I've got two old lathes, both about 80 years old. They have acquired excessive backlash over their lifetimes (rather than having it factory-fitted) but I think I'll try your fix. At worst I can put the original parts back on and I'll have learnt something. Thanks for posting this.
Agree! Great video, great skill, and imagination! I also saw a video on the " we can do that better " channel, and he made an anti-backlash nut that was really great.
Well done, sir. Setting aside the discussion over anti-backlash nuts, the bigger impact of this vid is showing how you can make quality parts with inexpensive hobbyist machinery. Lots of great work holding and tool holding tips in your vids. I'm a big fan.
absolutely agree! The actual replacement nut's design way a bit beyond me, but I learned so much about how the lathe can be used as a mill...really useful as a hobbyist who barely has place for a lathe :)
@@raelkoping5675Bit late to the party, so don't know if you'll see this, but whilst this is a great video (I've subbed on rbe basis of this video) and the content creator shows some great ideas, the way they hold milling cutters in the lathe chuck is a definite no-no. If you want to use the lathe as a mill (and if you don't have a mill, you should definitely give it a go; affordable benchtop mills are a relatively recent innovation and model engineers have been using their lathes as very workable makeshift mills for a long while), you should use a collet chuck. ER32 is a good choice; there are lots of reasonably priced ER32 collet chucks available for mini lathes. The reason for this is that milling cutters have hardened shanks, and lathe chucks have hardened jaws and so the jaws can't really grip the cutters shank sufficiently well. That lack of grip will, sooner or later, lead to the cutter slipping in the chuck jaws. You probably can get away with it for very light cuts but sooner or later it will probably bite you in the rear. At best you'll damage your cutter's shank, or the workpiece or both. At worst, because milling cutters tend to grab at the material, the tool could slip out of the jaws completely and end up damaging you! The only safe use of a milling cutter held in a lathe's chuck, if you absolutely don't have a choice, is using what is known as a 'slot drill' to either flatten the cone out of a blind hole drilled with a twist drill, or to make the flat bottomed blind hole in the first place. That's okay as the force back towards the chuck will prevent the cutter from ever pulling out. tl;dr: don't put milling cutters in lathe chuck jaws, get a reasonably priced ER collet chuck and a few collets in sizes you're most likely to use. 😉
Turned out excellent. I did a similar project to my Grizzly gunsmithing lathe about a decade ago. I simply bought a replacement nut and then cut it in half and inserted a spring of sufficient strength in between the two halves. The spring was held sandwiched in between the two halves of my new lead-screw-nut by four long-enough screws in each corner of the lead-screw nut. It too was adjustable but, my method was way harder to adjust than your design. I did it that way because I knew I could finish it in a day or less. It worked well for me though. I just couldn't afford to have my lathe down for any length of time.
It makes more sense to split the nut lengthwise. Attach one half to the crosslide and spring load the other half laterally. Because of the acme thread, provided there is tip clearance, this is like holding the half nuts closed with a spring force on a screwcutting lathe. This is how I got the idea: i noticed that my instrument lathe, which has a graduated dial on the screwcutting leadscrew handwheel, had no backlash on this (z) axis. (The halfnuts are held closed by a spring). Not much spring force is required because the Acme flanks are very steep (less than 15 degrees from square thread) which is less than the angle of friction. In the same way that a ladder will not slide regardless of weight if the angle is 15 degrees, the nuts will not open regardless of axial force.
Interesting design. I've been able to remove almost all the backlash in my mini-lathe by cutting a slot longways on the underside of the nut and squeezing it slightly in the vice. (Similar to the approach in Taig mill). But eventually this will wear, at which point I may try something like this.
Main take away here is how well you used what little you had, most people think you need machining equipment 100 times larger than what you have to make anything what we see you make in this video.
Like many here i absorb shit loads of shop & turning videos but I love lathe improvement, attachments, and upgrades the best. This project really shines because who the hell wouldn't love to reduce backlash to an absolute minimum? It's truly a rhetorical question. I'm going to investigate if I can adapt your excellent design to my old SB-H10 and Logan 920. Thanks for the very enjoyable and well done video. You did it with no monologue, yet the vid is clear enough to not suffer from its absence. Thanks! Wakodahatchee Chris
It’s people like you that make people like me, just like you. This is an awesome mod, specially for the Mini-lathe. Though, I would think with a bit of measuring it can be modified to fit any lathe that needed this. Great job.
I would do it so too. That way the central rotating part can be done in one piece with outside plate. There are "wavy" washers, which are ideal for this purpose.
Ten years ago I changed my minilathe to a brand new Chinese lathe with 255kg and 1,3 kw after I improved everything you could and It became a quite precise minilathe I sold it to a professional Vespa restaurator and he uses it to improve the cylinder and other parts of the italian motorbikes . I also changed the Nut into a sliced Nut with screws you could get the backlash down to a minimum...
Interesting way to do an adjustable backlash nut. I normally make a nut and bolt with a dowel an end plate on it then cut the thread. Then after this just take out the backlash with a shim and or grind. The problem with a rotating adjustment is it could move and lock it up. Its only a little lathe so it should be ok.
As usual you do really nice work. You got the backlash down to less than 0.001". Of course short of going to a ball screw, no one can eliminate backlash 100%, but instead only come close. It's why we always approach our work in such a way as to remove the backlash first. Even the most precise CNC operations are programmed to approach the work in such a way as to remove backlash.
Chinese mini-lathes like this one are metric, so double that assessment. ( I need to do this and more.. they are chatter machines after working steel for a while).
@@tsclly2377 Mostly due to a lack of mass but there is also the approach to machining required to minimize chatter. Very involved CNC machines will vary RPM on the fly for example to try and minimize chatter. Once chatter starts when machining a part you will of course have variations in the cut that are synchronized with the vibration requiring a change in RPM, and often taking a more aggressive cut for a short time to cut through those undulations in the material.
Chinese machinery is compromised in lots of ways. It's what makes them cheap. A waste of money if you ask me. I'd rather wait, save up and buy a decent used machine.
@@rollinrat4850 I've come to the same conclusion in hindsight. At some point I'll upgrade. I decided to spent a little more and buy a better milling machine than the Chinese offerings. I bought an A-head Bridgeport mill made in 1960. Loving it.
I've done lots of work on Bridgeports and big Cincinnati horizontals. My favorite lathes are Hardringe tool room lathes. Glorious, super accurate machines. I did some of my most accurate work on those, Vlieg horizontal boring mills and a big Cincinnati CNC boring/milling center. This milling machine had a programmable touch probe that allowed it to inspect its own work. Really cool! It was also laser calibrated every other month for highly accurate aircraft maintenance work. I actually enjoy using Matsuura mills the most out of all the CNCs I've used. Very heavy duty and long lasting. I really dig lots of Japanese made products. Especially bicycle stuff, fishing gear and Toyota 4wds. I actually enjoy manual machining the most, but CNC work was the most productive and payed the bills. I'm sorta retired now and build custom bicycles and wheels for fun and extra cash. That's my real lifelong passion. That and designing/machining my own bike junk to get exactly what I want.
I keep thinking there should be a way to make the Acme cross slide nut like an ER32 collet and holder, with a tension ring just behind the handwheel and dial. You could just turn it by hand to remove backlash.
I dunnoI watched this entire video and I still don't know what an Anti-backlash nut Chinese guy looks like! Humm, maybe I missed something. I enjoyed the video. Thank you, Sir!
7:05 well identified! the parting issue of the mini lathe. Seems like you've done quite a bit work with it to fix it that way. In general I'd rather install a ballscrew than this one, and make a new crossslide.
I can't begin to do precision work like the cuts you make. I have a cheap Jet lathe which has some kind of play which can make cutting tools dig in even on pretty small cuts.
You could also have taken the existing spindle nut, drilled two holes axially for hard pins (3 mm?) and then cut the nut in half so that you have two short nuts. Mill off both parting surfaces at 5 degrees and make a wedge at 10 degrees with a hole in the middle that is several millimetres larger than the spindle. If you then make a thread on the slide between the two holes for fastening the threaded nut so that you can press on the wedge with a worm screw, you could also eliminate the thread play in this way.
nice job did you lose any cross slid travel ? with the longer nut ? its open on the one end and from the look of the org. the mounting bolt hole to the front end looks about the same so ?
Bit late to the party, but I thought I should say that whilst this is a great video (I've subbed on the basis of this) and the content creator shows some great ideas, the way they hold milling cutters in the lathe chuck is a definite no-no. If you want to use the lathe as a mill (and if you don't have a mill, you should definitely give it a go; affordable benchtop mills are a relatively recent innovation and model engineers have been using their lathes as very workable makeshift mills for a long while), you should use a collet chuck. ER32 is a good choice; there are lots of reasonably priced ER32 collet chucks available for mini lathes. The reason for this is that milling cutters have hardened shanks, and lathe chucks have hardened jaws and so the jaws can't really grip the cutters shank sufficiently well. That lack of grip will, sooner or later, lead to the cutter slipping in the chuck jaws. You probably can get away with it for very light cuts but sooner or later it will probably bite you in the rear. At best you'll damage your cutter's shank, or the workpiece or both. At worst, because milling cutters tend to grab at the material, the tool could slip out of the jaws completely and end up damaging you! The only safe use of a milling cutter held in a lathe's chuck (ot a Jacobs style chuck too, for that matter), if you absolutely don't have a choice, is using what is known as a 'slot drill' to either flatten the cone out of a blind hole drilled with a twist drill, or to make the flat bottomed blind hole in the first place. That's okay as the force back towards the chuck will prevent the cutter from ever pulling out. tl;dr: don't put milling cutters in lathe chuck jaws, get a reasonably priced ER collet chuck and a few collets in sizes you're most likely to use. 😉
Although, wonderfully machine and a clever design.. it appears to me, that it is more of a slipper clutch or break-over friction clutch then a system that is designed to control lash gap in a controllable measure? Perhaps, maybe I missed something. But, yes, modifying this dinky little machines is so much fun, and affordable. I enjoy your content.
The nut on the cross slide has backlash adjustment built in, using the set screw between the 2 attachment screws you can take out the backlash by snugging up one end and turning the set screw in until the backlash is gone, then tightening the attachment screws
I think having minumum backlash is important especially on small machine or heavy cuts or turning square and etc.Would resist the tool getting pulled in especiallly your machine tool does not have sufficient gib lock or you are using that one axis for cutting.
Ultimately the screw will wear more in the region of the most frequent travel and then you have to leave some backlash anyway, otherwise the screw will bind up at either end. You must still develop the habit of making the last adjustment of the cross slide against the reactive force of the tool pressure.
Indeed. If I was going to this much trouble I would make a self adjusting nut which tightens wherever the feedscrew is worn and loosens where it is not. There are plenty of complicated ways this is done (eg on manual milling machines of high standard) but a very simple way is to use the geometry of the acme thread to your advantage. If the nut is split lengthwise (ie the split plane includes the rotation axis) the two halves can be spring loaded together. Because the spring (and adjustment movement) is orthogonal to the translation axis, the spring is not exposed to feed forces, so it need not be so tight as to promote undue friction and wear. I got this idea because my Myford lathe has a graduated handwheel on the leadscrew for precise moves of the carriage, and I noticed that if I applied a small downforce to the lever which closes the half nuts for screwcutting, and adjusted the stop which normally prevents the half nuts closing fully, the backlash is eliminated altogether. It is important when applying this idea to cut the nuts with generous clearance at both root and tip, so the the nuts close only on the flanks of the screw. I guess it is the lack of this clearance on most half nuts (at least, once they are worn) which prevents more people noticing the possibilities of this remarkably simple solution.
@@Gottenhimfella - seems valid, however an Acme thread is not a square thread. The flanks of the thread are indeed angled, and any pressure on the thread's axis will tend to open up a "spring closed" split nut. So, backlash can certainly be eliminated during no-load conditions using a spring, however as soon as force is applied to the tool, the cross slide/carriage would literally become rubber mounted and jump all over the place. It would take careful consideration anout the strength of spring required to be stable enough to cope with considerable and often intermittent loads from cutting tools. There is a certainly a mechanical advantage from the narrow 29° thread angle, but there would still need to be quite a lot of preload needed for the spring to cope with axial thread forces. That preload also applies at all times, whether the thread is loaded or not, so it might well be that the leadscrew becomes worn significantly more quickly by that continual abrasion.
@@johncoops6897 You're right in perfect (geometric) theory. A trapezoidal thread is not strictly orthogonal in the way that I (to keep the explanation simple) pretended. That would imply a square thread. However, real life materials, for once, come to the rescue, in practice, of something which should not work in theory. Geometric theory would tell the whole story if the split nut (and its guideways) were made of something as slippery as (say) wet ice. In the same way, a ladder made of wet ice would invariably skid out (particularly on a wet ice floor) at an angle of 15 degrees to the wall (which is about the angle, per side, of an acme thread). This "angle of friction" is exploited in mechanisms like Torsen diffs and Duff Norton jacks, to act a bit like a diode or a check valve. And that's what I'm doing here. It has to be a trapezoidal, rather than square, thread, or my proposal would not take up the backlash. But 15 degrees per side is about perfect (as I intuited from the fact that lathe half nuts do not try to open even when the carriage runs into the headstock)
It occurs to me that Myford lathes offer another example of this principle in action. I have an ML7 I inherited from my dad, and a Super 7 acquired decades later (when they finally became affordable). On the ML7, the tailstock barrel stays where you put it. On the Super 7, axial force can under some circumstances cause the barrel to retract if the clamp is not tightened. The reason is that the helix angle for the ML7 is well below the angle of friction. The Super 7 has a smaller diameter, three start thread, with a much higher helix angle, maybe four times as much. This is fantastic for drilling, being both quicker for withdrawing to clear chips and apply fresh fluid, and offering far better feel (important, for instance, with very small drills or reamers).
instant sub. but also, perhaps a spring system might have been better? that way you'd be able to use less pressure (less wear on the nut), it wouldn't come loose with vibrations, and it would work for a lot longer without maintenance because as the nut would get pushed forward as it wears.
@@CarlosSilva-gc8ny I don't get why you guys are being so defensive about it, it's simple feedback, a possible improvement on an already great job. And if I even had a lathe I would have already done it bud, and then I wouldn't be suggesting anything.
Отличная идея, скоро себе сварганю такую гайку, прижимные планки для продольной подачи уже сделал работают отлично, у меня вопрос по рпижимным планкам с клиньями, тебе не приходилось шлифовать станину снизу где скользят клинья или мне так повезло с моим *джетом 7*? Или эти станки все с одинаковыми болячками?
What stops the sleeve with the two tapers from screwing toward the main nut, undoing the "anti backlash" you achieved by "unscrewing" the sleeve away from the nut?
I'm surprised if you take all the backlash out of the nut that it will even let you turn the handle at all without a lot of torque. It will not take long to wear it till you have backlash again.
@@andrewwilson6085 - backlash is an annoyance, both on lathes and in vices, as well as anything else (eg: adjustable wrenches) where it means you can only sneak up from one direction. Minimising backlash makes all tools far nicer to use. It's unreasonable to expect to eliminate it entirely, however removing most of it just makes tool operation so much more pleasant.
Sehr schön gemacht. Das Umkehrspiel ist immer ein Problem, weil der Drehstahl oder der Fräser ich auch in die Maschine hineinziehen kann. Sie wissen nie wann und wieviel das Werkzeug gezogen wird. An der Drehmaschine tragen sie dadurch schnell zuviel ab, bei der Fräsmaschine zerbrechen sie das Werkzeug und mehr. Einstellbare Spindelmuttern mit Federn lösen dieses Problem nicht. Es muss ein fest einstellbare Spindelmutter sein, deren Gewindeflanken auf beiden Seiten tragen.
It was much easier, just give the existing nut a saw slot through the screw thread and then press with a screw that opens the saw slot, you have made a perfect backlash free nut that you can adjust. Costs you nothing and 10 minutes of work.
Nicer feel and also reducing the cognitive burden on the user. When you are in the last operations on a part, don't quite account for backlash and scrap that part, it's pretty frustrating. On a Chinese mini-lathe, even more so, because you'll have spent longer on all those operations due to the limits on the depth of cut you can achieve. 😉
Nice project! But on a manually operated machine backlash is not really a problem as you compensate for it by always feeding in one direction and make sure to take up any backlash before making a cut. And a anti backlash nut will just wear faster until it has just as much backlash again as the old one.
The anti-backlash nut is adjustable, so you just adjust it when it wears. Sheesh! The gibs on the lathe are adjustable. Do you leave them loose, because if you adjust them tighter they might wear so you'd have to adjust them? Do you not tighten your shoe laces because that would make them loose so you would have to re-tie them?
@@johncoops6897 Still backlash is not a problem on a manual lathe. Because you can feel the backlash and compensate for it by only dialing in one direction, or take out all of the backlash when changing the direction.
@@titter3648 - you are correct, that backlash is more of an annoyance than a serious problem. However less backlash and slop makes the machine feel a lot nicer to use... more "precise" even though is isn't really. My original comment is about the statements made in your first post. I am not saying whether such a modification is a good idea.
@@johncoops6897 You are assuming all the wear is on the nut. However the problematic wear is to the feedscrew, because it always wears more in the middle than at the ends.The nut cannot be tightened to compensate, because it will bind at the ends.
Great approach for minimizing backlash on a mini lathe! It looks like you've already addressed the leadscrew floating where it attaches to the handle, I removed 0.012 backlash at the nut and another 0.008" movement of the leadscrew. Scraping the dovetails took a 0.009" bow of it. Still need to do the x axis, thats going to be an incredibly long a boring job as I don't own a surface grinder. Over 0.030" to remove! Almost not worth it.
You should really use zero raked tools on brass, it makes a huge difference, backlash isn't much of a problem as your tooling does dive into the backlash anymore.
