Thread: Long Thin Spindles - pulling not pushing

Originally Posted by NeilS

Always happy to hear of other ideas, Jerry.

Also had a look at your video. Interesting! If I was going to be doing a lot of the same size I would definitely look at getting one of those 'pencil sharpeners'.

Have you found a use for those shavings?

If they are wet, I can get about an 8 foot strand. Someone doing Marquetry might have an interest, but they are way too thin. So far, just shavings to burn or toss........ Jerry (in Tucson)USA

The Vermec Threaded Live Centre arrived yesterday, with the customised thread added to the rear of the MT to take a drawbar and Loctite used when pressing in the live centre to (hopefully) ensure that it does not release under tension.

It was grandparenting yesterday and again today, so I haven't been able to use it yet, but did manage to go and get some allthread for the drawbar and give it a test reading to see if it will take the tension that I expect will be required without coming apart.

Dennis Gooding did his tension test at 55lbs (25kg). Stephen Kirby tested his rig up to 100lbs (45kg) but estimated that he was only using 15 to 20 lbs (7-9kg) on his spindles.

So, I split the difference and tested the Vermec up to 32kg (70lbs).

Kgs of tension test.JPG

Nothing let go at that tension. The modified Vermec may have taken more, but I couldn't see the benefit of doing so at this stage, especially if the required tension to get a desired benefit is substantially less than that. I should also mention that this was a static test, ie. the live centre was not spinning or any lateral forces applied, so under working conditions other factors will come into play. We will see.

I will probably get to try out some spindle work with it over the weekend to see if I can find any benefit from this technique and report back, one way or the other.

Until then...

Next step was to see what happens to the chuck in the tailstock live centre when stopping. If this is a problem it could be a party stopper. Stephen Kirby had reported that he was getting chuck oscillations at the tailstock end when stopping. If severe enough and the chuck heavy enough, these could damage a thin spindle.

Test setup: A 500mm length of 9mm dowel (from hardware shop) was use between two chucks of about equal weight and size (2kg and 100mm diam). The dowel was clamped firmly in the jaws of both chucks, but no pressure or tension was applied at the tailstock end.

First test was a running stop at 200rpm, which is not a practical speed for spindle turning, but a safe one to begin with. The previously reported oscillation was evident.

Next, how much more oscillation will there be with 3x that speed and what will be the impact be on the spindle. So, 2nd test, a running stop at 700rpm, ie, press the red stop button at full speed. Result: Almost the same as for 200rpm.

700rpm is still too slow for turning a 9mm spindle, so 3rd test, up the speed to 1500rpm. Press green button, wait until controller reaches full speed, stand back with face shield on and press red stop button. Result: Imperceptibly the same as for 200rpm and 700rpm.

Conclusion: The electronics is doing the all the work of gently slowing down the rpm until the lathe is almost stopped, then hitting the brakes at the very end with the result being the same oscillation happens across the above rpm range.

Unanswered question: Would the same apply at even higher RPMs? And, what would the impact be on a very thin spindle (eg down 2mm to 3mm); would it twist apart? What about a much longer spindle?

Anyway, here is a short video of the oscillation that I am getting on my lathe. It seems minor to me at this stage, but very thin spindles in some woods may object to that amount of twist.

Chuck oscillation on stopping from 1500rpm.mp4

So, all ready to go ahead now and start experimenting with this setup to see if it provides any benefits over standard spindle turning techniques.

Using the same length (500mm) of crappy dowel that I used for the oscillation test, I turned the left half under compression with about the same amount of 'push' I would normally use with a live tailstock centre and the right half with some tension (perhaps less than 10kg). Lathe speed was about 1500rpm.

500mm spindle.JPG

I took the 9mm dowel down at both ends to about 7mm in several places. Closer to the chucks there was no obvious difference. As expected, the further away from the chucks the more deflection and vibration occurred. At this point I started to increase the tension on the right side and with that I experienced the first noticeable benefit, while on the left side, the compression end, typical unwanted candy stick spirals and vibration noises were starting to happen. The more tension that was applied to the right half the better it went with perhaps (I can only guess) a tension in the low teens.

It was time to go thin and at the mid-point where the spindle is at its weakest; down to 3mm, no trouble!

3mm at mid-point.JPG

Just a small section down to that size, but a promising initial result. Of interest, the thinnest section held up against the stop oscillations.

I will probably take a few more areas down to 3mm (or so) to see what happens next.

Using the same rig as above, I got the following results turning down 480mm long x 9.5mm pieces of dowel to the longest and thinnest cross section I could manage, with and without tension.

Under tension:The tension applied was approx* 3kg (⅛ turn of handwheel). I managed to get down to 3mm diameter over a span of 250mm while learning a few things along the way.

With compression: The pressure applied was similar to what I would normally use with a live centre; ie, as light a possible. I got comfortably down to about 5 to 6mm diameter over a span of about 250mm, but as you do, I pushed on to see if I could do better, but without any success. The middle ⅓ didn’t want to know about it..

Here are the two spindles as far as I could go with both of them. It is obvious which is which.

Two spindles with skew.jpg

Two spindles - Lg.jpg

And here are closeups of the mid section of the two test spindles.

Two spindles c skew - detail - Lg.jpg Two spindles detail - Lg.jpg

And a brief video to give more perspective on the 3mm spindle.

Others, especially those who do a lot more spindle work, could probably do better, but the above results summarise what I could achieve with the two methods of holding the test spindles.

Note: I did not turn down to final size starting from the tailstock and working in small steps towards the headstock, as would traditionally be done. I took 1mm off at a time over the full length working in from both ends. There is little benefit in the incremental approach if you have a 2kg chuck hanging on the spindle end! A lighter chuck configuration on the spindle end would help overcome some of this. See tip below.

Observations

The concern leading up to these experiments was that the tailstock chuck would overrun on stopping (the oscillations previously reported), but the spindle failures that I experienced happened when starting up again after the spindles were down in the 3mm diameter range. They just twisted apart attempting to get the weight of the 2kg chuck at the tailstock end up to speed. There were no failures on ramping down, although at 3mm diameter over 2590mm the oscillations got up to about 120°of rotation (ie, twist, with this particular wood) if the following tip was not employed..

Tip from Tim Skilton: When stopping, turn the lathe off at the powerpoint when the speed has been dialled down to its slowest speed and then both chucks will coast down to a gentle stop, which avoids the VFD applying its stop brake.

Tip: If you must stop and start again, especially as the spindle becomes long a thin, pull the tailstock chuck towards you and give it a helping flick as you start up again. This reduces the severity of the twisting.

Tip: Once stopped, if you need to rotate the spindle (eg. to inspect 360°) hand rotate both chucks at the same time to avoid putting the spindle under unnecessary stress.

Tip: A lighter and smaller diameter chuck on the tailstock end, like a Jacobs chuck on a live centre (with drawbar), is expected to create less twist and oscillations.

Tip: Of course, use the VFD to slowly ramp the speed up and down.

Tip: Listen carefully to the vibrations coming from the spindle. As it gets thinner it starts to behave like a string on a musical instrument. I also thought I could see harmonics happening at different points along the spindle. These might be indicators for there needing to be more or less tension or a different speed.

*Note: The calculated tension used was about 3kg. This was calculated by measuring the force exerted on a set of scales placed between centres when the tailstock wheel was tightened with the same rotation (about 1/8th of a turn) as the reverse rotation used to get the required tension. Not precise, but at least an indication of how much tension I was using. I might try some other setups to see if I can get a reading that I’m more confident about.


Conclusion

I’m not likely to be turning spindles like these (I’m not a Trembleur type personality!) other than for the purpose of this experiment, but I’ve proven to myself that turning with tension is a more pleasant turning experience than the traditional way of holding the blank between centres. Your experience might be different!

The next time I do any straight spindle turning (multi-centre and eccentric pieces may be another matter) I will definitely be getting out this, or a similarly adapted, live tailstock centre, although I’m unlikely to be doing any more of the type of the spindle turning that I was doing 50 years ago. Yes, it has to be that long ago, as the spinner in the following photo is my wife, now in her 70s and grandmother of five…

Bron with spinning wheel - cropped.JPG

As always...

Interesting results!

(...and it always strikes me as odd how so many long-term turners' other halves are into spinning/weaving. )

I admire your determination on this one Neil and an interesting result. Somewhat like you, pepper grinders and tool handles are about the extent of my spindle turning.

Following on from my earlier experiments with the Vermec Live Centre (see previous posts) I ordered a 2" mini chuck from McJing, plus their live centre adaptor to match the 1"x8 thread on the mini chuck.

The purpose for this was to see if a smaller chuck (and matching lighter adaptor) would make any difference to the oscillations experienced with the heavier chuck on the tailstock end. Those oscillations limited the thinness and length of the spindle that could be achieved despite slowly dialing down the speed and turning the power off to let the lathe coast down to gradual stop (ie, no electronic braking) as suggested by Tim Skilton. In my previous experiments I found that the twisting in thin section spindles caused a twist fracture when I got below about 3 to 4mm on the test wood I was using. Starting up, which is more abrupt, probably had a similar twist effect, at least in one direction, but this was not as observable like the oscillations on stopping.

The items ordered from McJing arrived promptly (as usual) and the mini chuck is certainly small and light.

Mini chuck end profile.jpg Mini chuck side profile.jpg

Mini chuck weight.jpg

The mini chuck and the live component of its tailstock assembly weighs about one quarter of the Vermec with a 4" chuck attached (Nova and Vicmarc 4" chucks both weigh about 2kg). Here is a photo to show the relative sizes of the mini chuck assembly and the Vermec.

IMG_1590.JPG

As you can see the mini is a metal lathe chuck (for miniature engineering jobs), which I heavily fettled so that it wouldn't kill my fingers. I removed the jaws to do that. The only corners of the jaws I didn't round over were the inner holding faces that I intended to use to hold the spindle. I also eased back the first row of gear teeth that protruded beyond the chuck body when wound out that far.

Mini chuck - fettled.JPG

Next I drilled out the back of the Morse Taper ready for threading to take a draw bar. As advised by Turner Ted, I used the my 2" long nosed jaws to hold the MT, aligning it with a standard live centre to get some semblance of trueness. Unlike the Vermec live centre adaptor, which has an adaptor thread diameter (at least for a M30x3.5 thread) that is larger than the live centre housing, the McJing with the 1"x8 thread adaptor has a smaller diameter than the bearing housing and this allows the long nosed jaws to be used to hold the bearing housing on the MT shaft.

IMG_1592.JPG

I had to work up through small increments in drill sizes to get to 17/64ths, which is the bore needed for the M8x1.25mm thread I am using for my drawbars. The long nosed jaws held the bearing housing sufficiently to complete the drilling, but small drill size increments were necessary.

In use, the mini chuck ran true (at least sufficient for my purposes) on the McJing live threaded adaptor and with the tension I applied (probably about 2 to 3 kg) it didn't budge from its bearings.

I will report on the results from using this mini chuck attached to the McJing live threaded adapor in my next posting.

Using the mini chuck as described in my previous post and the same spindle that I was unsuccessful with in compression mode in my previous experiment; yes, the rough as guts effort that I only managed to get down to approximately 5 to 6mm diameter for part of its length, and without any success through its mid one-third.

First thing to report is that there was minimal oscillations with this mini chuck. So much so that I stepped up the belt position to max 3,000rpm, a better speed for thin spindles. I was also able to do less ramping the speed up and down until I got closer to my minimum thickness.

I turned away and turned away. It was almost pleasant, if spindle turning can ever be pleasant for a bowl turner!

Got down to below 2mm over a length of about 300mm. Yes, a useless exercise other than to prove a point. Here we are:

2mm spindle with mini chuck.JPG

The larger spindle (crossing the 2mm spindle) is the 3mm spindle achieved with the 4" chuck in tension mode during my previous experiment. Here are a few more photos as an indication of size.

IMG_1606.JPG IMG_1602.JPG

For me that settles it on the advantages of using tension instead of compression, at least on long thin spindles, and probably with any spindle.

On the chucks, there is not much difference between the cost of the Vermec with drawbar thread if used with one of your existing chucks compared to the McJing mini chuck combined with their live thread adaptor, although you do have to add the thread yourself for the drawbar with that option.

For 6mm diameter spindles and above, the Vermec + your own chuck will probably work well in most cases. For sub-6mm the mini-chuck comes into its own.

Either way, you may find spindle turning a more pleasant experience when pulling it.

And, as far as I know, you won't go blind from doing this...

Anyway, it's back to bowl turning for me.


And, as always....