Precisely Holding Long Square Bar in Lathe for Turning

[TimSParticipant @tims]

Thanks for even more replys i haven't had time to process all of them and as regards to accuracy thats a very good question, i hadn't actually considered that as this is my first really precise job… and i dont much like numbers so only tend to think of them as a last resort :/

using square bar was more of a thought exercise as i had planned on using round (which would seem to make the whole exercise a lot simpler) i usually over complicate things or overlook the obvious so i thought this would be the place to ask… and as usual i pick a question with multiple answers all of which spawn even more questions

however reading all this has brought to my attention that my problem could well be more psychological than technological; a hammer and centre doesn't feel precise… maybe thats just lack of experience but thats why i asked the question; i had never seen it in a book… well not one with bar that didnt fit through the headstock so guessed there was a good reason for this

thanks again and i will keep reading all your replys tim

[Steven VineParticipant @stevenvine79904]

Crikey Tim, that's a bit of a let down and feels like a bit of a time waster. I was following this thread closely and giving the square bar problem a lot of thought.

Steve

[jason udallParticipant @jasonudall57142]

Lathes are arguably the oldest machine tool.. (drawings for Egyptian verticle spindle devices)
The need for say two truely round portions / concentric and idealy parallel is as old as the wheel..
To an extent the section between could be as irregular as the tree it was cut from..
.
These would be spun between centers (simple points) and a tool held so as to shape the stock..
Scroll on a millennium or four..
We have bearings and a chuck with an electric motor spinning it…
( we even have watchmakers “turns” bearing a close family resemblance to our basic lathe)…thus we chuck up the stock or face plate etc. And turn..now to support the far end we use either a center or a steady..

In any case the prime purpose remains the same..rotate stock about a centerline and remove stock..
Jim ..to turndown a square involves an interupted cut which requires much firmer grip on the stock..
It really is a VERY different job.

[jason udallParticipant @jasonudall57142]

From experience
.turning rounds on “square”..is often..less than pleasing..the stock is not quite square..or is less than parallel. .and thus the blend from square to round is uneven..
As to precise. .center punching to at least the regularity of the stock is possible. .I would use the clock makers “trick” of a mounted drill bit instead of center punch and just spin between fingers to create mark..
A 1mm pilot centerdrill is my choice here
Check against scribed lines and ” adjust” if required..

[TimSParticipant @tims]

I'm sorry that this becoming a thought exercise is a let down but it has become very obvious to me that this is outside the scope of my machining abilities

a piece made from square bar would be much better than one made from round but i know i can solve any centre mark issues with the first cuts by turning the whole surface of the bar in the lathe, making a boss at the same time so I end up with an exact piece that i know will be both parallel and concentric (assuming tailstock alignment)

if i had stuck with using square then i would need to accurately machine the 4 faces of the cast iron bar to remove the scale in the mill, hope my machining skills are good enough to ensure that i end up with an exact square, centre mark the ends accurately, drill precise centre points and do my best to get a level of accuracy i dont think i can achieve, turn the boss; and hope all is accurate… i also dont have the kit or any experience with using precise measuring equipment to ensure that all is well with a square piece… round however makes all of that a lot simpler, and for me thats quite important; i am willing to sacrifice a far more ideal shape for the guarantee of precision… and precision is why i asked this question in the first place

apologies again if i have wasted your time

[Michael GilliganParticipant @michaelgilligan61133]

Tim,

I don't think there is any real harm done

At least this thread stands as a record of some useful ponderings.

… and it even stayed "on-topic"

MichaelG.

[MICHAEL WILLIAMSParticipant @michaelwilliams41215]

Hi Tim ,

Well – you can have a bit of my thinking time for the asking – I don’t mind .

If you want to – step back a stage and tell me about the basic design problem .

Regards ,

Michael Williams .

[Les Jones 1Participant @lesjones1]

Hi Tim,
I think any discussion like this is interesting as suggestions made could be useful in other situations.
What is useful when asking questions is to give more information about what equipment you have access to.
Also particularly as you could not give an indication of the tolerance required information about the item you are making would give a clue about the accuracy required. For example it could have been a shaft to fit a pulley that had a square hole with a bearing at each end running at a few hundred rpm or is could have been a shaft for a high speed turbine running at 100000 rpm. If it was the first case then plus or minus 10 thou would be good enough. If it was the second case I doubt if any of us on this forum could make it accurately enough.

Les.

[TimSParticipant @tims]

Thanks Michael and Michael; there isnt so much of a design problem as just an alternative; the same size square bar has more mass than a round of the same diameter but obviously makes the machining more challenging… and opens me up to a whole load of extra work and potential errors

essentially i can get by with one flat side so to produce this accurately i was going to:
1. turn the whole outside of the bar and boss at one setting (or between centres)
2. bolt some sacrificial pieces to my Mill and cut V's into them in situe
3. mount the round bar in those V's (thus fixing any alignment issues with my mill table and ensuring that everything runs true)
4. Mill the top face to size and do any other work to finish it; then remove once its finished (i figured doing it all at once is a good way to ensure everything is all on the same plane and i have none of the problems of setting it all square again if i were to remove and remount)

its probably now aparant that i dont trust my abilities that much but i wanted a way that would introduce as few potential errors as possible… and the only way i thought of ensuring that is to do it at the same time without the need to clock anything

if i have missed something that will cause me problems feel free to say as this is one of the only routes i knew that could ensure accuracy without having to measure each mouting etc etc. obviously i would clock it and not trust i hadnt made any mistakes in the setup… but its more confirmation than location

[jason udallParticipant @jasonudall57142]

Edit..crossed with above post

I hope I didn’t seem grumpy.

In the absence of your actual “requirements”..it is un surprising that my or anyone elses suggestions are off target..

Lets say you want a bar with concentric and parallel “journals” on either end. Of defined diameter.
Between those journals you need flats for convenience. .
You have a mill of such and such a scope.
You have a lathe of such and such center height and between centers

.
Me I might start with square

Mark centers.drill centers ..mount between centers on mill…mill flats flat at “required” locations..mount on lathe between centers and check flats..when happy with flats..
Turn boss as required..
You now have bosses concentic With each other and referenced to the flats. .

This still leaves the issue of blend

But if you do the above with round stock and turn all over you get the flats as required and a concentric part…

Edited By jason udall on 08/03/2014 09:30:10

[MICHAEL WILLIAMSParticipant @michaelwilliams41215]

Hi Tim ,

Observations :

(1) 300 mm of 70 mm A/F square bar is a big lump for lightweight machinery . If made of any ferrous metal it will be very heavy and most of the weight will serve no purpose .
(2) Consider square tube with inserted ends . Even if only 6 mm wall it is almost as strong as the solid bar in most applications .

(3) Precision square bar and tube can be bought from WDS and others . It’s very expensive but ready to use .

(4) Use round bar with flat plate(s) bolted or saddle clamped on .

(5) Make the whole thing a fabrication of several parts .

Regards ,

Michael Williams .

[Ian S CParticipant @iansc]

I was thinking of mounting the bar between centres on the mill with a rotary table at one end, and cutting the required round section with an end mill. Would that work? there is , what ever method used, the problem of an accurate centre, in this case possibly at both ends, unless the table end is a 4 jaw chuck. That's the trouble with thinking.

Ian S C

[Author]

Posts