South Bend 9C Headstock Alignment

[sdelivery]

.001 is fine , remember oil takes up space. That .001 is an oil bearer and likely when warm will be slightly less than.001.
If you have a problem after some time you can address it then.
Just monitor the oil as you become acquainted with the machine.
A picture sure would be nice.

 

[Jim F]

Sorry for being a bit confusing. I will try to answer everyone's questions.

I did 2 different 2 collar tests.
I did the 2 collar test in a 3 jaw chuck to check the initial taper and then also did the 2 collar test between centers to align the tailstock after I shimmed up everything. That was the last thing I did. I used a piece of 1.5" aluminum rod about 9.5 inches or so long so I could as close to an 8 inch separation as possible without getting chatter. Final pass was done with a 0.001" cut at a slow speed and using the power feed.

I did not think of using 2 shims on the V to cut the shim thickness in half. Thanks for the pointer.
After I added the shim under the headstock, I verified that the headstock was aligned with the bed using the backing plate with a bolt in it. This was done with all "twist" removed in the bed. Verified that there was no twist using a precision level.

I then did another 2 collar test in the chuck to verify that things got better and they did but it was still not perfect.
At this point I again added shims under one side of the foot at the tailstock end to remove the last of the taper.
The section where I mentioned shimming the tailstock end was putting shims under one side of the foot at the tailstock end of the lathe not shimming the tailstock itself.

Once I was able to get the horizontal taper where it should be, I also verified that vertically both collars were in alignment and they were less than 0.0005" different over the 8 inches. I did this as I was concerned that after adding the shim under the headstock that it would throw off the vertical alignment. Again, all of this was done using a chuck.

After all of this I then did a 2 collar test between centers to get the tailstock itself aligned with the headstock.

I am a little confused about the headstock bearing clearance though. Dave, you mention that the play should be closer to 0.0005" but The SB Bulletin H-4 on Alignment states that there should be 0.001-0.002 clearance and that less than 0.001" clearance can cause chatter due to reduced oil thickness. Mine is right around 0.001" so it agrees with the SB manual for cast bearings. Am I missing something?

A level will not tell you if you have twist....

 

[sdelivery]

A level will not tell you if you have twist....

I disagree. Instead of using your level straight across parallel to the cross slide use a straight edge and go diagonally.

 

[Jim F]

 

[sdelivery]

That sounds like Joe. I like his videos.
If you go diagonally instead of parallel and perpendicular.
The level would be from the headstock operator side to the tailstock rear way.
Then from the headstock rear way to the tailstock operator side.
This is how it has been done in industry for lathes, mills, planer mill , floor plates, steel plants , rubber rolling machines.

 

[danjp]

I have watched this particular video from Joe so I understand what he is saying.
With that being said, I am now confused on this whole leveling/twisting discussion happening here.
So here is my understanding. If not correct please be specific in what I am not understanding.

A SB 9 bench lathe only has 2 feet. One under the headstock and the other under the tailstock.
It has no ability to adjust the headstock relative to the bed aside from shimming or scraping the headstock mating surfaces.
So that only leaves the ability to "level" in both X and Y where X is lengthwise and Y is across the ways.
This is unlike the lathe in Joe's video where the headstock centerline has the ability to be adjusted relaitive to the bed.

Deviations in Y across the length of the bed is the twist of the bed. Deviations in X only affect the level in that direction and should not matter in determining if a bed has twist in it. That is why a lathe can still maintain alignment even when on a ship that is pitching and rolling as long as the bed does not twist from headstock to tailstock. So the bed can be out of level but not have twist which is what we want as a starting point just as Joe said in his video.

Using a level perpendicular or across the ways will tell if the bed is twisted if measured at each end of the bed. How it does not do that is something that I am not following in the discussion here. This does not mean the lathe is level as that implies that the lathe would be "level" in both X and Y or as noted by doing a diagonal or measuring in both X and Y. Or as Joe stated "it does not have to be level only true to itself"
So I was guilty of "abusing" the term level as I really meant to say I elimated bed twist.

So I made sure there was no "twist" in the bed before I did anything.

Right now all I know is that I shimmed the headstock as it's centerline was not parallel to the bed with no twist in the bed.
I also shimmed the outside of the foot at the tailstock end to get the remaining taper out of the cut.
i believe this was due to some degree of wear in the bed and/or cross slide.
I did try to measure the bed wear by mounting a test indicator on the carriage and having it touch the flat between the V's and then sweeping the carriage from one end of the bed to the other. I did see a variation of 0.0015" max which to me indicates minimal wear for an 80 year old lathe.

This combination of using a headstock shim and the shims under the one side of the tailstock foot has resulted in the lathe cutting straight over an 8" length when using both a chuck and between centers with less than 0.0005" deviation over the full 8" length in either case.

So what did I do wrong?

My original questions where and still are:

1 - Has enyone else ran into an issue like this and if so how did you fix it? What did you do to realign the headstock relative to the bed if you replace the headstock?

2 - Is it OK to shim the headstock to bring the spindle centerline parallel to the length of the bed while also making sure the spindle centerline is not pointing up or down? Is using shims an acceptable practice or should the headstock mating surfaces be scrapped to realign and if so what are some pointers on how to do that if that is the best way.

3 - Any ideas on how it got to be this far off besides it not being the original headstock.

 

[gard]

I think what you did is fine. A slight improvement would be to find the combination of shims that get the headstock parallel to the bed in both the horizontal and vertical plane with the bed level.
For the test you did with the backplate, if I understand correctly, you moved the cross slide so the indicator hit the bolt at the front and back? This is the test that is normally done to confirm that the cross travel is perpendicular to the bed after you know the head is aligned. It is testing both things.

I am in the process of rebuilding a 10L lathe. The bed was worn so I had it planned then I scraped the ways flat. Next step was checking alignment of the head stock to the bed. It was way off. I used a test bar with the bed perfectly level. The test bar results are
not impacted by cutting force. Plus its quicker to run an indicator along the top and side of the test bar then set up a cut. Next was a process of scraping the bottom of the headstock until it was in alignment to the bed. It took many iterations but finally got there. I did some tests along the way with shims, exactly like you did, just to verify where I needed to scrape on the headstock.

Next the the tailstock was scraped in for alignment to the bed and height of the headstock.

Hard to say how your lathe got in the condition it was in, perhaps a quality escape from the factory or something happened over the years.

 

[danjp]

I think what you did is fine. A slight improvement would be to find the combination of shims that get the headstock parallel to the bed in both the horizontal and vertical plane with the bed level.
For the test you did with the backplate, if I understand correctly, you moved the cross slide so the indicator hit the bolt at the front and back? This is the test that is normally done to confirm that the cross travel is perpendicular to the bed after you know the head is aligned. It is testing both things.

I am in the process of rebuilding a 10L lathe. The bed was worn so I had it planned then I scraped the ways flat. Next step was checking alignment of the head stock to the bed. It was way off. I used a test bar with the bed perfectly level. The test bar results are
not impacted by cutting force. Plus its quicker to run an indicator along the top and side of the test bar then set up a cut. Next was a process of scraping the bottom of the headstock until it was in alignment to the bed. It took many iterations but finally got there. I did some tests along the way with shims, exactly like you did, just to verify where I needed to scrape on the headstock.

Next the the tailstock was scraped in for alignment to the bed and height of the headstock.

Hard to say how your lathe got in the condition it was in, perhaps a quality escape from the factory or something happened over the years.

You are correct regarding the test with the backplate. I did that to determine if the headstock was parallel with the bed not that the carriage is perpendicular to the bed as I assumed that it was. I suppose the cross slide it might not be square but I really doubt that it is.

Fortunately my bed is not very worn. I do not think the lathe was used very much. I tried to measure the bed wear by running an indicator down the flat portion of the bed while on the carriage and I only seen around .0015" deviation so not bad at all. The little twist that I had to put into the bed after shimming the headstock I think was due to the bed wear. I did not have to shim it a lot so not too worried about that.

Right now I am pretty happy with the results by adding the shim under the headstock. When I cut a test bar I am seeing very little deviation between the 2 collars so I think I have it dialed in pretty good. I also checked the alignment using a precision rod from a printer to measure both the height and parallelism by running down the length with an indicator on the carriage and it was almost spot on in both directions.

My main concern was if it was acceptable to shim a headstock instead of scraping it. I was not sure if the head could move under the cutting forces with a shim under it. So far I have not seen any indication that it is moving but I have yet to cut something requiring a lot of force.

I have not tried scraping prior so if you can point to any good references maybe it's something I will try in the future.

 

[gard]

When I was messing with shims, I tried to find the combination of shims that so that each corner was supported. There are 6 corners, one side and end of the dovetail and each end of the front flat. I started with one or 2 shims like you did to get the test bar close, then used feeler gauges under the other 5 corners to see how much shim was needed there. Then tighten down the attachment bolts and see how much the test bar moves. Rinse and repeat. The thought was this would minimize how much distortion there is in the bed and head when the bolts are tightened. The shims show how much and where to remove metal to get a consistent fit on all the surfaces. There is a good argument that you have your lathe cutting strait now so don't fix it if it's not broke.

I have enjoyed learning scraping with a lot of reading and utubes in the machine rebuilding forum here and on practical machinist forum. Look for classes and advise from Richard King but there are lots of others also. Start by learning to scrape a surface flat by making one or 2 straight edges. You can buy cast iron unfinished straightedges and scrape them flat, I was happy with my purchase from Dennis Foster. Or start with a block of durabar cast iron. Next you learn to get a surface flat and in alignment. The old book "Machine Tool Reconditioning, and applications of hand scraping" by Connelly is the place to go to for details on which surface to scrape in what order with what goals to fully recondition an entire lathe or mill. You can get started with a modest investment in hand tools. The pros use power scrapers that are pretty expensive..