What are Reasonable Goals for Home CNC?

[Mid Day Machining]

I have a CNC in my garage because I needed something to do. As I said below, I had a real CNC shop until 2007 when I had a major health issue that could no longer be avoided.

I knew something was wrong, but I didn't know what, all I knew was I was tired all the time. So much so that if I carried in 3 bags of groceries out of the car on Sunday afternoon, I was done for the day.

I went to see my doctor and he sent me to a cardiologist. That led to an angiogram, and that led to cardiac bypass surgery, I had 5 cardiac bypass grafts and they replaced the aortic valve in my heart.

After surgery, my wife gave me an ultimatum. I could have her or I could have my CNC shop, but I couldn't have both. This was late October, 2007 and she told me I had till January 1, 2008 to decide. I sold the shop in January 2008 and made an attempt at retirement. That didn't work for me. Heck, I was only 63 years old and my original plan was to work till I was 70 and then decide what I wanted to do.

I made it about 6 months before I started pulling out what little hair I had left.

Believe me when I say you don't go from working 12 to 14 hours per day to not working at all and survive.

I have since decided that my funeral and my retirement party are going to be the same day.

I had a couple of part time jobs over the next year, but as the economy tanked, the part time guy was the first to go.

I tried unemployment, but that only lasted about 10 weeks, and then I was back to nothing to do and no money to do it.

I bought my Tormach PCNC 1100 in July 2011 and haven't regretted it for a minute.

Now I can work as much or as little as I want. I manufacture a line of running gear for remote control gasoline powered race boats, and I do take in SOME job shop work.

I don't have a lot of money invested in my Tormach compared with other machines. I paid $72,000.00 for my FADAL, and $37,000.00 for my Haas. I have less than $15,000.00 in my Tormach and it does everything I want it to do. Oh yeah, I just added a rotary axis to it, so now I have Just over $16,000.00 in it. And yes, it will do anything those other machines would do, and it didn't cost $50,000.00.

I know I said earlier that I don't take jobs that are over 25 pieces, but occasionally I get to look at jobs are 100 pieces. When that happens, I'll find a way to make 4 pieces at a time and that's like only running 25. It just takes a little longer to run 4 pieces than it does to run one, but that's OK because I have a 46 inch TV right next to the machine so I have NetFlix going all day when I'm working. And, when I get tied of Netlix, I have an internet connection there as well.

Now, back to the original question, DO YOU NEED CNC IN YOUR HOME SHOP? That depends on what you're trying to do. If you're doing engraving, then ABSOLUTELY. If you're making stuff that you only need to trim an end or drill a hole, then no you don't need it, but if you're trying to make a part with a shape or a surface, then, YES, you do need CNC.

I remember a time about 15 years ago, I told my dad if you don;t have CNC in your shop, you can't be competitive any more. His reply was OH, BS. I can do anything on a manual machine that you can do on one of those machines, and for a lot less money.

Now in defense of what he said, my dad was a tool and die maker, and a pretty damned good one at that.

Then one day he came back from my brother's shop and he told me "YA KNOW, IF YOU DON'T HAVE ONE OF THOSE IN YOUR SHOP, YOU JUST CAN'T BE COMPETITIVE ANY MORE". I thought, jeez, where have I heard that before.

I have said before, my dad had 2 male children, but he only had one son, and that wasn't me.

[SteveHGraham]

I am more confused than ever, and that's really saying something.

[Marty_Escarcega]

I am more confused than ever, and that's really saying something.

So, what do you want/hope to do with CNC? Do you have CAD/CAM experience? If not, be forewarned about that learning curve. Any machining experience? If not, you might grapple with feeds and speeds. I know I do!

I love working on/and converting the machines...which after my full time day job, leaves little time for learning the CAD/CAM end of things. I can just barely do it. But ya gotta keep doing it to get more proficient at it.

....dang day job keeps getting in the way of my hobby.....

Marty

[RET]

Hi Steve,

There is a lot of very good advice in this thread but you need some (preferably quite a bit) of machining experience to make sense of it. As others have said, you also need to know what you want to do with all this stuff once you've acquired it. I build live steam locomotives as a hobby, so I've always known what I want to do and over the years I've learned how to do it. Also, I'm a retired mechanical engineer and I find my training and experience help a lot in my hobby.

I started off in the 1960's with manual machines (South Bend lathe and a home made shaper) and I learned on those. Now I have a 13" South Bend and a "real" Bridgeport with a Sony 3 axis readout. Along the way I bought a Sherline turnkey CNC mill and learned CNC on that. While the Sherline was very good value for the money, it did have its limitations, so I rebuilt it into a bigger, more accurate machine with machine tool ballscrews and 1" dia. Thomson bearings & shafts for ways ($2,500.00 per axis for the purchased parts). Expensive, but as others say, you need to spend the money in the right places if you are going to have an accurate machine. I also made a rotary headstock for the CNC mill and that has turned out to be very useful since the Sherline software handles 4 axes to tenths.

Now for the parts I want to make, I'll use the machine that will do the job best. Often it will be a combination of manual machining to prepare the part and then CNC to do the curves and surfaces. Having the readout on the Bridgeport is a BIG help. With it I can locate holes accurately to tenths. The readout can go to half a tenth (one micron) but the mill isn't capable of that degree of accuracy. I used the Bridgeport to make the new version of the CNC mill.

One final word. Machine tools have to be big and heavy because they are designed to minimize deflection. The Bridgeport literally weighs 2,000 pounds and even with all that mass, if you put a dial indicator that reads tenths on the table with the indicator spindle on the quill and push the machine head sideways, you will see the needle on the indicator dial move a little. So don't expect the cheap, light machines to perform like the "quality" stuff (even Bridgeports are light for commercial work). The light machines just can't do it. CNC machines are even heavier built to further minimize deflection. You also need the experience on manual machines to be able to set speeds and feeds for CNC. Because its a hobby, I don't try to push the machines to the limit of their capabilities. Its easier on the nerves and the machines. For CNC, I use climb milling all the time, but I can't do that with the Bridgeport.

With sharp tools that are ground at the proper angles for the material being machined, the light machines can do a reasonable job using light cuts (five to 20 thousandths). They are adequate for hobby work, but they are totally unsuitable for commercial use. They are just too slow and not really capable of the accuracy needed.

I hope at least SOME of this helps.

Richard Trounce.

[RET]

Hi Steve,

I'm afraid what I said before may sound a bit daunting if you see it all at once when you are just starting out. The thing I want to emphasize is that all through my "journey" I had fun and I enjoyed what I was doing, whatever it might have been at the time. The fact that I learned a lot on the way, while helpful was secondary.

It doesn't matter if you are making live steam locomotives, or equipment to make your shop more effective, the important thing is to like what you are doing. That is the one thing that you will see that is common in all the forum divisions on this website, whether its Live Steam, CNC or whatever else your interest may be. There is a great feeling of accomplishment when you finish what you are making and it actually does what it is supposed to do! That is worth a lot and its what keeps all of us going.

The bottom line is "you have to enjoy what you are doing." I hope this helps. While it was germane, the previous post might have been too discouraging. Sorry about that.

Richard Trounce.

[magic9r]

If pure speed isn't your objective then home CNC can let you accurately produce interesting parts, I probably wouldn't have made this one if I hadn't had CAD/CAM/CNC.
I used a desktop mill to machine this foresight platform, the part replaced the screw-on compensator of an SVI Infinity and allowed the owner to fit the iron sights he wanted on the gun.
If you fancy a go jump in and scale that learning curve, it's a real buzz when you start producing nice parts,

- Nick

[SteveHGraham]

I am wondering what I can do to make my existing lathe more accurate. The backlash in the x screw is down around a thousandth, so I don't think it's that big a problem right now. The lead screw is where everything falls apart. I would not mind blowing a little money on a real screw if it would give me the capability to put the lathe to use for projects more difficult than tapers.

I spent a ton of money on the existing screw, but it looks like it's only good for projects where z precision is not important.

[magic9r]

Steve,
If you're using Mach3 then you have backlash compensation settings for the axes in the software, Mach3 backlash compensation works.

Have you tried the hot process for moulding your own nut in acetal/delrin to make a long nut & reduce backlash?

http://bbs.homeshopmachinist.net/thread ... e-easy-way

You could also add something to your Z allowing you to set the friction to hold position when the axis isn't being driven, using a separate bar and friction pads (think roller-coaster brake but with the caliper on the carriage and the bar fixed) rather than tightening up anything on the lathe will avoid increased wear.

Regards,
Nick

[SteveHGraham]

Thanks for the help.

I have used the compensation tool, but it's not that great.

The screw I have now is a very good, high-precision screw, and it has a nut that appears to be delrin. I bought the stuff, made the nut bracket and so on, and then found it wasn't accurate enough to justify the expense. I have read about the hot acetal method.

I can make curves and so on, but precision is another matter. I have read that the cheapest practical solution is the screws they sell at Home Shop CNC.

[hobgobbln]

When I converted my lathe I was having some issues with accuracy. Completely by accident, I changed the length of the pulse width for my X axis. The difference was amazing. I had to actually shut the backlash compensation off after that.

I have no idea if it will help you or not but its worth a try.

Griz

[DICKEYBIRD]

Steve I'd suggest that you focus on making some parts (hopefully simple ones) to learn more about the processes & software rather than focusing on the machine's problems. It should be able to repeat well enough that you can learn to compensate for the machine's errors as you go forward.

Most basic lathe ops will be repeatedly moving the tool into the work in the same direction so backlash isn't so much of a concern. Not knowing what you plan to make, I don't know what kind of precision you really HAVE to attain when turning fancy contours with direction reversals in mid-cut. A lot of them are done for aesthetics rather than a precise fit with another part.

As far as basic diameter precision goes, even "real" CNC machines many times have to be tweaked and you have make trial cuts to end up with a part that's on-spec. The material being cut, its diameter, the cutting tool and the SFM all have a noticeable influence on the final size, even if the machine is putting the tool tip precisely where you tell it to every time. The flexing of the machine under various load conditions affects the final size, hence the fiddle factor. It's just a fact of life.

I have a bunch of QC tools set up in the Mach tool table with all of them having previously cut test pieces and their exact offsets saved. When I start a new part, I go into the tool table and type minus .002" to .004" into wear DRO of the tool being used. That leaves the part a bit oversized so I can measure it and reset the wear number on the final pass to make it come out very close.

Precision just ain't very easy, for me anyway.

[Harold_V]

Precision just ain't very easy, for me anyway.

A lesson that can be hard learned.
I specialized in small, precision work when I was machining. I've always said that if you don't understand the procedure that leads to the desired end, you'll struggle endlessly. The skills required are practiced on a routine basis, so when they are called upon by need, they are well honed.

All of my work was on manual machines. The rules don't change when shifting to CNC. High precision comes only with greater effort, assuming one hopes to achieve that end beyond the circumstances of chance.

Your post was very informative. Thanks.

Harold