Pendulum Level

[tongleh]

Here are some pics of my pendulum level I've just finished. Now some of you may laugh, but it's accurate to within .0185 mm and the results are able to be replicated. Made from just some bits of steel lying around and a piece of S/S rule it was certainly cheaper than a $160 machinists level. I don't have the height in the shed to make it more accurate, but it's close enough for me.

[BRADFORD]

What have you used for the pivot at the top?
It would need to be quite friction free.

[shedhappens]

tongleh, I can't see why anybody would laugh, it's a great job :2tsup: and a handy gadget.

Close enough to 1/2 a thou over 6 inches is pretty bloody good.

I'd make one myself except the trigawhottomy would do my head in :no:

[Stustoys]

but it's accurate to within .0185 mm
I dont understand your numbers. 0.0185mm in ??

How tall is it? I cant see the top in the pictures.


certainly cheaper than a $160 machinists level
And certainly better if the $160 level doesnt have a flat base.(like mine used to be)

Stuart

[simonl]

I had never heard of a pendulum level but it makes sense to me now you mention it. I really interested in the concept. I better start googleing! Thanks for sharing with us.

Simon

[tongleh]

The hinge point is just a 6mm rod welded to the top of the pendulum and inserted into a 6mm hole at the top of the upright. Friction is not a problem, after mounting, it continues to swing for at least a minute or two, the slightest bump and it starts swinging again. I did toy with the idea of using a bearing or a knife edge hinge, but I think it would take all day for it to stop swinging using the bearing and keeping the pendulum from twisting side to side would require more thought than I was willing to put in. There is a guide rail at the back of the mounting apparatus to ensure it is square to the bed, if not the readings will be out. It also has a quick connect/disconnect system similar to the tailstock mount, minus the cam, although the original I made had a cam lock from a bicycle wheel. This pendulum level is just a copy of the plumb bob level, with a pendulum substituted for the string line, as posted in a previous thread. The math used to work out the discrepancies is the same, coming up at .0185 mm over the length of the bed (600mm). I tested the theory on a rough one I tacked together quickly, it worked so well I made a better one.

[BobL]

What about adding a laser pointer and a small mirror to reflect the image across the shed?

Just kidding - great job - I like it! :2tsup:

[tongleh]

I'm going to try and make the pointer sharper today, even with a loupe it's difficult to read under .5 of a mm. When I checked the yesterday it was reading (what looked like) well under.25 mm, Unfortunately I don't think there is a rule available that is marked in .25mm increments or less. The obvious option is to make the upright longer and cut a hole in the shed roof? :no:

[BobL]

Unfortunately I don't think there is a rule available that is marked in .25mm increments or less.
What about attaching a vernier scale to the end of the pointer?

[tongleh]

Here's some pics of a new improved pointer. I've welded a clamp on the end to take a pin and angled the rule a little for easier reading; unfortunately, I filed the groove for the pin a little off, so it's not sitting quite square but it still gives a good reading with a loupe. It's working quite well now so I'll try to straighten up the groove and neaten it up a little, but I think that should do it. I gave some thought to a vernier, but thought why ruin a perfectly good vernier, and of course, pins are quite cheap.
.

[BobL]

I gave some thought to a vernier, but thought why ruin a perfectly good vernier, and of course, pins are quite cheap.
.

There is no need to ruin anything. It is very easy to rule up your own vernier scale just take the distance of 9 or 90 divisions of your current scale and mark a piece of steel with 10 divisions within that distance.

[Michael G]

If you can find an inch rule marked in 100th's, that will give you 0.254mm per division.

Michael

[Vernonv]

:2tsup: I like it. Very interesting concept.

[BRADFORD]

Looks to me like that would be very accurate.
Tried the plumb bob on mine, lot of mucking about, but could not get an accurate enough reading, :( think I will give your method a try.
You have made a huge improvement on the original plumb bob concept.:2tsup:

[tongleh]

Here's a mock up of a vernier measurement system I made up for the pendulum level today. The top section is a S/S rule and the bottom section is piece of ali I marked off against the vernier using an eye loupe and a sharp blade. I tried a number of times, but just couldn't get the lines spot on, Unfortunately, no matter how carefull I was, some of the lines came out just a smidgin off. I also tried marking the graduations using a method employing a compass, but that didn't come out too well either. Unless someone has any ideas on how to mark off the graduations accurately, I think I'm going to have to buy a cheap vernier and canibalise it.

[shedhappens]

Just a suggestion but you can work out the finer details to it.

tongleh if you could put a sharp tool upside down in the lathes tool post and support the work at that height, by packing with something off the bed, then you could use the graduations on cross slide to accurately scribe the lines with the tool.

john

[MuellerNick]

I don't see how you could get the claimed 0.0185 mm / m with a vernier. It's getting difficult below 0.05 mm to read. Also, both graduations have to be very close together, not to say you will have friction making the setup useless.

A way out of that misery would be a digital caliper. They work contactless. Canibalize a cheap one. Cut off about 30 mm of the beam and mount it to the pendulum. The part with the display goes to the non-moving part of your contraption.
Still you will have to precisely adjust the instrument so that the gap is minimal (a few 1/10 of an mm work) and you never have parts that rub on each other.


Nick

[tongleh]

Well, shiver me timbers, that worked. Over 8 divisions it was spot on. Tomorow I'll mark up the full length and measure that up (wife won't let me work out in the shed at night - don't know why). Rather surprised the cross slide graduations are so accurate. Good idea, thanks for that.

[Stustoys]

I'm still struggling with the math behind the 0.0185mm number. If you are only reading to the nearest 0.25mm. For the 0.25mm graduations to be "0.0185mm a meter" apart, wouldnt you need a pendulum 13m long?
Stuart

[simonl]

I'm still struggling with the math behind the 0.0185mm number. If you are only reading to the nearest 0.25mm. For the 0.25mm graduations to be "0.0185mm a meter" apart, wouldnt you need a pendulum 13m long?
Stuart

Hi Stuart,

Does your mind ever go to sleep?! :U

Simon

[Stustoys]

Don't get me wrong I like the idea, I just think we are talking about different things with the 0.0185mm number. The math in it the video talked about checking to 0.005" a foot, with graduations 1/32 apart he needed a 74inch pendulum. For the younguns out there and so we are using one system I make that 0.417mm a meter for a 0.79 graduation with a 1880mm pendulum. I dont think tongleh's pendulum is that long but if it was as he is reading to 0.25mm the graduations would be 0.139mm a meter apart.

I think :D




Does your mind ever go to sleep?! :U
Only when I need it :D. Like currently, it cant even work out resistors needed for voltage drop in a simple curcuit! I used to know this stuff, it seems to have gone missing.

Stuart

[Bryan]

If my brain were an imaginary friend - The Oatmeal (http://theoatmeal.com/comics/brain) (Crude content.)

[tongleh]

Well, that's by my calculations (and I'm not a mathematician). I may be wrong, it wouldn't be the first time. I'm enjoying fooling around with it. Anyway, end of thread....

[BobL]

Well, shiver me timbers, that worked. Over 8 divisions it was spot on. Tomorow I'll mark up the full length and measure that up (wife won't let me work out in the shed at night - don't know why). Rather surprised the cross slide graduations are so accurate. Good idea, thanks for that.

Good stuff. Can you show us the final product?

[Stustoys]

I'm not a mathematician
Nor am I, not by a long shot. As I said I am trying to get my head around it and maybe we mean different things by the numbers.


I'm enjoying fooling around with it.
As I and I'm sure others have enjoyed watching. If my comments have been taken as something against the idea or your work then I'm sorry as that was not the intent.

Stuart

[simonl]

Nor am I, not by a long shot. As I said I am trying to get my head around it and maybe we mean different things by the numbers.


As I and I'm sure others have enjoyed watching. If my comments have been taken as something against the idea or your work then I'm sorry as that was not the intent.

Stuart

I certainly didn't read that way Stuart.

Hi Tongleh, once again, thanks for sharing. Well done on the project!

Simon

[Ben Dono]

Unless someone has any ideas on how to mark off the graduations accurately, I think I'm going to have to buy a cheap vernier and canibalise it.

Maybe a stiff scribe in the mill. Don't turn it on, just use the DRO or scales to do the dividing for you. Move the scale under the scribe to scratch a mark.

An old boring bar with a modified tip should do it.

Thanks for working this one out. I was sure this old-time method had some merit.
I do like the idea of a DRO scales. But unless you have some to cannibalize, a level would probably work out cheaper.
What about magnetic dampening on the scales?

[tongleh]

Unfortunately I don't have a mill and the graduations on the lathe cross slide hand wheel started going off after 8 or so graduations so I canabalized an old vernier I had. So far I've changed the top mount to a knife edge balance hinge with adjusters for moving it back and forward, there is also a plain bearing mount at the top where the pendulum joins the knife edge allowing the pendulum to hang correctly (vertical), regardless of column angle. I've mounted the ruler section on the bottom of the pendulum and added a S/S weight to slow down the pendulum. I've also made a sliding adjuster on the base to mount the vernier scale to for zeroing it in and the base now has a quick release mechanism from a bike seat post. I've tried a magnetic dampener (by accident) to the lower section of the pendulum but that didn't work at all, adding the weight on the bottom worked reasonably well, although the slightest movement, or simply breathing on it, starts it swinging again, a heavier weight should work well here. Painting day today and then wait a few weeks for the 2mm tap to turn up so I can mount the vernier scale properly. I've modified and changed every section of this project, over and over again, making slight improvements as I've progressed, this has been a very enjoyable project so far, I'll be sorry to finish. :no:

[Ben Dono]

Haha you really grabbed this project with both both hands! Good on you! I'm glad I stumbled across this YouTube video. Your doing a lot of leg work for everyone else and it's well documented. Thanks mate! :2tsup:

I'm surprised you found the magnetic dampener not to help. I have an old powder scale that is super sensitive with a magnet on each side of the the boom. I found it excellent at bringing the scale to a faster stop point. It too has a blade as a pivot point.

Let us all know how it worked out.

Did anyone come up with a definitive answer to the math and final post hight in comparison to an engineers level? I read all the posts a few times and got lost along the way.

[tongleh]

I have posted a new thread under 'Completed (almost) Pendulum level'. Along with some pictures of the completed unit. It also answers the question on my mathematics.