Thanks for this great technical approach. A small remark : when at 6:28 when you use the micrometer : you should use the torque limiter at the end of the handle to ensure a repetitive torque & measure
Thank you for your comments. The first micrometers I used didn't have the friction drive so I got used to measuring by feel. However, you are probably right!
I love how thourough your video is, I do have a question. Why would you machine a test bar without center support, or without the use of a steady rest, wouldnt that just introduce error due to deflection of the bar? Its just, because the bar is not supported at the end, the forces due to cutting would be acting on a much longer lever, than closer to the headstock. So, as the cut is made, the amount of distortion of the bar will increase as the cutter moves further away from the head stock. I understand machining the unsupported test bar is to see if the diameter changes from the starting point and end point of the test bar, allowing one to view the error due to miss alignment of the headstock with the bed. But that assumes the amount of distrotion in the bar from cutting, is not significant enough. Because if there is a significant amount distortion in the bar from the act of cutting it, wouldnt that just scew your results for judging how un-concentric the head stock is from the lathe bed? Im really curious, because I too am trying to alighn my headstock, and your video makes some pretty solid points that other videos didnt even delve into. Im just wondering if you were aware of this, and disregarded mentioning this in your video, because perhaps the amount of distrotion was almost negligible, due to limitations in measuring resolution to even detect said error. I just want to know if you had any suggestions for how this technique can be improved on for reducing or eliminating the error due to distortion of the un supported bar. Thank you so much, just subscribed.
Thank you, Jennifer, for your thoughtful response. Yes, the two lands need to be machined without tailstock or steady support to ensure that the degree of headstock alignment is revealed in the two diameters. Your point about cutting forces deflecting a relatively long bar is reasonable. In my setup I assumed that the cutting forces were insignificant for very light cuts with a sharp high-speed steel tool. It may have been better to make the test bar from 1.5in (38mm) steel stock to improve rigidity (I may do that next time). For the same length of bar the rigidity would be 5x greater since "I" is a function of D⁴ so that would be a viable way of eliminating that concern.
Yes, levelling does affect alignment. That is why before making any adjustment to the headstock it is necessary to ensure the bed is levelled (ie there is no twist and is the way it was originally made and set up). Hope that makes sense.
This covers alignment front to back. How would you check the angular alignment up and down, or in other words that the axis of rotation is parallel with the ways of the bed, that spindle nose wasn't pointed up or down?
Hello. Thank you for your good question. Checking the vertical alignment along the bed could be done. The real challenge would be to make adjustments. I assumed that this was correct at manufacture and wouldn't be susceptible to change through reassembly - unlike the horizontal alignment.
@@Workshopfriend for good quality american or british lathes I would agree with you. I'm more concerned with the dubious quality of today's chinese lathes. The headstock would no doubt need to be shimmed or better scrapped to correct it. I just can't seem to wrap my head around how to measure it. Or differentiate the error readings between that and left/right angular misalignment
I assume you would need to measure the height of a test bar above two unworn portions of the bed as close as possible to the headstock and tailstock. Or, you could turn up a long test bar to the same diameter as your tailstock quill and shim the headstock casting until the bar and the quill align using a DTI. In my homemade lathe I used shims to align the headstock axis in the vertical plane.
@@Workshopfriend that sounds like it would work! The only problem I have is the chicken before the egg issue. There must be a way to do it without a turned test bar... Because you must have an accurate lathe to produce a test bar
No, I don't believe you need an accurate lathe. For vertical alignment just turn a diameter as far from the headstock as possible of the same diameter as the tailstock quill and then align this with the tailstock quill. This will not correct for bed wear but it will align the headstock within the limitations of the worn bed. (Not valid for lateral alignment) @EitriBrokkr
Can this be done with a bench lathe of 500mm between centres? I also have the inadvertant taperon the machine which i only discovered later after buying the lathe and that problem annoys the heck out of me😅
It depends on the design of your machine. If the headstock is not separate from the bed the operation described here will not be possible. Even if the headstock casting sits on the bed there would need to be some provision for small movements as in this machine. Whatever, you should ensure that the bed is not twisted due to bolting down to a non-flat base - surprisingly easy to happen and can result in not turning parallel.
Hello Steve. Thanks for your feedback. I trained as an engineer and worked for a while in that field. Over the years, since I moved on, I have kept up my interest.
Thanks for this great technical approach. A small remark : when at 6:28 when you use the micrometer : you should use the torque limiter at the end of the handle to ensure a repetitive torque & measure
Thank you for your comments. The first micrometers I used didn't have the friction drive so I got used to measuring by feel. However, you are probably right!
Great video, I was waiting for it long time, finally will alignment my colchester student as should be. Thanks
Thanks for the feedback!
Good content, very classy and clean, I am glad to know you so that I can benefit from you
I am happy to hear that you found this video helpful!
I love how thourough your video is, I do have a question. Why would you machine a test bar without center support, or without the use of a steady rest, wouldnt that just introduce error due to deflection of the bar?
Its just, because the bar is not supported at the end, the forces due to cutting would be acting on a much longer lever, than closer to the headstock. So, as the cut is made, the amount of distortion of the bar will increase as the cutter moves further away from the head stock.
I understand machining the unsupported test bar is to see if the diameter changes from the starting point and end point of the test bar, allowing one to view the error due to miss alignment of the headstock with the bed. But that assumes the amount of distrotion in the bar from cutting, is not significant enough.
Because if there is a significant amount distortion in the bar from the act of cutting it, wouldnt that just scew your results for judging how un-concentric the head stock is from the lathe bed?
Im really curious, because I too am trying to alighn my headstock, and your video makes some pretty solid points that other videos didnt even delve into. Im just wondering if you were aware of this, and disregarded mentioning this in your video, because perhaps the amount of distrotion was almost negligible, due to limitations in measuring resolution to even detect said error.
I just want to know if you had any suggestions for how this technique can be improved on for reducing or eliminating the error due to distortion of the un supported bar.
Thank you so much, just subscribed.
Thank you, Jennifer, for your thoughtful response. Yes, the two lands need to be machined without tailstock or steady support to ensure that the degree of headstock alignment is revealed in the two diameters. Your point about cutting forces deflecting a relatively long bar is reasonable. In my setup I assumed that the cutting forces were insignificant for very light cuts with a sharp high-speed steel tool. It may have been better to make the test bar from 1.5in (38mm) steel stock to improve rigidity (I may do that next time). For the same length of bar the rigidity would be 5x greater since "I" is a function of D⁴ so that would be a viable way of eliminating that concern.
Thanks for the video. I was wondering if adjusting the leveling screws had an effect for the headstock alignment with the ways.
Yes, levelling does affect alignment. That is why before making any adjustment to the headstock it is necessary to ensure the bed is levelled (ie there is no twist and is the way it was originally made and set up). Hope that makes sense.
This covers alignment front to back. How would you check the angular alignment up and down, or in other words that the axis of rotation is parallel with the ways of the bed, that spindle nose wasn't pointed up or down?
Hello. Thank you for your good question. Checking the vertical alignment along the bed could be done. The real challenge would be to make adjustments. I assumed that this was correct at manufacture and wouldn't be susceptible to change through reassembly - unlike the horizontal alignment.
@@Workshopfriend for good quality american or british lathes I would agree with you. I'm more concerned with the dubious quality of today's chinese lathes. The headstock would no doubt need to be shimmed or better scrapped to correct it. I just can't seem to wrap my head around how to measure it. Or differentiate the error readings between that and left/right angular misalignment
I assume you would need to measure the height of a test bar above two unworn portions of the bed as close as possible to the headstock and tailstock. Or, you could turn up a long test bar to the same diameter as your tailstock quill and shim the headstock casting until the bar and the quill align using a DTI. In my homemade lathe I used shims to align the headstock axis in the vertical plane.
@@Workshopfriend that sounds like it would work! The only problem I have is the chicken before the egg issue. There must be a way to do it without a turned test bar... Because you must have an accurate lathe to produce a test bar
No, I don't believe you need an accurate lathe. For vertical alignment just turn a diameter as far from the headstock as possible of the same diameter as the tailstock quill and then align this with the tailstock quill. This will not correct for bed wear but it will align the headstock within the limitations of the worn bed. (Not valid for lateral alignment) @EitriBrokkr
Can this be done with a bench lathe of 500mm between centres? I also have the inadvertant taperon the machine which i only discovered later after buying the lathe and that problem annoys the heck out of me😅
It depends on the design of your machine. If the headstock is not separate from the bed the operation described here will not be possible. Even if the headstock casting sits on the bed there would need to be some provision for small movements as in this machine. Whatever, you should ensure that the bed is not twisted due to bolting down to a non-flat base - surprisingly easy to happen and can result in not turning parallel.
@@Workshopfriend thanks, no machine not bolted down
It is hard to see how your lathe isn't bolted down in some way. Could the lathe be very worn? How much taper do you see instead of a parallel cut?
Outstanding again. BTW are you a school teacher or lecturer by chance? Just going on the maths
Hello Steve. Thanks for your feedback. I trained as an engineer and worked for a while in that field. Over the years, since I moved on, I have kept up my interest.