All.

Well typical of me to start another couple of projects before I got the other big ones finished, but hell I need to amuse myself!

Sometime ago I picked up a few Boxford TCL125 training lathes being sold out of a local School.   Needless to say I bought the lot, just to save for a rainy day

I was intending to build a couple of bespoke machines for a job which never materialised.   But lately had a couple more ideas.

Only one of the machines rill ran, and all of them had been vandalised by the schoolkids to some extent.   They were all a bit rusty too.

So this week I pulled one apart, and started modifying the carcase of the machine to allow for some useful swarf (chip) management.  ie, cut a bloody great hole in the bottom!  I will build a custom stand to sit the machine on.

I also removed the bed and cleaned it up, and have temporarily bolted it to a MDF board for testing.   Steppers have gone, and in their place a couple of DC brushed servos.  With The new Leadshine servo drives running them.   These drives run great, and although a naughty Gecko ripoff, they have been redesigned and I fear now they may just be a better product 

At the moment running on the original stepper brackets and belts/pullies.   The x axis motor will be swung over to the right on an angled mount later on to make room for a gang too platen on the crosslide.   Both motors/encoders will be protected one way or another.

But this hashed up test works damn well.   The little machine rips along with 5000mm/min (196") rapids.

Many things to do but a good start and if it goes well the others will get the same treatment.

Wayne.....

Ok folks.  Here's a video of the little bugger in action.


<a href="http://www.youtube.com/v/jqdNUaM0hKo" target="_blank" class="new_win">http://www.youtube.com/v/jqdNUaM0hKo</a>
 

Good job fdos. Any plans for cublocs on these machines?

Brett

Brett

They may very well get a cubloc each. depends on how much automation I eventually put on them

They should end up with magazine barfeeders, and chip conveyor, so maybe a Cubloc will be pressed into service to have that stuff.  But have to get the basic machine able to cut yet

Intended use is to put all the Delrin work I do onto these machines.  I do a LOT of delrin   Plus a couple of small mild steel jobs I do regularly.

Probably wont need screwcutting ability, but would make a useful testbed for something Brian has in mind.

Wayne....

Wayne,
Very impressed with the speed!
Do you have a collet chuck solution? Any pointers would be helpful.
I have a toolplate on my TCL at the moment, it is an improvement on the 8 station turret and easier to set up.
The job it is running on is a brass collet blank. Cycle time is 32 secs, total G00 time is 5 secs.
Steppers are nema 23 2.5A and G00 speed is 1.8 M/min. This is the maximum speed with Mach3 at 25K, steppers stall if I go any higher, I think I may have to go to Geckos if I want a faster G00.

Ian

I do have a collet closer solution.   Sometime ago I designed an ER32 power closer for the Emco's we run.   This I need to redesign for the Boxfords, and make a few improvements.  The emco's are shortly to be retired!

I can actually run the rapids at 7000mm/min no problem and probably higher.   But we did a little test yesterday, and the difference between 7 and 5 m/min on that test program equates to 1 second over the runtime!   So we will when time permits do more tests and find out the optimum slowest rapids and stick to that.  Anything that gives no real benefit is just worthless on such a small machine.

We are seeing it hit the 5m rapids often during that program though, even on some short moves.  but then the servo's allow much more sever acceleration.

I like your tooling plate, similar to how I intend to run mine.   How well is Mach handling the effective from and rear toolposts?  Tool offsets being used?

Any video of your machine running that part?   I'd like to see it.

Wayne....

Wayne,

I hand code on simple jobs. With the bar extended, on Mach3 screen, I set X=0 on the OD and Z=0 on the bar end. Offsets are then picked up by jogging the tools into position, read off the values from the Mach screen and entered into code. (Quick and dirty!!!). Final code is tweeked for accuracy. Code for this is only 22 lines and that includes small 45 degree chamfers to break the edges. This is not a big production job, I do about 100 of these collets a year in various bore sizes.
I will do a video in a day or two.
Your collet chuck solution will be very interesting.

Ian

Ian

Only just noticed your toolplate is a Boxford product.  I didn't know that they had thought about gang tooling.   All mine had Dickson QC toolposts on them.

We have been reviewing the collet closer design today.  I want to make it simpler than it is at the moment.  I think I have come up with a more elegant design.  But these things always need very careful testing.

Wayne....

Looks good Wayne, shame about the music
 Was looking at getting one of these Boxfords a while back but ended up with the Computurn, not much difference LOL

Hood

Pffft Hood.    Devo are my hero's!   Besides I'm doing some work for em

So for your entertainment from now on ALL my vids will have a Devo soundtrack!!

Wayne....