Turning speeds re-visited

I forgot to mench, in using small home machines, I've found it useful to check several areas of the machine's rigidity to get a visual/mental image of what are reasonable expectations of performance. Performance, as in terms of expected accuracy relative to dial readings, anticipated cutting depth, feed rates and the other variables involved.
For starting out with a recently acquired machines, I like to start by placing a length of stock in the various workpiece and cutting tool holding devices, and applying finger pressure on the extended (bar of stock usually) to get a general idea of overall soundness and rigidity of the various holding devices/mounting points. These checks are just a bit of diagnostic evaluation to get acquainted with any issues that might result in poor performance, and to investigate any correction procedures.
I mention these checks since the topics are proceeding towards adding various accessories to your machine which may require the mill head to be raised away from the machine base and table. The larger the space in the work envelope gets, the less rigid the setup becomes. Keeping the spindle nose close to the table will likely result in the most rigid setup.
I think I recall some similar checks that you've performed with our new machine.
I encountered these issues with the 3in1 machine I've mentioned earlier. These combo machines have an unusually large work envelope when considering the fairly poor/minimally adequate rigidity of the machine (particularly the mill head when it needs to be raised on the column).
Many users probably need to develop particular habits (adjustments etc) of minimizing the adverse effects of movement/flexing associated with certain machine characteristics, to obtain the best results that can be attained for any particular machining operation.
Changes in the rate of feed (pertaining to maintaining a consistent speed while using a rotary table) in nearly all applications will effect the surface finish, and even more so on less than perfectly rigid machines.
Operator error/aahhhshit moments happen, but a chunk of metal really isn't ever totally wasted until it's all reduced to chips. Other methods can be used to add metal to a dimension or rework/resize it.. welding, brazing and knurling are a few, or one can sometimes adapt a mating part to fit.

I did this in a sort of random, unfocussed fashion. I leaned on things and watched the indicator move. It moved less than with the drill press but still moved.
A principle I should write somewhere on my wall in large letters.
There is one of these machines sitting in the local House fo Tools. Supposedly someone bought it and then they changed their mind. I have been looking at it a little more each time I pass. Just out of curiosity. The impression I had was that it had a more umphy motor than either of their mini-machines separately. The milling set up looked certainly unstable, but I wondered if the lathe part of it actually benefitted from the concept: A bigger swing, bigger motor etc. Although the spindle speeds are limited to 117-1300 rpm if you can believe the web-site info.
The question always is: Is it better or worse than the current method? Angle grinder with sand paper and then random sander. The first step is the difficult one as the marks from that often show up at the end when one is doing 400 grit finish. If they were regular it would not perhaps matter so much. I tried to spin the plate using an angle grinder on my "red-neck lathe". The speeds with which I can turn it are quite impressive not to mention scary but the results were poor. I borrow heavily from wood-turners: The part, just like their bowls, ends up different shape and size. I won't even go into rescuing the plates if the etch goes sideways...

Basically, the speed range is determined by the maximum swing (the maximum diameter which can be rotated over the bed). It has to be slow enough to produce a SFM (Surface Feet per Minute) rate within reason for turning the workpiece material with HSS (High Speed Steel). So this determines what is a reasonable minimum speed.
Maximum speed is determined by the smallest diameter workpiece likely to be used, and perhaps boosted by using carbide instead of HSS as a tool material.
As a result, large lathes tend to start with very slow speeds, and go up to speeds which are scary with the size of chuck which they will handle. (My Clausing will go up to 1600 RPM, which is quite scary with a 10" 4-jaw chuck.) It goes down to 35 RPM.
A secondary factor making slow speeds very useful is when single-point threading to a shoulder. While your reaction time can be improved with training and experience, it is safer to start with a very low speed with anything but the finest threads. (Perhaps 20 TPI on down to 224 TPI or so you can use higher speeds, but when you plan to cut a 4 TPI thread, and are turning towards a shoulder, you really need to be able to disengage the half nuts before the cutter crashes into the shoulder.
A lot of the small import lathes do not offer speeds low enough to handle coarse threading unless you already have good reaction times.
If it is mounted directly to the bed's T-slots, yes. If you make a mounting plate, you can handle a larger one.
BTW How much space is there between the vertical dovetail on the column and the back of the table with the table cranked as far back as it can go? This can determine what overhang you can tolerate.
A 3-jaw chuck can have the jaws expanded inside such a hole to grip while turning the OD -- but the trick is how to make that hole in the first place. If you use your older methods to turn the OD close to size, you can then use reversed jaws on the 3-jaw chuck to grip it by the OD while you machine the ID on the mill. A lathe would be better for this.
And -- with the plastic gears, at some point a larger cutting tool will result in striping the gears.
Get spare plastic gears *before* you need them or you may be out of service for some time while they are ordered.
O.K.
1) You can't use a fly cutter on one which is held by a 3-jaw (or 4-jaw) chuck without risking cutting the jaws. (Unless you set up a chuck with soft jaws, and mill a step shallower than the thickness of the workpiece so you can access the entire surface at once.
2) Using a fly cutter typically will show an artifact in the finish a short distance from the edge as the cutter "rings" from the impact as the cutter moves from air to cutting metal.
3) I would consider using a smaller cutter and making multiple passes at decreasing radii until the entire surface is cut to the desired thickness -- and then try the finish called "engine turning" -- an abrasive embedded in the end of a dowel, or an abrasive rubber spin in the chuck, brought down onto the workpiece surface (making a series of concentric rings) then lifted, and the table is rotated about half the diameter of the rod and it is brought down again. Repeat until you have a complete circle, then move in by perhaps 3/4 the diameter of the rod and repeat. It can make a high-tech *looking* finish, although you can do it with only a drill press and a rotary table.
I don't think that you will be able to get a better finish than that with a mill and a rotary table.
Every time I see you display thousandths of an inch in fractional form I get slowed down in my reading. The common way in machining to show such dimensions is as a decimal fraction with a leading "0" if there is no integer part -- so that would come up as '0.020"'. Among other things which slow me down is having to make sure that there are only three zeros after that one.
The leading '0' before a decimal point (if there is no integer part) is to make it more difficult to miss that there is a decimal point there at all.
You mentioned "howlers" in another article. This is not exactly a "howler", but it is awkward at best.
Fractional notation is normally used for powers-of-two fractions: 1/2", 1/4", 1/8", 1/16", 1/32", 1/64", 1/128" and multiples of those. 1/128" is the finest that I have ever seen used in machining, and that only with very old tools. And of course, since fractions are more difficult to work with (just listen to the Metric advocates), decimal fractions are preferred -- especially since the micrometers and modern calipers read in decimal fractions. (Some recent digital calipers have taken a step back -- approximating a reading as the nearest fractional size as an option.)
If the machine will not cut to tolerances may *require* buying a better one -- depending on how necessary the tolerances are. For your project making sundials -- no real problem I think. For someone making parts to repair or built a machine -- if it won't cut to tolerance, the part you make is just so much scrap metal.
Good Luck, DoN.

My initial point was that there is a discrepancy between the lowest speeds of the mini- and bench lathes and the swing. The speeds were too high to accommodate steel workpieces for the given swing. This has been addressed in at least some of the mini-lathes but by no means all. I know at least one person found a way to slow the minimum speed down in a mini-lathe by tweaking the electronics.
1.342"
Now you tell me :-)!
Good thought! Like the proverbial ostrich I have stuck my head in the sand and have not even enquired about spare parts availability.
I was thinking for the surface finish I could stick the whole plate to another plate (clamped to the table) by a double sided sticky tape like they do at MIT.
OK, I have not considered that. Or maybe just buy a lathe ...:-) But seriously, I have enough now to get going on *something*. One can theorize only so far.
Not a howler. Perhaps a faux-pas. I will endeavour not to slow you down.
As far as I am concerned, the whole imperial system of measurment can be sunk into a deep hole and covered by a thick layer of excreta. But I have to admit that on some days I do not feel as moderate...
Still, you have not lived until you grew up on a decimal money, then had to learn to give change in pounds, shillings, pence, half-crowns and farthings and *then* had to relearn the decimal system *using the same coins*. Nietszche was right!
Absolutely. Style over substance any day (pun almost intended...)

You'll know what you want the machine to do after you finish a job to taste by hand polishing.
And the "cylinder" was slightly tapered, right? I'd make two identical aluminum square plates and drill out their centers to a press fit near the large and small ends of the taper. I'd scribe a line across one plate first and drill on it, then use that line to align the spindle with the centerline of the cylinder. The work could be clamped in the vise by the plates but it would be more secure if held by the (parallel?) ends of the cylinder.
A step drill will make a smooth hole in thin aluminum and will hold it in place better than a twist drill when you loosen the vise to remove the smaller-hole piece, to drill the larger one by itself. If the cylinder is between step sizes don't drill all the way through, leave a thin tapered lip that can be pressed onto the cylinder. Or use a tapered reamer.
You could use extruded flat bar stock. If you can clamp the cylinder by the ends the plates don't have to be machined to exactly the same length and the holes needn't be centered, only the same distance from the lower reference edge.
That's a good idea of yours as long as the diameter matches a punch size. It would also match a lathe collet.
You can use regular pointed dividers. Set them very slightly large and observe how far the point can slide down the side of the cylinder. The point's taper magnifies the motion. The feel is better if you rock the dividers toward the edge to see if the point catches or slips over it.
Frugal Yankee here. I stretch two meals out of one can of soup with rice. Home-made boring head: can see the pattern my Toolmaker surface grinder leaves.
I'm not sure how much of this technology you are aware of, or exactly what you are or will be doing, so I briefly describe how various accessories solve problems. Plus I hope other beginners might find it useful, or real machinists disagree and suggest better methods. I watch others carefully and read a lot, but a fair amount of what I write here is home-grown, see above.
Yes, in time and materials lost. I consider it an educational expense.
Jim Wilkins

When I looked into this the people who sold the unfinished back plates maintained that they had to be finished on the lathe they were going to be used on. This made sense, but I guess for the purpose of simulating an indexer/RT it is not necessary.

OK. Or diametrical cross hole, to be sure, to be sure :-)?
Using the method described I just could not see how I could use a centre punch. I was hoping that a small centre drill would take care of things... Oh, yes. Big time! On my Z it is a full revolution.
Nice! Will try. I got one of them Y-shaped gizmos but found it worse than useless.
NO!!!
Mount some of the same Wow! That re-defines "umstandlich"! But if it works...
Pretty much. The fact that the two methods coincided makes me think there was another reason for the miss.
Ehm, the $4.95 one. The only one in store...
That looks like the answer.
Mammaries on a bull come to mind...
I Point taken.

"Proper" being in the eye of beholder :-) I see much experimentation in future.
I think the taper was minuscule. Less than 1 mm over 25 mm.
I am a bit confused here. The purpose of this is a) better workholding of a slightly irregular shape, b) easier centre finding, or c) both?
OK. Will try.
Coincidentally yesterday I found a DVD on making a boring head for "next to nothing". OTOH if I put the $40+ for the DVD towards a new boring head, I shall have a half of it already! Not to mention that the "next to nothing" concept probably does not apply to the skill level.
Did you design your own boring head or did you use somebody's existing design. What is the capacity (how big a hole can you bore)? Of course for the MT? taper - you need a lathe, right :-)?
Makes sense. I have seen a boring head in use both on a lathe and a mill. I have in fact thought of a project I would need one for. However, if you gave me one right now I would have to refresh heavily on the nitty-gritty of its use.
Maybe I should buy a few more DVDs anyway.

It is not exactly *possible*, so you do the best you can. If the register and face are turned with only one setup, and the hole for the peg to fit the hole in the center of the rotary able is bored in the same setup, they will be concentric. This is the best that you can do there.
Enjoy, DoN.

O.K. The mini lathes are not good examples of much of anything. :-)
And just running the speed down by slowing the motor loses torque which is most needed at large diameters and slow speeds. Serious lathes (assuming belt drive) have both multiple steps on the pulleys to get multiple speeds (or a continuously variable speed pulley, which tends to consume horsepower compared to a fixed speed belt) and a back gear, which reduces the speed from a given belt setting by something approximating 6:1, and boosts torque at the same time.
But *really* serious lathes (which I don't have) have gearboxes in the headstock to give all the speeds, and typically only one belt speed. These will produce more torque than most belt drives, which is both good (when needed for cutting large tough workpieces), and bad (when you have a crash, you don't have the belt slipping to minimize damage to the machine). For a starting user, belts which will slip are a better choice. :-)
Good -- you can add to the maximum size of the rotary table then. This helps counter the overhang beyond the diameter of the table. And measure the overhang needed based on an orientation for the rotary table which will point the crank towards you -- and so the crank won't hit the table when you try to turn it.
:-)
Right *now* -- open the top and count the number of teeth on the plastic gear. (If it is all metal, that is different, but if plastic, get replacements *now*.) Then measure the OD of the gear, and the width of the teeth (thickness of the gear) so you will have some information when it is time to order more if the maker no longer supports it. You may need to try your hand at cutting gears yourself -- which will need the rotary table or better an index/dividing head, the proper form cutter (you'll have to know whether it is a metric "module" gear or a US gear, and the pitch, and the pressure angle (determined by the shape of the teeth). Better if you can buy replacements without having to take out a second mortgage.
Hmm ... perhaps the grinding sounds when you were drilling was the gears reacting to too much torque requirement from the drill bit.
O.K. That could work -- if you stick to a fairly small diameter cutter. The larger the diameter, the greater the forces which will be trying to make the tape slip.
Good -- get experience. But don't include destroying the plastic gear (assuming that yours does have one) until you have spares.
[ ... ]
I suspect that it slows others down too. I just was one who was replying to your articles, and happened to think of mentioning it when I encountered it for the Nth time.
:-)
I grew up with it -- and most of my machine tools are only calibrated in that system, so a complete change would be difficult. I *do* make things to metric dimensions when it makes more sense, but it is more difficult (especially threading) except on the Compact-5/CNC which has a switch to go between modes.
Ouch! I know the basic units and their relationships, but a lot of the names for the various coins tend to slip away -- especially when the coin may be marked "N pence" or "N shillings", but be commonly known by a name such as a "Bob". :-)
I which of the things which he said?
Enjoy, DoN.

:-) I think that one or the other will do.
Unless you have located the spindle over the axis it probably won't. And with the tapered workpiece, it is likely to encourage skidding. With a totally horizontal surface (flat) you don't have as much chance of skidding off center, and the center dill will work quite well.
The other advice -- very shallow milling on the top surface gives you both a narrow flat, so things won't want to walk, and makes it easier to judge the true center. But your tapered workpiece with the ability to tilt under pressure is a problem anyway.
Ouch! (But then, you aren't measuring from the Z here.)
The same trick can be used on a lathe to check whether the tailstock is offset or not. Trap it between two hardened centers, one in the headstock and one in the tailstock, and an angle will tell you which way to correct. (You may be wondering *why* the lathe has the ability to offset the tailstock. This is one way of turning tapers, between centers. Lathes with taper turning attachments are significantly more expensive. But -- if you are doing multiple parts, you need to make sure that the distance between centers is the same for each one, or the taper will vary.
:-)
I've not yet tried it, because the ruler technique works well enough most of the time, but it is really a good way. If you are doing a lot of the same kind of part, you can either drill undersized, ream to just barely clear the drill bit and harden, or drill oversized to accept hardened drill bushings -- so the sides of the drill don't enlarge the hole as you work, introducing the possibility of more and more error.
Yes -- tilted surface and perhaps the spindle not at true right angles to the table, so a longer drill will touch down at a different spot than a stubby center drill.
Hmm ... what *store*? One which would stock Starrett tools, or one which would stock only the cheaper tools?
It is. I use them for a lot of things -- including scribing a line parallel to an edge of the workpiece.
Well -- I once saw an excellent answer to that sort of comment. This was in alt.folklore.computers, IIRC, and a young poster (who had his girlfriend in his dorm room at college, and was posting between sessions) commented that it was nice to have two more erogenous zones. :-)
But still get that first spilt-point drill -- just so you can see how much nicer it drills. :-)
Enjoy, DoN.

OK, just thinking aloud here: I get the unfinished adaptor. I attach it to the RT by means yet to be determined. I center it (and the RT) using an indicator. I start milling the periphery of the register disk carefully till it fits my chuck. Feasible?

Still, this sounds like a plan. Once the center is located it should not take much longer to change to a small mill.
Just illustrating a point. X and Y are not as bad.
Oh, good!
They won't let me into the one that sells Starrett tools :-)
What, more work?
I promise not to cry over it (sorry, could not resist it, I am sure you meant *split*) :-)

Clearly I need *two* lathes. A serious one and a comedy one...
In the end the answer will: Get hold of one and see how it fits. Not a simple proposition locally, but not insoluble. At least now I have much clearer idea of a) what I expect it to do b) what it can do and c) what to look for
I do not think so. I am pretty sure all the bad noise came from the workpiece. I know what the gear grinding sounds like - before I discovered the trick of changing the speeds :-)
1/2" - 5/8" end mill? Now here is a thing I do not remember from the DVDs: Clearly, if one has a 3x3" surface and only 1/2" mill (for arguments sake) one has to do this in several different mill positions. Each position may require several passess. So to get to an even depth, do you a) mill each position in turn, change depth and do subsequent passes in the same order or b) do you mill each position to the required depth by however many passes it takes and then move to a different position and repeat? I can see how easy it would be to get it wrong and have "steps" on the surface of the work instead of perfectly flat finish. Of course a fly cutter or a bigger indexing mill would solve the problem, but let's say "if"...
That is the sort of experience I am trying to minimize by doing due dilligence.
2 shillings was 10 pence. Half a crown (2/6) was 12.5 pence Sixpence was 2.5 *new* pence. And so on. Years of therapy... Most of them. The one applicable here I think is the bit about what does not make your head explode makes you stronger, or something to that effect.

That's a great looking boring head, Jim. If you have drawings, dimensions etc, the details would be an excellent addition to the dropbox.
When I had an interest in getting a boring head, I stumbled upon an eBay seller that had lots of the 3 primary parts (bodies, slides and screws) for sale, so I bought some of them. These were 2" models made in the USA by a company named Mesa from San Diego CA, that seemed to be liquidated finished parts of a discontinued product line.
The heads only needed a set of setscrews, a gib and a shank to be working tools. The only finished dimension that wasn't completed by Mesa was the 1/2" bar holes needed to be reamed/finished to 0.500" for an interference fit. The 40 tpi feedscrew heads didn't have the graduations marked, like a store-bought one does.

It was carved out of likely-looking scrap, the "plans" were a catalog photo. I guessed at the internal details, and as usual made each new part to fit the previous ones. The screw is 1/4-20 so the brass dial has 50 lines, the longer ones indicating 0.010" of hole diameter. The hex adjusting hole in the end was broached with an Allen wrench ground flat. The collar that forms one side of the bearing was brazed on. The rest of the bearing is on the back of the dial. The dial locknut is thick enough to use a standard screwdriver but a flush one might look better.
The integral gib probably should have been a separate piece held down by two cap screws. It required closer tolerance on the finished dovetails than I could manage, so I opened it up to accept a strip of shim stock, which works OK but slides out frequently.
Eventually I found a boring head on sale at Travers Tool with a B&S #7 arbor that fits my mill; the home-made one's arbor was MT2 padded with aluminum tape, removed now to use it as an offset tailstock in the lathe. They only list B&S #9 now, maybe I got the last #7.
The screw on the real one is 7/16-20 and threads into the body. A groove in the head of the screw engages a rib in the bottom of the slide. Since the screw is one piece there is room for numbers, but it sinks back into the darkness as the boring bar advances.
Jim Wilkins

OOPS, forgot the link; Jim Wilkins

Actually -- you first center the table on the mill using the indicator, then once the axes are locked down, you center the workpiece and clamp it.
But -- if you are centering the plate on the table, and are clamping it by the OD (using the T-slots), you have your clamps blocking the milling cutter.
And the ID hole is probably too small to reach through and use the T-slots at the inner end.
So -- you center it as best you can, clamping by the OD, then start drilling three or four holes at the smee radius (depending on the number of T-slots present), remove it from the table, and finish drilling through. Then you put it back on the table, with T-nuts in the slots below the holes, or with T-studs passing up through the holes, then lightly tighten the nuts and re-center the workpiece, and once centered, tighten them a bit more and then more until they are snug without shifting the workpiece on the table.
But -- when you talk of an "unfinished adaptor", you mean a purchased one made to adapt the chuck to a lathe? That will have a boss which is intended to go around the spindle and then a step back to a thinner section, so you won't have a flat bottom to bolt to the table.
I would suggest buying some cast iron of the proper diameter and starting from scratch.
Once you deal with the lack of rigidity from the central boss, yes. And starting with a cast iron disc, you will need to mill both sides flat, since the disc will have been sawn from a length of rod stock, and not really be that precise a surface.
Good Luck, DoN.

There is some doubt as to the necessity of the comedy one. :-)
At least you can make a cardboard layout of the table and the column at their closest approach to see whether things line up before carrying something home.
At least you don't have a large enough mill to have to beware the other end of the rotary table spectrum -- one which is too heavy to lift to your machine's table. :-)
[ ... ]
O.K.
With the workpiece held down by double-sided tape, I would consider 3/8" to be the maximum size for cutting the OD to shape. Perhaps 1/4" if you are going to be cutting a circular slot in a larger piece of metal.
For a slot -- you want a 2-flute cutter. For simply cleaning up the edge of the workpiece, the more flutes the better as you cut less material per flute, thus keeping the forces lower.
Plunge to a certain shallow depth, then cut the whole surface, then increase the depth and repeat. With the backlash in your Z-axis feed, I doubt that you could even come close to repeating the depth.
For a rectangular workpiece, cut along one edge, then shift axes to cut the edge at right angles to it, repeat until you return to your starting point, then move X and Y in to the next path (say 1/2 to at minimum 1/4 overlap with the previous, and repeat.
For circular, cut around the edge by rotating the table, then move in radially and repeat -- with similar overlaps. The 1/2 overlap will probably produce a nicer finish than the 1/4 overlap.
If the fly cutter is large enough to do it all in one pass. Otherwise, it increases the chance that a very slight out of-tram will produce noticeable steps in either the X or Y axis -- depending on where it is out of tram.
[ ... ]
Ouch! The old system I can remember (other than the common names for the various units other than the basic Pence, Shillings, Pounds, and Guineas.
Lots of duplications there -- including two adjacent with appear identical. I could see how various translations through time could produce some of the variations at least.
O.K. I had heard that (somewhat differently worded) many times, but not with attribution. :-)
Enjoy, DoN.