Page 1 of 2 12 LastLast
Results 1 to 15 of 17
  1. #1
    Join Date
    Jun 2010
    Location
    Canberra
    Posts
    1,322

    Default Drawbar not drawing straight :(

    If I mount my MT5-ER32 collet chuck in my lathe by just bashing the face with a hammer and a block of wood, I get 0.02mm runout on the inside face of the taper for the collet. However if I fit my drawbar first and use it to draw the chuck in and tighten it, I get 0.05mm runout.

    So if I want the best precision, it seems I'm best off hammering the chuck in place first, then securing it with the drawbar. Does this sound normal, or am I not tightening the drawbar enough, or is my drawbar not pulling straight?

  2. #2
    Join Date
    Jun 2007
    Location
    sydney ( st marys )
    Age
    64
    Posts
    4,887

    Default

    I would not be bashing it in from the front, I would be using the drawbar to hold it in the spindle taper from the word go.

    The draw bar should not have any effect on your chuck running out , not being able to know exactly how you use your draw bar I couldn't comment on the torque need to hold the chuck correctly in the spindle taper.

    As it is just a male MT5 going into a female MT5 it should read the same wether you bash it in and then use the draw bar or just pull it into the spindle taper with the draw bar.

    May be there might be an issue with the chuck.

  3. #3
    Join Date
    Sep 2012
    Location
    York, North Yorkshire UK
    Posts
    6,444

    Default

    Hi Guys,

    I would make sure that the inside of the spindle is clean and there are no scars !

    I use a 50 mm diameter pin spanner to tighten the ER32 collet chuck drawbar. Over tightening just makes it harder to release it.
    Best Regards:
    Baron J.

  4. #4
    Join Date
    Dec 2013
    Age
    56
    Posts
    1,416

    Default

    The drawbar should pull it in enough to seat it properly, the taper should seat on it's own as others have said.
    Using Tapatalk

  5. #5
    Join Date
    Sep 2012
    Location
    York, North Yorkshire UK
    Posts
    6,444

    Default

    Hi Dave,

    Quote Originally Posted by Dave J View Post
    The drawbar should pull it in enough to seat it properly, the taper should seat on it's own as others have said.
    I agree it should ! But if you are getting variations in runout, something is causing the change.
    Best Regards:
    Baron J.

  6. #6
    Join Date
    Aug 2008
    Location
    Melbourne
    Age
    34
    Posts
    1,075

    Default

    Do you get the same runout for each method every time, or does it vary? Perhaps try mounting the chuck in various orientations relative to the spindle and see what kind of repeatability you're getting. It may simply be compound tolerance errors from the 2 surfaces.

  7. #7
    Join Date
    Jun 2010
    Location
    Canberra
    Posts
    1,322

    Default

    Thanks to everyone who's responded - just wanted to check I wasn't getting some fundamental aspect wrong. Each time I've mounted the chuck, I've carefully wiped the inside of the spindle taper and the outside of the chuck taper by hand - both are pretty much immaculate - no dings or scratches. Yet.

    I made the drawbar from some 22mm round bar and added nut made with 2" round BMS, with a reduced section and flats to take a 36mm spanner. The drawbar has a standard M20 thread for the chuck and a fine M20 thread for the nut. With the drawbar and nut screwed into the chuck, there's a fair bit of mass hanging off the back of the chuck, making me think this might be the source of the runout. None of that mass is there if I just tap the chuck into place before screwing in the drawbar.

    That said, it's possible that measuring off the inside of the collet taper may not be the best approach. I'm wondering if I should mount some kind of smooth bar in a collet and take measurements off that.

  8. #8
    Join Date
    Dec 2013
    Age
    56
    Posts
    1,416

    Default

    Make sure your new drawbar theads are not a perfectly tight fit, as in class of fit to give a little movement.
    Never measure of the outside of a collet chuck, always mount a clean drill shank or end mill in a collet and put a dial indicator on that after it's tightened properly.
    Using Tapatalk

  9. #9
    Join Date
    Jun 2010
    Location
    Canberra
    Posts
    1,322

    Default

    Had another go this morning. Used the shaft of a 5/8" endmill to read off. The readings were better, but showed the same pattern. If I used the drawbar to seat the taper, I got 0.01mm runout. I did about 4 repetitions of releasing the chuck from the MT5 taper, rotating it a little and re-seating it with the drawbar, and never got better than 0.01mm.

    I then dismantled the drawbar, removed the collet and just tapped the chuck into place with a wooden block and hammer, then re-chucked the endmill and there was virtually no movement in the needle.

    While there is still a bit of slop in my drawbar's fit when fully screwed into the chuck, I might try threading it in less to allow more movement, just in case it is pulling it to one side.

  10. #10
    Join Date
    Aug 2008
    Location
    Melbourne
    Age
    34
    Posts
    1,075

    Default

    What are you making that you need better than 0.01??

  11. #11
    Join Date
    Jun 2010
    Location
    Canberra
    Posts
    1,322

    Default

    Quote Originally Posted by elanjacobs View Post
    What are you making that you need better than 0.01??
    Nothing really, but I can confidently get close to zero every time by tapping it in. The drawbar is less predictable, and I'm trying to work out why.

  12. #12
    Join Date
    Jun 2010
    Location
    Canberra
    Posts
    1,322

    Default

    So I went all BobL on my chuck and did a series of 10 insertions and removals of the chuck, measuring the runout each time, one series tapping in place with a wooden block and hammer, one series pulling in place with the drawbar. The numbers didn't really reveal much, as I found tapping in place was giving as much variation as using the drawbar.

    So I decided the wisest course of action was to throw money at the problem. I'd been using an ER32 collet so I could share collets with my mill, but that was proving to be a pain. I'd previously bought an ER40 chuck which I had no collets for, so I ordered a set of ER40 collets and a D1-4 ER40 chuck as well.

    IMG_2922.jpg

    My thinking was that the cam lock setup would be quicker to fit than a drawbar, and with a 31mm hole through it, I could use any length stock that'd fit in a collet.

    After adjusting the cams, I was finding it a challenge to get it to seat flat against the face of the spindle, and when I did, the best runout I could get was 0.04mm, which is probably fine given the advantages it offers.

    I then tried the MT5 - ER40 chuck that I'd previously got off Banggood:

    IMG_2921.jpg

    I used my M20 drawbar and measured off the shank of a 5/8" endmill, and it was back into the slight wobble of the needle territory. I removed it, rotated the chuck randomly and drew it back in 3 more times, and each time less than 0.01mm movement. The image above was a test turning some 12mm hot rolled down to around 9.8mm, flipping it and re-chucking in a smaller collet and seeing if the other end turned down to be concentric, and sure enough it was perfect (to the extent that I could detect).

    So I'm thinking the ER32 chuck I was using first might have something odd going on with it? Either way, very happy to have a collet setup that I can use with confidence.

  13. #13
    Join Date
    Jun 2007
    Location
    Victoria, Australia
    Posts
    114

    Default

    I liberally use brake or contact cleaner on the mating surfaces. This way I try to ensure no oil or whatever can affect the mating of the two surfaces. Works well in the tail stock.
    My lathe is a 16x40 2000KG and changing chucks is a bit of an effort. I typically have the 4 jaw mounted and have several collet holders which I can dial in as accurate as my ability allows. I also use the 3 jaw scroll and 4 jaw scroll inside the large 4 jaw independent. I guess these are the advantages of having a large lathe.

  14. #14
    Join Date
    Jun 2010
    Location
    Canberra
    Posts
    1,322

    Default

    Quote Originally Posted by topari View Post
    I liberally use brake or contact cleaner on the mating surfaces. This way I try to ensure no oil or whatever can affect the mating of the two surfaces. Works well in the tail stock.
    Cleaning contact surfaces has been a challenge, particularly deep in the MT5 taper of the spindle - I'm always finding bronze chips floating about the place. I've been using canned air to blow out chips where appropriate, but it's pretty anaemic. I might give brake cleaner a go.

    My lathe is a 16x40 2000KG and changing chucks is a bit of an effort. I typically have the 4 jaw mounted and have several collet holders which I can dial in as accurate as my ability allows. I also use the 3 jaw scroll and 4 jaw scroll inside the large 4 jaw independent. I guess these are the advantages of having a large lathe.
    I'm guessing, though, you'd be having to dial in whatever you mount in the 4 jaw each time? The attraction of the collet chuck is the idea that it's dead on once mounted, no adjustment required. Although I'm realising that in reality, I'll probably be checking with the dial indicator each time to make sure I haven't screwed something up. On the upside I'm definitely getting faster at setting it up

  15. #15
    Join Date
    Jun 2007
    Location
    Victoria, Australia
    Posts
    114

    Default

    Rusty, I use my indicator all the time. It takes little to muck up alignment. I have a top quality TOS 3 jaw, and I still use the indicator to check run-out. I find with different tension on the three cams, the run-out is affected. A friend asked why the indicator with the 3 jaw. You never know, the stock may not be seated correctly or something else has affected run-out. I have a commercially made 50mm ER32 holder, harden and machined flat on one side. The flat side is aligned with one of the jaws. Once it is dial in, it is locked in place and only the quality of the ER32 collets affect reproducibility. And if it is a problem, I can adjust the 4 jaw. I probably spent a full year only using the 4 jaw until it became second nature. Even the quality of the stock, the surface finish and maybe the odd burr can muck up your run-out.
    Collets vary greatly in price probably reflecting the quality. I like the saying from old ... "if it cannot be made accurate, make it adjustable".

Page 1 of 2 12 LastLast

Similar Threads

  1. A simple CAD drawing request
    By jhovel in forum GENERAL DISCUSSION / OFF TOPIC
    Replies: 4
    Last Post: 13th Jul 2017, 11:23 PM
  2. Machinists screw jack drawing
    By bollie7 in forum METALWORK GENERAL
    Replies: 3
    Last Post: 5th Jul 2013, 08:50 AM
  3. Need help with a Chuck Drawing
    By SurfinNev in forum METALWORK GENERAL
    Replies: 20
    Last Post: 25th Jul 2012, 07:49 AM
  4. Trouble attaching a drawing
    By neksmerj in forum METALWORK GENERAL
    Replies: 18
    Last Post: 12th Dec 2010, 07:17 AM
  5. technical drawing
    By tanii51 in forum METALWORK GENERAL
    Replies: 6
    Last Post: 1st Dec 2009, 06:56 PM

Posting Permissions

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