View Full Version : Still no 8tpi

06-24-2007, 11:03 PM
I hate to say it but here's my latest attempt at an 8tpi thread.:mad:


This was with the new boring block and life was wonderful until the compound depth reached .080. I thought we had it this time but as you can see.

The only thing left is to tear into the machine and see if the dummy who assembled it did anything wrong. I did my best to figure things out as I put it back together but it ain't like I knew what I was doing ......

Here's a a pic of the rear saddle gib.


This seems self evident. The screw in the center holds the gib down. You adjust the setscrews at the ends for suitable contact, lock the nuts, and you're done.

The front gib on the saddle is more questionable in my mind.


The 2 slotted heads, one's hiding behind the allen, pull up on a bar and if I tighten them up too far the carriage will lock up. So I run 'em just loose enough to get carriage movement and that's all I know about that.

Incidentally, I use the allen in that pic to keep the compound tied down with it's gib screws whenever I'm not moving it. It's so loose and grabby I've never trusted it fer nuthin'.

That leaves the cross. It's worn on the operators side some but I keep the gibs tuned to where I'm working. The 2 gib screws that are farthest away typically need the most attention as they hit the unworn area first. I run it as stiff as I can and still pull out of the cut when threading.

New ABEC-1's in the spindle which is the way it was originally designed and delivered. Chinese bearings of course but that's what Scott Logan sells for them these days too. I got mine from Kaman for a lot less money but that's beside the point.

Pre-load is accomplished by burying the take up nut against a spacer. Hard to mess that up. I have a spare spacer and have thought about shaving it a bit to up the pre-load, but I have no idea what I'm doing on something like this. More grasping at straws ....

And then there may be the fact that this poor old worn out thing just ain't capable of that coarse of a thread no more. I dunno. If that's the case I'll just buy a pre-threaded backplate and get on down the damn road! It'll still make lots of other useful stuff.

So whaddaya think guys? Give up? Try something else?


06-24-2007, 11:27 PM
What size boring bar are you using? It should be at least an inch, larger would be eaven better. Don't let it overhang more than you need to. How is your cutter sharpened? Do you have it set on center. The chatter in your threads may look bad but the threads will still work. Gary P. Hansen

Mike Burdick
06-24-2007, 11:58 PM

No, don't give up! Let's first start with the tool bit...

What is the clearance on your bit? The image below is for a square thread but the angle I'm referring to is explained in fig 150 below - this applies to ALL threads. Your bit should have that clearance or it will chatter.


Are you using thread cutting oil? If not, you can get some dark heavy thread cutting oil at Home Depot, Lowes, True Value Hardware stores, etc. Use lots of it; that will make a big difference.

If the above is okay, then we'll go to the next step...

06-25-2007, 12:30 AM
Just to get you further along with what you have, can you put a machinist's jack under the boring bar to help stabilize things? Either at the front or the rear of the tool post should pre-tension things a bit.

I've used a rubber bungie cord to help relieve the slop in my combo mill/lath on occasion to help fight the chatter.

06-25-2007, 12:38 AM
Face it, you have a light duty machine....

8 tpi threads are not easily cut in a light machine.

You might try opening the minor diameter way up.

06-25-2007, 01:06 AM
Try changing rpm while threading. Crank it up as high as you can while still maintaining control. As a general rule you get the best finishes on steel when the chips are smoking hot.

06-25-2007, 01:53 AM
Well thanx for the advice guys. This isn't a BBS, it's a support group for a buncha junkies!

I couldn't leave it alone. Went back out there and saddled up the carriage for another try. Got the saddle gibs tighter. Started locking the compound down with 2 screws. Had a sharpen run at the bit. After boring the old threads out I picked and picked at a new set of threads with more spring passes than I could count, and got this.


Still got a little vibration goin' on but I think we're getting in the acceptable range now. This is annealed 4140 I'm practicing on and I didn't realize how tough that stuff is until I wrecked a hacksaw blade cutting a piece of 1/2" hex the other day.

Cast iron for a backplate should be a breeze after this stuff. I sure hope so!

Thanx again.


06-25-2007, 01:57 AM
That is entirely acceptable.

Jim Hubbell
06-25-2007, 02:17 AM
I just recently threaded a backplate on my 10in Atlas. It was looking like the picture above. Even .0005" cuts chattered. When I was within 3 or 4 thou. I started pulling the chuck around by hand. Disengaging the back gears made it easier. At first the chatter could be felt and heard even that way. After several passes things smoothed out and I got the nicest threads yet. I put a little "blue" on the shoulder and screwed it on the spindle. It seats evenly all around on spindle shoulder. I used a 3/4" boring bar and a 1/8" bit. Compound set 29 1/2 deg.
I think setting the compound as close to 30 deg. as possible helps to keep the cutting forces down and therefore less chatter. Gotta' baby these old machines.

Mike W
06-25-2007, 02:26 AM
That chatter looks just like I was getting with a parting tool mounted upside down with the lathe running in reverse. Once I locked down the compound, it went away. It drove me nuts until I found what the cause was.

J Tiers
06-25-2007, 09:33 AM
You have a Logan 11", right?

1) the back gib seems WAY incorrect...... Is that per the manual for the 11"?

(you can and should get the manual from Logan)

The 10" and 11" are similar enough that I'd be really surprised if teh gibs were different, let alone like THAT "thing". I have parts from an 11" on my 10", and they fit and work, including a t-slot crosslide that I sometimes use.

What you show is WAY different from the 10", and also makes no sense. On the 10", the rear gib is held by all three screws, with tension set by the tightness. The extra piece is not present

With only the center one, the piece is unstable, not to mention the incredibly silly design that would represent.

Should likely look like this from backside

2) you are missing the carriage lock, it goes in the open hole at right end of carriage front.

3) the front gibs you DO tighten until they give minimum play and reasonable friction.

All the above would be clear if you had the manual from Logan. Cost you $25 I think.

At a guess, I would say that there may be more silliness involved in your machine, if the rear gib has been replaced with that useless POS make-shift. I would not doubt that the spindle preload might be wrong, the spacer might have been replaced by the previous owner.....almost anything could be messed up if the previous owner was that silly. Do they both measure the same?

You need the manual, and a good look through to make sure that at a minimum, the rest of the parts are present in correct form (other than carriage lock and rear gib, which we know are missing/wrong).

If you were closer, I'd take that "poor old worn-out thing" off your hands........ At a substantial discount, of course, as you have already described it as "worn out", and it has missing parts and substitute bearings............. betcha I could cut your thread on it....... not that you'd need it then.

I think you first need to get it to where it at least has the right parts in it......... THEN maybe you could complain if it doesn't do what you want.

06-25-2007, 12:07 PM
Try it with some hot roll or better yet some 1140 stressproof.

That 4140 would work fine on a machine that is more ridged.

06-25-2007, 04:42 PM
Will that plate screw onto your spindle?

06-25-2007, 08:34 PM
4140 is tough material and an 8 thread is a deep heavy thread.
With those two of those things together on the size of machine your using, the results your getting aren't uncommon.

90% of the parts I make at work are made from 4140 stock. Even on a CNC that feeds progressively down the 60 degree angle of the thread with each cut, an 8 thread will chatter sometimes. Difference there is once a CNC is in a threading process there isn't a thing you can do to intervene.

Cast iron should cut like chalk for you, but it is very abrasive on tooling.
I cut the M39-4 thread for a new cast backplate for my little Grizzly lathe.
4MM pitch is .157 lead so it's a little longer lead than your 8 thread.
That little machine isn't nearly as rigid as yours and it cut clean threads in the cast iron.

06-25-2007, 09:49 PM
JTiers, you're right. It's an 11' Logan.

And gosh, getting a manual was about the 1st thing I did. I took this whole machine apart and put it back together again and believe me, I'm not that smart! The manual was a necessity.

My particular manual has 2 Saddle Assy's in it. The LA-1097 is as you pictured, and then there's the LA-1060 which is what I got. That rear gib pic is kinda fuzzy, maybe I can explain. The manual calls the bottom piece a rear gib and the piece on top a gib bar. The rear gib bolts to the saddle from below like yours, but doesn't set the friction. The gib bar on top of it does that.

The gib bar is mounted in the center with a flat head countersink. The ends of the gib bar get forced up into contact with the lathe bed by means of setscrews at either end of the gib. You get 2 points of contact, both approx 1/2" wide, that you can set the tension on. That's my best guess anyway and that's why I put the pic up. I honestly don't see any other way to make it work. Just wanted to make sure I wasn't missing anything.

Good eye on the carriage lock too. You know your Logan's and I do have one. It just happened to be off at the time because I was studying the saddle gibs one more time. Would you believe I found washers stuck in there for a carriage lock? Those bent and mutilated things didn't have a chance of holding the carriage still.

Topct, that poor piece of metal has been my hands long enough that it will no longer fit anything else in the entire world! :D It started at 2.25 x 8 for my spindle. Wrecked that. Consulted the HB and stepped out to 2.5 x 8 for practice. Got better but lost again. This last attempt I just bored it smooth and got the threading bit out. If it fits anything it was an accident.

It's still in the chuck and I'm gonna keep fooling with it. A handslide like Jim Hubbell suggests could a good idea. I'll be trying that. I'm also partial to Evan's suggestion of a higher speed cowboy cut. :p Pure ignorance on my part but I've been known to do that .....

Thanx Guys.


J Tiers
06-25-2007, 11:23 PM
OK, I'm not familiar with that version, just the other one. It looked like there were no other holes in the strip. So it looked like it bolted on with one bolt only, which is goofy.

What you are saying is that it bolts on hard with bolts through other un-seen bolt holes, and what we see in the picture is only what affects the gib strip. That makes more sense.

It LOOKED like there was no screw in the middle hole, and that it bolted on one place. That would have been every bit as goofy as I suggested.

The other version as per the 10" is no big prize as a design either. But it is stable and works well enough.


If you don't have the carriage lock, you can't do the carriage lock test, which I was going to suggest as a general wear assessment.


Chatter is a sort of "stick-slip" phenomenon, like a violin when bowed. The cutter bends down until the forces are increased enough, or released enough, or both, that it can jump up while cutting slightly less depth. That gives the scallops as it repeatedly does the catch and cut routine.

The force increase is from the "wind-up" of the parts. The force release is from the cutter angle or the depth of cut changing enough to change how much force is required. Either may allow the cutter to cut a small scallop, before it gets back to the original situation and geometry, ready to repeat.

Chatter is greatly increased by either a resonance in the machine structure, OR a driving frequency.

A resonance may cause the scalloping to be greatly increased, by adding a resonant movement to the force-induced movement of the cutter. If "on resonance" the thing may about jump off the floor. Changing speeds will usually damp it, as will adding weights, etc.

A driving frequency, like the 120Hz pulsing torque of a standard single-phase induction motor, may provide a 'drive" that will turn incipient chatter into full roaring chatter. You can have a driving frequency at any harmonic of 120Hz, and possibly harmonics of 60 Hz, if the torque pulses are not all precisely equal.

So you can get driving frequencies of certainly 120Hz, 240Hz, 360Hz etc, and possibly 60, 180, 300 as well. That's a lot of choices and it would be surprising it none of them "hit" any machine or tool resonances, that they could "help out".

These torque-driven chatter types DO NOT respond as well to changes of speed, etc. None of the standard textbook stuff works as it should, it may just "partly" work, enough to make you think you are onto something, but never enough to fix the problem.

I do not recall if you have 3 phase or single phase.

I found a huge difference going to 3phase. I got far less chatter right away. I might still get a little noise, but suddenly everything that was SUPPOSED to change it, actually WORKED!

Might work for you too. Ok, that's a future thing if you don't have 3phase now.


Looking closely at the two pictures of the chattered part.........

The chatter marks are almost the same spacing on both. Were they cut at same speed?

Cutting in on top of chatter marks is tough, especially when you are at a finish pass, and not able to dig under them. it might help to change speeds, or lube etc.

Oh, yeah, and sharpen the bit until you can nearly shave with it.

06-26-2007, 01:28 AM
I'm glad my South Bend 9 doesn't chatter like that even though it has a single phase motor. In fact it hardly ever chatters. Maybe we have smoother power. ;)

06-26-2007, 07:17 AM

06-26-2007, 07:38 AM
Getting my 3 phase from a VFD driving what could be the original 3/4 horse Franklin-Morse motor. My machine was originally built with flat belts but along the way was converted to V's. Logan sold a conversion kit for them back in the day or someone coulda robbed them out of a 19xx series machine. Twist-lock poly-urethane belt.

I have had the impression the chatter is self feeding to some extent.

Both threads pictured were cut at 60 rpm.

No further experiments yet. Too whacked after work yesterday. Maybe tonight ...


06-26-2007, 08:12 AM
I have had the impression the chatter is self feeding to some extent.

Chatter is always "self feeding". It's a resonance phenomena where some part of the system including the tool and work are vibrating/oscillating at a natural frequency. All systems have some flexibility and will have some degree of resonance at certain frequencies. Once a chattered cut is established it is then nearly impossible to recut it smoothly without changing some parameter of the setup. Either the tool, the configuration of the tool holder or the speed must be changed as the chatter marks will act as a powerful exciter for more chatter. The easiest is to change the speed, either up or down. This will put the cutting action out of the resonance range where the chatter occurred. It doesn't guarantee it won't chatter again though but not at the same frequency.

06-26-2007, 09:42 AM
All of your photos look pretty clean, are you using any cutting oil? If so what? Dark sulfur cutting oil is still about the best there is for thread cutting.

My only other advice still is to ditch the 4140, it is quite possible that that is the best you will ever be able to accomplish with that alloy on that machine. It is probably the worst case combination you will be faced with. Until you try some other material, you will not know whether the source is the material or the machine.

J Tiers
06-26-2007, 09:57 AM
Yah, Evan has a bug for that power issue, he doesn't believe in it, I figured he'd comment. :D
All I can say is what has been observed here, on a flat-belt unit, at that. One quite similar to Evan's in general setup. But with your 3 phase it won't matter.

Do you have any data points on turning and chatter OTHER than this threading?

If you don't get chatter on most other work, then it is probably just a matter of the large cut when threading, combined with the 4140, which is harder to work with. That is reasonably normal and expected, at some point a cut is big enough to chatter on any machine within reason, although with some it may have to be huge.

In that case, looking further at the machine is probably not justified. Look at the setup, change the procedure, or tolerate the chattered part and get on with existence. It should work OK unless it is so rough that it can't fit consistently. Doesn't really look that bad, but we only see the picture.

If you keep messing with it, it may end up oversize.


If you DO get chatter on other work, when it seems you shouldn't, then there may be a reason to look further into the machine. There may be a remaining issue that is causing a problem, past the particular case.


Other things that strongly affected it included bearing preload. In fact that was the most important single item affecting chatter on mine.

You sourced your bearings elsewhere and not from Logan. When my original OEM bearings were replaced with recently made ones, the chatter was unbelievable. The new bearings were stupidly loose compared to the old. I measured the clearance, and it was *within spec*, but that spec had nothing to do with reality in terms of the machine. Preloading greatly improved the matter of chatter even with the 1 ph motor. But I was very surprised by the chatter that popped up with new bearings

Since yours is set by a spacer, it should be OK. But, is your machine one with the stack of belleville cone washers? if so, the way they are set up makes a big difference. I assume they are OK per the manual.

The springs take up the slack in a compliant way that compensates for small differences. But some units with double nose bearings won't "take up" the same on both bearings, unless the bearings are matched, which might possibly cause a problem.

And, if the setup is by spacer alone, are your new separately sourced bearings made to the same size down to the last tenth, as the originals? if not, they may not set up as intended. For opposed paired bearings (touching each other face-to-face), the slight offset of inner and outer races sets the preload when the nut compresses them flat together. If the bearings are set that way (and not by springs), but your parts are "standard" parts not actually made to pre-set that way, they may not have the right preload, if any, when the nut is set up as per the design.

A similar issue can occur when spaced bearing are set up by tightening a nut that forces them against a long spacer between them. Where the outer races bear on the housing, and the inner ones bear on the long spacer. It is the same basic deal as the opposed paired situation, but with the housing and spacer between.

And, opposed bearings only work right when they are face-to-face, so that longitudinal compression loads them in the correct way. otherwise they will be loose, although they may not seem loose to you. My preload-related chatter was disastrous even though the radial clearance was "only" 6 tenths. It was impossible to detect without a sensitive indicator, but the machine just roared with chatter.

But these matters should only matter if you have chatter in other more normal operations as well as this particular threading issue.

06-26-2007, 02:18 PM
pntrbl, try raising the cutter about .020"-.030" above center.

06-26-2007, 08:43 PM
JC, I'm using Tapmatic Extra Thick that's so brown and ugly it's staining the machine. Haven't a clue on sulphur content because no ingredients are listed. There's an MSDS somewhere I guess. I get it so sloppy wet sometimes the bits dripping when I pull it out. Sometimes I run it on the dry side just so I can maybe see how the chip's coming off.

JTiers, I have had success with an internal 1.75 x 20 tpi. Got a pic


And I can absolutely do 8 tpi externally on Alu. Here's another ..


That's cut to the mid-point of the Class 3A category as measured with my thread wires. I made it to test fit the internal 8 tpi I've been failing so miserably on.

Carld, I was thinking rake all day. Maybe the bit's too flat on top? Instead of getting under properly it's just bumpin' along? I will say this, I can do as much as 4 spring passes and still get metal sometimes. When I do that the next .001 increment won't catch much more than the last springpass tho. I dunno.

But I'd sure like to find out! :)


J. Randall
06-26-2007, 09:37 PM
One thing to try no one has mentioned, it to set your compound at 20 degrees instead of 29 1/2. It will distribute the edge loading a little more evenly but still cut a little less on the trailing edge. Works with carbide inserts if you are getting chatter, so it might be worth a try on HSS.

J Tiers
06-26-2007, 10:08 PM
Well, if you can cut other things, the machine is not "at fault" in any major way, and I wouldn't waste time looking for wear or problems that are not associated with the boring bar, holder, and the sheer size of the cut.

If you have stock left to cut from the part, WITHOUT going oversize, AND you can't work with the chattered part finish, I have one suggestion. (If you are "on" dimension and you can stand the chatter marks, don't mess with it)

Grind a threading bit, with a little formed area, one that will have relatively little cutting edge in contact at any time. Like a bit for doing 60 tpi, but on a "stem" from a larger toolbit to fit your bar.

Set the compound to the angle needed to move along one flank of the thread.

Now go ahead and pick up the thread with the tool, along the thread crest. Take a small cut depth wise, half or so of what you need to get off the thread at most.

Then slide down the flank about 2/3 of your cutter flank width, and do it again, repeating until you have the whole flank cut. You should have had no actual chatter, although it might have sounded rough as it cut through the chatter marks.

Set for and pick up the other flank and do again, aiming to end up "on" the pitch diameter, or slightly over.

That should clean up the chatter, leaving a good finish. But you can only do it if you still have stock to remove.

06-26-2007, 11:31 PM
pntrbl, you can raise the cutter and that will get neg rake or you can tilt the tool down to get neg rake but don't do both. Or you can grind neg rake on the top of the cutter.

The reason I prefer raising the cutter is I set the cutter on center square with the bore then raise the cutter what I want. If you tilt the tool it is hard to set on center and square. You can grind a slight neg rake on the tip and still set the cutter tip on center and square to the bore. Tilting the cutter down causes some set up problems and it's hard enough as is.

If you set the compound at 20 deg It will be a lot like a plunge cut. I have had best results with 30 deg as it only loads one side of the cutter. If you feel the urge to cut both sides then on the last cut plunge .001" with the cross slide.

06-28-2007, 10:34 AM
My 8tpi battles are continuing. I'd like to thank all of you who have responded with advice and help and apologize to all of you who are sick to death of hearing about this. While there is some discussion about about whether 8tpi in 4140 is exceeding the limits of my machine, there should be no doubt 8tpi in 4140 is exceeding the limits of the operator!

JCHannum and others have mentioned unless I try something besides 4140 I'll never know, and I have a "suitable for large bore 8tpi practice" piece of steel CCWKen sent me. (I hope he's OK down there in Texas BTW). It's alloy unknown and I sure can't assign a Rockwell number to it, but I've got a centerpunch and a hammer. I can definitively state it is softer than the 4140 and unfortunately it chatters up too.

I swear I'm getting so tuned in on this I can hear the chatter coming before I actually see it in the metal.

After boring a rather large relief area for the bit to run off into I did some 180rpm runs. As Evan suggested was possible all that did was change the frequency of the chatter. Was worth a shot tho!

Jim Hubbell's method of hand cutting to remove chatter definitely works. The amount of chatter I'm getting since you guy's prompted me to build a boring bar holder is so light it's really only a matter of cosmetics and my own personal stubborness. It's entirely conceivable I could rough the threads out with the motor and then make 'em pretty with a hand cut.

You could say problem solved but something I noticed when I was cutting with the manual method has me confused and concerned. When you're using your own two arms for motive power you get a real good "feel" for the chipload. I was expecting something halfway consistent and was surprised to find the load varied. I mean a lot!

For a thread or two it could be loose requiring very little effort on my part. And then a tight spot would show up. Back to normal for a couple. Etc. I got froggy and tried an .001 cut by hand and got it stuck! Had to back off and ram it thru.

Does that make sense to anyone? Would it possibly indicate a spindle issue?

I've had more "while sitting on the freeway" thoughts on bearing preload too. If all that excessive preload causes is premature bearing wear who really cares? I won't be asking this machine to run two shifts a day.

But then again, I've only been a lathe operator for 3 months and here I go trying to redesign Mr. Logan's lathe .....

Once again, Thanx to all for putting up with me.


06-28-2007, 11:36 AM
Hi There,

The gib bar is mounted in the center with a flat head countersink. The ends of the gib bar get forced up into contact with the lathe bed by means of setscrews at either end of the gib. You get 2 points of contact, both approx 1/2" wide, that you can set the tension on. That's my best guess anyway and that's why I put the pic up. I honestly don't see any other way to make it work. Just wanted to make sure I wasn't missing anything.

I have an 11" Logan (Powermatic version) and it had the same gib.
Personally, I didn't like this system because there was only two relatively
small points of contact on the underside of the bed way. I ended up
removing it and converting it to a conventional solid type gib. This gives
much more contact area and stability to the saddle.

Good Luck!
-Blue Chips-

06-28-2007, 12:49 PM
For a thread or two it could be loose requiring very little effort on my part. And then a tight spot would show up. Back to normal for a couple. Etc. I got froggy and tried an .001 cut by hand and got it stuck! Had to back off and ram it thru.

Ensure that the cutting edge of the tool is not approaching the work with ANY positive rake. In fact, a slight negative rake is preferred for inside work. Placing the tool slightly above center as mentioned has a similar effect as well.

06-28-2007, 01:04 PM
You're not using 4140 Pre Hard, are you?

That's a bitch to do internal threads even on a beefy machine...

06-28-2007, 07:20 PM
Bit geometry has come up several times and it's time to confess I'm still not understanding the difference between positive and negative rake. I'm assuming I've been using negative, positive makes no sense to me at all, and here's a pic of what my best guess is for an 8tpi internal threading bit.


Any critique is welcome.


06-28-2007, 08:35 PM
Your pic looks like it has a bit of positive rake.
For comparison sakes, here is my setup for M4x39mm internal threads. It's just a cross drilled shaft with a setscrew holding a cheap carbide insert. Negative rake works best for me. The material is 1144 stressproof. This is on a very lightweight 9x20 import lathe. I cut the threads at 40rpm.


06-28-2007, 10:32 PM
The "rake" is the portion of the tool that the chip flows across as it is peeled. On a lathe, that is typically the top, so we say "top rake". If the rake surface is tilted so that it wants to sorta slide in under the oncoming material, then that is "positive rake". If the rake is perpendicular to the tangential movement of the incoming material, it is "neutral rake". And if the rake is tiled away from the incoming material (as if the material movement had bent it somewhat past neutral) then it is "negative rake". Positive rake sometimes makes the cutter want to pull somewhat toward the material (depending on material and amount of rake among other things). Negative rake pretty much always pushes the cutter away from the material (which is why it needs a sturdy machine in general). There are exceptions though. For instance, you generally want neutral or negative with softer/gummy ductile materials like brass and some aluminum. Otherwise the pull in tends to cause over cut and tearing.

As such, the chip breaker on top of your bit would seem to indicate "positive rake". That's assuming you have the top of the bit mounted radial to the rotating material.

06-28-2007, 10:35 PM
Is that pic flipped? It just dawned on me that if not, it looks like the angle of the side relief would be indicating a left hand thread. Maybe it's just perspective? In any case, the vertical line formed by the intersection of the 2 side reliefs should closely approximate the lead angle of the thread.

Also be aware the chip breakers can be a problem for v-threading. That is because you are not cutting with the point, but rather along the sides. Dipping along the cutting edge changes not only the cut angle (which you want exactly 60*) but also the rake angle changes dramatically, and that could be part of your problem. A chip breaker for a v-threading bit looks more like a triangular dimple on the top rake so that the sides stay 60* and the rake angle stays positive at the same time. And to cut that dimple requires a steady hand and a small (relative to thread) carbide bur. When cutting 29.5* on the compound, I've also cut corners and just raked flat from the cutting side to the coasting side a tiny bit. This way it only affects your very light final cleanup cut and has a minimal effect (that you can compensate with the grind) on the angle.

J Tiers
06-28-2007, 11:20 PM
Chatter is really due to the things I mentioned earlier, the length of cut, which really means the FORCE needed to cut, the rigidity of the machine, and the geometry of the cut, whether the tool as it flexes cuts deeper or shallower.

if it cuts deeper when it flexes, you get a "dig-in" and maybe a crash.

if it cuts shallower, it either chatters, or adjusts the cut. A purpose made gooseneck holder will generally adjust, and a rigid setup that isn't quite rigid enough will chatter.

But that depends on having a cut depth that exerts sufficient force cutting to cause flex and stored energy that can be released when the resistance finally builds up enough to cut instead of "winding up". That makes the "scallop", and the next "wind up" starts even easier, because the snap-back may cut a little deeper.

The MATERIAL has less to do with it, except that harder to cut stuff may do it worse, particularly stuff that cuts pretty clean but hard. Gummy stuff is probably self-damping.

So it isn't surprising that the mystery metal also chattered with a similar set-up.

4140PH DOES have the tendency to cut easy and hard... it depends on tool sharpness, and also on local work-hardening from the last pass.

There may be places that you have to dig under with carbide. Been there, done that, with 10" Logan, single-phase, and all the added chatter "drivers" that exist with 1ph power.

06-29-2007, 12:54 AM
Some further news. I was boring a section out for a fresh test cut tonight and my spindle belt started slipping so bad the spindle stopped. It was only a .020 DOC and I've done .035 before. It's not as tight as it used to be.

Unfortunately the belt tension bolt was so tore up when I brought the machine home I had to shorten it. Seemed like it was gonna be OK but I got it bottomed out now and it's still slipping.

The tensioning bolt is a 5/16-18 and I can DO that, lol, so I'll make a new one, but do you suppose that slippery belt has something to do with my chatter issues?


06-29-2007, 01:57 AM


The situation is somewhat different when turning inside the work as compared to outside. Because of the change in geometry the work is approaching the tool from overhead and behind. This has the effect of reversing the effects of tool height vs center and exaggerates apparent rake angle. Because the work leaves the tool toward the tool the setup is very sensitive to being below center. If the tool is even slightly below center any deflection will result in a deeper cut and greater deflection. Positive rake exaggerates this even more. The tool also needs greater clearance below the cutting edge because the work is curving inward.

06-29-2007, 03:02 AM
Well thanx for that info guys. It's much clearer now and I've been using positive rakes the whole time. That negative rake doesn't make a bit of sense to me .... but I'm gonna try it anyway.


06-29-2007, 04:32 AM
Here is a diagram of what is happening depending on a couple of factors.

In setup "A" the tool (zero rake) is set directly on center and is also mounted on center in the boring bar with the cutting edge above the center of rotation of the bar (neutral axis). This is a common situation. When the tool is deflected by cutting loads it imposes a torque on the boring bar. The bar will both deflect down and rotate around the neutral axis under load. The circle that passes the tool cutting edge depicts the path the tool edge will take, increasing the depth of cut. This is an unstable system that will oscillate and cause chatter as the tool deflects more the more it is deflected. It's called positive feedback. A positive tool rake increases this effect.

In "B" the tool cutting edge is aligned with the axis of rotation of the bar. This can be achieved by grinding the tool appropriately, it need not be mounted as shown. Also, the tool is brought to the work above center. The resulting path depicted by the outer circle shows that under increased load the cutter will withdraw from the work reducing the depth of cut and the deflection as well. This is a stable system and will not tend to chatter. It's called negative feedback. Negative tool rake increases the negative feedback since as the tool deflects the cutting edge becomes less effective as the rake angle changes.


Note that a similar effect as "B" can also be obtained by simply rotating the boring bar and setting to the right height.


J Tiers
06-29-2007, 10:13 AM

I usually do the "similar effect" version, bar above center with tool slightly turned down, for boring.

Naturally, the turning up or down, and the grinding of a negative rake ALL change the shape of a "form tool" such as a threading bit. The resulting thread will not be exactly the shape of the tool.

So to do any but a small adjustment of that type, you need to compensate the form of the tool to avoid "blunting" the threadform to some degree.

06-29-2007, 11:12 AM
Yes, correct thread form is produced only with a zero rake on center presentation of the tool to the work. For large and deep threads such as 8 tpi the changes produced by varying the angle and center height are not so important. The effective shape of the tool when rotated is determined by the projected shape of the tool. This is the shape that is produced by a "shadow" of the tool directly below the tool. Turning the tool down, negative or positive rake and/or running it above center has the effect of changing the 60 degree included angle to a greater included projected angle with a shallower depth of cut.

06-29-2007, 03:18 PM
Just a niggling point, but particularly in faster threads, you would need to include the lead angle in your projection to get a correct form. Otherwise, the 90% projected angle will be correct, but the thread form will be narrow. This would, at least to some degree, compensate for the broadening effect of the third option, which is generally what I do most of the time.

06-29-2007, 09:20 PM
And to niggle your niggle, (:D) how fast a thread is depends not so much on the TPI but on the helix angle that results. In this case the helix angle is surprisingly small because of the large ID of the work. The helix angle (lead angle) works out to only 1.013 degrees. This can be safely ignored in all but the most critical applications.

06-30-2007, 12:26 AM

06-30-2007, 03:53 AM
I saw that post before, but don't know quite what to make of it. Seems like one of those fiddly projects that gathers dust except for rare occasions? I could see using it for this problem, just as you suggest, but also as suggested, how big a problem is it really? Enough to justify the time and materials to make, then setup? So far I haven't even gotten around to building a simple slotted 60* grinding block and have been quite satisfied with my hand ground 60* cutters fitted to fish gage via Mark I Eyechrometer. And on all except one, I haven't even accounted for the negative rake on ID, much less the lead angle I mentioned (other than for angling the relief intersection of course). Who knows, maybe one day I'll be caught up enough to give it a try. Wouldn't be the first time I found out that I shoulda' been doing it that way all along.

06-30-2007, 08:32 AM

06-30-2007, 10:08 AM
This 8tpi issue I've been having has been a long and bloody battle which is why I'm absolutely ecstatic to report that .... we've won! When setup properly this old Logan is completely capable of 8tpi in 4140. Look at these.


Good thing too. There wasn't much of that 4140 to practice on left!

It was a combination of "inexperienced operator" errors you guys steered me thru, but the primary one was the negative rake. I wouldn't have thought of that in a million years and Evan, I don't know how you do it dude, but that drawing with the circles made it so clear even I understood it.

And then you guys had a very timely discussion on the effects of bit height which was the next thing I was doing wrong. I've been in the Handbook and did some googling over compound depth at 8tpi and concluded .108 as the correct amount. Jeez, I couldn't get past .080 and the crests were getting too small!

That was due to bit height and the resulting wider than 60* projection. I just about climbed up that bore getting the bit slightly above center after that and got the full .108 DOC in the threads pictured.

So here we have a guy with 3 months experience on a 50 year old lathe that he put together himself doing 8tpi in a material that's so tough it wrecks hacksaw blades. That's almost unbelievable, but quite frankly, I don't see where I deserve a helluva lot of credit for it. Let's face it, without you guys I'd still be vibrating the turret off the end of the bed!

Thanx again. I can't ever say it enough.


06-30-2007, 10:43 AM
Well done! I wasn't sure the drawings were good enough. It was 2 am and I was becoming a bit tired.

06-30-2007, 11:01 AM
Nice job, Im glad you never gave up. :)

06-30-2007, 10:59 PM
"For a thread or two it could be loose requiring very little effort on my part. And then a tight spot would show up. Back to normal for a couple. Etc. I got froggy and tried an .001 cut by hand and got it stuck! Had to back off and ram it thru.

Does that make sense to a"nyone? Would it possibly indicate a spindle issue?"

No, I would look fot that problem in the gear train. Eight tpi puts a lot more strain on the gears than a higher thread count. On my South Ben 9N I cut all kinds of threads and never had a problem until I tried cutting 8tpi. I had a bad time until I noticed that there was a tooth missing in my 80 tooth gear. With that tooth missing the lathe would cut 16tpi all day long without a problem but it choaked on 8tpi. Check all your gears to make sure they have all their teeth. Also check to see that all the teeth are the same size and that there are no chips enbeded in between the teeth. Gary P. Hansen

06-30-2007, 11:15 PM
That problem was caused by the tool digging in because of the (incorrect) geometry of the setup. When being hand cranked it would be very sensitive to variations in speed and would dig in unpredictably. That was the key information I needed to suggest how to correct the problem as that sort of cutting action when hand turned is characteristic of a positive feedback geometry which is on the verge of oscillation.