What's new
What's new

A puzzling threading problem...!

Alley Cat

Plastic
Joined
Sep 1, 2008
Location
Cape Town
Hello everyone,

As you may have seen from my previous thread, I have recently purchased a Colchester Student.

I have been playing around getting used to threading with it but have run into a puzzling problem. (At least for me it is puzzling)

I have to thread with the half-nuts closed since I do not have a thread dial indicator.

For informational purposes it has a 6mm metric leadscrew and I am cutting metric threads.

Now I have done quite a bit of threading using my Myford and I have none of these problems - so, although not what I would call an accomplished threader, I also do not see myself as a complete beginner when it comes to threading.

The problem is this...: When I move the cutting tool in directly using the cross-slide, I get a good looking and appropriately pitched thread. Of course doing it this way doesn't necessarily give one as good a quality finish as one would like, nevertheless I could probably live with the thread, but I am looking at doing something perfectly here, not just getting it roughly right!

NOW...when I do what I have always been taught to do, and that is to swing the compund rest 29,5 degrees to the right, and move the tool in using the compound rest so that the tool only cuts on the one side, the thread is badly impaired - it almost looks as though the angle is changed so that the thread becomes somewhat wider and shallower. The thread is for all intents and purposes useless.

And yes...I am retracting the tool before I run the tool back along to the beginning of the thread to start the next run...!

I must be doing something wrong here that is quite simple to rectify. Please tell me what I may be doing wrong.

Regards,

Alley Cat:bawling:
 
External or internal ? Right or Left hand thread?

Ext RH - offset right*; LH offset left
Int RH - offset left : LH offset right

(Offset right = compound moves tool IN towards chuck. Offset left = compound moves tool IN away from chuck)

Is the Colchester's compound-slide protractor dial like the Hardinge, 90 degrees 'out' ?
I.E. you set the Hardinge to 59' rather than 29'
 
Ally Cat

The offset angle has to be measured relative to the top-slide at 90 degrees to the work-piece. Most lathe top-slide angle scales are calibrated so that the slide moves parallel to the work piece at the 0° position and perpendicular to it at the 90° position. Hence for 29.5° offset you put it at 60.5° on the scale. If you set it at 29.5° you get a wide misshaped shallow thread. If that's what you have done don't feel bad about it. I known for an absolute fact you're not the first to make the error. Beat-cha by a couple of decades at least!

By the way the usual presentation of the 29.5 degree angle off method has to be one of the most stupid "normal lathe work" methods ever promoted to the unsuspecting. Thread cutting is bad enough as it is, unnecessarily introducing trigonometry to calculate the thread depth is borderline insane. In my book trig is nearly as bad as advanced quantum physics. But at least quantum physics is interesting (my qualifications are as an atomic physicist so I hafta say that. Union rules.). Anyway who the heck can reliably and repetitively set exactly 29.5° angle off using the top slide scale

The easy version of the angled off top slide method was taught to me as the "zero to zero" method.
1) Bring the work piece is the correct diameter for the thread.
2) Set the top slide at approximately the right angle. Just has to be a bit less than the thread half angle and put the tool perpendicular to the work. (Mine lives at 27° so I can do proper British threads too.)
3) Bring the tool tip up to touch the work using cross and top slide movements as appropriate.
4) Set both dials to zero.
5) Withdraw the tool tip a few thou' back from the work using the cross slide and move the saddle so its clear of the end of the work.
6) In-feed the cross slide past zero by the depth of the thread to be cut.
7) Set the cross slide dial to zero.
8) Move back the top slide so that the tool tip is clear of the work then feed back enough for the first cut.
9) Use the cross slide to bring the tool clear of the work after each thread cutting pass. Return the cross slide to zero before commencing the next cut. Its all a lot easier if you have decent cross slide stops or a quick withdraw device.
10) Apply cuts as appropriate using the top slide. When the top slide index reads zero and no further metal is cut the job is done.

The great advantage of this method is that the lathe effectively does the calculation of the angular in-feed matching the perpendicular thread depth for the particular offset angle the top slide has been set to. No trig, thread depth straight out of the book and automatic compensation if the set-up is a bit out. Its also a doddle to cut things a bit slack or loose if necessary and its easy to figure out what went wrong if it all goes wahooie shape.

Clive
 
Last edited:
Alley cat, post a photo of the angle graduations on your new lathe's compound.

Some machines start them with zero straight in, others start them with zero being
with the compound parallel to the ways.

What is important, is that you start with the compound pointing straight in,
that is, with the compound slide pointing in the same direction as the crossfeed
slide.

Then swing the compound to the right through just about 30 degrees. On some
machines, after doing this the index on the compound will be pointing at "60"
which tends to be confusing.

Jim
 
A puzzling threading problem

First off, ignore the 29.1/2 deg bullshit. Set your thread tool by moving the compound through 30 deg to the right, from the straight position & forget what the graduations are on the compound ring as you might end up at 30 or 60 but the important thing is that you move the compound through 30 deg to the right & let the graduation line end up where it might. After moving the compound, set the thread tool correctly w/ a fishtail .Do this & you will be downtown w/ a good thread
 
Compound at 29 1/2 deg. from the "straight in" position to the right. Count! Reset you tool bit to 90 deg. relative to your work. And "yes" use 29 1/2 deg.
 
Anything between 29 and 30 deg will cut a good clean thread. How do I know, I have tried them all. I prefer 30 deg because it's easy to set. Some like 29, some like 29.5 and some like 30 deg but any of the three will cut a good thread. Actually, you can put the compound anywhere between the 29 and 30 deg marks and be ok. At 29 you will be chasing the right side more than you will at 30 and at 30 the thread looks as good as at 29.

It's really not that important, there are more important things to do when threading.

I too think he has the wrong angle and it is easy to do. The lathes I was used to had 0 at 90 deg to the work and when I set up to thread on a lathe that had 0 parallel to the work and I set the compound and looked at it I knew something was wrong. I used my protractor to check the angle and saw the error. From then on I was aware of that problem.
 
D'oh. I've just realised that there is an easy way to keep this offset angle business straight for a 60° thread.

Either

1) From the top slide travel parallel to the lathe axis position move through the whole thread angle. In this case 60°.

2) From the top slide travel at 90° to the lathe axis position move through half the thread angle. In this case 30°.

(Needs someone to come up with snappier wording).

Given the way most modern top slide angle scales are calibrated version 1 is easiest. However I've never seen it explained that way. Probably 'cos there is no similar simple rule for 55° threads. Would have saved me much grief and "have I got it right or not" thinking over the years.

Concerning the actual angle to set things at the threading tool is technically a form tool and so, by definition, must cut on both sides. Nothing about the cut being equal but to guarantee proper replication of the shape that trailing edge must cut. A half thou scrape is sufficient. Clearly if the offset angle is too great for the trailing edge to cut the thread form will be wrong. Which is pretty much where we came in! Hence the angle needs to be set a little shallow to get the trailing edge cut which ensures we get the thread angle as ground on the tool. If the trailing edge doesn't cut we get one side of the thread generated by the angle of travel of the top-slide which must, by definition, be different from that ground on the tool. Hence that pesky 29.5° recommendation to get a shaving cut.

As Carl says there are more important things to pay attention to when threading. The great thing about the zero to zero method is that it lets you ignore the angle question altogether beyond a simple "roughly right with a trailing edge cut" set up.

Leaving aside chip formation and crowding issues the big problem with the alternate straight in method is that you both make the cut feed and withdraw for the return pass on the same dial. Plenty of room for confusion as the cut setting varies every time. No sweat if you have a proper independent quick withdraw but otherwise its easy to get lost. The big advantage of the angle offset method has always been that it separates the withdraw and reset for a cut movements (cross slide) from the putting on cut ones (top slide). Its beyond me why the proponents won't accept this procedural benefit as sufficient justification and instead bang on about the, somewhat spurious, "better thread cut" advantages. Given a sufficiently well shaped tool used with sufficient skill on a sufficiently good machine the results from the straight in and the angular offset methods are indistinguishable. However if any of your personal skills, tool shape grinding, lathe condition or choice of material are a bit sub optimal its certainly a darn sight easier to get decent result by angular in feed that by straight in.

Clive
 
threads poor quality

i use a digital dial indicator and magnetic base reading tool post and magnet attached to carriage , i have found many lathes jump around a bit because of backlash and the gibs are loose. many lathe digital readouts do not have a readout for compound slide.

on 1 lathe i thought the gibs were tight and i overtightened one to the point i twisted the screw head off. turns out there is 2 screws pushing the gib one on each end. until the far one is backed off the near one cannot push it forward.

i like to keep cross slide and compound and carriage tight enough that it takes some effort to move them seen with digital indicator. i have seen vibration move stuff 0.005" or more which can affect thread quality and tolerances.

until i had digital readouts on compound, cross, carriage I had no way of measuring erratic tool movement other than random poor cut quality and tolerance control.
 
Hi,
Buy a threading center gauge. It will help you align your tool bit perfectly
to be 90 degrees to the center line of your work. I use the gauge along with a magnifier.
Mine is a Starrett EDP 51475. It costs a few dollars and is worth it.

If you have been eye balling your setups you can forget about having perfect looking threads.
 
Last edited:
Hello,

I always used 29.5 degrees off of parallel and a fishtail to set the tool. Thread DEPTH must be watched, but the angle and tool take care of the rest (I'm talking US threads).

Threading is intimidating to say the least when you start..and there are mistakes to be made along the way to be sure ( I have made more than my fair share for sure). Take your time, clear your head and thread something that doesn't matter how it turns out to get your practice.

With practice comes proficiency. How many of us remember learning to walk, yet years later we are great at it. Relax, you'll do fine.

Best of luck,

Roy
 
Hello everyone,

Bingo - problem solved. I think Clive hit the nail on the head. The graduations on my compound rest are different to what I am used to. The compound shows zero when it is perpendicular to the workpiece. So I was setting it at 60.5 degrees. Changed it, and things are just fine and dandy.

Sorry guys - never had this problem with other lathe - it never even passed my mind to think about checking to see if the graduations on my compound rest were the same as my other lathe - FOR PETE's SAKE - why can't they just standardise the stuff - how many hours have I just wasted trying to solve this ridiculous puzzle?

Best,

Alley Cat
 








 
Back
Top