|Ian L2||31/12/2014 16:12:59|
|106 forum posts|
Looking for some way to level my Halifax 10 inch lathe bed. Seen this item on ebay auction site "item No 291296211152" is it likely to be Ok or is it cheap rubbish?.
|Michael Gilligan||31/12/2014 16:38:35|
10622 forum posts
It's better than the average Builder's Level; but I think you could do much better.
Have a look at the specs for Moore & Wright, Hilger & Watts, or Cooke Troughton & Simms, before you decide.
P.S. ... Where abouts are you ?
|10 forum posts|
Check your PM inbox
2295 forum posts
No experience of that item Ian but the M & W 6 " precision level comes in at about £150. My own ( painful!) experience of life is that you get what you pay for!
It is worth looking for S/H as items like a level are usually looked after. I purchased mine HERE and it might be worth giving them a call to see if they have anything in stock.
PS I see the have a larger one in stock now.
Edited By NJH on 31/12/2014 16:49:45
|Graham Wharton||31/12/2014 16:50:30|
|147 forum posts|
The statement "Due to long time storage some sizes may have little bit rust (about 2%), please check photos." rings alarm bells for me.
|Bob Brown 1||31/12/2014 17:11:17|
961 forum posts
I think it is important to point out that having the "bed level" is the critical bit not necessarily the actual machine e.g. the swarf tray. On a ship the lathe can still be "level" even in a rough sea but hull flex may cause it to change that.
The two collar check should give a good guide as to how far out it is, this should help **LINK**
|Clive Foster||31/12/2014 17:15:57|
|1231 forum posts|
A level of that sensitivity will drive you nuts! Typically machinists precision levels have a nominal half a thou per foot sensitivity, around 10 seconds of arc, per division. The one in question claims 0.02 mm per metre sensitivity which is the same for all practical purposes bearing in mind that a level is a null indicator not a measuring instrument so the calibration is anything but exact.
Complete waste of money.
At this sort of sensitivity the bubble will twitch if you so much as breathe wrong and it will take pretty much for ever to bring the machine close enough to level to get the bubble visible anyway. To be of any use the base needs to be very accurate to at least an order of magnitude better than the level sensitivity. Usually inaccurate bases turn out to be convex so the thing rocks making it useless. Correction is possible but not for nomal mortals lacking high level scraping skills. Its implied that the thing is not properly set up so you will need to set it to level first. Not easy and a monstrously frustrating task. Yup! I've done it. Once!
In my view the best all round instrument for this sort of thing is the once common WW2 era vintage 6 inch (nominal) base gunnery clinometer featuring a 30 second of arc vial and a nice adjusting system covering a few degrees scaled to 1 minute accuracy. That is a measuring instrument and supremely easy to use. Couple of pictures, not very good I'm afraid, of mine.
Edited By Clive Foster on 31/12/2014 17:21:13
3691 forum posts
In other words having a lathe level just helps the coolant go down the pug hole. "Levelling" a lathe is something different and while using a precision level gets you some of the way easily which is why machine shops use one it isn't necessary or cost effective for a hobbyist.
|Robert Dodds||31/12/2014 18:08:50|
|241 forum posts|
|181 forum posts|
Well a mixed bag of views. The "don't bother" levelling your lathe view has always been a source of great amusement to me since I started visiting this venerable site. I repeatedly see fine examples of accurate, precision work carried out by people without formal machinist training. It always amazes and impresses me time after time I am impressed. I come from a toolmaking background, and then some years making special purpose machine tools and, wait for it......installing them.
If you want even a second hand machine, large or small, to produce consistent, accurate quality work then for goodness sake invest in a quality Engineers Level, and take time to get your installation as accurate as you possibly can. If you have a local Model Engineering Society close by join it, as they probably have a good level that you will be able to borrow. Ian, set you standards high in everything you do with your machines including the installation. Anything less should not be considered.
Other opinions are available but remember quality work needs quality installation.
|709 forum posts|
There is a need to differentiate between "level" and, yes, "level".
The lathe must be set 'level' to turn true to close tolerances, but that 'level' doesn't have to be level in the sense that a spirit level will produce 'level'. As has been stated above, a lathe on a ship can be 'level' and turn true even when the ship is at an angle of heel, or rolling, and the machine is anything but 'level' according to a spirit level!
It's more to do with lack of bed twist, and correction for it, that will enable a lathe to turn true, rather than level in the spirit level sense, as I understand it.
No doubt someone will shoot me down now, but it's New Year's Eve, so who cares! I do agree with the sentiment above though, that quality work needs quality installation; to my mind it is all part of the 'striving to work to the highest standards one can' attitude.
Hope that helps,
|Neil Wyatt||31/12/2014 20:36:06|
12600 forum posts
I've been playing telescopes and 10 seconds of arc is a long way
|Ian P||31/12/2014 20:42:35|
1832 forum posts
I am glad you allow other opinions, but opinions are just opinions so have no bearing on accuracy. What is high precision to one person might be rough bodge to someone else.
This whole lathe levelling business keeps cropping up and has been done to death from both sides. Lathes on ships comes up frequently too but to say that quality work can only be done after a lathe has been properly installed is stretching things a bit far. I don't think I have ever seen a watchmaker set up his lathe with a spirit level after taking it out of its mahogany storage box yet somehow it produces precision results! Don't get me wrong, machinery, especially in an industrial environment has to be installed in accordance with the manufacturers instructions but the requirement that home workshop engineers should spend a significant sum of money on a precision level is a bit excessive.
It would be a great annoyance to the operator if the lathe bed was tilted to the extent that the coolant did not run down it drain but it would not affect the accuracy one jot. A twisted bed would affect accuracy, but that could be corrected by test bars and test turning without adding an intermediate 'bed levelling' operation.
|Michael Gilligan||31/12/2014 20:54:01|
10622 forum posts
No shooting-down required, Chris
"Level" within itself is, indeed, what's required
It's just so much easier to do if you can set things Level [i.e. Horizontal] as a reference. ... The 'Engineer's Spirit Level' is a very efficient comparator.
|Bob Brown 1||31/12/2014 21:06:47|
961 forum posts
An engineers level on it's own is not the total story as a two collar check will give a more accurate reading of the lathes capacity to turn parallel, smaller/short bed lathes probably do not suffer as much bed twist as bigger/longer bed lathes. I will need to check my Boxford as it has been moved and will be machining a test piece.
|pgk pgk||31/12/2014 21:12:26|
|981 forum posts|
Just some thoughts from a newbie. There is an inherent accuracy in the lathe as per the manufacturers report - or perhaps an inherent innacuracy might be a better way of putting it.
Now it's not my place to criticise anyone who wants to go to the Nth in levelling their lathe but it strikes me that having the lathe level is more to do with preventing bed twist (and allowing a bed twist to straighten perhaps under its own weight) than it is about the relationship between ways and spindle and way mounted bits. It's that relationship which i see has to be the accuracy aspect.
Indeed it strikes me as possible that the cutting of test ars and realignment is more about forcing that relationship when it's inherently out. I quite like the theory of Rollie's Dads method which has a mathematical nicety about it although other sites poo-poo it.
What i would appreciate an explanation of though is how does one know when the tailstock is out of true causing a taper as opposed to a twisted bed except by cutting a test bar unsupported when its own weight has to bow it?
My new instalation was levelled with a builders level. However I did spend a lot fo time cross checking that level and I can detect a piece of paper under one end moving the bubble and the level i chose is as exact as my eyes whichever round way i place it. A bubble is a bubble.. the different fluid in an engineers level is thinner so the buble moves quicker and yes the level's underside is ground true. My builders level is new and has a machined ally V-base. While I accept it's not as good its probably as good as my patience and in any event the whole thing is going to need cross checking within 2 weeks of instal after the stresses of moving it and doubtless settlement of the cabinet stand. Sometime after that when i know what i;m doing I'll try cutting a test bar etc and with my inherent paranoia may well buy an engineers level
|Michael Gilligan||31/12/2014 21:46:18|
10622 forum posts
Sorry, pgk ... That's not quite the whole story.
Low sensitivity vials are often bent tubes; but the high sensitivity ones are straight tubes, internally ground to a large radius.
|Michael Gilligan||31/12/2014 22:20:28|
10622 forum posts
Straight question: How do you do that ?
... It would seem a good time to use an Engineer's Level; but I presume you have another way.
2545 forum posts
Interesting to note that early lathes like the Drummonds and Myfords (and my Portass) have "anvil" construction ie you can't easily twist them due to them essentially having only one foot. Not so realistic to design a large machine that way but with my Portass "S", I was never worried about having to level the thing as long as I only nipped the small tailstock bolts loosely.
For the Bantam (which like most lathes also has most of its mass at the headstock end), I try to make sure that the weight acting on each of the two feet at the tailstock end are roughly similar, so that there can be only minimal twist (torque) on the bed casting. Having resilient (rubber) mounts makes the height adjustment of the feet to achieve this easier (less critical). I don't see how I could do much better than this - if there was a slight twist in the bed without any external applied torque, I'm not sure I'd be able to detect it, nor be able to do much about it. However, as I have inherited a lot of farming genes, it's not likely to be a problem to me!
|Clive Hartland||31/12/2014 22:52:36|
2225 forum posts
OooH, Bubbles, I love bubbles. It was drummed into me from the start of my career as an Instrument tech, 'Never trust a bubble'. This I have found to be a true adage, heat from your fingers and breath will make a bubble run like mad and you are chasing rainbows trying to adjust it level.
With levelling, the fact that when the instrument with the bubble is turned end for end and the bubble stays central then that plane is level. A 20" bubble is adequate for all needs of levelling, You do not need 10" or anything like that.
Please login to post a reply.
Want the latest issue of Model Engineer or Model Engineers' Workshop? Use our magazine locator links to find your nearest stockist!
Love Model Engineering? Sign up to our emails for the latest news and special offers!
You can contact us by phone, mail or email about the magazines including becoming a contributor, submitting reader's letters or making queries about articles. You can also get in touch about this website, advertising or other general issues.
Click THIS LINK for full contact details.
For subscription issues please see THIS LINK.