Engineers level

Anyone ever been sent to the stores for a new bubble or sent a poor unsespecting apprentice?

Oh the days of my youth.

bucket of steam, long weight, etc

Thanks for the clarification, John

MichaelG.

Happy to learn. Happy to be corrected. At least my builders leve did show a change with one piece paper under one end of a 600mm

Looking back in my notes I see that the curve polished inside a bubble vial of 20" quality has a radius of 60 mtrs.?

I am just quoting what was told to me as I have no way of checking. We were recently quoted a price for a bubble that was 20" and no longer available from LEICA and they wanted a £150 and a quantity of 50 to make it viable for them. We declined and scrapped the Instrument.

Clive

Hmm...

After reading this I thought it is about time I checked my 60 year old Atlas clone for level (twist) having set it up in its present position about five years ago.

I put a length of 9/16 silver steel in the three jaw and a dial gauge set to centre height on the cross slide to measure in the horizontal direction. Next to the chuck there was 1 thou variation as I turned the chuck and I set the zero in the middle of that range. I then moved the saddle to bring the dial gauge to ten inches from the chuck and again rotated. This time there was 2.5 thou variation and the middle reading was only 1/2 thou off zero! I was expecting at least a couple of thou to need adjusting out.

The above method of checking for twist has, for some reason, become known as Rollie's Dad's method but I have used it for years - it just seems to be an obvious method. Using a level with 10" sensitivity would give about the same sensitivity but as most people have a dial gauge why bother?

Agreed, the Atlas won't take heavy cuts like a Colchester but it can certainly give accurate results.

Russell.

Edited By Russell Eberhardt on 01/01/2015 10:55:18

Goodness this thread has been dragging on since last year!

a couple more links for when this has been discussed before which are worth reading.

Lathe alignment.

New Lathe leveling

Coming back to the original question - the "what" rather than the "why" - and always a difficult thing on a forum!

I bought one of these levels (the 200mm version) a few weeks ago. I was also concerned about the "bit of rust" comment but mine was clean, no rust apparent. What was a problem was that the bubble had fallen out! The thing comes in a fitted case with a bit of foam packing, and was well wrapped in bubble wrap, but had probably been dropped or thrown in transit. The bubble assembly - the vial - is plastic, and it's glued into a piece of channel section pivoted at one end and bearing on a spring-loaded adjuster screw at the other. The glue had given way and the vial was loose. I fixed it back with a bit of PVA glue (my thinking was strong enough to hold it, weak enough that I could remove it if necessary). I tweaked it on my carefully shimmed surface plate so that it read the same when turned end-for-end, and it seems to be holding its calibration fairly well. I cannot detect any rock on my surface plate (a 300x300x20 slab of plate glass). This is not Moore and Wright, Hilger, etc, quality but it works. If I put my thinnest feeler gauge under one end, the bubble moves the number of divisions that I would expect. Again, not metrology lab standards, but a reasonable sanity check.

I bought it to help align the profile rails on a CNC router that I am building. It's a welded steel structure with steel box sections about 1800mm long carrying the rails, and due to manufacturing and welding distortions, there was a dip in the middle of roughly 1.5mm. I used an epoxy bed to get close to level, bolted down the rail, then used the "precision" level (with feelers as necessary) to measure height differences every 200mm. I'm assuming that this was fairly accurate as I found a +-0.1mm variation; after inserting the calculated shim thickness and remeasuring I was within about 0.05mm which is more than good enough for this job.

So, as a level it's not the best in the world, but it's certainly not the most expensive either. You get what you pay for, it does the job, and even with the Christmas post, I ordered it one day and it arrived the next.

I notice the level Norman pointed to has a sensitivity of 3.5 thou per foot. I am no expert but that does not seem to be very sensitive to me. This whole area can be a minefield since there are a number of variables which can influence alignment. The bed should not have a twist or a sag, the headstock must point down the bed within fine limits, the headstock spindle must be parallel to the bed and so on. Get all these right then start screaming when you try to machine something and find the results are not to your liking.

In case you are wondering, I don't have an answer, I'm not sure there is one.

Doug

.

John,

That's good to know

MichaelG.

I did find this on the internet **LINK** and goes to show that a level is not final part in levelling a lathe as the guy in the clip has actually twisted the bed very slightly (1 division  to get the machine to cut parallel to within 1/10 of a thou over 10" which in my book is very good.

Hi, am I not understanding something here.

The claim in the add is 0.02 mm/1000mm which is the same as 0.0007874 inchs/39.3700784 inchs near as dam, or a ratio 1/50000.

I'm no expert at maths but 1000mm is just a bit more 3 feet and 3.5 thou is quite a few tads more than 0.02mm.

0.2/1000 is quite sensitive, I have one that claims to be 0.0002"/10" and you only need thickness of the silver paper from a sweet wrapper under the very end of it to throw the bubble off the centre completly.

With levels of this sensitivity you will need it to be in the same area as the machine for at least a couple of hours for it to stabalize to the same temperature, before you start to use it and then you'll have to be carefull not to alter it's temperature with the heat of your hands. Even the temperature of the room it is in will alter the size of the bubble.

If your workshop has a wooden floor, or any floor that deflects when you walk on it, this kind of level is a waste of time, as you might as well go and chase rainbows.

Regards Nick

Thanks Guys Some interesting points. Thinking I will ask around at work see if I can find someone with something bit better than the brick layers level I have and just give it a go. Think one problem I have is mentioned by Nicholas Farr I'm in wooden shed with a wooden floor and home made wood bench so unless I do some more work and obtain a proper stand and remove some floor so it can sit it on the concrete underneath I could be chasing my tail.

But I might keep an eye out for reasonable second hand engineers level of known manufacture without spending silly money.

The level mentioned by Ian L2 does indeed have a resolution of 0.02mm per m, but the Moore &Wright one mentioned by NJH has the resolution of 3.5 thou per foot or approximately 0.3mm per m. Quite a difference in resolution between the two when they are both intended for a similar purpose.

Doug

Hi Doug, thanks for pionting out my mistake, just a case of assumtion on my part that you were refering to the OP.

Regards Nick.

I have a feeling that with all the best intentions there are likely to be other issues like movement of the floor be it wooden or even concrete. Taking the latter, foundations and concrete does move/bend all be it not a lot but possibly enough to the put the lathe out of level, if the sub soil is clay then this is a distinct possibility. Wooden floors are going to be more prone to movement as the moisture content of the timber changes with the seasons.

I remember back in the early 70's a wing jig for the production of wings for a twin engine light aircraft moving although it was set up on a concrete slab, it would move every 12 hours as the site was on an estuary and it was the tide causing the movement and had to be reset twice a day.

Bob

When I was earning a living as a machine tool fitter we had a problem with a multi spindle lathe which at certain times when out of true and work was scrapped, the floor it stood on turned out to be cracked and when a fork lift went over it it lifted one end of the machine twisting it, a pit was dug and a float was built for the machine.

Martin P