Thread: Lathe alignment question

Originally Posted by Stustoys

Not as I understood it. RDM method is for making a worn but level lathe cut parallel by twisting the bed. The name has since be applied(or misapplied) to just about every lathe alignment technique I have seen, including the "two collar method" which has been around since year dot.

Stuart

Hi Stu,

I think this is the original document on the "Rollie's Dad's Method of Lathe Alignment" from 2001. Before it was applied(or misapplied) to just about every lathe alignment..... The idea was to setup a lathe without owning a precision level and a test bar.

http://www.neme-s.org/Rollie's_Dad's_Method.pdf

I feel like I am having deja-vu..

Originally Posted by cba_melbourne

Hi Stu,

I think this is the original document on the "Rollie's Dad's Method of Lathe Alignment" from 2001. Before it was applied(or misapplied) to just about every lathe alignment..... The idea was to setup a lathe without owning a precision level and a test bar.

http://www.neme-s.org/Rollie's_Dad's_Method.pdf

I'm pretty sure the first mention of RDM I saw was from those guys. Though I recall a nice little story to go with the method. But as I cant find a link I guess its besides the point.

Originally Posted by .RC.

I feel like I am having deja-vu..

Now that you say that you may well be right. lol

Stuart

Originally Posted by .RC.

I feel like I am having deja-vu..

Unfortunately yes but this always happens on forums after a while.... PDW

Originally Posted by .RC.

I feel like I am having deja-vu..

And I feel like a fortune teller - I could see where this thread was going to go right from the first post

Originally Posted by Jekyll and Hyde

And I feel like a fortune teller - I could see where this thread was going to go right from the first post

Armchair experts like yourself really get me, i'm sorry i'm not an expert on lathe alignment and thought this forum would be a a good place to learn, not receive smartarse comments like yours.
Don't bother posting if your not posting something constructive to help out and thanks to the guys who have actually posted useful information.

if this has been covered before why don't you summarise the previous threads and get the mods to make it a sticky, rather than throw comments from the peanut gallery.

Originally Posted by bowseruni

if this has been covered before why don't you summarise the previous threads and get the mods to make it a sticky, rather than throw comments from the peanut gallery.

Because you end up with a full page of stickies which is a big PITA. You might have lucked out and not screwed your HS alignment, it's hard to say. However RDM is a *kludge* and is widely recognised as a kludge, not the technique you should use first up. There are copies of Schlesinger and Connelly out there on the net, it'd be a really good idea to download *and read* them before altering anything else on your lathe. Level the lathe, then turn a bar (big diameter, aluminium is good) doing the 2 collar test, then use a good micrometer reading to 0.0001" or metric equiv to measure the collars. If they're the same size or very close to it, buy a lottery ticket and don't screw with anything. PDW

Forget about the precision level.
Unless you want to scrape the ways its a waste of money.
All you need is a ground testbar with the apropriate morsetaper and a, or several in different length, round, true bar centerdrilled at both ends.

The purpose of a lathe is to make round and true non-tapered objects.
Unless your headstock bearings are shot all objects turned will be round.

If your carriage moves in a straight way, horizontal and vertical, all objects turned will be true.

If your carriage moves true to the center of the spindle there will be no taper.

Wear in the ways will affect trueness and taper.
Levelling the ways lengthwise and crosswise will not give you better trueness or taper.
It will not compensate for wear as wear will not be consistent over the entire length of the bed.

If a taper is to be avoided turn between centers and adjust the tailstock.

Due to wear in the ways the object will not be true.
Twisting the bed might by lifting one of the four corners of the bed make things better or not.

In turning not between centers the taper might be adjusted by adjusting the headstock in the horizontal and vertical.
As the wear in the ways is not consistent it will only be effective for the distance where the wear is somewhat consistant and not for other distances.
Again lifting one corner of the bed may make things better or not.

You use the testbar to find the relation between spindle and wear of the bed for any given distance.
Realignment of the headstock would have to be done for every given distance away from the chuck.
It is not feasible to do so every time over again.
It is feasible to do for one distance where most turning is done, mostly near to the chuck.
It wil not likely completely take out the taper but will make it less.

You will use the other bar(s) for alignment of the tailstock and finding out where and how much wear the ways have.
Twisting the bed may lead to better results.
The bar will give an idication to what is happening while twisting.

The final proof will always be a testcut where trueness and taper will be checked.

All the above goes unless you are willing to scrape.
While scraping the ways you will need a level.

The tailstock is probably the most important part that has to be adjusted.
Wear of the contact area with the ways will primarily occur on the front side of the tailstock.
Center-hight will drop.
This will lead to problems in centerholding and drilling.
Adjusting the height to the centerheight of the spindle by shimming mainly the front end will be necessary.
Problem will be that for each given distance away form the spindle the ways on which the tailstock slides will have worn differently.
So an adjustement will only be perfect for one given distance. But given that the wear on the ways is relatively small to the wear on the tailstock itself, it will only have minor influence.

Cheers Jan

thanks for that, my test bar showed up today, it has a taper to fit my tailstock and measures spot on along the length. Will have a play with it tomorrow or over the weekend depending how much free time I find.

I did manage to take some light cuts on a 1" MS bar, it tapered 0.003" over 150mm (larger dis at tailstock end) so there is work to be done. Pretty sure I have found a mate who has a decent level that I can borrow

Originally Posted by bowseruni

Armchair experts like yourself really get me, i'm sorry i'm not an expert on lathe alignment and thought this forum would be a a good place to learn, not receive smartarse comments like yours.
Don't bother posting if your not posting something constructive to help out and thanks to the guys who have actually posted useful information.

if this has been covered before why don't you summarise the previous threads and get the mods to make it a sticky, rather than throw comments from the peanut gallery.

The problem is everyone has a different way that may work for them and most threads like this end up getting locked.. Everyone has different expectations and needs and wants so what works for one may not work for another...

I suggest you go searching older articles on model engineering type places..

And remember US styled lathes and UK styled lathes are usually quite different in a critical design and thus have different pitfalls to catch the unwary.... That being the headstock alignment on UK style is adjustable and US type not adjustable..

Originally Posted by bowseruni

Armchair experts like yourself really get me, i'm sorry i'm not an expert on lathe alignment and thought this forum would be a a good place to learn, not receive smartarse comments like yours.
Don't bother posting if your not posting something constructive to help out and thanks to the guys who have actually posted useful information.

if this has been covered before why don't you summarise the previous threads and get the mods to make it a sticky, rather than throw comments from the peanut gallery.

Don't own an armchair, and not a great fan of peanuts.

My comment was in reference to the fact you used the phrase 'Rollies dad's method', no more, no less. If you'd like to see where your thread will end up, just search that phrase on this forum, and you'll see the argument about its purpose and validity is just getting started...

OK I will elaborate on my first post and get the ball rolling

How I would do it...

Since you said you have "realigned" the headstock it only leaves you with a few options to get it back right... Totally ignore any of these "two collar tests" and other like that... They are designed for US type lathes where the headstock sits on V ways and is unadjustable.. Like a Hercus for example..

In my opinion your only option now is to get a level and get any twist out of the bed through the levelling screws... Does not have to be perfect..

Since you most likely will not have a test bar for the headstock and no sleeving a test bar does not work good enough for me you are left with the cut and try method...

so a bright steel bar say 40mm diameter or more, 150mm or so poking out the chuck, make two collars on it and take very light test cuts using a sharp HSS tool ..keep adjusting until you get it within your satisfaction.... Do not try for perfection, the further most end from the chuck can be smaller then the chuck end. It will be frustrating as all hell to do...

Now align the tailstock in your usual fashion...

Job done, use the lathe..

After you have machined an undercut on the bar, set your depth of cut on the outer end, finish the cut and without moving the tool depth traverse to the end and cut that, then measure.

Make the undercut deep enough so you can do numerous passes

Originally Posted by pipeclay

After you have machined an undercut on the bar, set your depth of cut on the outer end, finish the cut and without moving the tool depth traverse to the end and cut that, then measure.

Make the undercut deep enough so you can do numerous passes

Wouldnt that be like putting two collars on a piece of stock?