Let’s say you recently purchased a lathe and installed it in your shop. Maybe you’ll even level it like a boss. You’re ready to make a chip, right? Well, not so fast. As a real machinist will tell you, you can use all the levels, lasers and whatever you want, but the proof is in the cut. Precise leveling can get your machine roughly level (the driver’s area is very narrow), but the final step in making sure the machine performs really well is trimming the test rods. This is a surefire way to remove the last traces of warping from your bed.
There are two types of test sticks. One is to check that the headstock is aligned with the guides, which is what we do. There is another type of tailstock alignment check, but that is a topic for another day.
First we sell some shares. You need something with a large diameter because we will have a lot of unsupported extensions, which is something you wouldn’t normally do. The equipment itself should be as rigid as possible. The longer the overhang, the more accurate the bed twist measurement, but if the overhang is so large that the workpiece cannot remain rigid when cutting, the test will fail. It’s a difficult balance. For the demonstration on my small benchtop machine, I used a 1-1/4″ diameter x 5″ long stock. For larger floor-mounted machines, a stock that is 2 inches in diameter and about 10 inches long is a good starting point.
Pull it as close to the four-jaw chuck as possible. The more failures we can eliminate now, the faster and easier testing will be. If your workpiece has a machined surface, this is ideal, but factory-produced cold-rolled steel will usually work. I’m using mild steel here, but a material like easy-to-cut 12L14 steel will make it easier to get a good finish (which helps with measurements).
The general idea is that we are making a barbell shape. We will make precision cuts on both ends, leaving a narrower area in the middle that can be easily missed.
Once the stock is dialed in, lower the release zone in the center of the rod, leaving about an inch of clearance at either end. We will only be measuring the two ends, so the middle will be in the way. Creating relief also minimizes tool wear between cuts (which affected our test results). Unloading 30-50 tons is quite enough. We need enough space to clear a few control cuts on both ends. Don’t release too much because we want the stock to be stiff.
Note that we are not using the tailstock as a support here. This is important because the tailstock has its own set of variables that affect alignment. We are only checking that the headstock is aligned with the rails, so we cannot use the tailstock. This means that we will have to make very light cuts since our stiffness is very low.
Once we’re done with the relief, we can make very light cuts in both measured areas. We need enough to clean the entire surface (so we know we’re within any chuck runout). I make the 2,000th cut here every time. Pass through the two measuring areas without touching the middle cross slide. Finally stop the machine and take measurements, then fold the bracket back and make another cut if necessary.
After making clean cuts on both measurement sections, compare the diameters using a quality micrometer. If they are different, the machine is cutting a taper, which means your bed is slightly crooked. Adjust or lightly adjust the machine’s tailstock leg and make another cut.
An enlarged tail end of the bar means that the right rake angle of the toolpath is too small (the tool head is moving closer to the workpiece during movement). If the end of the chuck rod is larger, the right front angle of the guide is too high (the cutting head moves further and further away from the workpiece as it moves).
How close you want to get these measurements is up to you, but for hobbyists, 5-6 thousandths of an inch will probably be enough. You can then grease the test stick with oil and save it for later use. After approximately 30 hours of back cutting, the same test rod can be reused several times.
Everything here belongs to him! Cutting a test rod is a simple hour-long project that will teach you valuable lathe skills and build confidence in your machine. Once you know that the machine can be trusted, you will realize that any future problems can only arise between the flywheel and the drawing*.
“If they are different, the machine is cutting a taper, which means your bed is a little warped,” or the fixed headstock is not aligned correctly. If you’re not sure, you can’t tell them. Therefore, you need to use fine alcohol to level the bed. It does not have to be completely level, the main thing is that the front and rear levels are the same. You can then cut the strip to check the head alignment. The procedure for moving the head is the same, just rotate the rod between centers as this does not take into account the fixed alignment of the head.
Right here. The most important thing is that when adjusting a machine, you need to be very aware of any assumptions you may make and make sure they are actually correct before proceeding.
Another source of potential narrowing is that the bolt-on headstock is simply not aligned with the bed. You _can_ use the tailstock when making barbell figures. But I’m not in a position to criticize, I haven’t even leveled the machine yet. Several years have passed.
Making a bar using a tailstock assumes that the tailstock is aligned correctly. It would be possible to use the tailstock to rough the shape, but when making actual test cuts it will only be held in place by the chuck in the headstock.
It is much safer to act without assumptions. I use the following steps: 1) Eliminate bed twist by checking each end with a machine level and adjusting one end (for example, using the built-in leg screws if equipped). This is even more troublesome to do with gaskets. I find it convenient to place the machine horizontally on top of the cross guides. 2) Turn the test rod between centers, then loosen the tailstock fixing screw and turn the side adjustment screw to reduce the error indicated by the cone. Retighten the set screws. Repeat the above operation until the cross guide turns parallel and locks into place. 3)_After_minimal adjustment, by all means try a parallel-turned test rod or precision-ground rod on your best chuck right now with a dial indicator attached to the cross-rail, but remember that you are measuring the chuck. The chainrings have the same runout and it is unknown if the narrowing is caused by a misaligned head. 4) It is best to secure the grinding rod in a new, clean collet whose taper matches the spindle taper. This may be sufficient even for fairly demanding hobby work.
If Andy can do this without having to level the lathe, then I think we’re not far off. (Because I _need_ to line it up before it becomes parallel – just centering the tailstock isn’t enough.)
Well, my lathe is pretty sturdy. I doubt they have many 10×20 class lathes with that much metal on the bed and base. The width of the bed is greater than the height of the center (which is common), but the depth of the bed is at least twice the height of the center, which is rare. https://photos.app.goo.gl/g5kvExAYDHpixrM86
Perhaps I should have made things clearer. What I mean is that the tailstock support can be used to rough the shape. Test cuts for this test must be made without support. It may be worth checking the tailstock bushing using a coaxial indicator. (This can be a quick way to return a misaligned tailstock to its original position after turning the taper)
There is little point in measuring a thousandth of an inch, marking it with chalk, and cutting it with an axe, as my father used to say. He has been using work drills for over 30 years.
Very good article. A slightly better method is to rotate the Morse taper to fit the headstock, drill and tap the narrow end of the tie rod, and rotate the collar. This will give you a bar that you can use to check the alignment with the dial indicator. A grinder with a tool holder is ideal, but sharp tools for fine feeds and light cutting will suffice. It was quite difficult to obtain tithes because heating became a major problem.
I tried making an 18″ strip between centers. This is very instructive. I still have this rod and will try it again eventually when I have a coolant overflow setup. Without coolant it’s nearly impossible, and hitting a tenth on a 6-8 collar over 18 inches can still be “extremely” difficult.
Thanks QD, I may have never owned a metal lathe… if I hadn’t read your series on lathes, I might have just bought a cheap (or crappy) lathe on a whim and never found out what needed for this. Basic tricks to make it really useful, leave it collecting dust in my store.
Don’t be too pessimistic. Lathes are a lot of fun to work on, and you don’t necessarily need the highest precision settings to do useful work. It depends on what you want to achieve – some jobs require more precision than others.
Also, accuracy is the sum of your equipment settings and your own technique, so as a beginner you don’t have to put a huge amount of effort into getting everything perfect from the start – you’ll just keep getting better and better. levels that can be explored in a very enjoyable amount of time.
It’s true that it’s rare to find a neat store that’s also a pleasure to shop at. Luckily, if you don’t mind not being able to fix your tractor or battleship, you can get useful work done with a small lathe. The oft-quoted advice to “buy the biggest thing you can afford” is true to a certain extent, but it also overlooks the fact that not everyone can save a lot of space or area, and many do amazing things that are done on small and rough devices. cars.
PS, fellow mods, please give the “report” button a “are you sure” because sometimes I accidentally click it when I mean “reply”. It’s a pity.
Well, buying a small lathe on a whim was the best thing I’ve done in years. This is the granddaddy of this hobby and I have been machining for several years now. Some of these things look/read complicated, but they really aren’t. If you can follow a process, recipe, or build a model, you can install a lathe or milling machine. The fun comes when you have different materials and you understand which one is best for what. HSS, Carbide – Bright, PVD, Tin, 304, 416, 4140. MY GOD. There’s a real art to figuring out what metal likes and what it doesn’t, and once you finally turn that perfect little cylinder, it’s the most satisfying experience. I never thought I’d like machining, but it was all over when I got my first little Seig. Since then I have upgraded a lathe and milling machine and worked on a CNC machine conversion project. Processing is truly one of the most fun things I’ve ever done, and it’s a completely satisfying artistic hobby. Give it a try. If you don’t try, you’ll never know.
But one thing: it looks like you’re wearing gloves. There seems to be a very strong belief among machinists that wearing gloves while operating a lathe is a serious safety hazard. Apparently they can get caught on the cartridge and drag your hands and arms into it. If I remember correctly, the US OSHA guidelines very clearly state to “instruct operators not to wear gloves” for most machines. The same goes for the UK HSE: “Wearing gloves increases the risk of entanglement and is completely unacceptable around rotating machine parts” such as hand-held lathes. I think most countries have something similar. Although blind health hacking Be skeptical about security issues, but the problem is serious. Cutting the chips is annoying and amputation is irreversible. Ladies and gentlemen, please be careful with your fingers.
I wouldn’t bet on it…the assembled nitrile glove is enough to pull your hand into the chuck, and then the flesh, tendon, and bone pull your hand firmly into it.
There are many stories on the Practical Mechanics forums about people wearing latex/nitrile gloves and aggravating their injuries as a result. Rubber holds up better than leather and takes longer to break than you think, sometimes even taking the leather with it. Apart from the obvious danger of getting your hands into the car.
However, even sometimes paranoid occupational safety organizations such as US OSHA, UK HSE and Dutch ARBO recommend not wearing gloves when working with hand-held metalworking machines such as lathes and milling machines. Nitrile gloves are tear resistant and durable enough to protect your hands under normal working conditions. But because they are softer and less rigid than leather/skin, they are easier to grip. Even the slightest jolt from a rapidly rotating chuck can be enough to cause serious damage. Don’t take any chances. If you do need to protect your hands from something like cutting fluid (contact dermatitis from oil can occur, so I’m not saying it’s impossible or unnecessary), keep a clean rag or gloves handy that Can only be used when the machine is completely stopped.
There are many barrier creams and sprays that can (in theory) prevent harsh compounds from being absorbed into the skin. Even a decent brand of skin moisturizer can provide some protection or at least minimize irritation.
Cream barriers work great unless you use a water emulsion (which, honestly, don’t do on a cart).
“Amputation” is what our security people at work called it a few months ago. Our manufacturer was sanding the shaft with a piece of sandpaper and the glove got stuck in the chuck and tore his thumb off. Once the initial shock wore off, those of us who understood the danger were left with only one question: Why was he, of all people, wearing gloves? Just a moment of bad luck and the car wins. . .
I have been servicing CNCs for many years and my rule is to never put your hands near moving parts. If you have to do this, your job is not safe :-[
If anyone likes to watch, read: Leveling this old lathe https://www.youtube.com/watch?v=THkb-x35fwc
Lathes are dangerous and wearing gloves is a very bad idea. About 15 years ago, someone posted a photo of a work accident on Rec.crafts.metalworking. A guy with long hair leaned against the cartridge, and the cartridge caught on his hair. After some time, he died. It’s not a pretty sight. The guy bent over the machine, covered in blood.
Honestly, trying to use a 3D printed plastic adapter to measure your case size seems like a stupid idea to me. As a fabricator/metal engineer, I suspect that the adapter is deforming (due to temperature and mechanical stress) beyond what you want to measure. You have a lathe and you can make it out of metal/brass 
Really? Considering that it is simply used as a stand, its location does not matter. The only thing that matters in this case is deflection due to pressure, every such indicator I have seen requires almost no movement of it, whereas any piece of plastic of any size must be larger than the spring force can provide. Even if she moves it, if it’s perfectly close to alignment, the spring force won’t change much, so the deflection should be pretty much constant.
I’d love for someone to actually crunch the numbers, but I just don’t see them being that serious about it. Of course it could be off, but I suspect it would be many orders of magnitude smaller than anything that can be read on the dial. Also, I suspect there will be little difference compared to a brass bracket. (If so, I’d like to know if I’m wrong.)
Why does every discussion about lathes turn into a rant about wearing gloves? (Or leave the chuck key in the chuck). People can choose what to do in their studio, just like they choose to base jump or cave dive.
I don’t remember the previous article in this series “devolving into a rant about wearing gloves”
I think it’s helpful to understand the dangers of these tools, especially for me, a hobbyist with no formal education or training. I pay less attention when it’s a friend’s machine, but they are no less dangerous.
Hi, I was wondering if there is any way to align the rails using the test rod method with an “offset headstock”. What I mean is that, assuming your bedside table is not parallel to the rails, now first level the bed using the test rod method (by test rod method I actually mean not using any level etc. lathe only), and then once If you are sure that the bed is level, you can proceed to leveling the headstock using the test rod method. So is this possible?
“Leveling” the bed essentially eliminates bed rolling—placing the bed rails on a flat surface. The available adjustments are typically nothing more than adjustment screws on the base legs, and only one end needs to be adjusted. (The bed does not have to be level, as long as its ends are parallel.)
So, if you have a large diameter ground test rod, one meter long, with parallel precision grooves spaced 180 degrees apart, a precision slide that slides along those grooves, and a lever on each side that each holds ” tenths” DTI, then the rotation of the bed can be measured from end to end, and if you have enough strength to lift it into place and it matches the length of your bed, you can make or buy one.
Precision machine levels are not pocket-sized items and can be manipulated with one hand, making higher resolution measurements much more convenient. It’s also damn cheap. It cost me less than $200 and I simply laid it on the cross guide, adjusted one end of the lathe to almost level, and then wrapped the guide from one end to the other. “Tool quality” lathes only measure +/- 0.001″ end to end and machine levels are measured better allowing for precise machine adjustments.
Although the article discusses turning a lathe on a slope, the next paragraph acknowledges that the test rod is only used to align the headstock and tailstock. But in reality he was measuring the runout of the cartridge. If you need any information about spindle to bed alignment, get rid of it and replace the collet that sits directly in the spindle taper. (In any case, if the lathe is of good quality, then simply adjusting the tailstock is sufficient.)
Post time: Dec-13-2023