Some questions about lathe

Home Model Engine Machinist Forum

Help Support Home Model Engine Machinist Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Nikhil Bhale

Well-Known Member
Joined
Nov 15, 2019
Messages
150
Reaction score
136
Location
Vidharba, India
Good day,

I bought myself a lathe test bar. I got it for about USD 15. I checked the bar along its length. It has a consistent dia of 22mm.
IMG_20220706_201824.jpg


I checked my lathe spindle runout with this bar. I am getting 0.04mm runout near the spindle and 0.07mm at the far end at about 250mm from spindle.
IMG_20220706_192147.jpg


IMG_20220706_191936.jpg


Are these runouts within limits?
My lathe is Sieg C4B Chinese make.
What other checks can I carry out with this test bar?

Regards
Nikhil
 
Hi !
How do you fix in this position ?
IMG_20220706_192147.jpg


Why don't you install the chuck or colet and do the test ?
 
Sorry , I missed it ( MT 3 )
A suggestion :
Checking the spindle bearing
If necessary, replace with tapered roller bearings

EDIT To be sure: Rotate the test bar at an angle of 90, 180 ...degrees from the old position and check again
 
Last edited:
I bought one of those through Ebay via India about 3 years ago. I didn't have high expectations knowing where it came from but actually it wasn't horrible, at least initially. The diameter was pretty consistent within 0.0002" down the length of the (~16"?) bar using my best micrometer. I detected a slight bow at the end of just under 0.001"which I marked with a felt pen. I tried different fit & clock positions in my spindle but it seemed to read that deviation quite consistently. My lathe is MT5 socket so was using a MT3/MT5 adapter sleeve which came with my (Taiwain 14x40) lathe which I assumed might have some influence.

Anyways, spring time forward to a month ago, I mounted it up again. The deflection now reads 0.0045" in the same position. No way I made a measurement error like that originally & could still see my original felt pen marking. So my only conclusion is the material is stress relieving. The diameter is still pretty good, but its significantly less useful as-is. I figure it costs me about the same as a stick of 4140 & it may turn into some object in the future, so my feelings aren't hurt. Anyways just be aware. My buddy says he has seen a few like that now, some right out of the box. Apparently a 'real' one like a USA/Suburban are like a grand or something crazy. I have seen some used ones but that can be dicey too. Thought I'd just chime in.
 
I bought one of those through Ebay via India about 3 years ago. I didn't have high expectations knowing where it came from but actually it wasn't horrible, at least initially. The diameter was pretty consistent within 0.0002" down the length of the (~16"?) bar using my best micrometer. I detected a slight bow at the end of just under 0.001"which I marked with a felt pen. I tried different fit & clock positions in my spindle but it seemed to read that deviation quite consistently. My lathe is MT5 socket so was using a MT3/MT5 adapter sleeve which came with my (Taiwain 14x40) lathe which I assumed might have some influence.

Anyways, spring time forward to a month ago, I mounted it up again. The deflection now reads 0.0045" in the same position. No way I made a measurement error like that originally & could still see my original felt pen marking. So my only conclusion is the material is stress relieving. The diameter is still pretty good, but its significantly less useful as-is. I figure it costs me about the same as a stick of 4140 & it may turn into some object in the future, so my feelings aren't hurt. Anyways just be aware. My buddy says he has seen a few like that now, some right out of the box. Apparently a 'real' one like a USA/Suburban are like a grand or something crazy. I have seen some used ones but that can be dicey too. Thought I'd just chime in.
Yes. That's one question that is nagging me. How much to trust the test bar?

Are there any other test besides the dia check that can be done on the test bar?

Regards
Nikhil
 
Nikhil !
" I checked my lathe spindle runout with this bar. I am getting 0.04mm runout near the spindle and 0.07mm at the far end at about 250mm from spindle. "


Just my thoughts.
With an error of 0.03 mm on a length of 250 mm
Usually, of the engines I've made, the cylinder is the longest part
For example , cylinder length 50 mm : 250 / 50 = 5 ,
So the error on the 50 mm length is : 0.03 / 5 = 0.006 mm , which is too good for me .
Crankshaft, camshaft is much longer but it is divided into small parts, when machining, just measuring each part is enough
Of course, you want the error to be as small as possible 0.01... or zero and that's fine, but it takes a lot of effort, money... and sometimes you just can't get what you want because it also depends on the accuracy of the entire lathe
With your lathe , I certainly make many types of engines : flame eater , stirling , 2 - 4 stroke , and " true " diesel engine .
 
Nikhil !
" I checked my lathe spindle runout with this bar. I am getting 0.04mm runout near the spindle and 0.07mm at the far end at about 250mm from spindle. "


Just my thoughts.
With an error of 0.03 mm on a length of 250 mm
Usually, of the engines I've made, the cylinder is the longest part
For example , cylinder length 50 mm : 250 / 50 = 5 ,
So the error on the 50 mm length is : 0.03 / 5 = 0.006 mm , which is too good for me .
Crankshaft, camshaft is much longer but it is divided into small parts, when machining, just measuring each part is enough
Of course, you want the error to be as small as possible 0.01... or zero and that's fine, but it takes a lot of effort, money... and sometimes you just can't get what you want because it also depends on the accuracy of the entire lathe
With your lathe , I certainly make many types of engines : flame eater , stirling , 2 - 4 stroke , and " true " diesel engine .

Yes you are right. Till the lathe delivers what I ask of it, I will not do anything. Chasing the last micron will be too expensive and the lathe itself is not built for that kind of accuracy.
I have built a running model Stirling engine and two PM Research engines on this lathe. Now I am building a Rider Ericson hot air engine from Myers. If it runs then it will prove that lathe has no issues
Anyways today I checked the spindle runout directly off the spindle taper. I get very small deflection of maybe half a division on my 0.01 mm indicator. So I think the lathe is good but the test bar has some issues.

Regards
Nikhil
 
Nikhil..

I have Sieg C6 lathe.

Make sure that there is no slack and impurities in the spindle that it can give measurement errors on the test bar.

Spindle bearings should be cleaned and greased with bearing grease every 5 years if there is no abnormal wear or noise. Abnormal wear or noises present: Replace to new spindle bearings with the correct part number. The 2 locknuts that are screwed into the other end of the spindle must be adjusted a little tight so that there is some rolling resistance. Run the spindle for 15 minutes and feel on the heat area of the spindle bearing that it should feel lukewarm hot. Too hot: Cool down and loosen the nut slightly and take a new test again. Cold: Tighten the nut a little more and take a new test again.
 
Nikhil..

I have Sieg C6 lathe.

Make sure that there is no slack and impurities in the spindle that it can give measurement errors on the test bar.

Spindle bearings should be cleaned and greased with bearing grease every 5 years if there is no abnormal wear or noise. Abnormal wear or noises present: Replace to new spindle bearings with the correct part number. The 2 locknuts that are screwed into the other end of the spindle must be adjusted a little tight so that there is some rolling resistance. Run the spindle for 15 minutes and feel on the heat area of the spindle bearing that it should feel lukewarm hot. Too hot: Cool down and loosen the nut slightly and take a new test again. Cold: Tighten the nut a little more and take a new test again.
Thanks for the tips. I will keep them in mind. Presently my lathe is working good and I will not try to find non existing faults and fix them.
But these tips will be useful for the future.

Regards
Nikhil
 
Hello Nikhil, very commendable to check the accuracy of your machine!
Whatever is precise enough depends fully on your expectations. For my horological components a runout on a testbar in a collet near the spindle of plus/minus 0,02 mm is fine. And very much more precise than the equipment that build excellent clocks in the past centuries. The run-out at some distance from the spindle is less relevant for my work as I do long work either at one end in a collet and the other end supported by a center or fully between centers. Personally I start to measure the run-out of the spindle-cone in two positions. Then I follow up by mounting the collet holder, a collet and a short precisionbar. First measurement is runout near the spindle. Second measurement is runout near spindle with the testbar remounted at 180 degrees. Third measurement with the collet re-mounted at 180 degrees. Fourth measurement with the testbar once again remounted at 180 degrees.
Your runout of 0,04 mm (total?) would be on the edge for my work. If the measurement of the inside of the spindle-cone would be over plus/minus 0,01 mm I would true-up that cone with a toolpostgrinder.
 
Thanks for the tips. I will keep them in mind. Presently my lathe is working good and I will not try to find non existing faults and fix them.
But these tips will be useful for the future.

Regards
Nikhil


Contaminants from turning work can get into the grease and interfere with the fact that it is a good maintenance to clean and re-grease every 5 years.
You may be surprised when you disassemble and check what's in there. How old is your lathe now?

Bearing manufacturers such as SKF and FAG commonly give a relubrication life period of about 40,000 hours under mild speed, load and service conditions. Also 40,000 hours = 4,6 year.
 
Nikhil..

Replace to new spindle bearings with the correct part number. The 2 locknuts that are screwed into the other end of the spindle must be adjusted a little tight so that there is some rolling resistance. Run the spindle for 15 minutes and feel on the heat area of the spindle bearing that it should feel lukewarm hot. Too hot: Cool down and loosen the nut slightly and take a new test again. Cold: Tighten the nut a little more and take a new test again.


That's how to adjust the tapered roller bearings ????????????
Are you SURE ????????????
 
That's how to adjust the tapered roller bearings ????????????
Are you SURE ????????????

Yes, the lathe has 2 nuts to adjust the bearings. See the picture..

Spindle adjustment1.png


Spindle adjustment.png
Spindle adjustment2.png


Can you imagine why we adjust the front wheel roller bearings until the roller bearings are slightly preloaded on vehicles? Yes, because the shaft and roller bearings get hot and expand until the roller bearings run without slack and less tight. If the was not preloaded roller bearings as we get slack wheels who are affecting the driving properties on the vehicle.

The same manner we do with the roller bearings on the spindle so that the spindle is not slack when the roller bearings and the main dock have reached working temperature. So we do not get measurement errors on the workpiece when we prevent the spindle wandering sideways with preloaded the roller bearings.

If you has chatter on the lathe while working, there is fault adjusted roller bearings, retightening the bearing or check the bearings is worned out..
 
Last edited:
Contaminants from turning work can get into the grease and interfere with the fact that it is a good maintenance to clean and re-grease every 5 years.
You may be surprised when you disassemble and check what's in there. How old is your lathe now?

Bearing manufacturers such as SKF and FAG commonly give a relubrication life period of about 40,000 hours under mild speed, load and service conditions. Also 40,000 hours = 4,6 year.
Hi
I have been through the process of adjusting bearing pre-load on my Denford Viceroy lathe.
Take a look at this link to see what I did Headstock bearing adjustment
 
Hi
I have been through the process of adjusting bearing pre-load on my Denford Viceroy lathe.
Take a look at this link to see what I did Headstock bearing adjustment

Information from SKF homepage..

Effects of bearing preload The primary beneits resulting from preload include but are not limited to:

-enhanced stiffness
-reduced noise levels
-improved shaft guidance
-compensation for wear and settling
-extended bearing service life

SKF tapered roller bearings are heat stabilized up to:

D ≤ 160 mm → 120 °C (250 °F)
D > 160 mm → 150 °C (300 °F)


Roller bearings can withstand so high up to 120-150 degrees Celsius and it becomes a great friction for the electric motor to rotate the spindle that it is not appropriate to have such a high temperature up to maximum temperature with such a large preload on roller bearings in the lathe.

In general it is between lukewarm to warm up to 50-60 degree celsius it is optimal we use in the lathe when the accuracy of the workpiece is good and there is no vibration present for example we use parting tool to cut the work piece who are most load on the work piece of all lathe tools.
 
before doing all sorts of complex stuff, read the runout of the taper itself, put your indicator on the ground part of the taper - if you can see .001, you know that your test bar will only magnify it over the length. what you are measuring is not the accuracy of your lathe, you are measuring the accuracy with which the morse taper was ground, and that shouldn't matter very much because normally you will use a chuck or you will put a dead center in the taper and turn a bar between centers with a dog.
 
I made a solid toolpost mount for my lathe from aluminium block.
I hope this will increase rigidity of the setup.
I plan to turn a bar on lathe now and check its diameter at different locations. Am I right to assume that if the diameter is same at all locations and orientation then the spindle runout is negligible?

Here is video of my solid toolpost mount


Regards
Nikhil
 
I think you will be disappointed. you actually cannot expect good results from what you are trying, even with a brand new $100,000 USD lathe, you can't expect good results. everything in your lathe, including the work piece, is a spring, literally evrything. to get a feel for this, put your test bar in place, put your indicator on the end of the bar, and push on the bar - how far does the indicator move? the force from the cutting tool can be at least 100 kg.

I have a larger lathe than yours - if I try to cut a 25 cm long bar, say 8 cm diameter, with this lathe, it will flex and I will get a taper. that is why you support work that hangs out of the chuck by more than a few inches at both ends, that's what the tailstock is for. you can still get flexure, but it's a lot less.

1657377800423.png
what you want to do is to put a piece of metal between centers, turn it with a sharp tool, making several passes without changing the cutting depth, until there is no more cutting, then measure the diameter of the bar along its length. if it is uniform diameter (say within a tenth of a mm) then you are good. if it tapers uniformly, your tailstock is not centered with the headstock in one or more axis, if it has a irregular taper that generally indicates wear on the ways.
 
I made a solid toolpost mount for my lathe from aluminium block.
I hope this will increase rigidity of the setup.
I plan to turn a bar on lathe now and check its diameter at different locations. Am I right to assume that if the diameter is same at all locations and orientation then the spindle runout is negligible?

Here is video of my solid toolpost mount


Regards
Nikhil

Greetings,

Turning a bar held in a collet or chuck at one end only will verify the lathe ways and the spindle center of rotation are parallel / coaxial / level different terms get used. It will not confirm the spindle run out. A perfectly aligned lathe, with the work mounted offset 10 thou from center, will still turn a "perfect" cylinder. A spindle with zero runout, perfect collet in the spindle bore, holding perfectly round straight stock, will turn a taper if the head stock and lathe ways are out of alignment. Two different tests for two different sources of error. Both are based on a bar of large enough diameter that other factors don't effect the result enough to matter.

Don't overlook that the machine and work are only somewhat stiff. A lathe bed isn't cooked spaghetti, but it will move around under cutting forces. The work will deflect, the tailstock can flex, all sorts of things come into play. A long bar will almost always end up tapered held from one end only, held between centers the center of the bar will not match the ends. It may be below what most of us can measure if everything works out nicely, or it may be enough to matter. If the bar is way too long for it's diameter, it may end up folding in half and beating up your lathe. Try not to be in the way. If we could always cut to exact dimensions tolerances wouldn't be such an important part of manufacturing.

Cheers,
Stan
 

Latest posts

Back
Top