Harrison L5 problem

[hthParticipant @hth]

[hthParticipant @hth]

Hi

 

My Harrison L5 displays a strange oddity. I have a new MT3 test bar , and the headstock to bed alignment seems to be near perfect when the test bar is inserted into the spindle bore , running a dial indicator up/down the bar indicates nil deflection/runout .

 

After putting the new 3 jaw chuck on the spindle , if I turn a section of bar 5" long, , there is a .003" – .004" taper over the 5 " length . Ive done this test a few times, the result is the same each time . The diameter is smaller at the chuck end .Mike

 

 

Edited By hth on 23/01/2016 10:52:24

[Ian ParkinParticipant @ianparkin39383]

What diameter are you turning the test sample to?

Are you using a steady?

[hthParticipant @hth]

Posted by Ian Parkin on 23/01/2016 10:55:04:

What diameter are you turning the test sample to?

Are you using a steady?

The diameter is 1" and no steady is used . The cuts are light so deflection isn' t the cause . I have tried the live centre in the tailstock to support the work but it makes no difference .

[Ian ParkinParticipant @ianparkin39383]

Have you a length of silver steel to mount in the chuck and redo your alignment test with the dial gauge?

also grasp the test bar mounted in the chuck with the dial on it to see if it will move under force

Edited By Ian Parkin on 23/01/2016 11:21:06

[hthParticipant @hth]

Posted by Ian Parkin on 23/01/2016 11:14:42:

Have you a length of silver steel to mount in the chuck and redo your alignment test with the dial gauge?

Ok I just did another test . I have used the 3mt ground test bar in the chuck , I turned the bar around to the plain end and mounted it in the chuck . The dial indicates that the bar is tapered ! Of course the bar isnt tapered as its a new ground bar . I think I have a bad chuck , it is a Chinese cheap chuck .

[Ian ParkinParticipant @ianparkin39383]

I cant imagine any chuck could let you turn a taper…if you tun the chuck slowly under power does the dial indicate wildly?

meaning the bar is held at an angle

 

sorry i worded that wrong if the chuck isnt holding straight it will turn a taper but bigger to the chuck end

Edited By Ian Parkin on 23/01/2016 11:39:08

[hthParticipant @hth]

I will do a test tomorrow – turning a bar between centres .If the test indicates the turned  bar is not tapered then This should tell me  the chuck is the cause of the  problem . Thanks for the advice Mike

Edited By hth on 23/01/2016 11:55:12

Edited By hth on 23/01/2016 11:56:03

Edited By hth on 23/01/2016 11:56:53

[Tony Pratt 1Participant @tonypratt1]

Posted by Ian Parkin on 23/01/2016 11:37:30:

I cant imagine any chuck could let you turn a taper…if you tun the chuck slowly under power does the dial indicate wildly?

meaning the bar is held at an angle

sorry i worded that wrong if the chuck isnt holding straight it will turn a taper but bigger to the chuck end

Edited By Ian Parkin on 23/01/2016 11:39:08

Whether a turned part is parallel is controlled by the alignment of the spindle to the lathe bed, the true running of the chuck or otherwise doesn't come into it.

Tony

[Nick HughesParticipant @nickhughes97026]

The lathe probably needs a final "Tweek " as in what's termed Leveling, where the bed mounting points at the tailstock end, are shimmed front or back as required, to eliminate the slight twist in the bed that is causing the taper.

Although depending on the machines use, it could also be caused by bed wear.

Don't turn between centres until the lathe turns parallel when you machine the bar as you've done so far (relive the mid portion of your bar to leave larger diameters at each end, this saves tool wear and time machining the full length).

When the machine turns parallel without tailstock support, you can then progress to aligning the tailstock.

Nick.

[John ReeseParticipant @johnreese12848]

Sounds like the taper is caused by deflection. Could be due to the part springing under cutting pressure. Try a larger workpiece. Make the final pass very fine, using a high shear tool.

Another possibility is the chuck. If it is not seated tightly against the shoulder that could be the source of deflection.

If the chuck jaws are worn bellmouthed the jaws will only hold at the heel allowing the part to deflect i the chuck.

[Alan Waddington 2Participant @alanwaddington2]

 

Posted by Nick Hughes on 23/01/2016 12:12:58:

The lathe probably needs a final "Tweek " as in what's termed Leveling, where the bed mounting points at the tailstock end, are shimmed front or back as required, to eliminate the slight twist in the bed that is causing the taper.

Although depending on the machines use, it could also be caused by bed wear.

 

Advice from Harrison was never to unbolt the bed from the stand, as they were shimmed by the factory.

But who knows, maybe someone has split them in the past.

 

Edited By Alan Waddington 2 on 23/01/2016 12:32:29

[Ian ParkinParticipant @ianparkin39383]

Posted by Tony Pratt 1 on 23/01/2016 12:01:17:

Posted by Ian Parkin on 23/01/2016 11:37:30:

I cant imagine any chuck could let you turn a taper…if you tun the chuck slowly under power does the dial indicate wildly?

meaning the bar is held at an angle

sorry i worded that wrong if the chuck isnt holding straight it will turn a taper but bigger to the chuck end

Edited By Ian Parkin on 23/01/2016 11:39:08

Whether a turned part is parallel is controlled by the alignment of the spindle to the lathe bed, the true running of the chuck or otherwise doesn't come into it.

Tony

If the work is held in a chuck at an angle then the work wont turn parallel at all whether the spindle is in line with the bed or not

It will turn a taper bigger at the chuck end and smaller at the far end

[duncan webster 1Participant @duncanwebster1]

the parallelism of the finished bar is governed only by the alignment of the spindle to the bed, inaccuracy of chuck doesn't come into it. Try repeating your test with the test bar reversed in the chuck, but rotate it slowly. No doubt the DTI will wobble, but if the mid point of wobble is the same both ends the alignment is good

[Nigel McBurney 1Participant @nigelmcburney1]

One way to prove parallel turning is you have a test bar which runs true,so turn up two mild steel collars which are a good fit on the test mandrel, secure them with super glue one near the headstock the other about four inches away from the headstock, then try taking a light skim off each collar and then check the diameters this will give a more accurate indication to see if the lathe is turning tapered as you know the bar when clocked is ok and a good fit in the lathe spindle, When test turning the most accurate results are achieved by being careful,lock the top slide tight,lock up the cross slide or tighten the gib adjustment make sure the tool is dead centre height ,a really sharp HSS tool with minimum tip radius is required to keep deflection to a minimum.Do all your tests using same speed and feed,if you have a removable gap piece check that it is in perfect alignment with the bed. Do not use the tailstock for support when alignment testing. The chuck jaws may be the problem when trying to turn parallel with the chuck ,if the jaws are loose in their slides they can grip better at the back,this leads to the work deflecting away from the tool making the diameter larger the further the tool is from the chuck.

[Martin ConnellyParticipant @martinconnelly55370]

There is nothing in the world that is truly rigid. If you put an indicator on your test bar and push the bar towards the indicator you will find it takes only a small force to give a deflection at the tailstock end equal to your taper. Think about how much force is required to push the cutting tool into the work piece and they are probably similar. More force is needed with relatively blunt carbide tipped tools than a freshly honed sharp HSS edge. What are you using? Also a suitable steady will push the workpiece to keep it from deflecting if you have one.

I think the test bar is telling you that the spindle is correctly aligned with the bed and the taper is caused by deflection as suggested earlier. Light cuts with a blunt tool are more likely to rub the surface than cut it if there is deflection.

If an out of true chuck caused a taper then we would all have binned all our 3 jaw chucks a long time ago. As discussed in another current thread 3 jaw chucks are unlikely to hold something perfectly aligned.

Martin

[Phil WhitleyParticipant @philwhitley94135]

Good post Martin! I think there should be a complete ban on lathe accuracy testing by the user until the user finds the parts they have made are "not accurate enough for purpose". If you look for problems you usually find some.

Phil

[Nick HughesParticipant @nickhughes97026]

Right, first off all of you, READ THE FIRST POST THERE IS NO CHUCK , OR TAILSTOCK INVOLVED, secondly WE DO NOT KNOW THE HISTORY/AGE OF THE LATHE so what the manufacturer recommended years ago is irrelevant NOW!

Ian Parkin, work held in chuck at an angle will not turn parallel, WRONG, THINK ABOUT IT!!!!!

Phil Whitley, THE USER HAS FOUND AN ISSUE, THAT'S WHY THEY ASKED FOR ADVICE!!!

Yes, use a sharp HSS tool to do the fine machining and use a reasonable diameter bar (probably at least 1"/25mm on an L5) BUT DO NOT USE ANY FORM OF STEADY.

Rant over .

Nick.

 

 

 

 

Edited By Nick Hughes on 23/01/2016 23:01:45

[hthParticipant @hth]

Hi All

Some thoughtful advice and it is appreciated .

The lathe is in good shape, it came off a ship , I believe a RN or RAN ship as it has MOD markings on the original motor , which BTW is a 240V DC brushed motor made in London .

Anyway the problem has been found . It turns out it was operator error by me , or a lack of checking everything out properly . I discovered that the cross slide gib was not tight enough , its a tapered gib with a screw adjustment. The cross feed lead screw and nut are worn to some degree ( are not all of them worn ! ) . The cross slide was wandering in and out about .010" at will . As I was turning, the forces were moving the cross slide , hence the strange taper . On the L5, you tighten up the tapered gib by turning a screw and locking it with a nut . Mike

[Author]

Posts