|
Why I started this list
Discussing the following project elsewhere is prohibited or unpleasant, so I decided to start this list. As the techniques are generally applicable to any machine tool, I have made the topic of the list more general than this one project. I bought a 7" x 14" Chinese mini-lathe for $527 delivered for the specific purpose of doing instrument maker class work. Obviously such machines are barely capable of doing crude work as delivered. But for very high precision and accuracy work anything I can afford would require an equal amount of work to recondition it. I bought the mini-lathe as a complete set of parts, assembled at the factory to make sure all the parts in the kit had been included. I plan to scrape the parts true to 0.0001", replace the deep groove spindle bearings with tapered hydrodynamic (aka plain) bearings to eliminate periodic errors and do whatever else is required. As plain bearings require less space, I'm currently evaluating making a new MT 5 spindle with a 1.5" bore and taper mount nose to replace the factory spindle. Changing from rolling to sliding contract bearings provides plenty of room for a larger spindle bore. I also have some larger machines to recondition though not to the same standards. The mini-lathe is partly a warm up project for those. All the surfaces on a mini-lathe are less than a single surface on the other machines. In the late 1940's David Broadhead, an amateur, rebuilt his 13" Southbend to such standards in preparation for making the screws for the ruling engine that John Strong, one of the world's most prominent experimental physicists was building at Johns Hopkins. That ruling engine was the first truly successful build since Rowland's machines built in the 1890's. A. A. Michelson had tried and failed from being too ambitious, though his engine was later rebuilt and successfully ruled 12" gratings, half of the 24" Michelson set as his goal. Broadhead's screws were tested by interferometer and accurate to less than a millionth of an inch. I would not dream of attempting to make such screws, but I am certain I can rebuild a mini-lathe to Broadhead's tolerances. Using such a machine to full potential requires very careful attention to many details, temperature stability being one of the most important. An accurate machine is only a small part of the problem. The user's skills matter more. Twenty-five years ago I reworked my Chinese bandsaw to cut very accurately by scraping the guides and fitting gibs. Not having done it since, I will have to reacquire the tactile feel for scraping. But it's like doing bench work with files or playing a musical instrument. If you don't practice, proficiency is lost and must be relearned. Your head may remember, but your fingers won't. Scraping and bench work with files are the means by which the original machine tools were built and are still important skills, though few possess them today. In addition to the bandsaw rework, I designed and made tooling to align my table saw to 0.001"/8" (30 seconds of arc) using only hand tools. I can measure the error more easily than I can get the trunnion to stay in place when the bolts are tightened. Alignment and reconditioning require nothing more than basic mathematics as taught in secondary school, algebra, geometry and trigonometry. Designing a new spindle is more complex and a proper job is a mechanical engineering exercise of considerable difficulty if high precision and accuracy are the goal. So while some threads will involve nothing more than basic mathematics, other threads may involve the solution of differential equations. The thing that all will have in common is the need for precise and accurate measurements. Have Fun! Reg
|
|
Thank you
Thanks for the ad, looking forward to these discussions. I'm not a big poster as I'm not very experienced, but I do attempt to make parts and make "improvements" to my little lathe and mill. -- Kelly Black, Utah USA HF 7"x10" updated with LMS bed extension. HF 44991 Mill
|
|
Questions about the Lathe rebuild
3
Reginald, I have a couple of questions related to your lathe project. To briefly introduce myself, I am a long time lurker on the Mini Lathe group but never posted there. And the way that you were treated turned me off from posting these questions there, although I did think about doing it. I have learned a lot from reading the messages on that group; however, I found that I learned the most from the messages that were a little bit off topic. I found your postings to be of great interest. In one of your messages, you mentioned making the parts for a level that would allow you to determine level to within 0.05 arc seconds (if I remember the number correctly). First question: what do you intend to do with a level having that precision? Second question: are you not concerned with the lack of stability of the cast iron parts of the lathe until they have been aged sufficiently? It would be truly sad to go to all the work of scraping the castings in and then to have them drift out. I don't have any personal experience with this, but in my younger days I was very interested in drag racing and building engines for drag racing, and I recall there being much discussion about mechanically stabilizing the engine blocks before machining them. The primary technique used was called "shot peening". I also seem to remember that a low budget approach was to beat on an engine block all over with a ball peen hammer for several hours before beginning to machine it.
|
|
Rudiments of the mechanics of materials
This is a follow up to the earlier discussion of shot peening and the stability of cast grey iron. Because the topic is vast and I have sufficient references to cover the topic fairly exhaustively, it will be very brief. There are several major factors, all of which are familiar to most people: elastic - if you bend it it springs back to *exactly* the same shape plastic- if you bend it it stays bent creep - this is a catchall for a variety of things viscoelastic - under pressure it slowly bends but slowly springs back over a significant period when the load is removed viscoplastic - under pressure it slowly bends but when the load is removed it does not spring back The flat spot that develops on an overloaded rubber caster which disappears when the load is removed is viscoelastic. The loss of contact pressure between a terminal and an aluminum wire is an example of viscoplastic behavior. In general the viscous behavior is temperature dependent. Richard Feynmann's experiment in the Challenger investigation meeting using a sample of O ring and a glass of ice water is the classic example of a thermoviscoelastic effect. Shot peening is generally done to accomplish two objectives: reduce the height of small asperities (high spots) so they don't cause stress risers which lead to cracking increase the stresses in the surface layer so that the material is harder and stiffer there is more to it I'm sure, but that's what I recall from a discussion in a book I read recently while looking for other information. There is a lot more to the subject and I don't want to go down a rabbit hole. I spent 5-6 years in the oil industry concerned about how the mechanical properties of porous rocks varied with fluid content and how the fluid properties varied with temperature and pressure. I have at least 3 ft of technical monographs on those topics and have spent a good bit of time discussing nuances of behavior and theory at the limits of human knowledge. Neat stuff, but not some place we want to go. I'd love to, but then I'd have to send you a very large bill for professional services ;-) In summary, for small machines viscoelastoplasticity is as deep as we should need to go. The movement of my lathe bed from one day to the next is classic viscoelastic behavior. My straightedge castings from Martin Model & Pattern were annealed after casting which is because of the thermoviscoelastoplasticity of the casting. That's almost "supercalifalasticexpialidocious" ;-) I expect that the electronically amplified high resolution level vial will allow measuring the effects. However, I fully expect that scraping the mini-lathe bed will demonstrate measurable viscoelasticity. Have Fun! Reg NB Most of the references for the above are quite expensive, however Dover has two books by J. P. Den Hartog who taught mechanical engineering at MIT, "Strength of Materials" and "Advanced Strength of Materials" which are inexpensive and I highly recommend both. I'm using "Advanced.." to learn about the behavior of irregular prisms (not circular) under torsional stresses so I can design a small crane mount. "Advanced.." was used for a 2 semester graduate level course at MIT. But it's nicely written and he doesn't require you to solve the differential equations yourself. But even with my background I'll have to read the chapter several times to be sure I understand it.
|
|
I made these last week from a pair of 1/4" long sections of 1.25" schedule 40 pipe. After cutting the rings, I slit them and straightened out ~1/2 of the pipe by bending it cold. For the hook end I had to use a torch and pliers to make the bend and then file the hook to fit the spindle nut. That was the only way I could form the hook sharply enough. I could not make a sharp enough bend using a hammer. Just not that good a blacksmith and no anvil that would fit the inside of the hook. I straightened the handles by squeezing them in the vise followed by a bit of cold forging with a hammer and some filing to round the edges. The pipe was very rusty and badly pitted so they are not pretty, but they work quite well. The paper is ruled 5 lines per inch. Have Fun! Reg
|
|
[email protected]
I have started a related group for discussing things not related to machine rebuiiding such as making a cylindrical precision reference square from a piston wrist pin which is accurate to ~0.0001" over the length of the square. For a 3" wrist pin that's under 10 arc seconds and sufficiently precise that thermal expansion from handling is the dominant source of error. I've made one which appeared to be accurate to 20 arc seconds, but I want to refine the technique and tooling before I post it. Have Fun! Reg
|
|
This project is *quite* ambitious. The original motivation is the desire for some gauges for testing RF connectors which have 0.0002" tolerances. I just assembled an electronics test bench which is fairly comprehensive up to 3 GHz. Good quality RF connectors are shockingly expensive and cheap ones can destroy good ones the moment you connect them. I have a traveling set of voltage and resistance references (c.f. the EEVblog.com USA Cal Club 2 thread in the Metrology section of the forum for details) arriving with which to calibrate my voltage standards and my HP 34401A and 3478A DMMs. Aside from doing that I need to find and fix a bug in an Arduino based temperature logger that is part of the cal kit. So that work will impose some interruptions. The most difficult part of the project is accurate measurements. The coefficient of expansion of iron and steel is about 6 millionths of an inch per degree F. A 16 degree temperature difference eats the entire 0.0001" tolerance. Heating from handling parts and measuring tools, even if one is wearing thick gloves, becomes a major source of error. This is not a project for the impatient. Things are much easier if your goals are more modest, say 0.0005" instead of 0.0001". I am almost certain to work to no closer than 0.0005" on the first iteration. I've never done this before, so I am assuming that I will encounter numerous challenges. Except for the spindle bearings, I have completely disassembled and cleaned my machine. Some observations: Mine was dropped 4 times in transit. The cross slide screw was broken off, the tailstock ram clamp stud bent, the chip tray and backsplash bent and the front bearing race was dented. If you receive a machine with damage to the wooden crate do *NOT* unpack it. Have the shipping company come and pick it up. UPS left mine in my driveway with holes smashed in the front, top and end of the plywood crate. I just discovered the bearing damage a couple of days ago and am still trying to get UPS to replace the damaged parts. The standard UPS answer is "repack the machine so we can pick it up and inspect it". That can be a problem if the crate falls to pieces when you unpack the machine. And it is unlikely that their inspector would know how to check the machine. Under no circumstances should you accept a machine which has visible damage to the shipping crate. Everything was covered with abrasive and cast iron grit. Lots of it. The machining and fitting is indifferent at best. The headstock casting needs 0.018" removed from the bottom in order for the vee to seat on the bed prism. This was measured by inserting a pair of 0.012" feeler gauges in the gap between the prism and the vee groove. The tailstock needs a gib and a positive means of setting its alignment with the spindle. The tailstock ram hangs out the front of the tailstock too far, so the tailstock will need to be lengthened. The holes in the spindle face cause significant imbalance. So a dynamic balancing system will have to be designed and built. Fortunately, the cost will be low. Despite these issues, I think the design is quite impressive. A great deal of thought went into making it inexpensive to produce. A fairly modest effort will reward the owner with a very capable machine and a good education in using it. I have a 10" x 20" Clausing 4902 lathe and a 6" x 24" Clausing 8520 vertical mill both in good condition. However, I intend to do all the work using the mini-lathe if at all possible so that someone not so equipped can replicate the process even if they don't wish to pursue such high precision. Most of the work required is manual bench work anyway. Have Fun! Reg
|
|
Machine tool testing
I've uploaded a PDF of a major reference on the subject to the files section. I'd forgotten I had it, but was saving some stuff related to electronics metrology and saw it in the folder. For those who know what a Talyvel is, I'm investigating a build of a variant of that. The Talyvel is a pendulum level sensor accurate to 0.1 arc second with a settling time of a few seconds. The Talyvel uses an LVDT (linear variable differential transformer) which is expensive, though not that hard to build. A bubble vial that sensitive takes quite a while to settle. The concept I'm considering is a round disk pendulum on a single wire flexure made of music wire with a pair of opposing sensors to test the position of the disk. Both capacitance and several optical methods are possible. The Talyvel is single direction only whereas this would test both directions at the same time. I've started a thread on the general topic of dimensional metrology here: https://www.eevblog.com/forum/metrology/dimensional-metrology/msg2560830/#msg2560830 Most things can be more easily built with some electronic elements. And quite a few of the EEVblog members have machine tools and other metalworking capacity. I have been too busy doing calibrations on my HP 34401A 6.5 digit DMMs to write about the pendulum sensor concept yet. I'll finish a 24 hour data acquisition run on an LTZ1000 based voltage reference at 10:30 pm this evening at which time I'll start a run on a 2nd voltage reference with the intent of getting the calibration kit on its way home Thursday following a couple of shorter runs on some precision resistors. I have been reading the list, but just not had time to comment. There is a long list waiting on the Cal Club kit and I've felt obliged to spend all my time on that. Have Fun! Reg
|
|
Mini lathe crossfeed nut backlash and height adjustment (more than you ever wanted to know)
You may say that this goes into far excessive detail on a cheap mechanical system which does not merit such attention. You are probably right. But I think it is in fact an excellent way to demonstrate the reasons this mechanism is cheap and suboptimal. No mechanism is too poor to be educational. The actual instructions for setting the cross slide aren't long and can be skipped to, but reading the other stuff at least once would be good to understand why the procedure is the way it is. One thing I want to emphasize is that eliminating backlash at all costs is not an effective goal here. If reducing backlash makes the handwheel hard to turn within the backlash area, you are putting force on the screw journal or even canting the screw in it, which doesn't actually reduce backlash, it only weakly forces the slide into a middle position (if at all) while causing excess wear in the journal and nut. It results in a "soft" backlash which is very bad for accuracy and feel. Eliminating any cant in the crossfeed screw bearing (no I don't mean the nut) will stiffen the backlash so that you can feel a clear start and end to the backlash. The backlash may appear to be larger than before, but in practice the difference between the tool position when set and the tool position when under pressure, is more important and will be improved. The backlash in the crossfeed comes from two sources. One is the nut, and the other is the screw bearing flanges. It is easy to reduce the nut backlash just by canting it against the screw using the adjustment screws, but the backlash of the screw journal bearing is set in steel and needs to be shimmed or redesigned, which I haven't done and won't address here. However getting the screw parallelism and nut height just right will allow the cross slide backlash to be minimized without causing excessive wear or introducing inconsistency along it's travel. Consistency improves the operator's judgement as to what is happening on the machine. That is what good handfeel is needed for. I see some problems with the common method of adjustment as described here: http://www.mini-lathe.com/Mini_lathe/Tuning/tuning.htm#xbacklash 1: It allows setting the crossfeed screw up or down at an angle, which misaligns it with the screw journal bearing and can cause excess wear. 2: The angle of the screw changes as the nut comes closer to the bearing, which changes the angle of the screw in the nut, so the backlash of the screw against the nut is inconsistent and creates a situation where minimum backlash is not obtainable without creating excessive resistance at the furthest back position (towards the screw handle). On the other hand if this resistance is minimized out of concern for the longevity of the lathe, there is excessive backlash at the furthest forward position. 3: The continual back and forth adjustment to find a "goldilocks" zone between these two extremes takes too much time. 4: This also introduces some cosine error which increases as the nut comes closer to the bearing, if you see the screw as a level, the bearing as the fulcrum and the nut as the handle. In practice this is probably irrelevant as it's extremely small and the lathe already uses a metric screw with an imperial dial, resulting in a 0.0013" error per turn. 5: Wear becomes more inconsistent along the cross feed screw. 6: The operator may due to ignorance end up deliberately canting the screw in an attempt to reduce backlash to below 0.005" or so, but this only conceals the backlash in the feel and mechanism, while maximizing uneven wear and inconsitency across the travel of the cross slide. The extra backlash is due to the screw journal flanges, not the nut and so the only correct way to deal with it is by shimming the flanges or installing a bearing mod. So the aim of this method is to set the cross slide nut to be at the same vertical level as the screw, so that the screw is kept level at all points of travel. To put it simply, we do this by adjusting the middle adjustment screw to find the limits of vertical play of the slide screw and then turn t
|
|
Toolposts for lathes
3
I thought I would compile some of the available toolposts I found while looking for options for the mini-lathe. It was surprisingly hard to find some of these. I think it is important to note from the outset that many lathe users want to use a large toolpost thinking it will be more rigid but in fact a larger toolpost means more tool overhang and this is much worse for a mini lathe where the weak spot is the gibs and saddle hold down plates. The first most obvious one is the OXA toolpost engineered in cooperation with Tormach and Littlemachineshop.com. https://littlemachineshop.com/products/product_view.php?ProductID=3112 This what I got and I still think it one of the best budget options for a mini lathe. The wedge type mechanism has a mechanical advantage over plunger type mechanisms and to me seems to be about as rigid as one could hope for. However I do wish it was smaller or designed somehow to reduce overhang of the tool over the cross slide (admittedly difficult to do in a design this well optimized). Another one that looks very interesting to me is the Boxford style toolpost. The mechanism is an eccentric bushing which pulls the toolholder onto the dual V rails. On principle I would still prefer the wedge over an eccentric bushing, but it seems to work fine for most people. It is considerably wider than the OXA toolpost and so the overhang would probably be an issue on mini lathes, unfortunately. However I think of the box shaped toolposts the dual vee type like this probably has the best repeatability, whereas the OXA wedge type toolpost may drag the toolholder up or down when the wedge locks down. https://www.crenolandwilson.com/products/toolposts-accessories/ For threading there is an advantage if you can accurately set the toolpost to a 60 degree angle. This way you only need to grind one side of a threading tool. There are some toolposts which allow this. The coveted Multifix toolpost: http://www.chronos.ltd.uk/acatalog/Lathe_Quick_Change_Toolposts.html It looks like the smallest in this series of toolpost might be comparable to the OXA toolpost in terms of overhang. There seems to be a lack of cheaper options for toolposts with a built in 60 degree angle. I came across triangular toolposts which seem ideal in terms of overhang and angles. But these are all vintage toolposts and don't seem to be made anymore except by one swiss company. http://anglo-swiss-tools.co.uk/tripan-tool-posts/ /g/7x12MiniLathe/topic/24144854#100828 If one were to make a DIY toolpost it seems a triangular type may be easier to machine as the upper left corner only needs to be an angle and you only need to make 2 dovetails. It's possible to turn a triangular prism using a 4-jaw chuck which neatly takes care of all 3 corners, but I don't know of a good way to make this mountable. /g/7x12MiniLathe/message/100793 Another DIY toolpost that can be very easy to make is the Normal style. It simply used a round post and the toolposts are essentially split rings which lock onto it. However it has no built in angles, has low repeatabliity and the toolpost will easily slip around the inner post if it is not tightened well. However it should be capable of relatively low tool overhang. https://www.toolsandmods.com/lathe/mini-lathe-qctp For larger lathes here are some other interesting options: The Lardi/Mecanizados Huesca (Mexican) or Boni Feldback (swiss) or commonly known as Dickson quick change toolpost probably has the best position repeatability of toolposts aside from the multi-fix. Sliding wedge type toolposts can drag the toolholder down with the wedge. These toolposts have dual V-grooves and a pull lock mechanism. http://mecanizadoshuesca.com/productos/torretas_portaherramientas_lardi Lardi is a toolpost manufactured by Mecanizados Huesca, although some of them have the latter name written on them. The toolpost configuration is different between Lardi and Boni types, I don't know whether they are interchangeable but in principle it's possible.
|
|
Links to various small lathes
3
The Craftex link made me realize that a table of links to photos and information about various small lathes of any origin would be very nice. I'm not sure how to work out the permissions for a single file yet. But I'd like to develop one in the Files section. For now, if you have links to things like the Craftex, please create a file and post it to the "Small Lathes" directory.
|
|
Yeah, the old base was pretty bad. I think it came very roughly scraped and concave so that it might sit flat on the bed ways. The front face of the registering key on the top was the same way, and combined with the divot from the set screw on the back of the key it was difficult to adjust. But my new one and my old one have the same size and shape base. They’re both fitted to the 7x16 bed now, and I copied the 75° dovetail and gib from the new one when I redid the old one. They both clamp the same, though with thicker plates and fitted so they don’t tilt or skew as they tighten. The captive left-hand threaded clamp handle on the new one works okay but I got used to keeping track of the wrench. I rebored the casting and made a new barrel for the old one, and it fits as well as the new one. It’s 22.5mm diameter instead of 25mm but it isn’t slotted or bored out for a Morse taper. The new barrel made a big improvement, but the old tailstock is still flexible compared to the new one. Once they’re clamped in place, I’m not sure what to point to other than the structural design of the casting. It looks like newer, non-nesting cam lock versions shown on the lower price lathes on Amazon and Ebay correct that. Looking at pictures of other, better small lathes I think it’s interesting they seem to have changed from tailstocks like my new one where the barrel is longer than the main casting to ones with shorter barrels that work more like the old one used to.
|
|
Chuck grinding
2
Chuck grinding is one of those subjects that tends to incite fear. The chuck is sort of a sacred object it seems. I don't believe it is necessary to fear chuck grinding as long as you do your homework. In practice I found there is one good reason I know of to regrind the stock chuck on 7x mini lathes. First I want to point out something I discovered after I ground my stock chuck after the fact which I would have liked to have known beforehand. The main source of eccentricity on the stock 3-jaw chuck is the scroll wheel which is responsible for holding and clamping the jaws. The scroll wheel must be perfectly centered or the work will not be held on center in a 3-jaw self-centering chuck. The typical runout spec is 0.003". If you grind the jaws on the chuck, your runout at the position they were ground will be zero. But when you turn the scroll wheel 180 degrees, the runout will then be the maximum runout of the scroll, let's say 0.003". Now if you were to somehow deliberately grind the chuck 0.0015" off center in the opposite direction of the runout, then instead of a chuck with runout that bounces between zero and 0.003", you would get a constant runout of 0.0015" but no positions where the runout is zero. When I ground my chuck I achieved zero runout in one position but now have to live with larger runout than before on other positions (it is still about 0.003"). If the scroll error is cyclic and in sync with the scroll revolutions, then it may be possible to regrind and sleeve the scroll ID/OD to eliminate runout. That is a big if and could be the subject of some future endeavor. The stock jaws are not ground on the chuck at the factory but on a separate fixture. The jaw tooth surfaces are flat rather than circular which they would be if they had been ground on the chuck. They are perfectly perpendicular to the jaw guides. In practice this is an issue, because when clamped the jaws are never perfectly parallel to the jaw guides. They toe out unless you have a bar all the way through the chuck. This means the tips of the jaws often don't touch the work, increasing the effective stickout distance and allowing the part to deflect (as seen by the jaw impressions, tri-lobing and chatter of cut threads). And when you do have it all the way through, the jaw guides are perfectly parallel which means the jaws are very poorly constrained. When I ground my chuck jaws all I did was spin the chuck fast and allow centrifugal force to pin them back with a similar toe out as when holding work in front of the chuck face. With the jaws ground in this working position, toe out is now accounted for and workholding was improved. I was able to hold a 1' bar with no detectable runout at the end, and jaw marks on stock extended to the tops of the jaws, showing full contact. This method gets a lot of doubt but in an ironic twist much of that doubt is purely speculative and I had good results. This is not a perfect solution because the contact points on the jaw guides change depending on work diameter, which changes the toe angle. However it is certainly not a step backward. The centrifugal force on the jaws is probably comparable to the force used to hold a thin tube, but lighter than the force you would use on a steel bar for roughing. So for roughing work the jaws still toe out slightly. There are youtube videos showing various fixtures people have used to apply full force to the chuck jaws for grinding. This makes the jaw parallelism ideal on solid stock but then slightly less ideal on things like thin walled tubing. So you are now at the limit of what the chuck can do and you must make a choice to make your fixture representative of the most important use case. But in either case it is likely an improvement over stock.
|
|
To start this thread you should really start reading the posts here as this is where it started and provides the context: /g/machine-tool-rebuilding/message/13 A horizontal force on the gib will produce a force normal to the surface and a downward force. I don't see an issue with the gib being in contact with the base of the male dovetail. That's a precision surface. The problem is that if the gib can't preload itself against the gib screws using the downward force, you don't know where the screws will seat in the gib bores. Yes, you can put points on the gib screws but this is self-defeating as it creates high stress at the screw seats so they deform quickly with use (confirmed, I tried it). Well then you say, use a ball and socket shape for the screw and bore, or make the screw tip and bore exactly the same diameter, or use a conical tip going into an undersized bore so that finally, we will have a gib-screw interface which seats correctly every time and never deforms. But no, you still have a problem because the screws themselves have play in the threads and move around easily under stress. So maybe you fix that somehow, but also did you consider that the screws tips may not be perfectly concentric, and so adjusting them always introduces errors? So then you find a way to fix that. Then what you're left with is a very finicky mechanical system where numerous small things all have to be working perfectly and is basically impossible to maintain in optimal condition and is always degraded after a repair. The form taken by this error is also particularly bad. What happens is that when horizontal force pushes the gib against the dovetail, defects in the system flip a coin to determine whether the slide is deflected down against the slide or up away from it (or both alternately during a cut like a toggle switch). The latter is bad because the compound suddenly becomes very compliant. This means that if you clamp the gib screws in an attempt to rigidize the compound, you may actually end up pushing the slides slightly apart, and then you lose accuracy as well as rigidity. I considered this stuff when I wrote my webpage. In my solution the preload between the gib and screws serves not only to rigidize the gib but to stabilize the screws and screw seats. Preload here provides us with room for error, room for wear in, and a system that doesn't need to be too reliant on everything going well. BTW I found confirmation that there should be clearance between the gib and the upper surface of the female dovetail in a short article about making gibs in one of Guy Lautard's "Machinist's Bedside Reader" volumes. On reflection, that makes sense. You would not want the upper face of the gib to bind against the clearance face of the female dovetail. That requires that the lower face of the gib be forced into contact with the horizontal surface of the male dovetail I'm interested in this article. A solution would be to shim the gib enough that the upper left corner is below the male dovetail corner. Another solution would be to make a groove in the left face of the gib to provide clearance around the dovetail corner using a Dremel mounted something like a router. Anthony, as you have done a good bit of work in this area before, would you mind posting a bit about measuring the deflections of a dovetail way in various positions? The real focus here should be, "How do you determine what the problem is? How do you measure it? How do you fix it." With links to projects-in-metal for how to make things you don't have. You determined that the gib was the problem, I simply accepted your premise as I had experienced it. I posted some pictures of my measurement setup. I wrote a webpage explaining how to fix it. I didn't have to build anything to do the measurements, I just used a random pipe, spring scale and tenths indicator. Depending on the state of your gibs, fixing it might be as simple as adding a shim. I was mainly concerned about sideways forces on the cutter, either from facing or trepanning or from the cutter being extended past the edge of the cross slide so it acts as a lever. I quantified this by using a tenths indicator to measure the movement between the slides close to the gib. I put a pipe over the toolpost mounting bolt and used a spring scale to apply a known force and extrapolated the force on the gib according to the ratio of lengths from the gib and scale to the fulcrum which is the left side of the cross slide. You can see my setup in this picture and I also included the force deflection chart.
|
|
I decided I'd finally bought enough electronics test gear and so I unpacked the mini-lathe to start work on it again. The hiatus proved useful as I completely rethought the 3 point mounting system for the lathe and this should work much better. The first photo shows the parts. It's 4" and 6" pieces of 1" x 1/4" HRS, three 3/8" swivel feet and three 3/8" bolts. I still need nuts for leveling the feet. Also I've not yet marked out and drilled to bolt the HRS to the lathe. The 2nd photo shows the supports assembled (minus nuts) The third photo shows the lathe set up on the legs. By using 3 points and ball joints at each support, there is no way to apply twisting force to the bed of the lathe. For someone working in a cramped space, the bench top could have shallow counterbores to take the swivel feet. That would allow storing the machine easily, but still having it securely mounted to the bench when in use. All without concern for whether the lathe is actually level, etc. Have Fun! Reg
|
|
Well, at least this is why mine isn't. It's probably a common defect, but there are so many ignorant bodges that attempt to fix the problem with shims that actual standard lathe building practices have never appeared in any mini-lathe discussion I've read. The problem here is that the flat portion of the headstock to the left of the vee needs enough metal removed to let the vee fully mate. IIRC I measured it as 0.003" using feeler gauges on both sides of the vee and a bit of trig. So this is a few hours of bench work with files, scraper and surface plate to fix. Easy to do, just time consuming if you're not proficient at it and have to feel your way along for the first time as I shall have to do. Though I do have the advantage that I regularly practice my bench work instead of setting up machine tools for a trivial task. It's a finger tip skill just as is trowel work or playing a musical instrument. You've got to stay in practice. One could bodge it by inserting a shim along the side of the vee, but that would not correct any lack of spindle parallelism in the vertical plane. So a 2nd shim would be needed and probably another to fix the errors that introduced. Have Fun! Reg
|
|
Some context on my current lathe rebuild project
UPS dropped my 7 x 14 at least 4 times before delivery. I foolishly opened it. If you buy a mini-lathe and it shows *any* damage to the packing on delivery, refuse it. Any other course of action will result in getting screwed by UPS. Aside from the broken cross slide screw which did *not* match the parts that Small Machine shop sells and had to be supplied by the seller, dropping it upside down after the initial warm up drops of 3-4 ft spanked the inboard bearing. I'll get replacements eventually, but for now I simply moved the damaged bearing to the outboard position. I've got a single shot rifle to build and I don't need better than 1-2 thou for that. But I'll eventually be out the $70 or so for replacement bearings if I don't change over to conical bearings. I'd already completely stripped the machine, removed paint and filler, burrs, etc. Then I got much too serious about buying electronic test gear and just making basic checks and a few repairs on four 5 ft stacks of gear forced the mini-lathe into storage for a while. Then I had the luck to get a very nicely refinished Page-Lewis Model C Olympic .22 LR single shot rifle. The Page-Lewis is one of 3 rifles I know of which were built from flat steel stock and by far the better design. This triggered a 40 year dream of building such a rifle from scratch. Sadly, the original ejector and spring got lost so I had to make a replacement. Doing that in a 10" x 20" Clausing 4902 using a 3 jaw chuck was rather scary as I had to file the workpiece with about 1/2" of stock sticking out. I made a working replacement though I'm not happy with it as it was made from a 16 p nail. It works and I have no desire to work that close to the 3 jaw again. So time to get the mini-lathe going so I can use an ER-32 collet. All of this is somewhat driven by having spent 18 months or more deeply immersed in FPGA logic and analog filter design which make it seem like a nice break. Especially as I do not plan to attempt the really fiddly stuff where you work for a few minutes, let everything cool off for a few hours, measure and repeat. I'm just going to address obvious faults and get it to basic ASME lathe standards for now. I have created 3 related lists on groups.io: machine-tool-rebuilding projects-in-metal single-shot-rifles while some overlap is inevitable, I'm trying to separate my posts so that the design aspects of making a single shot will appear in that list, but actually making the parts will appear in projects-in-metal. Machine modification will appear in machine-tool-rebuilding. Eventually accompanied by the repair of my Craftsman table saw as well as an Armor horizontal mill and some other projects Dad left me to finish. Have Fun! Reg
|
|
Looking for a reliable source for mini-lathe bearings with rubber seals
When UPS dropped my lathe they spanked the bearings. I measured 0.0005" jumps using a tenths indicator on the inside of the spindle taper this morning. The jump takes place over a few degrees and is completely repeatable. This appears to be the cause of the excessive TIR at the spindle that I measured yesterday. The Little Machine Shop listing is for the 7206-B-TVP which lacks the rubber seals. I've emailed them asking for clarification. I'm skittish about getting burned by Chinese counterfeits. FAG's designation is 7206-B-2RS-TVP for what I want. I've found listings from $220 to $15. Obviously the latter are Chinese. I'd like to get US, German, or Japanese made name brand (e.g. SKF, FAG, Timken, etc) bearings but at sensible ($30-50) prices. Does anyone have any recommendations? Have Fun! Reg
|
|
Mini-lathe spindle alignment fun
12
FWIW I reassembled the spindle and headstock today and have been struggling with aligning the headstock to the bed. I think it might be helpful to others to describe the experience. I have a 1/2 thou indicator mounted with a magnetic holder on the cross slide which is just sitting on the bed held by gravity. My major problem has been the fine adjustment on the indicator stand. Despite my disapproval of the use of shims to align a lathe headstock I am using them because I need to make some small parts and *really* don't want to make another such part using the 3 jaw chuck on my 10" x 20" Clausing. A portion of the part I need to make is 0.375" x 0.085" diameter and the only way I could find to turn it to dimension was with a file. Even the shallowest cut bent the workpiece. So I had to work with my fingers *very* close to the jaws on my Clausing. I play guitar. My left hand is very precious and the Clausing has the power to rip a finger off in a heartbeat. I used a 0.025" feeler gauge to force the vee into closer contact and then 0.002" and 0.003" gauges at the left and right side of the headstock to align the horizontal axis with the bed. I was going to file and scrape until I started to mark off the cut line with my height gauge and discovered that the line fell in the chamfer at the edge of the headstock. That makes taking off 0.018" and maintaining correct alignment challenging. I estimate a day or two at my level of skill. Obviously much quicker if that's all you do all day. This has been moderately successful. I've got about .001" or less TIR at the headstock end of the test bar and about 0.011" TIR 9.5 inches from the spindle. I've got about 1/2 thou of horizontal and vertical error along the test bar *iff* I have the bar in a neutral position and slide the carriage along the bed. The TIR equates to a 4 minute of arc misalignment of the spindle taper and the Chinese radial contact bearings. All in all for a $550 lathe that seems pretty decent. I'm hopeful that a better set of headstock bearings will eliminate that and that what I've got is good enough to allow me to make conical plain bearings for the mini-lathe with the mini-lathe. The price for a higher ABEC grade of radial contact bearings is almost as much as I paid for the lathe. Conical bearing seem a rather popular change, but extensive study of rolling contact bearings shows that the mini-lathe is well served by angular contact bearings. The spindle loads just aren't that big. I've been pleasantly surprised at how accurate the mini-lathe can be made with minimal effort. But the use of shims is *very* fiddly, so the traditional scraping method seems well worth the trouble. And grey iron is very pleasant to scrape, Steel is not. I shall eventually remove the shims and scrape the headstock, but right now I need to get some work done. Long term I do not trust the shim approach to stay in alignment. One of the things that has become quite obvious is the least speck of dust or lint in the test bar to spindle taper interface will produce large measurement errors. In fact, just reseating the test bar can make a large error go away. The 0.025" vee shim does not provide consistent alignment and requires pushing the test bar to get the correct front to back alignment of the headstock axis of rotation before tightening the headstock hold down screws. No matter how tight I make the screws, I doubt that the alignment will hold for long. The major lesson in all this is that how you mount your test indicator is *very* critical when you get below a thou. I've been using various generic magnetic mounts, but I now see a clear need for a bespoke mount made for the purpose which screws into a threaded hole on the cross slide. It's just too easy to set the generic mounts improperly without realizing it until you take measurements that are completely out of wack relative to the last set. That's a disaster if you're scraping. While I'm not happy with the current alignment, it is good enough to make the 1/2" long part I need to make and I have a full ER-32 collet setup for the mini
|
|
The rack is 16.5" long. The carriage has a maximum travel of 15" when the tailstock is at the end of the bed. But when I get close to the headstock, the gear and rack disengage about 1/2" from the headstock. It's easy to reengage it, but annoying. It's also a potential nuisance as I plan to use fixture plates which puts me closer to the the headstock than than a chuck does. The milled surface for mounting the rack extends to the left of the rack by slightly more than 1/2". This has me contemplating moving the rack over by drilling and tapping new holes in the bed. In either position the extra holes will be covered by the rack and only visible from the rear of the lathe. So it seems a pretty innocuous modification. You can, of course, use the top slide to move the tool farther towards the headstock without disengaging. But it seems to me the rack and gear shouldn't disengage in any carriage position. Is there something I'm missing, such as a reason to disengage the hand wheel? As there is no gearing to drive the rack from the leadscrew I can't see a reason to allow it to disengage. If there were, it would prevent crashing the carriage into the headstock under power feed. But on these the power feed is always done by the leadscrew threads. Do other mini-lathes disengage the rack near the headstock? Reg
|
1 - 20 of 29
Image Name
Sat 8:39am
Sat 8:39am