Setting-up the PM 1236-T Lathe

[Titanium Knurler]

I apologize, for some reason a diagram and a photo did not show up in the last post. You can only see, “view attachment 288059” and “view attachment 288158”. I will attempt to cut and post the last post and insert them. Hope it works, sorry about that, pretty frustrating.

Why get rid of the twist in the Ways? One reason is that a twist in the lathe bed or Ways will cause the cutting tool to rise or fall from it’s starting height as the carriage travels down the Ways. Since the workpiece is round, as the tool rises or falls from the center of rotation(COR) the tool will remove less from the workpiece causing a taper, or if the cutting tool starts above or below the COR and then crosses the COR a spindle shape will occur with the thinnest spot, or waist, at the point the tool crosses the COR. So, not only might you have an undesirable uniform taper formed but you could also have a part whose diameter does not vary linearly along its length. Here are some diagrams I showed earlier that illustrate the effects of a rising or falling tool:

In the above diagrams the tool starts at height “a”(the COR) and descends along the Ways to point “b”, removing less stock as it travels resulting in a uniform taper. This is also true if you start at “b” and go to “a” or if the tool ascends or descends.

In the diagrams below the tool again starts at point “a” but this time it is above the COR, it descends as it travels down the Ways and crosses the COR at point “x” where it removes the most stock. This results in a spindle shape with the thinnest spot at point “x”:

So, the first reason for untwisting the Ways is to prevent taper and spindle formation.

The second reason for untwisting the Ways early-on in the alignment process is so that you can accurately adjust the Headstock and Tailstock. For example, if you try to align the Headstock or Tailstock before untwisting the Ways by using a dial indicator or performing a two collar test you might find taper that is attributable to twisted Ways and incorrectly assume it is from a misaligned Headstock or Tailstock and make unnecessary and incorrect adjustments of the direction of the Headstock or the position of the Tailstock. By the way, this is one of the reasons you will see in the upcoming sections that I used a dial indicator with a ground flat tip rather than a spherical tip. If there is any residual twist in the Ways the ground flat tip will minimize any error that might be introduced by a spherical tip riding up or down the test rod as you you move down the Ways indicating taper where there is none.



So, to make a long story short, the reason for untwisting the lathe bed prior to aligning the Headstock and Tailstock is to get rid of any taper that could be mistakenly attributed to Headstock or Tailstock misalignment.

So how does one untwist the lathe bed? The first thing is to make sure you are in the right frame of mind; make sure you have plenty of time to carefully make repeat adjustments; often very small adjustments. If you have mounted your lathe in the conventional way, that is, directly to the stand, the steps are as follows:

1. Place a precision level on a ground block with a notch to clear the V-Ways or the carriage and zero the level at the spindle using the leveling feet of the base.
2. Carefully slide the level to half way down the Ways and take a reading, then slide all the way down to the tailstock end of the lathe and take another reading.
3. Loosen the bolts securing the tailstock footing, place a shim under the low side footing, tighten down the tailstock bolts and repeat the steps until you are satisfied with the results. What you do at the Tailstock end will almost certainly affect your reading at the Headstock so you may need to re-zero the bubble in the level if goes off the scale.

If you have mounted the lathe on a plate with jack screws, like I have done, then the process is similar except that it is a bit easier since one can make adjustments quickly with the jack screws, rather than using shims, and the adjustments made using the leveling bolts to re-level the stand at the Headstock will not affect the lathe twist. This is because the lathe/baseplate unit is unaffected by the adjustments of the stand leveling bolts. Here are my results:


As you can see It took eight tries to get it where I wanted but I did go back later and was able to reduce the twist to zero according to the precision level. It is a tedious process but now I can proceed to the next steps confident that any taper I find is not due to twisted Ways.

Aside: since I am still waiting on supplies to complete the following sections I thought I would torture you with putting a finer point on Dr Jackson’s formula for determining the change in the depth of cut caused by a tool dropping or rising due to twisted Ways. If you recall he determined that:

Change in the depth of cut = (Distance the tool drops x Distance the tool drops)
(2 x Radius of the Workpiece)

Since (2 x Radius) = Diameter, the formula can be re-written to:


Change in the depth of cut = (Distance the tool drops x Distance the tool drops)
Diameter of the Workpiece

What kind of taper would I have found with the lathe prior to untwisting it? If you look at the chart above, the out-of-the-box twist for my 1236-T was 4.5 lines of twist on the precision level. Each line represents .0005”drop/ foot. So, 4.5 x .0005”=0.0023”drop/foot. The lathe bed is three feet long so the total drop in the tool height would be 3 x 0.0023”=0.0069” over the length of the bed. Now, if we use Dr Jackson’s formula the change in the depth of cut over three feet for a workpiece with a diameter of 1” will be: (0.0069”x 0.0069”)/ 1”=0.00048”. That is about a half a thousandth of taper over three feet. Not much, and I think it is probably even less than what Dr Jackson’s formula predicts. Why do I think it is less? Well, if you look at the photo of the twisted rails at the end of the last post you can see that when the right rail is lifted up not only does it go up but since it is tied to the other rail(as are the rails of the lathe) it moves to the left or toward the other rail as well. If the right rail is analogous to the front rail on the lathe then as the front rail is lifted it moves inward toward the workpiece and would decrease the loss of depth of cut that a tool that is simply moving upward would experience.

Well, is all of this worth a half a thousandth over three feet? Is it all much ado about nothing? 0.00048” taper over three feet is not nothing but it’s darn close. I will let you decide if it is worth it. However, I can at least move to the next steps, Aligning the Headstock with the Ways and Aligning the Tailstock with the Headstock, knowing that any taper that I find is not due to a twisted lathe bed. Surely, that’s worth something.

Which step should be done next, Aligning the Headstock to the Ways or Aligning the Tailstock to the Headstock? I believe aligning the Headstock next makes the most sense and I will tell you why I think so in the next post.

 

[Firstgear]

I bought some aluminum strut channel from Zoro here recently as I ordered a bunch of stuff when they had their 25% off flash sale. It came with free shipping. I have to laugh, I bought a 5ft piece of strut and it arrived in a tube that was 9ft long! I kid you not, 9' long! I realize they get a discount for the volume of stuff they ship, but I wouldnt want to pay for a 9' long shipping tube!

Amazing! Even my wife said if that was me I would have cut the tube down to 5'3" and saved on the shipping charges...amazing!

 

[P. Waller]

I was correct, at 60 my life expectancy is not long enough for this saga to end.

If you do not mind please answer a simple question.

If you desire such accuracy and build quality why did you not start with a Hardinge tool room lathe or similar?

This would be nearly what you want and likely take less then a decade to make "PERFECT"

 

[Firstgear]

I was correct, at 60 my life expectancy is not long enough for this saga to end.

If you do not mind please answer a simple question.

If you desire such accuracy and build quality why did you not start with a Hardinge tool room lathe or similar?

This would be nearly what you want and likely take less then a decade to make "PERFECT"

Just for the record, I appreciate his documentation as I am new to machining. I was always on the other side doing design work, but not making the pieces parts. TK as well as others who spend the time documenting aren’t doing it for themselves, they are doing it for people such as myself.

I get that you already know all this stuff with your 30 years of machining, but those of us who are new to this hobby don’t know the details. His information helps a lot of people. I hate to say this, but if you have nothing constructive to add to the discussion, just move along.

 

[P. Waller]

Just for the record, I appreciate his documentation as I am new to machining. I was always on the other side doing design work, but not making the pieces parts. TK as well as others who spend the time documenting aren’t doing it for themselves, they are doing it for people such as myself.

I get that you already know all this stuff with your 30 years of machining, but those of us who are new to this hobby don’t know the details. His information helps a lot of people. I hate to say this, but if you have nothing constructive to add to the discussion, just move along.

I appreciate the fact that someone takes the time and effort to explain how to do a particular operation or set up a machine for use.

However I find it disingenous to teach new people in the game that success is dependent on doing everything a particular way.

Do whatever job that needs to be done using a method that works, this may be the opposite of conventional wisdom at times.
The only way to learn this is to do a good deal of it, there in lies the experience part, it does not make one smarter merely less prone to doing something that does not work more then once.

 

[Titanium Knurler]

I was correct, at 60 my life expectancy is not long enough for this saga to end.

If you do not mind please answer a simple question.

If you desire such accuracy and build quality why did you not start with a Hardinge tool room lathe or similar?

This would be nearly what you want and likely take less then a decade to make "PERFECT"

Hey, P. Waller, happy to see you are still “with us” and as curmudgeonly as always; things were starting to get a bit boring without you.

As far as your question, “...why not start with a Harding tool room lathe...this would be nearly what you want and likely to take less than a decade to make “PERFECT””. Where is the challenge in that?!

 

[Titanium Knurler]

I appreciate the fact that someone takes the time and effort to explain how to do a particular operation or set up a machine for use.

However I find it disingenous to teach new people in the game that success is dependent on doing everything a particular way.

Do whatever job that needs to be done using a method that works, this may be the opposite of conventional wisdom at times.
The only way to learn this is to do a good deal of it, there in lies the experience part, it does not make one smarter merely less prone to doing something that does not work more then once.

P. Waller, to be honest, I don’t understand what you are trying to get across with your last post, so I won’t address it, but let me just say that the reason I put together this thread on setting-up the 1236-T is to help other HM members who have, or who are consdering, a similar lathe and to create a collegial dialogue with other like-minded individuals; experienced and inexperienced. It appears as though you have many years of valuable machining experience from which we could all benefit. I am sure you have created many chips over your thirty years of machining, however, it appears as though some of those chips have landed on your shoulders. I believe it would benefit everyone if you could brush those off and contribute to this thread and the Forum in general in a positive way. I would much prefer this because you seem to have a lot to offer, however, if this is not possible for you then just simply go elsewhere; you are damaging “The Friendly Machinist Forum”.

 

[Titanium Knurler]

Step 3. Aligning the Headstock with the Ways-sorry so long between posts, sometimes real life interrupts; in this case her name is Grace, 6lbs, 12ozs, all doing well.

I thought I would start this section with my thoughts on why the Headstock should be aligned with the Ways before Aligning Centers. Then follow with: how critical the Headstock alignment is; some initial readings of the Headstock alignment; how to adjust the Headstock on the PM 1236-T; some non-cutting and cutting checks of alignment and then finish with interpreting the results, which, by the way, were not what I expected.

You may recall that in an earlier post I mentioned that the PM Manual lists aligning the Headstock with the Tailstock before alignigning the Headstock with the Ways. The Manual indicates that the Headstock can be “wildly out of square” and still machine parallel between centers; I do not doubt that this is true, but why align centers before aligning the Headstock with the bed? To align centers with a Headstock that is out of alignment one has to move the Tailstock as indicted by the red arrow(added). It seems to make more sense to align the Headstock to the Ways then align the centers so everything is lined-up and you can go from turning between centers to turning in the chuck without making adjustments. Can you imagine what the taper would be with the Headstock at this angle if you were turning in a chuck!?



Well, I am glad you asked. Imagine we have the above misalignment of the Headstock. Let’s say it’s 5 degrees. What would the taper on a 36” piece of stock be if it were in a chuck rather than between centers. I know you would never do this but let’s just see what it is:



Some of you may recall the formula: sine of an angle=opposite side(red arrow, error) divided by the hypotenuse (yellow arrow, stock in chuck). We can rearrange this formula to solve for the opposite side which would be half the taper, or opposite side= hypotenuse x sine of 5 degrees(opposite side=36” x .087=3.14”). The taper would twice this or 6.3”! And this is an underestimate of the taper since the hypotenuse is actually longer because it extends back to where the Headstock rotates. Now, this isn’t a real-life example because I don’t think anyone would turn something 36” long in this way but it does illustrate how sensitive this system is to Headstock misalignment when turning in a chuck. It also illustrates that, unlike Dr Jackson’s formula for error with vertical misalignment, the longer the workpiece the worse the error. Let me give you an extreme example: we live in Milwaukee, we like to boat, if I were to take our boat across Lake Michigan east to the beautiful shores of Michigan, about 100 miles, and my compass was five degrees off I would miss my destination by almost nine miles! I will give you some real life machining numbers soon when I give you the initial Headstock numbers for 1236-T. By the way, can anyone guess what I was trying to get across with the railroad picture at the beginning of this section?

The key to checking for Headstock alignment errors and making adjustments is to make sure your test rod is as concentric with the spindle axis. Once this is accomplished one then runs the dial along the length of the rod looking for misalignment of the headstock. To do this you need a smooth, round, straight and untapered test rod(s), a good dial indicator and a large dose of patience. If you have not done this before I think you will be surprised how sensitive this system is, so make your adjustments in very small increments. I used non-cutting and cutting techniques using a MT5 test rod, Thomson shafting and drill rod in my adjustments.

MT5 Test Rod- I purchased a couple of MT5 test rods from eBay; one for using just with the indicator, the other for cutting. I started with the MT5 test rod because it takes out any pointing errors that could occur with a chuck:



After checking it for roundness, straightness and taper as well as I could, I carefully cleaned the MT5 end of the shaft and spindle, inserted it into the spindle and then gave it a good whack with a rubber mallet to seat it. I then marked it along its length at 0, 6 and 12 inches. I also marked 0, 90, 180 and 270 degrees around its base. I then mounted a Starrett model 25-211 dial indicator with a flat footplate to the carriage with a magnetic base. I then checked the TIR at 0, 6 and 12 inches. Next, I slid the dial indicator along the length of the bar to check for Headstock misalignment. I am going to call this LIR(longitudinal indicator runout). I rotated the bar in 90 degree increments to check to make sure the LIR was the same. If not, then the bar is not concentric with the spindle.



The drawing below shows the test bar in a chuck but it is otherwise similar to what I did with the MT5 test bar:



Here are the results from before and after I adjusted the Headstock:



You can see that the TIR ranged from .0002” to .0004” before and after the adjustment and that the LIR average improved from +.0016” to +.0003” over twelve inches. The positive sign on the LIR indicates the bar was angled toward the front of the lathe so I rotated the Headstock counterclockwise to move the bar away from the front of the lathe. To do this on the PM 1236-T one loosens the set screws on the Headstock and then you tighten the appropriate jack screw. In this case screw “A”. Be very careful with this. I adjusted it the smallest amount I possibly could and even with that I almost overshot. This diagram shows my situation with the Tailstock end forward:



This is how you make the adjustments to screw “A” on the PM 1236-T:





I decided to stop there and this is why. Using the formula above for determining the taper for a certain angle one can do the reverse, that is, find the angle of the Headstock error for a certain taper. When this is done the starting Headstock error was .0076 degrees. After the adjustment, the headstock error is in the same direction but only .0014 degrees. I felt that with any further adjustment I would overshoot, so I stopped.

Next I decided to do a cutting test on a MT5 test bar. It is made of EN31 steel alloy. I contacted Seco and got their recommendations for inserts and speed and feed data:




I used exactly their recommendations except the speed of 308 RPM seemed best for a smooth finish. Chips came off in nice, tightly coiled spring-like shavings as per Mikey’s chart. Interestingly, some of the shavings were many feet long. I used CC22 cutting oil and a follower rest(more on this later):





Here are the results:



At six inches the diameter was +.0003” larger and at twelve inches the diameter was .0009” larger than at position zero. The taper is in the opposite direction than I anticipated since the Headstock is tilted slightly forward at the tailstock end. Not sure why but I have an idea what might be happening; more on this later. Does anyone have any thoughts on this?

I am going to stop here for now. In the next post I will give my results for the 4JC and 5C collet chucks. In addition I will make an attempt to explain the taper that seems to be in the opposite direction of the small Headstock rotation error. Again, happy to hear any ideas on this.

 

[Mitch Alsup]

At six inches the diameter was +.0003” larger and at twelve inches the diameter was .0009” larger than at position zero. The taper is in the opposite direction than I anticipated since the Headstock is tilted slightly forward at the tailstock end. Not sure why but I have an idea what might be happening; more on this later. Does anyone have any thoughts on this?

I did not see where you measured the test rod for parallelism with respect to the ways.
That is: the headstock could be pointing down towards the tailstock.
This, too, can cause taper.

 

[Titanium Knurler]

Hmmm, I had not not thought of that. I will check it this weekend and let you know. I am not sure what I can do about it (shims, maybe?) but it would be nice to know.

As a Newbie this was a great exercise for me: it made me think and gave me the opportunity to spend lots of time in front of the lathe. I learned a lot, but to be honest, I can see a person going crazy chasing down that last .0001". To be honest, I am not sure what an acceptable taper is but .0003" at six inches is as good as it is going to get for me for now.

Thanks again, Mitch Alsup