• January Turning Challenge: Thin-Stemmed Something! (click here for details)
  • Conversations are now Direct Messages (click here for details)
  • Congratulations to Gabriel Hoff for "Spalted Beech Round Bottom Box" being selected as Turning of the Week for January 6, 2024 (click here for details)
  • Welcome new registering member. Your username must be your real First and Last name (for example: John Doe). "Screen names" and "handles" are not allowed and your registration will be deleted if you don't use your real name. Also, do not use all caps nor all lower case.

Balancing lathe

KEW

Joined
Jun 9, 2005
Messages
340
Likes
1
Location
North Metro Atlanta
What are some methods to ensure the weight of the lathe is balanced between the diagonal pairs of legs?
It is easy enough to adjust the foot pads so they all contact the ground, but with the weight of a lathe and the length of the legs, I have to believe it is likely that two diagonal legs may be supporting much of the weight allowing the lathe to vibrate more easily in the other direction.
Thanks for your thoughts and comments!
 
I think it's more important to have the lathe level in all directions. I would think that to get it level you would need all the pads touching the floor. Since the headstock is heavier than the tailstock one end would be heavier anyway.
 
All pads are touching the floor, but my thinking is that, say 70% of the weight could be carried by two of the diagonal legs with the remaining 30% on the other two. The reason for mentioning "diagonal" legs is the load should be fairly even between the diagonals if every thing is balanced.

For example, I'd bet I can adjust a foot pad two revolutions from the point it contacts the floor to the point it lifts another foot off of the floor (as the frame and bed flexes). Those with wooden floors would have even more range of contact as the wood gives.
Thus, the foot I am adjusting could carry 1 pound (just as it touches the floor) or ~300 pounds (as it gets close to lifting one of the other legs off of the floor).
In either of these extremes, the lathe would be more susceptible to resonance.
Another way of saying it is I am looking for ideas on how to ensure the two headstock legs are carrying close to the same load and the same for the two tailstock legs.
 
If there's no gap or flex there's no motion. All points grounded, regardless plumb or level, will take away the gap. Rigid stand components will deny flex. One of the reasons why I wonder why someone would put rubber feet on a lathe.

You can Weeble your lathe with low weights or make it rigid with broad stance to compensate for the angular momentum of that heavy spot coming over the top. Since I don't care to store sand, I use the latter method.

It is important to keep the ways straight, though most have enough structural rigidity to hold their own even if they do have an unequal static load.
 
Well you could put a strain gauge on each leg and then actually measure how much is under each foot. If you buy a lathe for 6 grand it should come with something like that.
 
On my oneway, I put some isolation rubber in between the foot pads and the concrete. They help absorb some of the shock and actually do allow for a bit faster rotation on the same given piece of wood.

That said, level isn't important as the same amount of (weight distribution) contact on each pad. That is why Oneway shows to level the feet (level isn't zero degrees to horizon or whatever) while an out of balance piece is spinning. A little harrowing if it is too big since you have to move from foot to foot with the allen wrenches while it is spinning.
 
^^^^^


All of the above is well worth considering, but there is one thing that can be done that hasn't been offered on this thread.....yet!

Use anchors to solidly mount your lathe to a concrete floor.

This is easy to do, and is the one thing, above all others, that will do more for stabilizing out of balance conditions on the lathe.

Think of this for just a minute: Weight alone will do some for dampening vibrations, but the source of vibrations is the centrifugal forces generated by all that moves......ie: your workpiece, spindle, motor, belts, etc. Those forces MUST move the lathe itself for there to be a problem that can be overcome. If those forces cannot move the lathe, then vibrations are dependent solely on the give and take of those lathe parts, and components that move. If the lathe is built robustly, heavy, rigid, and solidly mounted to something really heavy.......like a concrete slab......you've done the best thing you can to help eliminate problematic vibrations.

You can bolt your lathe to the shop floor in less than a couple of hours, but you'll need a hammer-drill to drill holes in the cement.....to seat the anchors.

If you buy a cheapie lathe in the first place, you'll forever be trying to solve problems that are pretty much unsolveable.......unless you stick with very small turnings exclusively.

otis of cologne
 
Last edited:
Thanks for the comments.

Odie, the anchor bolts are definitely a great option, but I hate to put holes in my basement slab. The housing boom was in full swing when my house was built. Building Inspectors were understaffed and Contractors were focused on moving on to the next house as soon as possible. I worry that my basement slab might be only 1/2" thick and crack on me.

I bought a Nova 3000 a bit over a year ago. When I built the stand, I decided to use one of these for leveling.
http://www.woodcraft.com/family.aspx?FamilyID=3986

I only used one because I really don't like the idea of using ~3/8" threaded rods as stable support as well as concerns about the rigidity of the joint at the threads. By putting it on the back of the tail stock end it carries the least load (my stand is asymmetric from front to back so I can get right up to the lathe). Under the other three corners I put pieces of non-slip carpet grip. This stuff compresses down under the weight of lathe with stand (550#), keeps the lathe from drifting on my concrete floor, and protects the painted wooden stand from any moisture from the concrete.

I adjusted it by loading a blank on the lathe and adjusting the pad one way until it vibrated then adjusting the pad the other way until it vibrated, then splitting the difference for my final setting. Not exactly precise, but it seems to work well.
However, now that I have my PM3520, I was wondering if there is a better way.

Tonight, I gave it some more thought and did some experimentation.
Again, I only adjusted the back tailstock leg. It seems like you could spend a while chasing your tail trying to adjust all 4, and I sincerely believe one leg is all that is needed to bring it into adjustment. I slid the headstock and tailstock toward the headstock end for this adjustment.
I found it was between 3 and 3-1/8 revolutions of the foot pad adjustment between the wrench position where I could slide a sheet of paper under the pad I was adjusting and the wrench position where I could slide a sheet of paper under the front tailstock pad.
Note - I have a 1/8" thick sheet of polyvinyl under my Powermatic. Though it is pretty dense, I'm sure it has some give and if you are directly on concrete, it would take less than 3 turns adjustment between the extremes.

While the front pad was slightly off of the ground, I turned on the lathe (with a rough turned bowl) and adjusted the speed to the speed I would turn at based on vibration.
This was 400.
I then adjusted the foot pad 1/4 turn and (my comfortable) turning speed went to 542.
1/2 turn yielded 630.
3/4 => 668
Between 1 and 1-3/4 turns the speed was 680 to 690. I really could not discern a difference I'm willing to swear to in this range.
Over 1-3/4 turns, the speed started dropping back down.

From this sequence I would believe that it does matter how much load is on the feet, but there is a pretty broad "sweet zone" of ~3/4 turns. Also, simply doing the paper test and splitting the difference generally seems close enough.

After this, I put a magnetic bubble level on the frame at the tailstock end. The bubble vibrated within the vial and I found I could use the level to narrow the "sweet zone" by half to 3/8 of a turn. I set it to the middle of this range and am ready to call it good.

I really wonder if I should get rid of the leveling bolts. I am 6'-2" and have the lathe pretty high.

Again, any thoughts and comments are welcome. I think this is a sound approach, but could be missing something.
 
Last edited:
Well you could put a strain gauge on each leg and then actually measure how much is under each foot. If you buy a lathe for 6 grand it should come with something like that.

LOL. The bad thing is I taught an engineering lab on mounting and testing with strain gauges 25 years ago. Now I don't even remember what the instrument was which read the output from the gauges!
 
I did it the old fashioned way. I adjusted the lathe for level. then I went back and put a wrench on each leg and felt how much tension it took to move it 1/4 turn. On one leg I went about 3/4 turn until I felt the same tension. I assume this leg wasn't setting as firmly. It didn't seem to change the level any.
Bolting my lathe down isn't an option. I have to move it occasionally.
 
I've balanced a few drawbridges in my time, and I think "balancing" a machine with whirling, unbalanced pieces is a fool's errand. No amount of strain gages will solve the problem. Even bolting to the floor could be asking for trouble.

Think about ballast. I have two 4-gallon buckets of railroad spikes on the lower shelf of my lightweight HF34706. I haven't weighed them, and don't intend to. I have some scrap lead that I can place between the buckets when/if needed. Sand could also be used, and a bunch of water bottles (say 2.5-gallon empty jugs from pool chemicals) would also offer some dampening. Ballast works quite well for empty cargo ships.

Joe
 
I cringe every time I hear about bolting machinery to a concrete floor.

I'd rather the equipment have room to vibrate instead of cracking by a weld.
 
Maybe I should not have used the term "balance".

My intent is that the load be spread as evenly as possible between the legs. Obviously, with the weight of the headstock on one end, you will not get equal loading, but the diagonal pairs of legs should be seeing close to the same load.
If they are not, then you are loading your lathe bed with some twisting force.

"Level" is often used, but it doesn't seem so important that the lathe be level as it is that the lathe be evenly supported.
 
I took a closer look at the rest of the thread, and splitting the difference between just contacting the floor, and lifting the neighboring leg, is a sound approach. Significant twisting forces in the bed will occur at the two extremes (opposite twist at each extreme); in between, less significant and possibly zero. I think you've misled yourself by assuming that diagonally-opposed support reactions must be equal to prevent torsion in the bed. With four legs, the lathe structure is "statically indeterminate." That just means that there aren't enough equations of equilibrium to compute internal forces, and deformations must be considered. Deformations can be limited by adjusting the legs as you've done.

(Yes; I'm a structural engineer.)
Joe
 
Bolts/No Bolts

Tying one's kangaroo down has been often discussed. It is generally regarded as a bad idea because the net effect seems to be to transfer all the vibration and centrifugal[sp] force in the running machine directly to the two moving surfaces, the spindle bearings, causing rapid wear. This is why few industrial machines are actually bolted down rather than being put on vibration isolator pads or feet. It's better to allow the machine to oscillate at its own frequency than try to kill all vibration.

If you'd really like to dampen the vibration on your lathe, I suggest you look over John Williams' solution at the link. It's certainly not as pretty as a Birds Eye Maple box filled with sand or lead under your lathe, but it WORKS.

http://www3.sympatico.ca/3jdw8/antivibration.htm

For a lathe to operate properly, it is more important for the spindle, the bed, and the attachments to all be tuned on the same axis. Beds, whether cast iron or steel will all twist, especially if the frame is not properly set. If you wish to have your machine tuned to optimum, there are several steps:

First, (as John said) level your bed (not the under-frame) to be dead-nuts level in both directions with the feet "firmly" on the floor surface. Use as long a level as you have to reduce any error in the tool, and check the vials at both ends to see if they "agree." This may involve both adjustable feet on the legs and shims (metal flashing is great) in the bed-to-leg joints.

Second, check your female morse tapers (head & tail stocks) to make sure they are clean, accurate and without damage. Galls, bumps, dirt and burrs can be carefully removed by hand with an MT reamer.

Third, check your male morse tapers, drive spur and live center, and gently remove and correct any burrs.

Fourth, make sure the points on your centers are straight and true and the bearing surfaces of the bed, tailstock (and for sliding heads, the headstock) are clean, and smooth.

Fifth, insert the centers and bring the points together with the TS locked down and the ram fully retracted into the casting. If the points align precisely, you're done. If however, they are off, you have more work to do.

For lateral misalignment (front-to-back), you will need to torque the bed (not the legs) in the direction opposite to the divergence. Thus, to align a TS point which is off towards the rear, you add a shim(s) to the rear bed/leg joint on the tailstock end, and visa-versa.

For vertical misalignment, the task is tougher because you must change the pitch, front-to-back of either the tailstock or the headstock. For fixed-head lathes, you can shim the attachment joints (inboard or outboard) to raise or lower the spur point a "smige." For sliding headstock lathes you're probably looking at machine shop work to regrind the two surfaces on the head's casting that meet the bed rails. Same thing would be needed for the tailstock.

Last step is to recheck your level. If the bed adjustment has thrown the level off, you re-level using the adjustable feet only so you don't undo the alignment work.


mm
 
Last edited:
Tying one's kangaroo down has been often discussed. It is generally regarded as a bad idea because the net effect seems to be to transfer all the vibration and centrifugal[sp] force in the running machine directly to the two moving surfaces, the spindle bearings, causing rapid wear. This is why few industrial machines are actually bolted down rather than being put on vibration isolator pads or feet. It's better to allow the machine to oscillate at its own frequency than try to kill all vibration.


Sometimes it's advantageous to think of the law of conservation of momentum, and realize that the bearings are always loaded by in imbalance, whether or not isolation or shock mounting is employed further downstream. Shock mounts on machines are generally designed to diminish the amount of energy transferred to a secondary structure. Easy to understand the advantage when the machine is located on a second story, but machines are shock-mounted as well to foundations so as to reduce the possibility of damage to the more brittle component.

The weakest link will always find a way to fail, which is why bearings and isolation mounts are designed to be easily replaceable.

The reason you don't want the lathe moving is that the operator generally doesn't move at the same rate and in the same direction the way a cutting tool secured (without shock mounting) to the machine can. Makes control more difficult, and in its most severe form, where a free-standing toolrest is used, is an invitation to disaster.
 
KEW,
I know what you mean about wanting each leg to carry its load and that being level (with in reason) wasn’t as important as sitting equally on all four feet, heck, I’m half a bubble off myself. My concrete slab is flat, but not flat flat, so when I move the lathe (3520) I have to reset it. I’ve gone the spinning the out of balance peace, crawling around, eventually figuring out that only one leg at one end needs to be adjusted. To some degree I think this works but I thought, there has to be an easier way, so this is what I came up with, this seems to work for quickness, repeatability and ease, although I’m not sure about the quickness or the ease part. The feet at the headstock end are locked in place, banjo and tailstock slid to the headstock. I use a small bottle jack centered on the shelf support bracket to left the tailstock end, now it’s a tripod, I wiggle the lathe to settle everything, usually it’s pretty solid, I think the wiggle part is just for my benefit, I use a peace of ¼ inch plywood as a feeler gauge, it’s a peace of scrap, about 1â€X8â€. The paper you’re using as a feeler gauge might be more accurate and give a better feel. I use the bottle jack to get the short leg close and then adjust them both using the ¼ inch feeler gauge, getting the “drag†about even, cinch the nuts, check again, lower the lathe and give it another wiggle. Usually it doesn’t. So far this has worked, when I first started, I was using a lever, a fulcrum and a peace of wood in place of the bottle jack, but the jack made things a lot easier. If you try this, add to, take away, change something or come up with a completely different idea, let us know. I know I’m not obsessed, really, I’m not…I’m not....
cc


Mark, I just read your post… now I have to break out my level…
Also for this;
“For lateral misalignment (front-to-back), you will need to torque the bed (not the legs) in the direction opposite to the divergence. Thus, to align a TS point which is off towards the rear, you add a shim(s) to the rear bed/leg joint on the tailstock end, and visa-versa.â€
If my points don’t align, couldn’t I just use the leveling feet, instead of a shim. I haven’t looked, so I haven’t tried. But I’m going to…
cc
 
If my points don’t align, couldn’t I just use the leveling feet, instead of a shim.

CC
The mechanical joint between the bed casting and the base or leg sets will give you finer control over the amount of twist you induce in the bed because it is closer to the plane of the bed. I've seen more than one lathe out of line simply because of uneven tightening of the bolts at the 4 corners and a slight defect in those mating surfaces. You could drive yourself nuts trying to torque the bed into alignment using the lathe feet. Once you have the lathe aligned "internally," it's a simple matter to re-level it without changing that internal alignment.

But, before you go through the shim thing, make sure that any misalignment is not caused by the taper. I have a Stubby 1000. After I received and cleaned it on arrival from Down-Under, I found that the points were off by a few thou's. The centers were new and flawless, so my next target was the tapers. The headstock taper was flawless, but when I went to the TS ram's taper, I discovered a very slight bump in the surface about 1/2" in from the lip. I have a #2 MT reamer, so all it took was a couple of turns to remove the defect. My points then met perfectly.

Your points can be thrown off by a piece of dirt or wood fiber stuck in the taper. I use, and highly recommend everyone to use, one of the green urethane rubber Tapermate [link] cleaner tools sold by Packard every time I remove or insert a center or other taper-use tool.

http://www.packardwoodworks.com/Mer...=117902&Category_Code=lathes-acc-tapacc-tapmt

mm
 
Back
Top