A Couple Of Lathe Tips, Any Others?

[Dave Opincarne]

Here are a couple of tips some on this list may find useful. I bet others have some they think are common knowledge or not worth mentioning but others will find useful.

I put this one in the unusual tool thread, but I’ll repeat it here. I scored a good deal on Kennedy tool chest at a local pawn shop a couple of years ago. The price was good but not great. What made it worth while was it came with a drawer full of M2 lathe bits. As a toolmaker (aerospace composite) and patternmaker I routinely use machine tools, but not on a continuous basis, also, in commercial operations tooling is almost exclusively carbide so I’m not as accomplished at grinding single point tools as I’d like to be. These bits are all unused and look like the examples in a text book. Whoever ground them knew exactly what he was doing. There is no hesitation marks or indication of an unsteady hand on any of the various geometry. I haven’t used any of them and don’t intend to. What I do use them for is as readily available examples of what the bits I grind should look like. For that, they’re invaluable. Having an example to work to is a great help. I do this with fishing flies too. I’ll buy a quality example of a professionally tied fly and use it as an example to work to.

My second tip is one that I came up with the other night. It’s probably not new, but that doesn’t mean that I didn’t invent it again yesterday. (Ala Star Trek) In setting up and tuning my lathe in a new location I’ve mounted to studs in the floor and accurately leveled it, but wanted to tweak in the alignment of the tailstock. Here’s what I came up with: I chucked up a piece of readily available bar stock and turned a 60* point on it. With the dead center in place I moved the tailstock very close to the point I just turned and locked it down. Holding a thin parallel (any short, thin flat piece should do) between the centers, I advanced the tailstock feed until the parallel was lightly held in place between the centers. Sighting down from the top, I could see the parallel was slightly askew from perpendicular to the ways. I then adjusted the tailstock by a few thousandths until the parallel was perpendicular to the ways. No indicator was required as a very small change in the tailstock produced a significant change in the parallel’s alignment. I’ve used this same principle to place the point of the lathe bit on center indicating the parallel until it’s vertical (not an original idea) and adapted it to positioning round stock on center with a mill or drill press spindle. All that’s needed is a pointed tool (won’t work with a flat contact area). It also works better with smaller diameter workpieces since the error in picking up the tangent point is minimized.

So what are you’re setup tips?

[yf]

One thing that has become a habit for me is to make multiple passes when turning, at the same setting. (without advancing the tool)
On a small lathe this can be almost necessary because of deflection, not so important on a massive machine.
(you would be suprised how many passes it keeps cutting on a Taig especially if you took a heavy cut)
I always do this before taking a measurement, keep moving the toolbit back and forth till it wont take a chip anymore.
This gives me a better chance of geting the size right on the next pass, after infeeding.

Also I almost always finish turning with a file.

[docsteve66]

Yf, on a precision part, I do same thing. crank in some amount make a pass and measure. Do it again with no changes. write it all down so I know how many passes i can make until the cutting stops. then when the "last pass" is due, go enough larger than finished so that i "sneak up" to finished with a couple of passes left it things failed to cut as before. use the rubber band effect to get precision that i would have trouble getting if i had to advance the tool and back up to remove back lash and go again. That trick plus a dial indiactor instead of reading those ever smaller "hash marks" (as some one else calledthem) keeps me out of trouble
Steve

[docsteve66]

Yf, on a precision part, I do same thing. crank in some amount make a pass and measure. Do it again with no changes. write it all down so I know how many passes i can make until the cutting stops. then when the "last pass" is due, go enough larger than finished so that i "sneak up" to finished with a couple of passes left it things failed to cut as before. use the rubber band effect to get precision that i would have trouble getting if i had to advance the tool and back up to remove back lash and go again. That trick plus a dial indiactor instead of reading those ever smaller "hash marks" (as some one else calledthem) keeps me out of trouble
Steve

[Oso]

There was an article in HSM/MW about that, how to bore using the deflection and get right on without spring passes.

Not that I personally plan to do that. It needs a consistent feed rate etc, and I refuse to use the leadscrew for feeding.

I don't have the logan automatic apron, so the only pick-off power is for cross-feed. Feed is with half-nuts

[Dave Opincarne]

My technique is a little different. I feed half of the amount shown over the final dimension up to the maximum cutting capability. This has worked well for turning, milling, boring, as well as milling wood through a planer. The nice thing about this method is there is an automatic progression from roughing to finishing and it helps to avoid taking to much off.

Here's an example: On a small lathe turning up a .500" stub on a .750" shaft. My maximum cut is going to be about .050" so I'm taking .100" per pass of the diameter. After the first pass at .050" I should show a little more than .650" or .150" over. On my next pass I try to take off half of what I need so I feed in .035 on a direct reading cross feed which results in a little more than .575". This pattern repeats itself several times with diameters of .538, .524, .512, .509, .504, .502, and .501 up to the point I can finish off with polishing or I’m within tolerance. During the final passes I am doing little more than making a second pass with no infeed. Because each pass is lighter than the previous one the spring back of the cutter is progressively lessened. Also, because the quality of finish is partially determined by the preceding surface quality and each pass is progressively lighter the surface finish improves after each pass, setting up a smoother finish for each subsequent pass. This may take a few more passes than other methods, but using it I have never taken off too much material chasing that last little bit or screwed up a job because I had a brain fart when I confused the diameter and radius.

[Dave Opincarne]

Thought I'd share this too. I made a no-spill dispenser the other day out of a plastic soft drink bottle. I cut the top third off and inverted it and glued it in place over the bottom third using silicone adheasive. A chip brush fits nicley through the hole and if it's not overfilled fluid will not spill out when it tips over.

[darryl]

I never liked using the power feed, and I didn't like using the 4 inch diameter handwheel for the longitudinal feed, so I replaced it with a section of aluminum pipe 1.5 inches dia, about 4 in long. Of course, I had to make an adapter to get it attached to the handwheel shaft. Now I get much better feel of the cutting forces, can reverse feed instantly, and can use both hands easily to keep the feed going smoothly. (the extra torque required with the smaller grip helps keep the feed smooth). I've never wanted to go back to the original, and have thought about making an even smaller 'slide out' handle that can fit within the new tube, so I can get faster traverse, with greater feel. If I have to torque too much on this hand control, it generally means a dull cutter or I'm doing an operation I shouldn't be doing. I wouldn't even consider going back to the original handwheel, in fact I modified it for use on a different machine.

[Dave Opincarne]

darryl- I'm trying to picture this, is this a screwdriver type grip? What kind of machine is it on?

[Rustybolt]

Most of you guys probobly know this, but Ill include it anyway in case some one does not.
When drilling deep holes on a lathe use a center first. When the drill you've chosen hits the bottom of the center let it dwell there for 1/2 a second then beguin drilling. Drill the first pass no deeper that three times the drill diameter. The second pass, twice the drill diameter, and after that no more than the diameter of the drill. When you've gone past the flukes remove the drill often to remove chips and use a lot of coolant.
I've noticed on larger drills that when the fluke get loaded the chips will start cutting the bore, ruining the chance to clean up with a reamer.
On very tiny drills(smaller than 60 wire size)Its a good idea to peck drill the hole with no lubricant at least in steel. Don't dwell at the bottom of the hole with these tiny drills either as that wipes away the cutting edges.

[darryl]

Dave, it's a screwdriver handle, basically. It sticks out towards me, horizontally, almost as far as the crosslide handwheel, and is perpendicular to the bed, parallel to the crosslide. It's on an 8 x 18 lathe. I find that I don't get as much error using it as I did with the original handwheel, which, if you crank it, pushes and pulls on the carriage, in two axis, up and down, and side to side, as well as it's intended use, to traverse the ways. I have the same problem with the mill, when I crank the x axis, the table rocks slightly unless the gibs are tight, which causes another problem, more force required just to move the table, resulting in less 'feel' of the cutting action. I won't be changing the handle there, though, as it wouldn't he handy to use.

[darryl]

As long as we're on the subject of lathe tips, here's a brief account of my lathe motor mod project. The bearings were starting to get quite noisy in the motor, so I replaced them, cleaning up the motor and mount area at the same time. The new bearings didn't last long till they were as noisy as the ones I replaced. I went by the book replacing them, using the dished washer and the spacers to get some preload, per the design. The belt tension is set with a spring loaded idler, so I didn't overload the bearings that way. While I was still in 'motor mode' I went ahead with an old idea, that of removing the motor from the lathe, and mounting it separately, using a length of shaft and a pillow block to relocate the driving pully. Ok, now I could hear the phone ring when the lathe was running, the noise level was much lower, but I still had vibration, just different. I ended up removing the ac motor, and I now
have a dc pm motor mounted to the cement wall, belt coupled to a jackshaft, belt drive to the spindle. The original motor is history. I'm using a variac feeding a bridge rectifier/capacitor to a reversing switch, to run the motor. This variac is also isolating, no connection primary to secondary, for safety. Now I can hear the phone, the doorbell, the neighbors, etc., with the lathe running. This has made a difference in the surface finish I get, and gives me a continuous speed variation from about 10 rpm, to about 6000 rpm, which is too fast for the spindle, I think. It's a pleasure to use the lathe now.