3 link Posted January 5, 2017 Share Posted January 5, 2017 (edited) Hi John, I hope you feel chuffed, those sleeves look well tidy and you wouldn't be able to buy them off the shelf that exact size and spec. Ok only a small exercise but it's a start and once you get use to that lathe the sky's the limit. ATB, Martyn. Edited January 5, 2017 by 3 link Link to post Share on other sites More sharing options...
Nimbus Posted January 5, 2017 Share Posted January 5, 2017 Pultra ended up with Smart & Brown at Biggleswade. As a student, my college class had a guided tour of the factory in the late '70's. Last item on the itinerary was an opportunity to drool over a stand-mounted Pultra with all the trimmings, inspection completed, and ready to be sent on its way. We were told it had been ordered by a lady for her husband ! The Nim. Link to post Share on other sites More sharing options...
bertiedog Posted January 5, 2017 Share Posted January 5, 2017 (edited) In case the method is not known or taught these days, any chuck back has a ground reference standard bore and a flat back, that are the only basis to mount a chuck. The nose of the lathe is fitted with a new or recovered back plate, and it is faced true, then machined back to leave a boss to just fit the chuck bore, no play, but a firm fit The bolts are then fitted just to secure the body to the back plate, they do not affect accuracy or fit, just retainers. A test piece is placed in the jaws and measured for run out and should be better than half a thou for a home lathe. This assumes the bolts done tight and even, In no way must the bolts be altered or sleeved to achieve a low run out, but in the case of a half thou, then the bolts could be eased and side pressure applied to the chuck to try to get it to run true. But if still out then all you can do is redo the back plate. If still out then the jaws are suspect on an old chuck, if new the machining of the ground hole is suspect and must be pro re ground. Most new chucks are accurate and can be trusted to fit the backplate first time, the troublesome ones are old slightly worn ones, with dodgy jaws that need attention all round to use them, Even with an old chuck you can usually trust the ground hole, it never wears, so you may have to trust the hole and fit, then test and then re tip the jaws by turning or grinding. Never put packing behind the chuck back or try to add shims, you will fool yourself into a false security of a seemingly true run at the tips of the jaws only to find a straight cut makes a taper! Once paired with the back plate the chuck remains with it for life. Stephen. Edited January 5, 2017 by bertiedog Link to post Share on other sites More sharing options...
AndyID Posted January 6, 2017 Share Posted January 6, 2017 (edited) A test piece is placed in the jaws and measured for run out and should be better than half a thou for a home lathe. Good luck with that. I believe you are doing well if you can repeatably get it down to a couple of thou. Also, you really need to check it with a range of different diameter test pieces. The scrolls on "automatic" chucks are not terribly accurate. The runout might be pretty good at one diameter and nowhere near as good at a different diameter. Concentricity can always be obtained with an independent four-jaw chuck although it takes a lot of messing around to get there. (I remember reading once that if you can only afford one chuck, buy an independent four-jaw.) Collets are a better solution of course, but tend to be a bit spendy. Edited January 6, 2017 by AndyID Link to post Share on other sites More sharing options...
bertiedog Posted January 6, 2017 Share Posted January 6, 2017 A Pratt Burnard chuck was mentioned and the bore at the back and the flat should in industry set the body to under a tenth of a thou or better, that is the body and has nothing to do with the jaws or the scroll state. Once the body is true and checked the jaws can be fitted and a 1 inch test bar fitted to give an assessment of the jaws. If new they should grip true to about half a thou or better, If worn they may be up to 5 thou or more out, and need remedial grinding of the tips or turning by boring , the jaws tips, done by setting at half the max movement the scroll can manage, A disc is placed at the back of the jaws and tightened in place to strain the jaws to a nominal work position for the grinding, where the disc touches is ground away afterwards, The jaws are then tested on a 1/2th inch bar and a 2 inch gauge. All should be within a thou, but worn older chucks may be poorer. The final test is a 12 inch test bar put in the chuck and tested for run out at each end. finally a worn chuck that is eccentric will still turn true!!! As long as the work remains gripped, not turned or second held. So never condemn a chuck, they can always be serviced or used within the limits of the wear. I had to make BS test equipment on 50 year old lathes with worn chucks, and yes the gear was to test the BS standards, and had no tolerances at all, just practical limits, measured afterwards in microns. It does not need super accurate lathes it needs the operator to allow for the condition of the lathe. When the wear gets high fit soft jaws to the tips and machine the jaws for every step through a sequence, perfect precision.....no run out at all worth measuring for home use. All these tips assume a well adjusted main bearing cluster or taper bearing, well oiled etc., and no silly things like cracks in the bed or headstock! Link to post Share on other sites More sharing options...
jrg1 Posted January 6, 2017 Share Posted January 6, 2017 A Pratt Burnard chuck was mentioned and the bore at the back and the flat should in industry set the body to under a tenth of a thou or better, that is the body and has nothing to do with the jaws or the scroll state. Once the body is true and checked the jaws can be fitted and a 1 inch test bar fitted to give an assessment of the jaws. If new they should grip true to about half a thou or better, If worn they may be up to 5 thou or more out, and need remedial grinding of the tips or turning by boring , the jaws tips, done by setting at half the max movement the scroll can manage, A disc is placed at the back of the jaws and tightened in place to strain the jaws to a nominal work position for the grinding, where the disc touches is ground away afterwards, The jaws are then tested on a 1/2th inch bar and a 2 inch gauge. All should be within a thou, but worn older chucks may be poorer. The final test is a 12 inch test bar put in the chuck and tested for run out at each end. finally a worn chuck that is eccentric will still turn true!!! As long as the work remains gripped, not turned or second held. So never condemn a chuck, they can always be serviced or used within the limits of the wear. I had to make BS test equipment on 50 year old lathes with worn chucks, and yes the gear was to test the BS standards, and had no tolerances at all, just practical limits, measured afterwards in microns. It does not need super accurate lathes it needs the operator to allow for the condition of the lathe. When the wear gets high fit soft jaws to the tips and machine the jaws for every step through a sequence, perfect precision.....no run out at all worth measuring for home use. All these tips assume a well adjusted main bearing cluster or taper bearing, well oiled etc., and no silly things like cracks in the bed or headstock! If you are unsure of the machine accuracy, it is best not to reset the work in the chuck-i.e. do all the operations possible without moving the work in the chuck. Link to post Share on other sites More sharing options...
Guest Isambarduk Posted January 6, 2017 Share Posted January 6, 2017 I agree with all the advice given, particularly re fitting the backplate. It is not reasonable to expect a self-centring chuck to be truly self-centring at all diameters because so much reliance is placed on the internal scroll. Machining all at one setting solves the problem but, if the job has to be reset, then a chuck with independent jaws (normally a four-jaw) is the solution - or a collet, which is also ideal for working with accurately sized (ground) stock material where the original surface is to be used (eg for a shaft). David Link to post Share on other sites More sharing options...
jrg1 Posted January 6, 2017 Share Posted January 6, 2017 A four-jaw chuck can be very useful-such as for setting up castings for machining. For accurate setting of machined diameters a micrometer clock is needed. Link to post Share on other sites More sharing options...
AndyID Posted January 6, 2017 Share Posted January 6, 2017 A Pratt Burnard chuck was mentioned and the bore at the back and the flat should in industry set the body to under a tenth of a thou or better, that is the body and has nothing to do with the jaws or the scroll state. Once the body is true and checked the jaws can be fitted and a 1 inch test bar fitted to give an assessment of the jaws. If new they should grip true to about half a thou or better, If worn they may be up to 5 thou or more out, and need remedial grinding of the tips or turning by boring , the jaws tips, done by setting at half the max movement the scroll can manage, A disc is placed at the back of the jaws and tightened in place to strain the jaws to a nominal work position for the grinding, where the disc touches is ground away afterwards, The jaws are then tested on a 1/2th inch bar and a 2 inch gauge. All should be within a thou, but worn older chucks may be poorer. The final test is a 12 inch test bar put in the chuck and tested for run out at each end. finally a worn chuck that is eccentric will still turn true!!! As long as the work remains gripped, not turned or second held. So never condemn a chuck, they can always be serviced or used within the limits of the wear. I had to make BS test equipment on 50 year old lathes with worn chucks, and yes the gear was to test the BS standards, and had no tolerances at all, just practical limits, measured afterwards in microns. It does not need super accurate lathes it needs the operator to allow for the condition of the lathe. When the wear gets high fit soft jaws to the tips and machine the jaws for every step through a sequence, perfect precision.....no run out at all worth measuring for home use. All these tips assume a well adjusted main bearing cluster or taper bearing, well oiled etc., and no silly things like cracks in the bed or headstock! I think it is very misleading to say that people can expect half a thou TIR with a self-centering chuck on a home lathe. In my experience plus/minus 5 thou is more likely. This link confirms that - https://en.m.wikipedia.org/wiki/Chuck_(engineering) Link to post Share on other sites More sharing options...
bertiedog Posted January 6, 2017 Share Posted January 6, 2017 I have set up hundreds of chucks on lathes and I did say brand new, and if a Pratt body was unable to run truer that ,5 thou on test gear it would have been thrown back to them, or I would have re ground it myself, With a new chuck the Pratt is accurate at all positions on the scroll, and the grip tru type is adjustable. Most Taiwanese and Chinese chuck bodies are more accurate than a tenth, the scrolls are not as good on most, so repeatability is poorer, but no worst than a thou at extremes As said repeatability is the problem with all three jawed chucks, you cannot grip. machine reset and regrip with any degree of accuracy, so work is always done to suit the chuck. A typical Sieg mini lathe chuck can just about hold things repeatedly enough for non critical model engineering, but to do an IC engine would need a 4 jaw and collets, or work set between centres. Stephen 1 Link to post Share on other sites More sharing options...
AndyID Posted January 7, 2017 Share Posted January 7, 2017 I have set up hundreds of chucks on lathes and I did say brand new, and if a Pratt body was unable to run truer that ,5 thou on test gear it would have been thrown back to them, or I would have re ground it myself, With a new chuck the Pratt is accurate at all positions on the scroll, and the grip tru type is adjustable. Most Taiwanese and Chinese chuck bodies are more accurate than a tenth, the scrolls are not as good on most, so repeatability is poorer, but no worst than a thou at extremes As said repeatability is the problem with all three jawed chucks, you cannot grip. machine reset and regrip with any degree of accuracy, so work is always done to suit the chuck. A typical Sieg mini lathe chuck can just about hold things repeatedly enough for non critical model engineering, but to do an IC engine would need a 4 jaw and collets, or work set between centres. Stephen I don't understand what you are saying. Are you talking about the concentricity of the chuck body itself, or the concentric repeatability of a test piece mounted in the chuck? It seems you are now saying the concentric repeatability is not very good which contradicts your original point about "a half thou or better". Repeatability is what model engineers are interested in and it's likely to be at least an order of magnitude greater than "a half thou". If it were possible to achieve that sort of repeatability we wouldn't have to mess around with independent four-jaw chuck and collets. Link to post Share on other sites More sharing options...
raymw Posted January 7, 2017 Share Posted January 7, 2017 Hi Andy, It's a Pratt chuck, high quality, not a cheapy, and a new one which gives the higher accuracy mentioned. The cheapies/worn ones are not so accurate, which is also mentioned. fwiw, you can get better relocation accuracy by tightening each location in turn, not just wellying it up on one location. If you have to remove and replace the workpiece, it helps to mark (felt tip pen?) the location of number one jaw on the workpiece. Alternatively use a split sleeve, bored in situ. If I could be arsed, I'd run a dial gauge indicator over mine, but I expect it's way out, but for what I now do, it doesn't matter. You learn to work with the errors, that's the skill part of manual machining - it's not all digital in the real world . 2 Link to post Share on other sites More sharing options...
AndyID Posted January 8, 2017 Share Posted January 8, 2017 Hi Andy, It's a Pratt chuck, high quality, not a cheapy, and a new one which gives the higher accuracy mentioned. The cheapies/worn ones are not so accurate, which is also mentioned. fwiw, you can get better relocation accuracy by tightening each location in turn, not just wellying it up on one location. If you have to remove and replace the workpiece, it helps to mark (felt tip pen?) the location of number one jaw on the workpiece. Alternatively use a split sleeve, bored in situ. If I could be arsed, I'd run a dial gauge indicator over mine, but I expect it's way out, but for what I now do, it doesn't matter. You learn to work with the errors, that's the skill part of manual machining - it's not all digital in the real world . Hi Ray, Yes, but PB talk about .003 TIR and .001 repeatability using "nominating pinion". I can't find anything on "nominating pinion" but it sounds like a load of techno-mumblespeak from their marketing department. (As an author of a lot of techno-mumblespeak I can recognize it when I see it.) Cheers! Andy Link to post Share on other sites More sharing options...
PatB Posted January 8, 2017 Share Posted January 8, 2017 Hi Ray, Yes, but PB talk about .003 TIR and .001 repeatability using "nominating pinion". I can't find anything on "nominating pinion" but it sounds like a load of techno-mumblespeak from their marketing department. (As an author of a lot of techno-mumblespeak I can recognize it when I see it.) Cheers! Andy I suspect they're talking about there being a preferred chuck key position (out of the three available). Is one of the key sockets marked in some way? Link to post Share on other sites More sharing options...
M Wright Posted January 8, 2017 Share Posted January 8, 2017 Hi, "nominating pinion" is just a posh way of drawing attention to what is normal practice in a workshop. To get the best from any chuck it is best to always use one of the key sockets for that final tighten. (often one of the sockets is marked by the manufacturer). All my chucks were marked or have been as have the mounting flanges on the lathes and the backplate rims. It is part of trying to keep everything the same. Chucks are therefore always placed on the spindle in the same position their backplate was machined - tightened finally in the same key slot. It is all part of trying to get repeatability and therefore know if something is amiss it is you, not the machine! None of my chuck bodies or backplates have any noticeable eccentricity on a 1/1000" clock gauge other than the slight vibration due to the ball tip detecting tiny surface imperfections in the surface. Therefore I would think the guy who says 0.0005" is right. 1 Link to post Share on other sites More sharing options...
bertiedog Posted January 8, 2017 Share Posted January 8, 2017 For a 5 inch chuck Pratt show .0005 TIR , it only increases on 20 inch plus sizes Link to post Share on other sites More sharing options...
bertiedog Posted January 8, 2017 Share Posted January 8, 2017 All chucks should close with a preferred pinion first, better are so marked, then followed with the other pinions to be as tight. this helps with repeatability, up the max the wear and tear has grown to with use. For complete multiple grips in a sequence, a four jaw or face plate is required, where each reference surface can be set to a datum or the previous surface machined. But three jawed chucks are not that bad these days at repeat settings for Home mechanic uses, especially when gripping bar that goes right through the jaws, which helps keep the jaws in line with the chuck body slots they run in. The troubles come from gripping with the tips, which add the tip wear into the eccentricity of the object being held. but again it does not affect work held on a stub of material where all vital dimensions are machined without release of the jaws. For the class of work in model railway no lathe is going to fail, as long as the limits are understood. Any lathe can produce very high precision in the right hands, and a lot of amateur turners do not realise that the final cut in the lathe is the final finish, there may be several other process to refine accuracy and fit, and also finish, honing, lapping, grinding, precision polishing. An old hoar is the statement that a good lathe operators final cut is the finish, oft repeated and completely false! It is true only when a factory is making a determined effort to cut the number of operations on a part, and can afford to leave a part as it comes from the lathe. CNC tends to encourage a final finish from the machine, as it is part of the tight repeatability offered. Stephen 1 Link to post Share on other sites More sharing options...
Grasshopper John Posted January 8, 2017 Share Posted January 8, 2017 Thanks everyone for all the advice so far My chuck was a bit out +-5thou so I loosened the bolts where this was happening and it seems to have cured it, beginners luck me thinks. You keep referring to TIR can you tell me what it is might make the penny drop a bit more if I knew. Thanks Grasshopper Link to post Share on other sites More sharing options...
bertiedog Posted January 8, 2017 Share Posted January 8, 2017 (edited) Total Indicated Runout, the max detectable running out from a nominal correct position, or from a datum. in other words the permitted eccentricity total from min to max. With the body of the chuck to the spindle it should be zero, but that's not measurable! so from about one tenth of a thou upwards on a dial gauge with a lever extender to read 10ths. Most read in thous min, but you can estimate half thous on most dial indicators Edited January 8, 2017 by bertiedog Link to post Share on other sites More sharing options...
PatB Posted January 8, 2017 Share Posted January 8, 2017 One of the wonderful things about a lathe, any lathe, is that it can be used to make another lathe which is more accurate than it is itself. Were this not the case, we'd still be about here 1 Link to post Share on other sites More sharing options...
RMweb Gold Enterprisingwestern Posted January 8, 2017 RMweb Gold Share Posted January 8, 2017 Pedantically, wouldn't we be here? http://www.historicgames.com/lathes/ancientlathes.html Mike. 1 Link to post Share on other sites More sharing options...
Horsetan Posted January 8, 2017 Share Posted January 8, 2017 Pedantically, wouldn't we be here? http://www.historicgames.com/lathes/ancientlathes.html... If it's there you want to be, I wouldn't be starting from here. 1 Link to post Share on other sites More sharing options...
Guest Isambarduk Posted January 8, 2017 Share Posted January 8, 2017 "One of the wonderful things about a lathe, any lathe, is that it can be used to make another lathe which is more accurate than it is itself." Oh yes, indeed! My late father build a watchmaker's/instrument lathe from scratch when he was a young man ... well he was still a boy when he started the exercise. See here for a brief description in captioned pictures: www.davidlosmith.co.uk/WatchMakersLathe.htm David Link to post Share on other sites More sharing options...
bertiedog Posted January 8, 2017 Share Posted January 8, 2017 Lovely design, far to good for just making models!!! Stephen Link to post Share on other sites More sharing options...
Deano747 Posted January 8, 2017 Share Posted January 8, 2017 Ahhh!!! Now we know where you get you skills from, David!!! Regards, Deano. Link to post Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now