Time to jump in the water.....

Greetings,

I chatted with a few people over at the practicalmachinist site awhile back. I know Big John was one of them. They steered me in the direction of LinuxCNC. It is time to actually start using it.

I have a lot of experience building the mechanics of various control systems. I have never done the actual controls. I now have three projects I want to CNC. I am looking for advice on my which should be my first project.

One is a manual Birmingham 12X36 Lathe. I don't expect this to be a volume production machine. I just want something better suited for my prototypes. I tell it where 0,0 is, take off this much, this far, and this fast, and it does it. Threading and tapers would also be very nice. Just having a DRO would be nice. A gang setup may work, but I would probably get a better lathe. I am thinking I can do this for $1500 with VFD.

One is retrofit of a 2-Axis Bridgeport Eztrak. This would involve changing out the controls and adding a Z-axis. I believe I can sell most of the old working components, but maybe not.

On both the above, I would like to keep the manual handles but this may not be needed with the right control software. Eztrak is not the right software.

The last one is 3-axis CNC engraver. I do not have a huge need for this as I use the mill, but I thought it may be a good first project. The idea would be to use a high speed spindle from ebay and Nema23 motors to move it around. This would be used for general part marking. I think I could do the electronics on this for $500, but it may be less than robust components.

I do not want to start on the mill as the first project. I just cannot afford to have it down that long if there are mistakes. The lathe is a little more forgiving since there are two axis. I figure I can work out the controls off the lathe and install when I am comfortable. The lathe would give me the most payback. The CNC engraver is a nice to have. I see it as a minor challenge mechanically. It would allow me to get my feet wet slowly. Personally, I don't mind jumping in. Drowning is not fun.

Thanks for any advice,

Bill

Welcome to linuxcnc Bill .
dont worry there are plenty of us with Lifebelts to throw in your direction .

take a good look around and get comfortable then just ask any questions that your unclear at etc

i would suggest you look at JT's pages at www.gnipsel.com, and in particular his 7i77 tutorial , along with his BP conversion files etc , but no doubt you have already been
pointed in that direction

andypugh wrote:

willysnowman wrote:

I do not want to start on the mill as the first project. I just cannot afford to have it down that long if there are mistakes. The lathe is a little more forgiving since there are two axis. I figure I can work out the controls off the lathe and install when I am comfortable. The lathe would give me the most payback. The CNC engraver is a nice to have.

I think I would suggest starting with the engraver, as it is least critical and will get you used to the process. Your second conversion is likely to work a lot better than your first.

It took a few iterations to get my lathe/mill combo working. You don't want to be trying to decide whether to make parts or upgrade the CNC of a working machine.

Yeah, I use the lathe several times a week. It would be tough to be without it for a week. I forgot how I was able to get by without one? On the other hand, it would be great to have it CNC. There are lathe jobs I pass up because of no CNC. I just design my own stuff simpler which is actually a benefit of manual lathe. You find ways to simplify the manufacturing. It just does not look as bling.

I will have to do a little more planing and compare cost and time.

Can you take a look at the overview below and tell me if this is correct? I can then figure out my components.

For the mill, I will be adding a servo. For the lathe, I will add steppers or servos or combo. For the engraver, it will most likely be steppers.

yes you have that correct Bill ,
but just to clarify a few points , the breakout board for steppers is a 7I76 and for servos is a 7I77
if i may suggest go for Leadshine stepperdrivers or geko's ) rather than the all in one chinese drives , far more reliable
and of course you dont have the problem if one drive packs up they all do ! .
it actually makes it easier to setup and control .

if you use these across the board it will make it easier to swap out in times ( hopefully not ) of any problems

also go for the drives and power supplys capable of 48v - 60v rather than the 12 - 24v range
and where needed a 12 - 24v supply for contactors or low volts relay switching

p.s glad your using solidworks ! , i have hsmworks also
although i tend to use camworks more often

Dave

There are lots of free apps too. Here are a few

www.hongkiat.com/blog/25-free-3d-modelli...you-should-not-miss/

And mesh cam is a decent freeform cam app for $250 or so.

No reason to break the bank before you decide this is for you or not.

But really you should buy a book and learn to program your own. Its not that hard and your programs are much more controllable and SMALLER than any cam produce would produce for you.

jimmy111 wrote:

There are lots of free apps too. Here are a few

www.hongkiat.com/blog/25-free-3d-modelli...you-should-not-miss/

And mesh cam is a decent freeform cam app for $250 or so.

No reason to break the bank before you decide this is for you or not.

But really you should buy a book and learn to program your own. Its not that hard and your programs are much more controllable and SMALLER than any cam produce would produce for you.

Thanks for the info.

I have been using SW since '97 and already own a copy so no money spent. I tried others an quickly give up. I am too entrenched.

The HSMWorksExpress is a free program. It actually works pretty well for what I do (2.5D). I would highly recommend it. I do have to change my code every program since I do not have a Z. I also have been modifying them because of my lack of knowledge of HSMWorks, but I am doing this less and less.

I have read the "CNC Cookbook" (book not website) many times. I would recommend this as an easy read with a good general overview.

For me, it is a matter of improving my business. I already know I love machining. I just need to make the machines which fit in my basement more productive.

Bill

I started to pick a computer this morning. I thought I would just search for what others are using and copy. I keep finding most of these CPUs/motherboards from just this year are obsolete.

I have found the latency list: wiki.linuxcnc.org/cgi-bin/wiki.pl?Latency-Test Do I sort this by least max, jitter and go from search for a source for the top ones?

A lesser option would be to buy one and run a latency test. If it is no good, ship it back. Is this SOP? I have a small list of some which may work??

I guess this also would be a good time to at least think about my power supplies and communications. This will be a stand alone computer dedicated to linuxcnc. I will need a way to talk to it (transfer G-code files) with my windows computer. I will also need a way to control linuxcnc at the machine (touchscreen,joystick, computer keyboard?). I am sure this will become clearer as time goes on, but it would be good to have it included in my power supplies spec. I would like to go MicroATX or smaller, but size is not a driving issue except on engraver.