DIY inside grinding attachment for cylindrical grinder, what design? - Page 3
Close
Login to Your Account
Page 3 of 3 FirstFirst 123
Results 41 to 55 of 55
  1. #41
    Join Date
    Apr 2016
    Country
    UNITED STATES
    State/Province
    Texas
    Posts
    119
    Post Thanks / Like
    Likes (Given)
    67
    Likes (Received)
    61

    Default

    That's for active cooling with direct airflow. Passive (convection) cooling is where the 10in2/W comes in - a little airflow makes a huge difference! Also you don't have to dissipate the total system power (300W?), only thermal losses. So 10% or so for your power supply, same for your motor and some (5%? I've no idea) for the bearings in your spindle. If you're running at 100% load, that's around 75W. Plus you only need to keep it in thermal equilibrium (for size change issues) and below the lowest max temp of your system. If that's 85C for example, then keeping it at 80C all day long will be just fine.

  2. #42
    Join Date
    Apr 2010
    Country
    UNITED STATES
    State/Province
    Wisconsin
    Posts
    6,177
    Post Thanks / Like
    Likes (Given)
    4238
    Likes (Received)
    3859

    Default

    Quote Originally Posted by mattthemuppet View Post
    That's for active cooling with direct airflow. Passive (convection) cooling is where the 10in2/W comes in - a little airflow makes a huge difference! Also you don't have to dissipate the total system power (300W?), only thermal losses. So 10% or so for your power supply, same for your motor and some (5%? I've no idea) for the bearings in your spindle. If you're running at 100% load, that's around 75W. Plus you only need to keep it in thermal equilibrium (for size change issues) and below the lowest max temp of your system. If that's 85C for example, then keeping it at 80C all day long will be just fine.
    I know that you are saying really intelligent stuff and I am trying to follow along to learn something, here. But all I am getting is Charlie Brown's teacher in my head.

    Teacher - "Wah, wah. Wahwah wah, wahwahhhh."

    CB - "Yes, Ma'am."

    Teacher - "WahWAHwah. Wah wahWAH wah wah."

    CB - "Yes, Ma'am."





    Can you dummarize it for us meatheads? Thanks.

  3. #43
    Join Date
    Apr 2016
    Country
    UNITED STATES
    State/Province
    Texas
    Posts
    119
    Post Thanks / Like
    Likes (Given)
    67
    Likes (Received)
    61

    Default

    I'm flattered

    It's like a car engine - the cooling system only needs to get rid of the heat that doesn't go towards moving the car. So if the engine is 30% efficient (30% of the energy in the fuel does some work) then the remaining 70% is heat and needs to be transferred to the air via the coolant and radiator. The engine also doesn't need to be kept at room temperature, so the fan only comes on when the engine temperature gets above a certain set point.

    Similar deal with the grinder - most of the energy used does work in the form of turning the spindle and removing metal with the wheel. Only the "leftover" energy (electrical conversion losses, friction, stuff like that) is converted to heat and needs to be removed. And only enough so that the components don't melt or weld themselves together (no kidding, I've had LEDs melt their solder and fall off before).

    That's pretty much it really, hope I didn't condescend.

  4. #44
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    3,054
    Post Thanks / Like
    Likes (Given)
    1990
    Likes (Received)
    1124

    Default

    Lower clamp:



    Here are the parts:



    From left to right are:
    handle M12
    cover/retaining plate
    steel bushing locked into two dimples with two opposing M6 setscrews
    spring
    all riding on an M14 threaded shaft

    When the handle is loose, the bushing and cover plate prevent removal/loss. When the handle is tightened, the M14 thread should be gently "bottomed out" The steel bushing presses against the spring, which in turn presses on the cross bar, with about 50 lbs of force (250N). I prefer this to a direct connection which can easily be overtorqued by the operator.

    By the way, here's the procedure for centering the grinding spindle holder (only needs to be done once):



    I put 6mm ground pins in the workhead and grinding spindle, and spun them by hand to be sure they were centered. Then adjusted the two stop bolts of the swivel plate until both pins were at the same height over the table to within a few microns. It was easy to do and the bracket consistently repeats the height -- a good sign!

  5. Likes mattthemuppet, sfriedberg liked this post
  6. #45
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    3,054
    Post Thanks / Like
    Likes (Given)
    1990
    Likes (Received)
    1124

    Default

    I have enough of this finished that tomorrow I am going to try it for the first time. I have not yet built the enclosure or done proper wiring. Here's what it looks like now.

    Front view:


    Back view:

    (power coming in is 48 VDC + earth ground, on a 3-pin removable "aviation connector")

    Only "new" feature compared with the photos above is the motor mount:


    which can be adjusted in the up/down or left/right direction to accommodate spindles of different length and with different diameter pulleys. The motor carrier is guided by a horizontal slot visible in this photo, with two eccentric guiding pins. Adjusting the eccentric pins "tilts" the motor & pulley slightly to help find the ideal alignment.



    I also spent some time doing alignment. The spindle axis is now within 0.02mm (about 0.001") of being at the same height as the workhead axis at the nose, and it's parallel to within 0.02mm along 200mm of length in both the horizontal and vertical directions.

    After clamping it into place with the lower clamp handle and the upper clamp on the swivel bracket ears, I tested the entire assembly for rigidity by pulling the spindle nose with a luggage scale while watching an indicator riding on the tip. Applying 9kg = 90N = 20 lbs of pulling force horizontally deflected the spindle nose 5 microns (0.0002"). That's better than I had expected.

    The first thing I am going to ID grind is the taper section on the nose of an 5c collet adaptor. This was made in the far east and the taper section has about 50 microns (0.002") of runout. I'm going to use a fairly coarse wheel, about 18mm diameter and 10mm wide. It has a 6mm hole through the center for a mounting screw. Should I epoxy it to the screw and/or arbor to help keep it rigid?

    My plan is to align the axis, shift the table 10 degrees with gauge blocks, then dress the stone and go at it. I'm going to do this initially without coolant. If that does not work well I'll put a plastic bag over the ID grinder and then do it wet.

  7. #46
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    3,054
    Post Thanks / Like
    Likes (Given)
    1990
    Likes (Received)
    1124

    Default

    It's been a year and a half since I started this thread, and I'm not sure if anyone is reading anymore, but I finally did my first ID grinding today, and IT WORKS!

    I was cleaning up the 10-degree (20-degree included) taper at the front of an MT5/5c collet adaptor which was made in the east and poorly ground. I fixed the cylindrical section some time ago on a mandrel, but the internal taper still had about 50 microns (0.002") TIR. Now I can't see any runout.

    Here's the setup: a 50mm Fischer grinding spindle with an 18mm-diameter aluminium oxide stone 12mm long with a 6mm through-hole. I picked the coarsest one I could find, epoxied it to a screw mount, and gave it a fast open dress. I ran the motor at 9000 rpm corresponding to a spindle speed of about 17000 rpm, or 15m/sec surface speed.


    First photo was during a pause to dress halfway, you can see the swing back dresser in the foreground. At this point part of the taper is cleaned up, and part of the taper is not yet touched.





    Here's grinding:


    And checking the result, which came out better than I had expected:


    I ground without coolant, because I wanted to see what was going on the first time, and because I don't yet have the enclosure finished. I removed about 12x36x0.03 = 15 cubic mm of material, and the part never even got warm. But I'm sure the finish (and my lungs) would be better with coolant.

    I ground on the operator side not on the back side, because I used the swivel table and gage blocks to generate an exact 10-degree angle. That wouldn't work in the other direction.

    Earlier in this thread we've had quite a bit of discussion about cooling and heat. I am already convinced that for my type of use, no fans or cooling are needed. I ground two of these tapers today, and was running the ID grinder in total for perhaps 15 or 20 minutes. But that was not all at once, because I was interrupting to measure the part, dress the stone, and adjust the setup. So neither the motor nor the spindle even got particularly warm.

  8. #47
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    3,054
    Post Thanks / Like
    Likes (Given)
    1990
    Likes (Received)
    1124

    Default

    I'm almost finished with the cover. I originally thought about making it from film-faced plywood (light, good vibration damping) but decided that to get decent cooling I wanted a metal cover. I ruled out sheet metal because I wanted more weight and better vibration damping. So I have made it out of 10mm and 12mm aluminium plates screwed together. When all is done I will epoxy it together, fill the holes, and have it powder-coated.

    One thing I did not count on was that the cover is heavy (6.3kg, 14lbs) and I needed to control its motion. This was accomplished with a gas-spring strut, which holds it fixed in any open position. Total range of opening angle is 135 degrees.

    Here is the inside grinding attachment, swung down into the operating position:

    A nice benefit of the cover is that it reduces the motor/spindle noise quite a bit.

    Here is the attachment swung up out of the way:


    The cross-hatched region above is where I can add a 60x60mmx20mm cooling fan in the enclosure to draw in cold air if needed. If I do this then I will make exit holes for the air at the top, on the back side at the left. This will prevent coolant and grit from getting inside in either the working or parked positions, and will also direct the air always upwards as it warms.

    Control panel. On the top this is easy to reach and see but also well protected against grit and coolant. This is also a shielded location when the attachment is swung up and out of the way.


    The rotary 0/1 knob on the left is power on/off. The green LED shows power on. The switch at the bottom right is motor on/off. The potentiometer dials motor speed from 0 to 12000 rpm.

    Here is a view when open showing the gas spring, which is partly inset into the cover. I had originally thought about mounting the gas spring outside the cover, to allow 180 degrees of opening and to provide more space internally. But after some careful fitting I was able to get it inside, where it is better protected.


    Wiring and internal layout:


    The small toggle switch above the spindle to the left controls the rotation direction of the motor.

    I have been having issues with the motor controller/drive and plan to swap it for a higher-quality one.

    I have monitored the motor temperature during half an hour of non-stop operation. During this time it goes from room temp 20C to about 60C, which is well under the max operating temperature (75-80C). So for the moment I won't add any cooling, since in real use it runs only for brief periods. The total weight of metal is considerable: spindle 8kg, spindle holder 6kg, backplate 2kg, cover 7kg, in all probably ~30kg (66 lbs) so there is quite a bit of heat capacity there.
    Last edited by ballen; 04-21-2020 at 07:48 AM.

  9. Likes CharlyDE liked this post
  10. #48
    Join Date
    Feb 2010
    Location
    ch
    Posts
    2,501
    Post Thanks / Like
    Likes (Given)
    215
    Likes (Received)
    329

    Default

    where did you get the fischer spindle? new?

  11. #49
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    3,054
    Post Thanks / Like
    Likes (Given)
    1990
    Likes (Received)
    1124

    Default

    Quote Originally Posted by dian View Post
    where did you get the fischer spindle? new?
    That's funny! Do you have any idea how much these cost when new???

    I've been slowly collecting used Fischer UJ/JR spindles and mandrels. I've got two 40 x 250, one 50 x 250 and one 60 x 250 (all right handed). I think I might sell off my other grinding spindles and "standardise" on those.

  12. #50
    Join Date
    Feb 2010
    Location
    ch
    Posts
    2,501
    Post Thanks / Like
    Likes (Given)
    215
    Likes (Received)
    329

    Default

    i do. i was wondering how rich you were. collecting: ebay?

  13. #51
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    3,054
    Post Thanks / Like
    Likes (Given)
    1990
    Likes (Received)
    1124

    Default

    Quote Originally Posted by dian View Post
    Ebay?
    1 x Ebay
    1 x Ricardo
    2 x friends and acquaintances

  14. #52
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    3,054
    Post Thanks / Like
    Likes (Given)
    1990
    Likes (Received)
    1124

    Default New motor controller

    The motor controller has been problematic, and there was no way to tune the operating parameters. So I've replaced it with an Electromen EM-346-48V brushless DC motor/Hall effect sensor feedback controller. I had to add a separate speed display, a YM5135-FR 5-digit tachometer. Cost of the two parts was about 100 Euro.

    Here's what it now looks like. In the back you can see where I have extended the handle for adjusting the cross slide. On the top panel you can also see where I patched the hole from the previous control:



    The rotary 0/1 knob is spindle on/off. Green LED shows power on. Tachometer shows motor speed, in this case 10,760 rpm. Gray knob on the right is spindle speed, and the red LED is a fault indicator. This lights up if the motor current exceeds 9A for more than 250 msec or if the controller overheats.

    Photo below is the controller itself. Input power, and the 8 motor leads come into the bottom. Top wires go to the control panel above. The part that looks like a TO-220 transistor is actually a 3.9 ohm 20W braking resistor.

    The toggle switch at the bottom, below the controller, is to select spindle rotation direction (left/right). I've oriented the board so that vertical air convection carries airflow through the heat sink:



    Here is an inside view of the top panel where the controls are mounted. I've socketed all the components (even the LEDs) to make disassembly simple and quick. On the left is a 48V->12V buck converter. The 12V powers the tachometer and is also provides power for cooling fans, if I decide to add those later.



    The control cables are all thin, and tied down above the hinge and below. So hinge movement flexes the cables but not the connection points. This should be good for tens or hundreds of thousands of open/close cycles.

    Here's a final photo:



    The gray 5-conductor ribbon cable poking out the back lets me tune and adjust the controllers parameters from a PC. I'll remove it when I'm done with that.

    The controller will drive the motor up to 15,000 rpm, although the manufacturer only rates the motor for 12,000 rpm. I've set the max speed for 14,000 rpm because the motor seems happy enough at that speed and does not get uncomfortably warm.

    The motor drives a 65mm pulley, and my smaller Fischer 40mm spindles have a 25mm pulley. So this motor speed gives me a max grinding spindle speed of 14000 x 65/25 = 36400 rpm. With an 18mm spindle pulley I can get up to 50k rpm at the spindle.

    I have lined the large front plate of the cover with some 2.5mm bitumen-impregnated damping material. With the cover closed the attachment is remarkably quiet and free of vibration.

    Still needed is some filler and paint, but I'm going to use it for a while to do some grinding before putting more time into the construction.

    Cheers,
    Bruce
    Last edited by ballen; 05-02-2020 at 04:47 PM.

  15. #53
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    3,054
    Post Thanks / Like
    Likes (Given)
    1990
    Likes (Received)
    1124

    Default

    A few more bits finished. I have integrated the power supply (400W, 48VDC) into the machine. Here is the cabling connecting the DC output of the supply to the ID grinder. This uses GX-20 3-pin "aviation connectors"



    and here is the front panel control that switches this power supply on and off (fourth red lever switch, bottom right). A friend from the Netherlands sent me the red lever, and I did the switch and the plate to match the other three.

    Last edited by ballen; 05-13-2020 at 04:58 AM.

  16. #54
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    3,054
    Post Thanks / Like
    Likes (Given)
    1990
    Likes (Received)
    1124

    Default

    I needed to align the axis of the inside grinding spindle to be parallel to the long traverse axis in both the vertical and horizontal directions. Since this required disassembly and careful alignment of several components, I took the opportunity to make a couple of modifications. I added 5mm steel bearing balls to the M8 adjusting screws to get smoother and more repeatable motion:



    I also added a positive "latching" lock mechanism for the "swung back" position. This consists of a curved aluminium plate with a "lead in" bevel:



    The plate has an 8mm hole (visible in the photo above) which engages a spring-loaded pin. I've had that commercial spring-loaded mechanism sitting in my junk parts collection for at least a decade, I don't even remember where it comes from. To release one just pulls on the black handle visible here:



    (The gray ribbon cable visible in the two previous photos is not part of the permanent installation, but was installed so I can connect a PC to the BLDC motor controller, to tune its parameters.)

    I've also added some 3mm thick bitumen-based vibration damping material inside the cover. This might reduce the cooling somewhat, but it's not been a problem so far.



    The cutout is to provide clearance for the locking knob inside.

    I'm very happy with how this is working. Here's a bit of recent face-grinding:


  17. Likes CharlyDE liked this post
  18. #55
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    3,054
    Post Thanks / Like
    Likes (Given)
    1990
    Likes (Received)
    1124

    Default

    I decided to order a custom-made DC brushless motor for this. The minimum order was five, so I now have four extras. If anyone wants one, drop me a note, they are 100 Euros each plus shipping.

    A drawing is below. These are 48VDC, 500 Watt (about 5/8 HP) 12,000 rpm with Hall effect sensors, dynamically balanced, and an 11 x 40mm ground shaft with a 12mm shoulder, tapped for an M5 retention screw. They can be run either CW or CCW, and either face mounted or clamped around the body. These would be a good choice for driving a variable speed ID grinding spindle or a lathe toolpost grinder. I have a 65mm pulley on mine, which is the equivalent of a 260mm pulley on a 3000 rpm motor.







    Motor controllers cost about the same, and can be had from many suppliers in the USA and Europe. I'm using an Electromen EM-346-48V:
    Electromen :: EM-346-48V BRUSHLESS DC-MOTOR DRIVER 24-48V 10A

    Cheers,
    Bruce
    Last edited by ballen; 06-16-2020 at 08:26 AM.


Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •