Viceroy T.D.S 1/1 G.B - Spindle refitting

Read what other Denford users are doing with their machines and software.
Educational resources, ideas and courseware.
If you have a Denford related success story, please tell us about it here.

Moderators: Martin, Steve, Mr Magoo

User avatar
clarkea1
Posts: 23
Joined: Fri 01 Sep , 2006 18:48 pm
Location: Cardiff, UK

Post by clarkea1 » Mon 27 Nov , 2006 12:40 pm

Dazz,

Firstly, what a nice job you're doing - the engineer in me appreciates your cunning idea of monitoring spindle bearing temperatures as a clue to correct setting etc.

Do you have a catchplate and dog for working between centres - if so, try a test of cutting between centres as this would eliminate any problems with the chucks.

Also, I'd be very surprised if it were the single phase motor vibrations causing the problem - I've got an old Harrison which was fitted from new with a 1ph motor, and don't get any vibration problems on the machine. Whilst my Harrison may have a bit more metal in it, your Viceroy looks chunky enough that vibrations from the motor shouldn't cause any trouble.

Can you beg/borrow/steal some inserted tip carbide tools as this would eliminate tool grinding as a variable?

Good luck, and keep us posted.

Alastair

User avatar
dazz
CNC Expert
CNC Expert
Posts: 243
Joined: Sat 19 Aug , 2006 11:31 am
Hardware/Software: Viceroy TDS 1/1GB lathe, Imperial, stripped down and rebuilt.
Viceroy Taper/Tracer attachment, Shop made cross slide tracer attachment, VSD.
Location: New Zealand

Post by dazz » Tue 28 Nov , 2006 4:29 am

Hi

Just to clarify, when I refered to wobble (W) in the previous post, this was the axial error measurement on the face of a chuck, or whatever I had fitted. Eccentricity (E) was the radial error measured on the cylindrical surface.

Chatter has reduced as I relearn tool grinding. It's been 20 years since I last used a lathe and I still have my old tools. I now have some 1215 steel so I will cut some test pieces and see what I get. I have a driver plate but no dogs. The lathe came with lots of accessories but was missing a lot of the basic and essential stuff.

At high spindle speeds, this lathe rumbles. I can feel it through the concrete floor. A VSD may, or may not make any difference. Adding a VSD would allow me to eliminate one belt drive. I could fit the 3phase motor in the cabinet directly under the lathe.

I am in a catch 22 situation. Carbide tooling and VSD are reasonably expensive. They'd be a good investment if they improve things, but a waste if they don't. I need to do a few more tests before I consider making modifications. Then I'd need fiscal approval from the wife!!!
Regards

Dazz

User avatar
davidimurray
CNC Expert
CNC Expert
Posts: 194
Joined: Thu 24 Aug , 2006 9:45 am

Post by davidimurray » Tue 28 Nov , 2006 9:55 am

Hi Dazz

I've only just caught up with the thread - so apologises if theses have been covered already.

Basically what I understand is that you are having chatter problems on the lathe you rebuilt.

Firstly can you give us some idea of the spindle speed, bar diam, dpeth of cut etc. What is the shape of the tool you are using?

Does the chatter occur regardless of whether you use power feed or not? Have you tried locking the saddle and using the compound slide to cut. Have you re-checked the gibs to ensure they are tight - they may have settled down.

Once you have turned the part have you run a DTI around the diam at varius points along the length - i.e. are the parts round? Also you might want to try putting a DTI against the back of the work as you are cutting to show the actual deflection as a result of the tool.

I seriously doubt that the VSD would make any difference. Can the lathe wobble? You might want to try packing under the corners so it is rock solid. Unfortunately most lathes will wobble a bit at high speed - none of the lathes at work are fixed down (Colchester Students and Smart and Brown) and I'm always worried they are going to wander off!

Don't worry mate - you will get there - any questions give me a shout.

Cheers

Dave

User avatar
dazz
CNC Expert
CNC Expert
Posts: 243
Joined: Sat 19 Aug , 2006 11:31 am
Hardware/Software: Viceroy TDS 1/1GB lathe, Imperial, stripped down and rebuilt.
Viceroy Taper/Tracer attachment, Shop made cross slide tracer attachment, VSD.
Location: New Zealand

Post by dazz » Sun 03 Dec , 2006 2:02 am

Hi

A good point. I aim to get tools right on centre. Maybe slightly below, but never above. With cutting-off tools, I normally aim for a few thou south of centre.

I tried a carbide parting tool today. I got the same results as with the HSS version.

At present, very light cuts, and heavy cuts work OK. There seems to be an intermediate operating area where chatter happens.
Regards

Dazz

User avatar
Triac whizz
CNC Expert
CNC Expert
Posts: 238
Joined: Mon 17 Jul , 2006 21:48 pm
Location: France

Post by Triac whizz » Sun 03 Dec , 2006 11:03 am

I put my tools at centre, maybe a gnats above as cutting forces will deflect them down, probably more so with a parting tool, though if its a insert one read the manufacturers instructions, they spent ages breaking tools to get it right :)

What speed are you running at and diameter turning? I have a point on the cnc where if I go too fast I get chatter on lightish cuts, Tipped tools are at least 200m/min (of top of my head) though that contradicts my previous thoughts!
Self Catering Lodges in Central France with covered pool & large grounds
www.la-coterie.com

User avatar
dazz
CNC Expert
CNC Expert
Posts: 243
Joined: Sat 19 Aug , 2006 11:31 am
Hardware/Software: Viceroy TDS 1/1GB lathe, Imperial, stripped down and rebuilt.
Viceroy Taper/Tracer attachment, Shop made cross slide tracer attachment, VSD.
Location: New Zealand

Post by dazz » Mon 04 Dec , 2006 10:59 am

Hi

I aim for around 20m/min cutting speed.
My tool grinding is never going to be accurate enough to worry about very precise tool height.

I never go above centre because a certain range of tool deflection can increase depth of cut and cause chatter. Greater depth means more deflection which is positive feedback. Positive feedback is bad because it promotes instability. All chatter is caused by unstable positive feedback. It's only the source that differs.

With a tool starting at centre or below, tool deflection should reduce the depth of cut. Less cut, less deflection and less chatter with negative feedback. Negative feedback is good and promotes stable operation.

For my lathe I suspect that light cuts with large radius tools start the spindle assembly ringing like tuning fork. High loads seem to apply enough force to keep the spindle from moving around. More pre-load might reduce this tendency. I notice the later Viceroy designs had the linked belt running over the top of the spindle (mine runs over an intermediate shaft). This change in design would help suppress oscillation like placing a finger on a tuning fork. There is no possibility of me making this modification on my lathe.

I tightened the drive belts today. I have been intentionally running them slack as a safety valve while it explore the limits of the machine. This adjustment made a big difference to high-speed smoothness. With loose belts, the machine was rumbling at high speed. Tight belts have fixed that and it now runs with a smooth whurr of belts and gears.

I cut my first job today. I made a copy of a missing piece to the taper turning attachment. It's made from 1215 steel. It's a small simple job but it shows the lathe does do it's job. The picture shows the original and the copy.

Things are looking better. Lathe performance is incrementally improving as I make adjustments. Using decent steel, adding bearing pre-load, tightening the belts, re-learning tool grinding, and exploring the machines limits have all helped. As time permits, I will continue to tinker with settings to extend working limits of the machine.
Attachments
IMG_0435.JPG
IMG_0435.JPG (142.18 KiB) Viewed 14147 times
Regards

Dazz

User avatar
dazz
CNC Expert
CNC Expert
Posts: 243
Joined: Sat 19 Aug , 2006 11:31 am
Hardware/Software: Viceroy TDS 1/1GB lathe, Imperial, stripped down and rebuilt.
Viceroy Taper/Tracer attachment, Shop made cross slide tracer attachment, VSD.
Location: New Zealand

Post by dazz » Sat 10 Feb , 2007 11:15 am

Hello

I finally got a chance to turn up a test bar. The finish achieved is improving with good steel and improving tool grinding technique. The finishing tool I used to make the cuts had a large nose radius. A prime candidate for chattering, but everything was smooooooth and rock steady.

I did two tests:

Firstly, with the 35mm diameter bar held by the chuck and tailstock centre, I adjusted the tailstock offset to get a parallel cut over a length of about 270mm.

I then measured the bar diameter at 50mm intervals to check if the tool could cut a parallel bar. Any displacement error of the saddle from a straight line would be multiplied by two. I measured with a Mitutoyo micrometer with 0.0001" graduations (10ths of a thou). I was pleasantly surprised to find that the bar was dead parallel. This was too good to be true so I rechecked all the measurements and got the same results. Exceptionally good for a machine over 20 years old.

It's probably worth noting that these measurements were the result of quite a lot of time and effort. Getting the tailstock aligned and obtaining a high quality finish were itterative processes. It took a bit of practice to get repeatable measurements from the micrometer. I have lost count of how many cuts I made before getting the final results.


For the next test I simply withdrew the tailstock so the bar was only supported by the chuck. I mounted a finger type dial gauge on the saddle I then took a measurement at each end of the bar. I rotated the bar 180 degrees, took a measurement at each end again, then averaged the measurements at the respective ends. The 180 degree rotation and measurement averaging zeroed out any runout error as a result of the 3-jaw chuck not holding the work exactly in-line with the spindle. It was relatively easy to find the positions where 180 degree rotation produced no change on the DTI.

This test took a few goes to get consistent results until I mounted the DTI on the saddle rather than the cross slide. The measurements showed that over a distance of 250mm, the free end had an error of 2 thou inches away from the tool. This means that the spindle is not quite lined up with the bed. This is a small error over a relatively long distance. It doesn't take much of a force on the end of the bar to produce a 2 thou displacement. The weak link is the three jaw chuck.

I am not too worried about this error. There is no way that I would attempt to turn anything 250mm unsupported from the chuck. Tool forces would create much greater errors than 2 thou. So in practice, I think this error is immaterial.

Applying a few kilograms pull to the tailstock (with centre withdrawn from the bar) corrected the error. This force was enough to twist the bed to bring the bar end closer to the tool. As an exercise in pedantic perfectionism, could this error be eliminated by adjusting one of the bed mounting bolts to twist (untwist) the bed. Is this a technically "correct" method of eliminating this type of error.

The images below show the bar being cut. The other shows the finger indicator being used to measure runout at the free end of the bar.

So these and the previous measurements show that the lathe is still accurate and nothing is bent. The small twist in the bed is insignificant compared to tool force displacement errors. The only other test that I think I could do would be to face the surface of a large disk. This would check the accuracy of the cross slide. I have a surface plate, but I don't have any suitable steel for a test piece, so I won't be doing this.
Attachments
IMG_0547.JPG
IMG_0547.JPG (91.52 KiB) Viewed 13711 times
IMG_0545.JPG
IMG_0545.JPG (122.22 KiB) Viewed 13711 times
Regards

Dazz

User avatar
dazz
CNC Expert
CNC Expert
Posts: 243
Joined: Sat 19 Aug , 2006 11:31 am
Hardware/Software: Viceroy TDS 1/1GB lathe, Imperial, stripped down and rebuilt.
Viceroy Taper/Tracer attachment, Shop made cross slide tracer attachment, VSD.
Location: New Zealand

Re: Viceroy T.D.S 1/1 G.B - Spindle refitting

Post by dazz » Sun 24 Jun , 2012 11:34 am

Hello

I has been close to 6 years since I started this thread. I have noticed a few links to this post so I thought I would provide an update.

In that time my tool grinding has improved significantly. I know that any chatter or rough finish is my fault, not the lathes.

I have been increasingly using carbide and cermet insert tooling. They work great only if the are run at high cutting speeds and depth of cut (DOC). Usually I run at max rpm with enough feed and DOC to get the drive belts smelling hot. Only then do I get the type of smooth finish seen in the glossy brochures. They only work properly when properly loaded. The chips need to be a dark blue colour. When used correctly, cermet provide a very good finish, almost mirror like.

At these feeds and speeds I often get showered with hot chips. I find that I need to position my hands so that the hot chips don't settle on the skin and burn. A chip guard is on the agenda for a project.

The work holding has to be just right. I don't use driver dogs at carbide speeds because of the unbalance and resulting vibration. I use a 4-jaw chuck at one end and a carbide tipped dead centre at the other. This ensures perfect balance and strong work holding.

For HSS I have found top rake and tool sharpness makes a big difference to finish. For MS I use up to 40 degrees for finishing. I hone the edges using grinding paste on a flat copper plate. I end up with a mirror finish on the cutting edge. This gives a mirror finish on aluminium. I also find that the DOC should be at least the radius of the cutting tool point.

This has proved to be a great lathe doing sterling service.

Dazz
Regards

Dazz

Post Reply