[MagillaG]

Greetings! I have been surfing the forums on here for a few weeks, and I have learned quite a bit. I am working on converting my Grizzly mill to CNC. Budget is a huge issue, but I hate buying stuff twice. But, a few Keling steppers and a G540 isn't in my price range. So, I will layout all of my assumptions and describe my plans. I am hoping I took into account alot of the vast knowledge on here. Any feedback would be appreciated- along the lines of "yes this will work," or "no way, dummy, you forgot XXX," or "This is Ok, but this may be an issue."

Basics: I am planning to use Sketchup for CAD/CAM and use the phlatboyz tool sto write the Gcode. I have an old Pentium PC witth more than enough power for running LinuxCNC. I am keeping the stock leadscrews (for now), and doing a 2 axis conversion (X and Y). Z will be done later. Ball screws will be later. My interest is more having a reliable, accurate system than a fast system. Given the mill I am starting with, It won't take much to be more accurate with teh electronics than the mill will support.

Hardware:
2 X NEMA 23 (23HS8630B) motors from here:
2 X TB6560 1 axis Drivers here:
1 X 5 Axis Breakout Board here:
1 X 36V 9.7A PSU here:

I think I calculated the Vmax correctly (~40v), and the motors are 3A, as are the drivers. Feedback? Will this work, or willI be miserable with it?

Thanks!
Ali

[baithog]

The power supply is adaquate, providing they weren't overly optimistic when they speced it. I am a retired EE. We generally used power supplies that could do 125% peak. So if your driver current limits at 3A, I would get a 12A power supply. And that is only if you are dead certain that you will never get a rotary table and need to drive a 4th axis.

I have one of the TB6560 3-axis boards that everyone was having problems with a couple of years ago. Mine has always worked acceptably driving 2A. I am close to finishing my X2 conversion, so it will be moved to driving 3A motors. If it chokes, I will be getting single drivers that can handle more voltage and currents up to 6 or so amps. The thing is, our machines tend to get bigger over time. If you can afford to spend the money on drivers up front, then you won't have to buy new ones later. I would get that board only if I was short of funds.

I looked at the motors you reference. 270 oz/in or so motors are what I believe Hoss used on his mill for X and Y. He went up to something like 320 for the Z because it is a heavy bugger. I did not buy that particular motor. I went with a US seller that has been around for some time. My Keling motors are only a few dollars more... of course they have a little shipping added too. On the plus side, I can call them on the phone and talk to a native English speaker who understands what they are selling. I have had very good luck with products from the far East, and some very bad luck at times. What has been consistent has been very poor communication and technical support.

Larry

[MagillaG]

Thank you, sir!
As an update, I have decided to shift gears a little. I am looking at spending more on the motors (as these are the most expensive part) and taking a risk with the drivers. Worst case, I have to upgrade the drivers later, but those are less than new motors, and can be replaced one by one. I'll update as I actually commit to buying stuff.

[baithog]

Since you haven't committed funds yet, may I offer some opinions?

I have seen quite a few start their conversion project by buying the electronics. I think that is a mistake. The hardest part for most hobby machinists is the mechanical execution. It requires a lathe, or access to one, as well as the tooling to go with it. You will also need a functioning mill and its tooling. You cannot fix a gotcha if the mill is in pieces on the bench. You also need the skill to effectively use the machines and the ability to over come errors, if not the foresight to anticipate them. I used the knee mills and lathes in the company model shop for years, but when I started my conversion, I was retired and didn't have access any more. I got a 7X14 lathe and cleaned and adjusted it. After hours of machining the Z-axis parts, I discovered a grievous error. I assumed that my 3-jaw chuck was accurate. It wasn't. The Z-axis looked like a stripper doing a pole dance when I ran it. I managed to salvage it at the cost of a whole lot of unanticipated time. A fellow named Murphy is the constant companion of people who build things. You will have at least one visit from him when you go about the mechanics. Be honest with yourself and buy a kit if you need to. If you go ahead with the mechanics by yourself, Then start with the Z-axis. It is the easiest one to retreat from.

People think the electronics are hard. Its just connecting some wires and run the program. It will work. On the other hand, poorly executed mechanical will leave you with a herky, jerky machine that is useless.

Larry

[MagillaG]

Ah, Yes.... Mr. Murphy is my constant companion. I do have the mill functioning and a lathe. My intention was to not render either one unusable (Isn't that always the goal?). My plan was to create the motor mounts, couplings, etc and dry fit everything along the way. Perhaps my naivete is preventing me from seeing it, but I am imagining the Z to be more challenging, given the rack and not a leadscrew. Perhaps you can enlighten me on the likely mechanical mistakes?

My thought is that the motor needs to be mounted with the motor shaft in-line with the end of the leadscrew, and a coupling to lock the motor shaft to the leadscrew. Is there more than that? Limit and home switches?

[baithog]

The Z first mainly because the X and Y still work while you are fitting the modification.

Do not expect that your dry fit will go on the machine right the first try. Some things can't be dry fit while the mill is in one piece.

You could just mount motors on the ends of the stock screws. You will find that you are fighting backlash and doing adjustments frequently. You will also have a machine that runs horribly slow. I see a lot of upgrading to ball screws by frustrated stock screw/nut users. I would think that you would at least want to fit a 2-start 1/2-10 and anti-backlash nut. I just priced out acme against import ball screws and the ball screws are nearly the same or cheaper. It doesn't make sense to me to do the upgrade 2 or 3 times when I can do it right the first time.

Forget the rack. It is way too coarse and sloppy for CNC. You will need to fit an acme or a ball screw to move it. There are a number of ways to go about that. Motor mounted to one side driving a screw through a nut attached to the side of the head. Another is to remove the rack and cut a slot in the head to clear a screw. A motor is the mounted above the column driving a screw that runs where the rack used to be. And another is to use a rotating nut above the head to move a screw that is fixed to the top of the head. The side mount pulls off center and either blocks the gib adjusters or the requires the counter balance be removed. Using a rotating screw above the head require disassembling the head to cut the slot. Retreating from that one also would require the rack system to be installed. The rotating nut method designed by Hoss is the most complicated to to machine, but it pulls from the center and leaves the rack in place until you are absolutely certain everything works fine. Even if everything goes together without a hitch, it is nice to be able to easily retreat. Motors and circuit boards fail at the most inopportune times.

Of course, this is all just my prejudice and opinion. I'm not a first class machine designer. You should read the thread on by Hoss. This one Hossmachine Homepage and this X2 Finished

Larry