Sherline steppers

Home Forums TinyG TinyG Support Sherline steppers

Viewing 15 posts - 1 through 15 (of 18 total)
  • Author
    Posts
  • #9230
    TarHeelTom
    Member

    Have an old (maybe 10 years) Sherline mill. It is the version which does not include a PC, but has the stepper controller in a separate box, (perhaps 6″ square). This box has been a constant source of trouble. You mill for a while, then the box just quits driving the steppers. Power off, wait a few minutes, power on and it seems good to go for a while.

    It’s been back to Sherline, but it never does this in their shop.

    Considering putting the tinyG in place of the Sherline provided box.

    Has anyone here used the tinyG on a sherline mill, and does it provide enough power to the Sherling provided steppers?

    Thanks

    Tom

    #9231
    cmcgrath5035
    Moderator

    I have not seen references to Sherline Mills here.

    Is this the conversion kit you have:

    The motors don’t look like anything special, controller looks to be a Mach3 setup.
    Are they 4 wire motors?
    Any markings at all?
    What are the motor shaft diameters?
    NEMA17s are usually 6mm, NEMA23s 8mm or 1/4″ typically.

    Either way, they would appear to be tinyG candidates, that is obviously a guess. Given that the machine is ball screws, the torque requirements should not be huge.

    #9232
    TarHeelTom
    Member

    That looks like the kit I’ve got. From what I remember, they are either 4 wire or 6 wire motors. (Don’t know that I’ve ever had the wire connection cover off) There appear to be 5 wires going to the DIN plug, but the fifth wire seems to go only to ground, possibly a shield. Sherline’s web site says that they are NEMA23 steppers, but they drive acme screws, rather than ball screws. The original Sherline just doesn’t have room to install a ball screw(s). Everyone seems to want ball screws, but I’m only aware of one user who has ever found a way to install ball screws.

    >>The motors don’t look like anything special, controller looks to be a Mach3 setup.

    All the Sherline cnc products come with LinuxCNC (actually before they changed the name). The version they supply today is an ancient version. But that box is quite simple inside. It drives up to 4 stepper motors, but has no provision for limit switches, eStop, or anything else.

    One tragic thing is that no one at Sherline has any knowledge of cnc, and Joe, the owner, died recently. He was the only cnc expert. The website actually tells you up front that they offer NO cnc support at all.

    #9234
    cmcgrath5035
    Moderator

    LinuxCNC output to the Parallel Port Mach3 interfaces.

    A good guess would be standard 4 wire bipolar NEMA23s with a shield connection.

    I would continue to guess tinyG would work well, you’ll have to experiment a bit to get your ACME screws set up, but that is just a parameter setting.

    • This reply was modified 8 years, 10 months ago by cmcgrath5035.
    #9237
    Vex
    Member

    I’m actually in the process of converting a Sherline mill to Tinyg. If you have any questions feel free to ask.

    Sherline uses a 4-wire NEMA 23 stepper motor for axis control. From my machine it’s a double shaft on X- and Y-axis with a single shaft on the Z-axis. I’m working on swapping out the motors for slightly stronger ones and double shafts all the way around. This should allow me to use the machine manually without having to run the TinyG or even allow me to back off the work piece during an emergency stop.

    The cables running to the standard Sherline motors are 7-wire type+shield. 3 just aren’t normally used. I imagine if you wanted to run encoders it would make things easy. The wires are aluminum though; and if you want to eek out more efficiency I would suggest converting them over to copper–but that all depends on how involved you want to go.

    Similarly you don’t normally have limit switches on x- and y- (mine didn’t), and even then it only had one limit switch on the z-axis. It looks to be for a G53 Z0 command (home). I would highly recommend modifying your sherline to have limit and home on all axis. This will allow you to take full advantage of what TinyG has to offer and reduce the risk of crashing your mill.

    TinyG will run the Sherline mill, but it will require some forethought on your part. Currently integration of spindle control (using a 1/2 hp universal motor and the standard spindle head), is what’s going to cause the biggest hiccup for me I think. On my machine they use a PCM2000 (I think, it’s been awhile since I looked at it) to convert the AC voltage to a 0-48V DC voltage. I’m unsure if I can adjust the dash pots on that board to get the 0-96V DC voltage I would require to achieve the 10k rpm the motor is rated for. The rub is that it’s an industry standard controller and as such required a 0-10v DC input signal for CNC control (you can use the knob to control manual spindle speed, but has no feed back for RPM control). This means you’ll need to figure out a circuit or connection to alter the PWM output from TinyG for spindle control to a 0-10V DC signal. Alternatively you could get rid of the PCM2000 board and pick up a PWM motor controller and hook it up that way (or even get a VFD type motor and run that way). The field is really wide open for spindle motor integration and how much time and money you want to throw at it.

    #9238
    TarHeelTom
    Member

    The plug on the controller end of my stepper cables is a 5 pin DIN plug, but I don’t remember ever having the cables loose from the steppers, so really don’t know how many wires are in the cable. But it appears that only 4 plus ground are used at the controller end of the lashup.

    My Sherline has no limit or homing switches on any axis. But I would like to add them. Probably Hall effect switches. Not possible with the existing controller, as there are no inputs at all, only the stepper outputs and the parallel cable from the computer. Joe didn’t bother to add limit switches or homing switches on the Sherline as he was under the impression that running into the physical limits on the Sherline would not hurt anything. And he’s generally right.

    Spindle control is of interest, but not something I’m going to worry about. I have provisions for it on my Bridgeport, but it just never has gotten anywhere near the top of the stack. But getting 0 to 10 vdc doesn’t sound that hard. Seems to me that a pwm signal looks a lot like a DC signal with lots of ripple. What if you just put a diode on the pwm output, then a capacitor, then a drain resistor. Wouldn’t that look just like a 0 – 10 VDC signal? I may try breadboarding something like that this evening.

    But my question is still a simple one. Do the stepper drivers on the tinyg have enough ooomph to run the Sherline steppers. Once you get yours hooked up and running, let me know.

    Thanks

    Tom

    #9239
    Vex
    Member

    Re Stepper Motors: Yup, sure do (at least on mine). I would recommend mounting a fan for direct airflow on the ground plane of the TinyG. I haven’t really found many NEMA 23 steppers that are rated higher than 3amps… but I didn’t look all that hard. As way of clarification, my sherline started life as one boxed up in a spectralight (case and larger enclosure box); so there might be a slight difference in motors–still a bi-polar (4-wire) type.

    Mounting of TinyG

    • This reply was modified 8 years, 10 months ago by Vex.
    #9241
    TarHeelTom
    Member

    Glad to know that they will pull the load. Neat deal with the fan. Do you have it (the tinyg) inside of any kind of enclosure? I had hoped to use the existing enclosure which came with the Sherline supplied drivers, mostly because I can continue to use the existing cables and DIN plugs.

    Interesting side story here. Bought a CNC Bridgeport, an early one (1960) which had the early computer boards in it. They tell me that you just keep a bucket of replacement transistors nearby, as you get to change them frequently. Anyway, gutted both big boxes on the Bp, and threw the computer box away. Installed everything in the old electrical box, which is now still mostly empty. Went to great pains to add lots of heat sinks, fans, etc. Using the Gecko 203V’s and have a box with two computer fans which blow across the heat sink on the Geckos. Kept the 8″ fan in the cabinet door, and added a thermostat to turn it on when needed. It has never once come on. Putting my hand in above the Geckos, in the air stream, I’ve never felt warm air. Disconnected one of the fans. No change.

    Are you sure you really need the fan under the drivers?

    #9242
    Vex
    Member

    Under the hood
    You can see the TinyG, power supply, and the spindle controller.

    I also modified the controls a little bit:
    mod

    As for the fan under the drivers: it’s recommended by synthetos. I do note the drivers get warm during normal operation, but I haven’t really loaded them up yet, or run them for an extended period of time. The fan was a spare casefan I had lying around. The plexiglass mount with recess was made to reuse as much of the hardware as possible that was within the machine to begin with. If you’re planning on running 3A or more it is recommended to have forced cooling on the large copper ground planes. For that reason I have considered mounting the board upside down, but that comes with it’s own dilemma.

    • This reply was modified 8 years, 10 months ago by Vex.
    #9245
    TarHeelTom
    Member

    >>>Spindle control is of interest, but not something I’m going to worry about. I have provisions for it on my Bridgeport, but it just never has gotten anywhere near the top of the stack. But getting 0 to 10 vdc doesn’t sound that hard. Seems to me that a pwm signal looks a lot like a DC signal with lots of ripple. What if you just put a diode on the pwm output, then a capacitor, then a drain resistor. Wouldn’t that look just like a 0 – 10 VDC signal? I may try breadboarding something like that this evening.

    Started to build a mock-up of the PWM to DC converter, but then decided to google it first. Here’s a link to a very simple way to do it. Of course, this simple circuit only gives 0 to 5 vdc, but there are ways around that.

    http://www.instructables.com/id/Analog-Output-Convert-PWM-to-Voltage/

    Tom

    #9246
    cmcgrath5035
    Moderator

    Yup, what you need to do is convert the Spindle PWM output to a 0 – 10V pulse train with a non-inverting level shifter, then build the low pass filter.
    I might suggest converting to a 0-12V pulse train, since 12V is already available on tinyG at the fan connector.
    You can then calibrate your PWM signal to provide 10VDC at your chosen RPM using the $p1* parameters.

    What sort of small signal transistors/FETs do you have in your junk box?

    #9248
    Vex
    Member

    For further reading on this subject, here’s an old thread:

    Spindle: Spin/Sdir/PWM

    #9249
    TarHeelTom
    Member

    Vex
    What are you using for limit/homing switches? Can you show me how you have them mounted?

    Thanks

    Tom

    #9250
    Vex
    Member

    Tom,

    I don’t have the limit/homing switches mounted yet as I’m waiting to upgrade the steppers before mounting. I plan on mounting brackets to the Y-axis portion of the table such that it trips on the saddle. Mounting the x-axis to the stepper mount and the z-axis base to, again, trip on the saddle. I am going to mount the z-axis at the stock location with maybe a little bit better design and bracket mount off of the base to trip on the spindle carriage.

    As for limit switches, I bought these type:
    http://www.parts-express.com/parts-express-spdt-miniature-snap-action-micro-switch-with-roller-lever–060-614

    I think from Amazon, but it’s been awhile.

    #9255
    TarHeelTom
    Member

    Be interesting to see how repeatable these are. I’ve bought several types of micro switches from different manufacturers, as I’ve had one bad switch (sticky) on my Bridgeport. Finally found a source (Allied, I think) for the original Honeywell switch and bought one. $160 Oooouch. But, it works, and works well. My best indicators only read to 0.001, and the Honeywell switch is repeatable to less than that. The repeatablility is important to me, as I have a fixture bolted down to the table, and if I can home to a precise position, it saves me having to dial in the position with an edge finder, which is a slow process.
    But the price is right on these, $0.99 each, so I ordered some.
    Overall, I think I like the Hall effect switches better, especially if I can find a way to mount them where they are protected from swarf. They are, of course, magnetic, but I rarely cut any steel on the Sherline, so that’s not a problem.

    Thanks again.

Viewing 15 posts - 1 through 15 (of 18 total)
  • You must be logged in to reply to this topic.