Hi,
I've had a lot of inspiration from this forum, so i think it is my turn to give something back. I've had an idea of building a large VMC for many years(decades?) and now i've finally come to the point where i'm building one. So this thread will contain the buildlog for my project, hopefully it can inspire others to follow through on their ideas. It is important to note that there is no major endgoal for this project. I have no vision of me making a living or whatever of this machine somewhere in the future. This project is all about the process, the learning, the development and the challenges. If this project should have a tagline, it would be "It's not the destination, it's the journey".
I've previosly made a gantry-styled router for full sheets:
... and here i've had my first experience with gcode, servo motors, endmills etc... It uses 2005 ballscrews, 25mm rails, clearpath servos and a Centroid Acorn as controller. I learned a huge amount on this project, among other things the importance of rigidity and mass distribution... I also had the plan to use this machine to help me making of the the next machine...
So i've spent a lot of hours drawing and deciding on a "knee type" or a "gantry type" mill. And i ended up with the knee-type working best for me. The rough design looks something like this:
Key information about it something along these lines:
- Height 196cm (my garage port is 200cm high)
- Weight: ~2500kg (~1700kg epoxygranite, rest is mostly steel)
- Controller: LinuxCNC
- Axis servos: CTB - 2,2kW on all axis
- Spindle servo: CTB - 7,5kW
- Spindle: Fuka FKS1201 with ATC and through-spindle coolant
- Rails: 35mm Hiwin RG rails
- Ballscrews: Hiwin 3010 ground C3
The decision making process for arriving on the above components are not based on clever math or elaborate spreadsheets. Rather i've been looking at how Haas or other professionals have done things and how other DIY projects have turned out. I started with the spindle and worked my way "backwards" through the mill. I did some calculations on the ballscrews, bearings etc, but nothing fancy. Then i've tried to make the best compromises for the project with a balance of "endresults" vs. price vs. time..
I've received the spindle and tested out the toolchanger and making sure i understood how the ATC and CTS works. Here is a picture of the spindle itself and my testsetup:
I'm yet to have to servos running, because i'd like to make use of EtherCAT, but i'm having a hard time getting it working in LinuxCNC, i'll get there, for sure, but it is a steep learning curve:
I'm almost done with the molds for the EG. I'm using phenolic plywood("formply"):
Similar Threads:
Very nice mate, you got a lot of work ahead of you, looking forward to seeing the end result.
Sent from my SM-G781B using Tapatalk
Watching with interest. Looks like a great build.
Axis travels?
That frame design is usually called a "C frame"
Knee usually refers to part of a Bridgeport mill design.
How much $$ was the spindle?
Thanks! I hope there is a lot of work in the project;-)
Ah! Yeah, i learn something new every day on this project :-)
Travel is going to be X: 650mm, Y: 450mm, Z: 580mm (max spindlenose to table: 680mm).....or thereabouts :-)
The spindle itself was 1600USD, pneumatic cylinder was 130USD, CTS rotary union(LX86V-244) was 440USD and shipping was 650usd. I talked to a Tracy Ding at https://ntphoenixtree.en.alibaba.com/ and she has been very helpful and professional through the process... I've been doing some minor checks on the spindle itself(runout etc) and verything seems well within spec, actually quite a lot better than spec as far as i can tell... i have yet to make chips with the spindle, but besides that i would recommend them...
Looking really good !
Do you plan to have double piece base ? there is visible separation line in base that is why i ask
Hello,
This is a weird but serious question: Would you be interested in making a smaller size casting for me? I would pay you of course and handle shipping to Portugal.
I have never done anything like that ; i am not a machinist (yet).
Why am I looking to have something custom made?
Because the floor of the only space I have available is not strong enough to handle heavy machinery and I am not allowed to make any changes to it.
So... the only solution is to make somethin small and light(ish)...
Thank you,
Jorge
Hello, Jorge
I have a couple of Epoxy Granite Machine Bases that were made by Intelitek for their Benchman and Prolight machines. The machines were designed for small prototype parts and for education in trade schools. The base weighs around 110KG without any components mounted. I have 2 assembled machines with the same base and the axis travels are X=305mm, Y=152mm, Z= 228mm. I purchased the bases in order to make a micro high speed milling machine from micro components but I only need one for that project and have an extra base if your interested. It’s a 1 piece base with precision ground steel inserts for mounting your rails onto. Message me if you are interested.
Regards,
Jose
Progress has been slow the last couple of weeks. The kids, the wife and me took turns being sick(not Covid, just a cold, but still) the last couple of weeks...
Have been struggling a little with EMI on EtherCAT when i enabled all the drivers, so i took the decision to make the "high voltage" cabinet for the 3 axis and the spindle. I've just put it on the wall for testing purpose:
...it helped a lot on the noise...actually it eliminated it completely and i was able to power on all drives without the EtherCAT bus acting up... so it was defently a EMI issue i had earlier...
Next step is to get LinuxCNC to run all the drives as i'd like...and later this week i'm going to finish the last inserts for the EG casting...
Yes, the frame consists of a lower part and a top part. This is the only way i can get the steel inserts machined afterwards. The inserts will be milled and then ground. After that the two pieces will be assembled and aligned...and then glued into place using a special epoxy...
I have the servo's running! LinuxCNC has a somewhat steep learning curve, to say it at the least! It took me a while to nail the whole EtherCAT thing i LinuxCNC, but now it works! It works! I'm high as a kite right now, this is AWESOME :-D
Hi Thmas- Well done. Looking fwd to your journey... Which Silimix where you going to use? Peter
Notes for my own reference and if others choose CTB servo and drivers and find this via Google or whatever. I had some trouble getting EtherCAT to work in the first place. With these driver settings i was able to get it working...
1. Remember to order the drives with EtherCAT modules(i had to order these afterwards and install them myself, easy peasy, but they might aswell install it during assembly)
2. Here is a list of registers and what i've found to work(some of them might be default):
A1.00 = 1 (enable advanced features, nessesary to set the registers below)
A1.02 = 2
A2.00 = 2
Bn.06 = 0
Bn.13 = 1 (depending on your cycletime in the controller. I've setup 1ms in LinuxCNC)
Bn.21 = 0 (depending on controller, but the driver in LinuxCNC behaves like Beckhoff)
Hn.30 = 0
Hn.33 = 0
3. To clear any encoder alarms after connecting cables, set Pn.05=0
And again for future reference, here is the manual + EtherCAT reference document.
Where did you ordered waycovers ?
Sent from my iPhone using Tapatalk
With these guys:
https://www.alibaba.com/product-deta...834288030.html
...haven't gotten them yet, so can't vouch for quality etc.... it is always sort of a gamble with these suppliers, but we'll see when they arrive...
Interesting, are those linear rail cooling bits on the z axis? I'm interested on what your idea is on how to make these. I have a slot in them and plan to partially squash a copper pipe on one end so its round on one side but flat on the other, like you'd do with heatpipes.