall" ...
What we have here is a yoke mechanism with way too much travel for my taste, eye-estimate of 2-3cm.
The "yoke-swivel" is a 7mm pin pressed into a round disc beneath, this then has a roller bearing and a hardened sleeve on top.
The front end also only accepts the thin "sawszall" type blades.


I think I had this apart in 15 different pieces at one point.
After I had taken out the "yoke-spinny-thing", I managed to knock the 7mm pin out and drill a new hole ~5mm from the center, should give me a throw of around 10mm.
Drilling this was no easy task, HSS drills were struggling a bit. I'm so spoiled from always working with construction steel where you can almost cut it with a butter knife.
I drilled the hole to depth with a 6.8mm drill, brotched ~25% of the depth to get the pin started and then pressed it in the last 75% to make sure it's not going anywhere.
Optimally I would've had a 6.95mm brotsch but this is not the kind of mini-project I'm going to order expensive tools for.
Gave it plenty of new grease as I started reassembling everything.


With the 'scrape length' problem solved I turned my attention to the mounting point for the scraper itself, this turned out to be even more of a nuisance.
The whole thing clamped in a clotted which was convenient, but trying to get a cut into it... Horrible.
I suspect I managed to work-harden the material because eventually it completely stopped cutting and I had to resort to manually going at it with an angle grinder..


Eventually got a proper slot in there and just barely managed to get a through-hole drilled, I was sure my drill bit was going to explode a few times.
Material was definitely "too hard" but I rather sacrifice a cheap HSS-drill than spend a bunch of time grinding up a masonry bit.
I did clean off a majority of the grease here before I started reassembling the panels, no point in having too much here that will just catch dust and act as sanding grit.


Sacrificed my Sandvik hand scraper since I hated how stiff it was.
Also really hard steel, carbide at my usual speeds and feeds were throwing sparks, but the finish was amazing.
The fit in the slot is a mild push fit and then a M5 screw to clamp it together.


Starting to really get somewhere now.
Definitely a steep learning curve how to use my frankenstein-power-scraper but the time and effort it saves is absolutely insane...
I need to try bending the shaft down to imitate the angle of 'real' power scrapers and also give the carbide a good sharpening, pretty sure it's quite dull so would be interesting to test it freshly ground.
Since pretty much entire right side was light on ink, and a big void top right, I changed method to "only" scraping high points along the middle and right, I figured it'd be less work to basically "tilt" the surface rather than "lower" the entire surface to pick up the surfaces that were missing.
This definitely sped things up during roughing but will now probably start lowering most red sections now to get a more even coverage.


Hopefully I can get this finished during the weekend so I use it on the lathe.