Diagnosing eccentricty issues
NZreg | 20/09/2018 02:40:48 |
18 forum posts | Hello, My recently aquired S7 had spindle thread problems when I first aquired it. The 3 jaw and 4 jaw chucks fitted but I was unable to fully screw on the faceplate, collet chuck and driver plate. I purchased a die from Chronos and ran this over the thread which then enabled me to screw on the mentioned items. Problem is now that of the three chucks I have 1 Burnerd 6”4 jaw, 1TOS 5” 3 jaw and 1TOS 4” 3 jaw the two 3 jaw chucks hold round stock that measures .2 mm eccentricity at around 75mm from the chuck. The measurement nearer the jaws is only slightly better. The collet chuck shows exactly the same eccentricity. A test bar mounted in the internal morse taper gives a reading at the very end of the bar of .05mm by my reckoning about 2 thou. This would indicate to me that the mandrel is not bent. Measurement of the land show around the same eccentricity I am aware that all threads must be very clean, no dirt swarf etc. There are no detectable dings or marks on either the thread or land. Is it possible that thread distotion could cause this issue. I welcome any constructive ideas. The lathe had supposedly been restored to new condition but considering the work I have had to do on the lathe makes me sure it was not much more than a nice paint job, stiil I am generally pleased with my purchase but would like to solve this problem, even if it means a new mandrel Thanks in anticipation Reg Edited By NZreg on 20/09/2018 02:42:14 |
Speedy Builder5 | 20/09/2018 06:58:55 |
2878 forum posts 248 photos | The last thing I would do is to muck about with the spindle with dies etc. Once you have removed metal, you can't go back. Chuck backplates etc are much cheaper. The final location of a chuck is not normally the thread, but the collar behind it ?? or is the Myford different ? |
David George 1 | 20/09/2018 07:15:32 |
![]() 2110 forum posts 565 photos | Hi Reg the location on your lathe is controlled by the location diameter and face on the spindle and if the thread is a little loose it doesn't matter if you have a damage maybe caused by swarf if foreign body trapped it can throw out the chuck etc. Inspect and check the location diamiter and face and check with a quality dial indicator for any run out or distortion. If you find the face is damaged it is possible to turn it to true it but only remove a few thou but don't touch the diamiter. Put some engineering blue on the chuck and see if if beds corectly and not held off by anything as well. Have you checked the spindle bearings for play by putting a clock on to a bar held in the chuck whilst pulling the chuck sideways and up and down. David Edited By David George 1 on 20/09/2018 07:17:49 Edited By David George 1 on 20/09/2018 07:23:11 Edited By David George 1 on 20/09/2018 07:23:58 |
roy entwistle | 20/09/2018 08:33:55 |
1716 forum posts | You should probably have bought a tap and used it on the faceplate driver plate and collet chuck Roy |
Hopper | 20/09/2018 09:56:23 |
![]() 7881 forum posts 397 photos | +1 on what DG1 said above. Chuck location depends on the outside diameter and face of the plain register collar on the spindle and the matching surfaces on the chuck. Thread does not provide location, only retention. Measuring runout at 75mm from the chuck is not telling you much. Could be bent bar, flex etc. Measure up as close as possible to the chuck jaws. On these old chucks, the jaws can be worn and distorted so the job runs out even if the chuck body is running reasonably true. You could put a dial indicator mounted on the bed and with the plunger bearing on the unthreaded locating collar surfaces of the spindle, measure actual runout of the outer diameter and face of the spindle collar area. All this of course is assuming your headstock bearings are in good nick and adjusted nice and tight, which is a bit of a trick on the S7, but has been covered in various threads here before that should turn up in a search. |
not done it yet | 20/09/2018 10:21:02 |
7517 forum posts 20 photos | Without any confirmation as to how the chucks worked before the thread was chased, we are clueless as to whether the chasing caused the problem or if the fault was present previously. I reckon it was dodgy all the time. Apart from that, Roy has made the most fitting comment. |
Mike Poole | 20/09/2018 11:09:13 |
![]() 3676 forum posts 82 photos | I would check the register faces very carefully, if you have run a die down the thread you may have touched the edge of the register and raised a minute bump / burr. I would leave chucks out of the nose diagnostic process and after very careful inspection and cleaning mount the faceplate and check it runs true. It is most important that the nose and thread is clean and no swarf is embedded, the same goes for the register and thread in the chucks and faceplate. A useful tool to clean the internal threads is a piece of coat hanger type wire bent so the two ends spring into the thread and file the ends to the thread profile. Screwing this through the thread should clear any debris in the thread which can then be brushed out with a toothbrush, preferably not the one you use at the moment. The nose register faces should run absolutely true and unless something catastrophic has occurred should not need to be touched. If something serious has happened then a new spindle may be required. Mike |
Neil Lickfold | 20/09/2018 12:56:23 |
1025 forum posts 204 photos | Hi Reg, The face of the spindle (where the chuck and face plate seat against) should be zero with a dial indicator. If this is indeed zero, then the next area to look at is the concentricity of the thread itself on the nose spindle. Setting the lathe to screw cut the thread, 12TPI Whitworth I think, but check. Then place a DTI onto one of the flanks, and as you turn and the carriage feeds at the same time, check the concentricity of the thread form. If it is not concentric, it will cause all sorts of inconsistent mounting problems. The thread can be quite loose and still be effective. You can recut the form with a 55 deg included threading tool to correct this error. To reset all the clearances etc on the lathe and re scraping or getting the surfaces reground, can take quite a bit of work and time, and attention to the detail is a must. Neil |
NZreg | 21/09/2018 02:36:43 |
18 forum posts | Many thanks for the good advice re the spindle problem. I am slowly working my way through the diagnostics and will report back any results found. My feeling at this point is that the thread is the culprit as the form seems to be misshapen. For those who suggested I obtain a tap and use it on the items that would not screw on I failed to mention that I purchased a tap along with the die however i did not use it as it occurred to me that the 5" chuck and the collet chuck were both brand new yet behaving in exactly the same manner i.e. gross run out even close to the jaws. Time will tell. Thank you all again for the helpful suggestions. Reg Edited By NZreg on 21/09/2018 02:37:04 |
Hopper | 21/09/2018 06:36:45 |
![]() 7881 forum posts 397 photos | Posted by NZreg on 21/09/2018 02:36:43:... ...the 5" chuck and the collet chuck were both brand new yet behaving in exactly the same manner i.e. gross run out even close to the jaws. Perhaps therein lies the problem? Two new chucks running out. Are they correctly mounted to their backplates? Were the backplates turned concentric to this lathe's spindle in situ before the chucks were mounted? Are the backplates correctly machined to bear on the register collar surfaces on the spindle? How does the faceplate run if mounted on the spindle? Runout on the face? Runout on the outside diameter? If it runs reasonably true but your chucks are out by a large amount as stated, the problem might be something in the chucks rather than the spindle. |
Brian Wood | 21/09/2018 10:14:58 |
2742 forum posts 39 photos | Another possibility, but unlikely with new chucks I have to admit, is that the chuck register itself is faulty. I have corrected an old chuck with damaged register (someone had dropped it onto a hard object and bruised it) by clamping it using its own jaws [which were good] onto freshly turned bar so that the rear of the chuck faced out towards the tailstock and then carefully re-machined the chuck register whilst so mounted. A new register to suit the changed dimensions was then turned, in situ, on the chuck backplate and the two were united again. It was a salvage that worked rather well. Regards Brian |
Ian S C | 21/09/2018 11:30:37 |
![]() 7468 forum posts 230 photos | I had problems a few weeks back with a little Box - Ford A, first was the saga of removing the four jaw chuck, that was found to be caused by embedded swarf, that was cleared out using a hand chaser (we got a big box full of them along with the lathe), then I found that the three jaw didn't fit up to the register, so I cleaned that thread out using an internal chaser. Every thing fits like a glove now. Ian S C |
Harry Wilkes | 21/09/2018 14:20:56 |
![]() 1613 forum posts 72 photos | A point worth mentioning I recently purchased a 100mm back plate machined to fit a Super 7 when I came to try it on the spindle it would only just start due to it being under size Other than the bore I was happy with the back plate. Edited By Harry Wilkes on 21/09/2018 14:22:02 |
NZreg | 28/09/2018 18:58:08 |
18 forum posts | Spent time over the last week checking out the suggestions made. The mandrel seemed to be ok, no discernable movement in the bearings and only a low run out with a Mitutoya dial indicator. I removed the 5” chuck from its backplate and again obtained a good result on the face of the plate. After following the good advice posted it seemed to me that the mandrel thread deformation must be causing at least some of the problem so I obtained a replacement mandrel from Myford (ouch) fitted same and rechecked. This time the chuckbackplate showed runout but after taking a light skim it now runs much better. The collet chuck which is a Myford one that screws directly onto the mandrel nose now runs spot on. I can only think that in some way the deformed thread was tilting the chuck to give the poor runout figures. The results are very pleasing. Thanks to those folk who helped me with this problem. Reg
|
Brian Wood | 29/09/2018 16:41:45 |
2742 forum posts 39 photos | Reg, I've sent you a PM. Regards Brian |
Please login to post a reply.
Want the latest issue of Model Engineer or Model Engineers' Workshop? Use our magazine locator links to find your nearest stockist!
Sign up to our newsletter and get a free digital issue.
You can unsubscribe at anytime. View our privacy policy at www.mortons.co.uk/privacy
You can contact us by phone, mail or email about the magazines including becoming a contributor, submitting reader's letters or making queries about articles. You can also get in touch about this website, advertising or other general issues.
Click THIS LINK for full contact details.
For subscription issues please see THIS LINK.