[defrag010]

I bought a HF 7x10 mini lathe earlier this year and use it for all kinds of stuff. I cut everything from polyurethane to titanium on it and have had good results.

I'm an automotive machinist by trade, and i'm wanting to mess around with different shapes of valve heads that will work good with my porting... but to do this I realy need to convert to cnc. I will need to be able to do dynamic radii on the valve heads (starts out as a certain size radius and gradually changes to one of a different size in the same cut) and undercut the stems at the head. No way I'm going to be able to do this manually. Take a look at this picture of a valve and you will see what I'm talking about. the plan is to use a section of delrin rod to make a valve with test profiles and see how they react on the flowbench, and once I have found my ideal profile start to cut them into actual valve blanks to build the heads with.

I am apt enough to get the mechanical part of my cnc conversion down, but I'm pretty much clueless as to the electrical/control part of it. I've been reading these forums for a few weeks and have picked up alot of good info, but am still too "green" to really go anywhere with it. I don't know what my ideal setup would need. I need it to be robust enough to machine stainless and titanium and the like. I've got a ballscrew setup laid out for the Z and X axis.

I'm having trouble completely grasping the meaning of the following: steppers, servos, dual shaft, bipolar, gecko, timing belt vs. direct drive, among others. My learning curve is steep once I get the basics figured out, but the basics are where I am having problems.


Basically, what kind of motors, drive, and control do I need ?

Thanks for any advice/help.

[RotarySMP]

Hi Defrag010,

I did my 7x12 a few years ago.

http://www.wrathall.com/Interests/interests_index.htm

For the controls I would recommend you use

2x Gecko G251 (these were not made back when I did mine)
1x Keling KL- 4875 Power supply
2x Keling KL23H276-30-4A
1x Keling C10 breakout board.

You could probably fit all that into the space left by the Gear train.

If you buy a Homann speed controller, you can also control the spindle from the CNC software.

You can get break out boards from lots of vendors, also Vexta motors could be used, and a PSU you can make yourself, so I have just used Keling as an example here. This stuff is pretty cheap, so rather than mucking around trying to save the last couple of bucks, making one order with an outfit like Keling will probably save you a bunch of time.

For software, I use TurboCNC as I am running it on an ancient Pentium laptop. If you have space for a desktop PC, then Mach 3 seems to be the most popular control software. EMC also has a strong fan club, but will need a bit more computor skill to set up.

[defrag010]

Thanks!! That's exactly the kind of information I have been looking for!

That spindle speed controller sounds Great. Which model would work with my mini lathe? there are a bunch to choose from on their website but I have no clue which one would work for me. Is there a way to wire one of those up to where I could still use the spindle speed pot to manually control the spindle if I was doing something manually?

I forgot to ask about limit switches also.. I understand their purpose, but where do they fit into the grand scheme of the conversion?

[RotarySMP]

I don't have limit switches on my mini lathe. These low power steppers will simple stall in the case of a hard crash, they don't have the torque to strip a ball screw.

As I use my lathe for all purpose hobby engineering, I am constantly swapping between the three jaw chuck, the four jaw independant, the face plate, the ER-32 collet chuck, etc so a limit switch would need to be adjustable for all these different -Z limit positions.

The DC-2XL is the one that was designed to interface with the chinese DC controller. Yes it has connections to control the spindle with a manual Pot.

I am also a mechanical guy, but usinga break out board with Gecko's make wiring things up a no brainer.

Some people buy double ended steppers and then put hand wheels on the back of the motor so they can still do manual machining. I didn't and am glad of that. My Mill is not yet converted, and I find turning the wheels such a pain after just telling the lathe where I want it to go.

One of the really huge advantages of a CNC is that you can put the thing in an enclosure and keep all the crap from flying everywhere.

Because I didn't have a mill when I first did my conversion, I went with timing belts in a 2:1 reduction. This made motor alignment much easier (and since I only had a 2A driver back then, my motors are not that powerful). Direct drive probably matches these more powerful Keling motors better to the load (Your rapid would be faster).

If you have access to a mill make up some ball screw bearings and mount the motors like Steve did (http://www.5bears.com/cnc04.htm).

That being said, I like how my X Axis motor is folded down out of the way.

[RotarySMP]

I meant to ask, How are you planning to do the X Axis ball screw? I designed a ball screw installation around rolled ballscrews, and one installed on Z, but decided there wasn't enough space to install the screw I bought for X, and that a rolled screw without an anti backlash system is no significant improvement on the metric trapaziod (ACME style) screw I am currently using.

Bellmouths here http://www.wrathall.com/Interests/interests_index.htm I turned up on the mini lathe.

[RotarySMP]

How are you planning to do the X axis ball screw?

[defrag010]

I'm going to ditch the stock location and mount the X axis ballscrew on the outside of the chuck side of the carriage. I know I won't be able to bring the carriage all the way past the chuck, so that dead space between the carriage and gear housing when the carriage is as far left as it can go is where I'll mount the setup. I am going to take off that piece where the stock X-axis screw is attached to the crank (dont know what its called) and make my own piece with alot more relief so I can slide the cross slide farther towards me and have a larger diameter capacity.

I'm also going to can the compound slide and just make a spacer block to mount in its place for simplicity... and because my compound slide moves around alot. I'm going to do the Z-axis just like blades did in his project but my X-axis will be totally different.

I was kind of leaning toward a direct drive on the X-axis, but I didn't know well it would actually work.


So, you can use the cnc software to "jog" each axis by a certain incriment with a key stroke? Like, press (whatever) key to jog X axis in/out at (whatever) incriment? If that's the case, then I guess I won't need to add hand controls or a manual spindle control.

[RotarySMP]

When I replaced my cross slide with a T-slot table, I also increased the X travel. Can be really handy.

With a CNC you don't miss the compound slide at all. A block of steel is far more rigid

I put my Z axis motor at the other end, as then you can have all of the wiring going to the head end of the lathe (First photo below). After I burnt out the cheap chinese DC motor, I re-engined my lathe with a 1hp three phase motor running off a single phase to three phase invertor. I wired this up to the original direction/off switch and speed pot (The Homann speed controller weren't made back then). I have been meaning to get a speed controller and integrate this into the controller. The Chinese DC motor controller is in a box somewhere, the controller cover on the front of the lathe is currently empty.

Direct driving the X will work fine, it is just not quite so compact that way. Since I only have a corner of my inner city balcony as a machine shop, compact was important for me. Those modern Keling motors with 3A Gecko G251s will easily direct drive both axis.

The CNC software offers multiple chooses to jog. You can use the MDI (single line of code) interface and tell the machine where you want it to go to:

G0 X25; Machine will head of at max speed to a diameter of X 25mm.

You can use the jogging interface to move using keystrocks either for as long as you hold the jog key down, or a fixed distance per stoke (where the fixed distance can be easily changed with a single button push between say 1", 0.1", 0.01", 0.001", 0.0001"
(This is implemented in TurboCNC, so I am sure Mach 3 and EMC will have at least this and even more sophisticated options).

You can also implement an MPG jogging wheel (see here http://kelinginc.net/BreakoutBoards.html) so you can precisely manually control the movement of an axis with a wheel rather than a jog button.

Putting handwheels on a powered axis has the following disavantages:
Danger, spinning thing to hit your hands, pick up swarf etc.
Knotchy movement when using steppers.
Every time you touch it, the machine controller loses it's position, as steppers are open loop systems.

I personally feel no disadvantage of making the machine CNC control only. YMMV.

[RotarySMP]

There is a good point made in the Blade Thread. You can't use the stepper bearings to control the leadscrew thrust, as they have not thrust bearings.

Use a pair of AC bearings on one end of the leadscrew to control thrust. The link to 5 bears bearing block how to is in post #4. I didn't use AC bearings in mine, just normal deep groove ball bearings. Just one of a less than optimum features in my first CNC build

[RotarySMP]

Oh, and you can never have enough QCTP tool holders. I bought mine from Shars.

[Crevice Reamer]

Hi defrag. welcome to the Zone!

I recommend these electronics for best performance and easiest installation:

$289 1) Gecko G540
$098 2) KL23H2100-35-4B 281 oz motors
$050 KL-300-48 48V/5A PSU or $60 KL-350-48 7.3A PSU
--------
$437

You may find some of this info helpful:

http://www.cnczone.com/forums/showthread.php?t=78713.#3

CR.

[RotarySMP]

Why pay $289 for a G540 when you only run two axis? For 2x $69 for G251's and a $30 you have exactly the same functionality for half the price.

[Crevice Reamer]

Why pay $289 for a G540 when you only run two axis? For 2x $69 for G251's and a $30 you have exactly the same functionality for half the price.

You can of course do it that way, but I think it's false economy. Using the G540 gives you ALL of its unkillable features, built in speed control, provided connectors, a built in breakout board (and BOB PSU) with most of the wiring done already and much easier installation.

The G540 does not even require a separate box to hold all of the components--It can be mounted INSIDE the computer case, or in an out-of-the-way location. You get all of this for just a little more than what the individual parts/shipping would cost--AND it comes with two free $61 spare drives.

G540, $289. Getting your machine up and running quickly and reliably--Priceless!

CR.

[blades]

Hey defrag, welcome to the zone! Glad to see another 7x conversion!

Just a note on the below point: while it is certainly a valid point about the thrust bearings, you have to keep things in perspective with regards to size and presumed actual applied force to the leadscrew. I have zero measurable end play in the motor shafts, and that's pushing/pulling quite hard on them. Zero.
Longevity is yet to be seen, but I think a bearing supported leadscrew on my 7x project would have been redundant, especially for my intended use. I have not had the motors apart (nor do I intend to), so I really can't say for sure how they are supported with regards to end-forces, but it would be interesting to find out. Too bad Keling doesn't supply full data sheets with their motors.

There is a good point made in the Blade Thread. You can't use the stepper bearings to control the leadscrew thrust, as they have not thrust bearings.

[blades]

You can of course do it that way, but I think it's false economy. Using the G540 gives you ALL of its unkillable features, built in speed control, provided connectors, a built in breakout board (and BOB PSU) with most of the wiring done already and much easier installation.

The G540 does not even require a separate box to hold all of the components--It can be mounted INSIDE the computer case, or in an out-of-the-way location. You get all of this for just a little more than what the individual parts/shipping would cost--AND it comes with two free $61 spare drives.

G540, $289. Getting your machine up and running quickly and reliably--Priceless!

CR.

Which is exactly why I intend on using the G540. Plus, you have built-in expansion capability for whatever you can dream up to use an additional output for (tool indexer, powered tailstock, etc.).

BTW: where is this $289 price coming from? Last I saw it was $299 on their website. Would love to catch this puppy on sale!

[Crevice Reamer]

Which is exactly why I intend on using the G540. Plus, you have built-in expansion capability for whatever you can dream up to use an additional output for (tool indexer, powered tailstock, etc.).

Well said blades! There are even a few desirable features that I forgot to mention--like Idle current limiting.

BTW: where is this $289 price coming from? Last I saw it was $299 on their website. Would love to catch this puppy on sale!

G540 4 Axis driver, $299, $10 discount
when you buy with any components $289

http://kelinginc.net/GeckorDriver.html

Or $290:

http://cncrouterparts.com/index.php?...ffn19khecp85l2

Or $285 if you buy 2 from Gecko.

CR.

[RotarySMP]

Crevice reamer certainly has a good point. $150 is really nothing in the big picture of playing with machine tools, and that G540 would save a lot of work. I kind of like playing around wiring things up, but it is penny wise.

At the end of the day, the 7x lathes are never going to turn into a silk purse. The Tailstock is ****e, the motor is weak, the cross slide dovetail is pretty roughly machined, and the whole thing is a lightweight. I really like mine, but if I had space, I wouldn't have considered it.

If those Valves you want to turn are larger than 20mm diameter, you are always going to be stuck with one offs.

[Crevice Reamer]

At the end of the day, the 7x lathes are never going to turn into a silk purse. The Tailstock is ****e, the motor is weak, the cross slide dovetail is pretty roughly machined, and the whole thing is a lightweight. I really like mine, but if I had space, I wouldn't have considered it.

If those Valves you want to turn are larger than 20mm diameter, you are always going to be stuck with one offs.

Mark makes a really valid point.

I believe the Grizzly 10 x 22 is the most-bang-for-the-buck/useful lathe to CNC. Better to spend the money to CNC one of these:

http://grizzly.com/products/10-x-22-...al-Lathe/G0602

CR.

[blades]

Given the option, I would have taken a 8x12 over the 7x10, but space and weight are real issues. I have to be able to pick up and move the lathe myself, preferably without causing a hernia. If someone put enough time and effort into the 7x it could be vastly improved, but to paraphrase what Mark already said, you can only polish a turd so much.
When I used to hang out on the bench mill section here on the zone, I was dumfounded on several occasions at how much $$$ some guys would throw at an X2 mill when converting it! The X2 mill is about on par with the 7x lathe as far as comparable size and level of quality (I use the term loosely). Not bad for hobby stuff, but certainly not something that merits several times the cost of the machine to convert to CNC.

It's just too bad there are so many other misc. items in life standing in line that require $300 (or more), the G540 has been on the back burner for some time now. I think it won't be too much longer though.....

[defrag010]

Alright, I am about to order everything. I'm going with 2x g251's, 2x KL23H276-30-4A 's, a KL-4875, and a C10.

I've been reading the manuals and schematics and have a few questions.

-Will one C10 breakout board control both of the g251's with steppers? or will I need to get one C10 for each G251? I am going to power the C10 with PC USB power.

-Will I need to put a 3k ohm current set resistor on each G251 since the motors are only 3A and the self limiting current of the unit is 3.5A? Would this cause the motors to overheat?

I have pretty much everything down, except I'm unsure how to completely hook up the C10 board as the documentation isn't as thorough.