Can anyone suggest a suitable material for a Foucault pendulum platform?

[lesaid]

I'm about to pick up again on a project started last summer to build a Foucault pendulum that will fit on, or at least, above a desk in an ordinary room. This was described, with photographs in https://www.physicsforums.com/threa...or-physics-project.890835/page-3#post-5845824

The pendulum (about 1.2 m in length with a 1.5 kg bob) pivots on a 1mm ruby sphere at the tip of a 'probe', that rests on a hard platform. The hardest surface I have found so far is a piece of glazed dinner plate - but it seems this is not hard enough. After some time running, the platform is pockmarked with tiny pits, which I suspect will be influencing the swing.

Does anyone have any ideas for a material that is

• very hard (harder/stronger then glazed china)
• very smooth
• readily available without being too expensive
• can be cut or broken into a piece that provides a roughly square/circular platform around 3 cm across with a very flat top surface and not more than about 1 cm thick
• Will support a weight in excess of 1.5 kg on the 1 mm ruby sphere

I have read of 'ceramic glass' but not so far found a source that was both affordable and from which I could create the necessary size and shape of platform.

I had the pendulum working well last summer, running for nearly six hours before the (unpowered) swing decayed too much to establish its direction, and showing rotation at the expected rate for my latitude. I then disassembled it to make some modifications but on reassembly, it no longer worked (swing settled into a preferred direction over three or four hours). I think the problem lies in the structure that supports the pivot, so my first task will be to design a more robust (very rigid but finely adjustable) version.

Other thoughts or ideas welcome!

 

[ChemAir]

Does anyone have any ideas for a material that is

• very hard (harder/stronger then glazed china)
• very smooth
• readily available without being too expensive
• can be cut or broken into a piece that provides a roughly square/circular platform around 3 cm across with a very flat top surface and not more than about 1 cm thick
• Will support a weight in excess of 1.5 kg on the 1 mm ruby sphere

Tungsten Carbide comes to mind as a possibility. I don't know that it is much harder than glazed china, but it is used for machine tools, so it is fairly durable. It can be bought in bars, rounds, etc., with machine tolerances. It is not terribly expensive, but not what I'd call DIY cheap, either. Someone may have a scrap odd size here or there.

 

[gmax137]

I'm about to pick up again on a project started last summer to build a Foucault pendulum

I enjoyed the previous thread from last september. Looking forward to your further adventures.

but it is used for machine tools, so it is fairly durable

carbide cutting tools are quite brittle and chip easily.

 

[lesaid]

Tungsten Carbide comes to mind as a possibility.

Thanks for the suggestion. That hadn't occurred to me. I have had a look around and not (so far) found a sensibly priced piece of an appropriate shape though - as you implied, buying it as a sheet is prohibitively expensive. So I'm looking for some tool - e.g. a knife blade - that is thin and has a flat, smooth surface. I'm sure there will be something out there, though I imagine I'll have to break it (don't suppose it will be easy to cut) into a piece of a useable size.

Its hardness is quoted as not far short of that of my ruby pivot, so I guess that's as hard as would be useful.

Though on reflection, what I think I'm really meaning here is not so much absolute hardness, as resistance to permanent deformation, which might not be quite the same thing.

Not a clue how 'hard' my piece of dinner plate is, but this is worth a try!

 

[lesaid]

I enjoyed the previous thread from last september. Looking forward to your further adventures.

Thank you! I'm looking forward to them as well! Once I have the 'basic swing' back on track, and ideally have the whole affair mounted on a desk/worktop, the next thing will be looking at the electrostatic driver. My target is still to have this contraption demonstrate a full 24 hour rotation.

carbide cutting tools are quite brittle and chip easily.

Fortunately, this doesn't matter to me - the hard platform piece is sitting on a flat brass platform so has support, and isn't subject to impacts. What I need is hardness, resistance to permanent deformation due to the localised high pressure at the pivot point and smooth flatness.

Dinner plate certainly works, at least for a while, but if the pivot is sitting in an indentation that quite possibly won't be symmetrical, I can well imagine that it will influence the swing - the pivot arrangement seems unbelievably sensitive to the tiniest asymmetry or lack of rigidity, as is the top of the pendulum. Whereas, contrary to my original expectations, the bob-end of the pendulum seems not to be so critical. I have yet to discover any effects caused by air currents, or problems due to starting the swing just by giving the pendulum a tug with my fingers.

Fascinating !

 

[gmax137]

Oh so the material you are looking to replace is the china plate; the thing upon which the ruby pivot rests? This might have what you're looking for in carbide. The small pieces are reasonable but large pieces get quite pricey.

https://www.mcmaster.com/#tungsten-alloys/=1cvaq9y

 

[ChemAir]

Oh so the material you are looking to replace is the china plate; the thing upon which the ruby pivot rests? This might have what you're looking for in carbide. The small pieces are reasonable but large pieces get quite pricey.

https://www.mcmaster.com/#tungsten-alloys/=1cvaq9y

Mcmaster-Carr is where I looked first as well. They have what could be used, but $$. After a little Amazon searching, there may be a gauge block ($60), or a https://www.amazon.com/dp/B077SNLK31/?tag=pfamazon01-20 that would fit the bill for $20 or so, but the grade of tungsten carbide used in jewelry might be suspect. I would have little confidence that any of them could tell specifically what kind of tungsten carbide was in their jewelry, but for a DIY try, it may be worth it, and it may not matter.

These guys have some pretty good property information about the different types used in tool making.

 

[lesaid]

yes - it is the platform I am looking to replace. A small piece is all I need - the price of a small thin bar isn't too bad - I only need a thin piece, as it is sitting on an already flat brass base. I am now hunting around for a UK based supplier. Most of the time when I try to order something from the USA, I run into problems, either with prohibitive postage costs, or with refusal to accept a credit card without a USA zip code in the address!

If I were to get a bar of rectangular cross section, maybe a half or one inch across, six inches long and 1/16 inch thick, would I be likely to be able to break off a piece around half to one inch long from the end of the bar? Is it possible to snap it in that way? I imagine it is next to impossible to cut without specialist tools?

I am still looking for 'ceramic glass' too, but it seems even harder to get than tungsten carbide.

I was looking at hardness categories - and I think I do need the harder, more brittle types. I haven't found any jewellery that could form a flat platform, and tungsten carbide blades tend to be only edged with carbide, not made of it. These small bars are the only feasible thing I have seen so far!

Also wondering about a jewel 'cup bearing' which might do the job, though not sure of the radius of cup that would be best with my 1 mm ruby sphere on the pivot. Just started looking at that.

 

[NTL2009]

I would think that you would only need a very small piece of carbide for the jewel point. Look at local hardware /tool stores. I know that it is pretty easy to break off a carbide tip from a saw blade. Biscuit joint blades are only about 3 inches diameter maybe mount the whole blade with a tip where you need it?

 

[Tom.G]

I am still looking for 'ceramic glass' too, but it seems even harder to get..

Try searching for 'Corelle', they make tableware, dishes and plates, from Ceramic Glass. The brand name here is Corelle Ware. If you find some, check the smoothness and flatness before buying, you may have to select a good one. I only have the small breakfast bowls which aren't very smooth and there is not any flat area on them.

Another possibility is the screen from a smart phone. I seem to recall there were replacement screens or screen protectors that used the advertising phrase "Ruby Glass".

Also wondering about a jewel 'cup bearing' which might do the job, though not sure of the radius of cup that would be best with my 1 mm ruby sphere on the pivot.

Probably won't work. The Ruby pivot point is essentially a ball rolling on a surface. As the pendulum swings it will try to climb up the side of the cup. If they are both good bearing surfaces, the pivot point will slide back down wasting energy.

Cheers,
Tom

 

[256bits]

I was looking at hardness categories

Dinner plate might be as low as only a five ( Mohr scale ) depending upon the glazing, compared to your ruby ( corundum which is a nine ).
Where did you obtain the mm ball of ruby? - they don't have a selection of a small flat piece.

Quartz is a seven, some steels can be as hard as 8.

the radius of cup

As Tom.G says not very usefull as is.
Have you checked the geometry of the ball when on a concave, flat, convex surface?
Would it have an effect upon the movement of the pendulum - something to consider.

Look at this.
https://www.thoughtco.com/what-is-gorilla-glass-607863
Apparently, the Mohr hardness is similar to your little ball tip.
You might have some already!

Since the ball and surface contact at a very small area, there is a large compressive stress induced in the materials.
Both ball and surface will deform somewhat, and if a plastic deformation occurs just under the surface, then pitting and flaking will occur.( as you have seen ).
Small area fatigue failure as the pendulum swings the ball is pretty much going over and over the same area.
The ball ended up with no failure - you didn't mention any cracking.

I would try a hard knife - even a dinner knife could provide a hard surface with little deformation ( and energy drain ), and provide a high enough yield strength so as to avoid plastic deformation and failure of the surface.

You have to realize that going up to harder and harder materials, the area of contact reduces, and the compressive stress increases.
The mm ball could itself fail. Something to think about and investigate and calculate.

 

[lesaid]

Try searching for 'Corelle', they make tableware, dishes and plates, from Ceramic Glass. The brand name here is Corelle Ware. If you find some, check the smoothness and flatness before buying, you may have to select a good one. I only have the small breakfast bowls which aren't very smooth and there is not any flat area on them.

Another possibility is the screen from a smart phone. I seem to recall there were replacement screens or screen protectors that used the advertising phrase "Ruby Glass".

Thanks for those ideas - I'll be on google this evening !

Probably won't work. The Ruby pivot point is essentially a ball rolling on a surface. As the pendulum swings it will try to climb up the side of the cup. If they are both good bearing surfaces, the pivot point will slide back down wasting energy.

That is why I went for a ball and flat platform in the first place! My thinking about a 'cup bearing' was that if the radius of the cup was substantially larger than that of the ball, there should be a sufficiently flat section in the centre in which the ball shouldn't slip, and it would tend to self-locate in that position. It should only slip around until it slides down to the central section. That would help the general pendulum alignment with the base (the dial, and where I hope to have an electrostatic driver arrangement), rather than the present setup where I position the pendulum more or less anywhere on the flat platform, and then move the base under it to get things centred. It would also give me a hard surface designed as a bearing.

However, that would mean that the ball would be rolling around on a slope and would not be precisely aligned to the horizontal point. Intuitively, I think that the ball sitting on a slight incline - especially a curving one, could introduce an asymmetry into the swing, but I'm not sure. When I think about it - the problem seems non-trivial - at least at my level of maths (undergraduate). I had a go last summer at the maths to model such an effect on the pendulum but got bogged down in algebra and calculus that I found hard to solve! I'll have another go at that in the next few days and see if I do any better! I am keen to model what I am doing and not just rely on trial and error !

 

[lesaid]

Where did you obtain the mm ball of ruby? - they don't have a selection of a small flat piece.

The ruby pivot is one of these : http://www.renishaw.com/shop/Product.aspx?Product=A-5000-7808
I will check their site, but I doubt if they would have flat pieces - they are specialising in instrumentation probes which happen to use ruby, rather than ruby products.

I have tried hardened steel, and a piece of titanium - neither of which were as good as the dinner plate (judging on how quickly and how badly they become pockmarked).

Have you checked the geometry of the ball when on a concave, flat, convex surface?
Would it have an effect upon the movement of the pendulum - something to consider.

By eye, under a small microscope, the ball tip 'looks' spherical. and for some time, with that ball tip, this pendulum worked well, many runs where it maintained a steady, straight swing for five or six hours that rotated around the dial at the expected rate until it decayed too much to measure. So in principle, I think I've demonstrated that the ball on a flat surface works. I have tried to model the effect of a slope of the platform on the swing and got bogged down in the maths - this was last summer - I'll have another go! See comments in my reply to Tom.G.

The ball ended up with no failure - you didn't mention any cracking.

Good point. I'll disassemble thing tonight and take a look with a microscope - see if any problem is visible on the ball.

Thanks for the Gorilla glass tip - I'd never heard of it. I might indeed have a cracked phone cover that could usefully be replaced - worth a try!

I would try a hard knife

I'm not convinced that they would be harder than what I have already tried, but could be worth a try. It takes some effort and time to mount a new surface though, so I am trying to pick the best candidates first - intuitively, something like the phone cover seems perhaps a better bet to begin with? Though - I have read of ceramic knives ...

The mm ball could itself fail. Something to think about and investigate and calculate.

Yes indeed. I don't know how to approach calculating the likelihood of failure though. Concern about this is why I'm only using a 1.5 kg bob rather than something heavier. I tried to guess at a weight that felt like it would probably be all right while still being as heavy as possible. Those ruby probes are too expensive for destructive testing ! With that in mind, I have been intending to keep the hardness of the platform slightly less than that of the ruby - in the hope that if something would take damage, it would be the platform rather than the ruby.

Like so many things, I guess it is a trade-off of cost/risk against precision.

I should add - I don't have evidence that pockmarks on the platform are actually causing a problem just now - and I know the pendulum can work well with the current arrangement. But increasing pockmarking must sooner or later cause erratic symmetry problems so I'm trying to minimise/avoid it now!

 

[NTL2009]

The ruby pivot is one of these : http://www.renishaw.com/shop/Product.aspx?Product=A-5000-7808
...

Interesting, and yes $49 is a bit much to destroy for a personal experiment.

You mentioned ruby, and I started thinking that many phone stylus designs use a ruby or diamond (even harder, right? - ahhh, google says diamond is 10, ruby 9 mohs). I googled, phono styli come in spherical tips, are cheaper and readily available in the UK.

Here's one: https://shop.mantra-audio.co.uk/acatalog/Teac_stylus.html

Teac LPP1000 STL103 Stereo/Mono Diamond Stylus, Sanyo ST05D Ref 391D

Total Price: £11.00 (Including VAT at 20%)

Stylus Profile: Spherical Diamond, ... diamond styli provide for a significant longer stylus life (300-400 hours) than alternative ruby/sapphire (50-100 hours) replacement stylus types.​

Is there any reason a phono stylus would not work for you?

 

[lesaid]

Is there any reason a phono stylus would not work for you?

I can think of no reason why not, if it has a spherical tip. And I imagine that a smaller radius tip will be an advantage (less distance traveled as it swings), so long as both it and the platform can take the pressure!

But - the reason I went for that ruby probe is that the ruby is already securely mounted on an extremely rigid column that terminates in a convenient screw thread, so I don't need to fabricate anything beyond mounting an appropriate threaded sleeve onto my pivot assembly. I haven't given it any thought yet, but I am wondering how I would mount a phono stylus on the end of a rod of some sort, such that it will be symmetrical, not risk falling out and getting lost, and support the pendulum assembly, which is over 1.5 kg. The arm it will already be attached to won't be designed for that kind of thing!

I have found that the slightest 'play' anywhere in the pivot assembly or upper pendulum will almost always introduce an asymmetry that causes the swing to evolve into cyclical elliptical patterns - with a cycle time typically of half an hour to a couple of hours.

Actually - on reflection - I wonder if a larger radius ruby might actually be beneficial - by causing less deformation on the platform it could lose less energy and cause less damage! And with a maximum swing offset of maybe two degrees from the vertical, it still won't travel very much (maybe maximum 20 microns each way, if my quick calculation is correct, for a 1 mm diameter ruby).

 

[NTL2009]

Can you post a drawing or picture of this pivot? I'm having trouble understanding why you can't just epoxy that phono stylus to a plate on your pivot.

Maybe you are being confused by the plastic carrier part? The 'business end' of the stylus just pulls out, and is the diamond mounted to a metal tube (glued or crimped?). I'd just cut most of the tube away, and epoxy that diamond to a plate on your pivot. It looks flat at the nd where the diamond is mounted:

Actually - on reflection - I wonder if a larger radius ruby might actually be beneficial - by causing less deformation on the platform it could lose less energy and cause less damage! And with a maximum swing offset of maybe two degrees from the vertical, it still won't travel very much (maybe maximum 20 microns each way, if my quick calculation is correct, for a 1 mm diameter ruby).

I'm curious too. I'm not a mechanical or materials engineer, I have no idea how to estimate what the contact area actually is, but it seems that 1.5 kG on the contact point of a 0.1 mm tip would be a BIG number! OK, if I pull number out of the air hat might be in the order of magnitude, I'll just guess that 1/100th of the diameter of that ruby is in contact (knowledgeable people can laugh at my guess, that's OK!), I came up with ~ 6,000,000 Kg/mm^2, which is 8.5e+9 PSI? Does that make any sense?

 

[lesaid]

Can you post a drawing or picture of this pivot? I'm having trouble understanding why you can't just epoxy that phono stylus to a plate on your pivot.

Photos, diagrams and design description in this post from a thread last summer

https://www.physicsforums.com/threa...or-physics-project.890835/page-3#post-5845824

But on this precise point - the current probe has an M2 threaded termination that screws into a matching brass hexagonal sleeve. The pivot frame is based around pieces of M4 threaded sleeve (chosen because of the size of the sleeve sides matched the brass bar that the frame is made of). So - the M2 sleeve is soldered onto the head of an M4 screw, to match the sizes. With that description, the photo and diagram might be clearer.

So yes - I guess I could epoxy that stylus to another M4 screw head and a bit more sleeve to make the sizes right - and swap it with the current arrangement. I wasn't thinking properly!

I'll just guess that 1/100th of the diameter of that ruby is in contact (knowledgeable people can laugh at my guess, that's OK!), I came up with ~ 6,000,000 Kg/mm^2, which is 8.5e+9 PSI? Does that make any sense?

I haven't had a chance to look yet, but it wouldn't surprise me if one of the hardness scales would give me the means to estimate what kind of indentation might result for a given radius of sphere and given weight. Without that, any number for pressure is going to be no more than a wild guess I think. But yes - very high! I'm more and more thinking that I'm wrong in looking for the smallest feasible radius, and should instead be looking for some 'sweet spot' at a larger radius.

 

[gmax137]

If you are still thinking about carbide, you could get one of these brazed-carbide lathe tools cheap. The cutting part (the carbide) is the grey sort of semi-circular part at the far left in the photo. The carbide is usually about 1/4 inch wide and maybe 3/8 long. This one is chipped, if you lived nearby I'd give it to you. If you know anyone into home shop machining I'm sure they have discards like this.

 

[lesaid]

If you are still thinking about carbide, you could get one of these brazed-carbide lathe tools cheap.

Thanks. I'm still thinking about carbide, along with all the other suggestions that everyone has made! I was starting to give up on cutting blades though as most of the ones I've seen online have only been carbide tipped. This seems different :) And I do know someone that might have something like that!

The most attractive idea I've seen so far is perhaps the 'gorilla glass' - but who knows how it will turn out!

Lots of ideas to take on board and try :)

 

[jim hardy]

gmax's suggestion is what came to my mind too.

A local machine shop probably has a bucket full of broken lathe bits

But i wonder - is an anti-friction bearing the best solution? Might a flexure work as well? An inch or two of music wire...

old jim

 

[lesaid]

gmax's suggestion is what came to my mind too.

But i wonder - is an anti-friction bearing the best solution? Might a flexure work as well? An inch or two of music wire...

old jim

Thanks for the ideas! I have already tried that and failed! I tried with non-stretch fishing line (which still stretched!) and fine tungsten wire. Both failed - I believe, for two reasons.

1. because the attachment at the top needs to be very precisely symmetrical. It cannot be a knot on a hook! I tried by feeding the wire though a hole drilled as closely as possible to the diameter of the wire. Unfortunately, drilling such a hole is (for me) seriously hard, and having drilled it, it is next to impossible to feed the wire through it! I then tried with the smallest hole I could get the wire through, and putting a blob of epoxy at the exit point to hold it in place. However, getting a symmetrical 'blob' so that there was nothing to make a swing slightly easier in one direction than another i.e. the epoxy surrounding the wire was not precisely level and smooth - one part being very slightly proud of other parts of the circumference! Also was difficult keeping the surface of the wire free from contamination with stray epoxy (affecting its flexibility asymmetrically) as it entered the blob. The behaviour of the swing rotated with the orientation of the pivot.

The pivot arrangements also have to be very precisely level - otherwise, there are 'uphill' and 'downhill' sides which is similar to one side being proud of the other.

2. I also found, as soon as I moved to a rigid pendulum on a ball pivot point - that the swing would last for much longer before decaying too much. With the current setup, it will swing for just shy of six hours. I never had more than half that with a flexible pendulum - I'm guessing because of energy lost in flexing the support. The amplitude of the swing, even at the moment of starting, is at most, three or four centimetres (over a length of 1.2 metres), and the swing is 'useable' down to under 1 cm - sometimes as little as 0.5 cm.

Another benefit of a rolling ball bearing is that, if I can use a spherical cup bearing, precise levelling of the pivot point should be unnecessary - the ball should find the lowest point and roll around that without slipping. Even with a flat platform for the rolling ball, it seems that precise levelling is not too important - though it is level at the precision that an ordinary spirit level can determine. I think intuitively that this is reasonable, since the pendulum does not in practice rotate on the bearing over time. If the platform is not precisely level, I think it will rotate slightly in opposite directions on each half-swing, but I don't think this has been a problem so far. I have tried to model the effect of a slightly sloping platform with some maths, but without success so far - that problem seems beyond my current maths skills!

My visible progress is slow just now - I also need to redesign the arrangements to support the pivot as a whole so as to be more precisely adjustable and predictably rigid. As it is now - if I disassemble and reassemble it - it is hit-or-miss whether the pendulum will work afterwards. I have to sort that too before spending a lot of time evaluating different pivot materials. Existing setup is made of a mix of wood and brass - with screw adjustments in brass sleeves to adjust levels. I am wondering if a cup bearing will remove the need for such adjustment at all and let me use a rigid, fixed support made of angle brackets, or soldered brass bar or some such. That would be a lot simpler!

 

[jim hardy]

Thanks - you mentioned some things i'd not thought of.
I of course am familiar only with the flexures for clock pendulums(pendula?) which swing only in one plane .

RC modelers use a lot of clever mechanical parts. Might something like this help you level it up ?

https://www.towerhobbies.com/cgi-bin/wti0001p?&I=LXK091&P=M#technotes

https://www.towerhobbies.com/cgi-bin/wti0001p?&I=LXD912&P=M

https://www.towerhobbies.com/cgi-bin/wti0001p?&I=LXGWE1&P=7

 

[jim hardy]

here's the smallest self aligning bearing i could find.

http://www.vxb.com/108-Self-Aligning-Bearing-8x22x7-Miniature-p/kit7041.htm

Interesting project , keep us posted ?

old jim

 

[lesaid]

RC modelers use a lot of clever mechanical parts. Might something like this help you level it up ?

Thanks for those ideas. I'm not sure those however are quite what I'm looking for. The pivot support arrangement I have just now (which works at least sometimes, but I suspect, is not robust or rigid enough) consists of two pieces of reasonably stout pieces of wood bolted together into a 'T' shape - the crossbar about two feet long, the long arm, about three feet. The long end is mounted on a crude pivot bolted into a solid wall near the ceiling. The cross piece is matched with another two-foot long wooden bar mounted rigidly to a solid ceiling. At each end of the cross piece(s) is a screw height adjuster (threaded brass sleeves sunk into to the fixed piece, with non-threaded sleeves sunk into the 'T' piece - and brass bolts that can be tightened or loosened to provide a fine adjustment of both the height and level of the 'T' crossbar.

The pivot platform is mounted roughly half way along the long arm of the 'T' - so by adjusting the two bolts, I have a very fine control of level both lengthways and across the support. Maximum range of adjustment moves a spirit level bubble over three or four millimetres, spanning what the spirit level says is level.

This has been demonstrated to work reliably for a few weeks - then following dismantling and reassembly for a minor change, it has not worked since! I think the use of wood and the crude mounting arrangements are too affected by chance factors, and are not reliably rigid enough. I have found that micron-level lateral movements of the pivot platform are enough to upset the behaviour of the pendulum, as evidenced by a short piece of brass bar that supports the pivot platform itself not being rigid enough, and needing a second bar soldered along it to give it a 'T' cross section before things worked. I measured the spring constant of the brass bar (with the aid of a USB microscope to measure the (sideways) flexing of the bar under various weights, and calculating the tiny lateral forces on the pivot plate from the swinging pendulum. There's a bit more on this in the discussion in this forum referenced in the first post of this thread, if you are interested (along with photographs of the setup).

So - I don't quite see how I could use the RC devices for this level of fine adjustment, where I want to get the level exactly right and then lock it into place - and support a weight of 3-4 lb (combination of bob and the brass support assembly at the top of the pendulum). I'm currently thinking of replicating the existing design, but using rigid metal bar, or angle supports or some such instead of wood.BUT - I am still not sure that all this effort to get things precisely level is actually necessary. The pivot is a rolling ball on a flat surface. Given that the forces in practice do not cause the pendulum to rotate about its own axis (to my surprise!), and assuming that the ball does not slip as it rolls (it seems not to move over a full six hours of swinging so long as the amplitude of the swing is kept down to a few cm), then it is hard to see how a slope would do other than introduce a constant downhill torque. A constant torque that is independent of the angle or direction of swing, intuitively, should not influence the behaviour of the swing, other than to introduce a tiny, constant offset into the 'zero position' of the pendulum bob?

And, in spite of all my measurements and the sensitivity of the pendulum to tiny movements of the pivot, I have yet to see any convincing evidence that the precise level of the platform is particularly critical even though it feels like it should be! But the platform's rigidity is super-critical!

If my guess is correct, then all I need to do is find a way of fixing a rigid platform that is reasonably, but not precisely level, and make it very solid. But if I'm wrong, I'll spend a lot of time chasing my tail to find out!

I also wonder about the possibility of a cup bearing - it would help with self-centering - but the ball would then be rolling around on a slightly curved surface - which makes me wonder how that would then affect the direction and rotation of the pendulum (about its own axis). Suddenly this feels akin to the 'parallel transport' that I have encountered in my beginning studies of relativity and Riemannian geometry - in which case I'll quickly get out of my depth!

So before I decide where to invest my effort, I'm going to have another go at modelling this mathematically. Suggestions and ideas welcome!

 

[lesaid]

here's the smallest self aligning bearing i could find.

http://www.vxb.com/108-Self-Aligning-Bearing-8x22x7-Miniature-p/kit7041.htm

Interesting project , keep us posted ?

old jim

Will keep you posted - keep ideas coming !

I'm not sure that that kind of bearing will work well in comparison to the 'rolling ruby ball' - because it is hard to believe that any ball race, no matter how well lubricated and slick, will not lose noticeably more energy than a single ruby ball that is rolling on a hard surface without slipping. I also suspect that these things are designed primarily with low friction for high speed spinning loads, rather than for the 'alignment' movement.

The other aspect of 'universal bearings' that I am wary of is any difference in angular momentum, effective pendulum length or friction of the setup in different directions - the kind of thing that causes those nice patterns in sand pendula! But as far as I can see from the design spec, that might not be a problem here.

I am amazed by the quoted static load rating of 560N - can such a small bearing really support up to 56 kg load without problems? An order of magnitude more than I need !

It would be interesting to find out how they compare - if the this kind of approach is in fact good enough, it would remove all issues around levelling the pivot and be a lot simpler (and cheaper than ruby pivots, if anyone else wants to build one!). So I think I'll get something like this and try a few experiments!

It'll be two or three weeks by the time I've got something and have it set it up, but I'll let you know. Thank you for the idea.

 

[Merlin3189]

Probably a bit late in the day, I'll suggest zirconium dioxide ceramic as sold in the form of kitchen knives. I don't think as hard as WC, but probably harder than your glaze or steel. These knives are available quite cheaply now with blades up to about 4 cm wide..

 

[jim hardy]

Ahhh i FINALLY found your photos in original thread. What an impressive piece of work !

Finally i understand your top support bearing.. Took me a while to grasp why the frame doesn't have to rotate , but that's my plodding nature.

Do i understand correctly that your ruby tip actually rolls a very small distance across the ceramic plate as opposed to sliding ?
And it's pock-marking the plate?
I have some carbide lathe tool tips , triangles about 1cm on a side but they've a hole dead center for mounting.
i just tried and it definitely scores glass(mason jar) and ceramic(old white porcelain mug)
PM me an address and i'll drop one in the mail. Perhaps you could turn it on edge.

They're available with no hole .try a search on "brake lathe bits"
http://www.gwrauto.com/phoenix/brake_lathe_bits2.htm
a local brake repair shop might have some worn out ones cheap.

Fascinating project,
and i apologize for earlier silly proposal of model car parts , i hadn't quite yet understood your setup.

old jim

 

[NTL2009]

What about a floating bearing? My wife has a set of large stainless steel mixing bowls sized to nest in one another. I'm picturing a large one set on the frame, with some light oil in it. Float a smaller bowl in the oil, and attach the pendulum to that smaller bowl. It would be self leveling, and I would think self-centering due to surface tension.

Your pendulum is 1.5 kg, so the bowls would need to be large enough to displace that weight, so ~ 1.5 Liters of water, a bit more volume with an oil that is lighter than water. Large, but maybe not unmanageable?

I recall reading that some old lighthouses floated a heavy lamp on a fluid mercury bearing, the density of mercury allowed it to float some large objects. Though mercury isn't so dangerous in that form, you probably still don't want to use mercury around students.

 

[jim hardy]

What about a floating bearing?

What a great idea.

I remember an instrument we had with a floating bearing supported by compessed air.

http://www.pi-usa.us/blog/using-spherical-air-bearings-in-satellite-testing/

overkill , but food for thought...

 

[NTL2009]

What a great idea.

I remember an instrument we had with a floating bearing supported by compessed air.

http://www.pi-usa.us/blog/using-spherical-air-bearings-in-satellite-testing/

overkill , but food for thought...

The air bearing is interesting too, but I didn't see specifics on the amount of air required (a generic < 1 SCFM), which I think is decent sized pump to run for many hours? A water pump might work too, but of course water could get messy unless you had a clever design to catch and drain the water and return it to the pump. But a small submersible pump might provide enough power to keep a liquid boundary layer between the sphere and a close fitting cup.

Our machine techs at work had large metal mats with tiny air holes distributed on the bottom for moving/placing heavy machinery. They'd hook it up to standard compressed air in the factory (~ 45 ~ 60 PSI, IIRC), slide the mat under the machine's legs (jacked up for access), and could gently push around a heavy machine (kind of a reverse air hockey table, or a hover craft). And yes, it was tough to get them to not play a "Magic Carper Ride" game on them, but the safety people weren't too keen on that!

 

[lesaid]

The air bearing is interesting too, but I didn't see specifics on the amount of air required (a generic < 1 SCFM), which I think is decent sized pump to run for many hours? A water pump might work too, but of course water could get messy unless you had a clever design to catch and drain the water and return it to the pump. But a small submersible pump might provide enough power to keep a liquid boundary layer between the sphere and a close fitting cup.!

I like that idea though I don't think it is top of my list for the pendulum! A practical problem with pumped system of air or water, apart from complexity - could be whether the flow of fluid can be sufficiently symmetrical as not to impart any torque or unbalanced forces to the system, even if they are tiny. This setup is exquisitely sensitive to any asymmetry!

Complexity and risk of unwanted asymmetries is also why I'm not at present contemplating a magnetic bearing. I am trying to avoid anything that involves introducing energy from any source for the first version, though I hope to introduce a driver (probably electrostatic and without electronic control if possible) once the passive pendulum is reliable and consistent.

Our machine techs at work had large metal mats with tiny air holes distributed on the bottom for moving/placing heavy machinery.

Yes indeed! It was quite a long time ago now, but I remember a guided tour of a nearby industrial site, one unit of which was being used to refurbish railway coaches (from the London Underground if I remember rightly). Those were on platforms that were slid around a prepared (very smooth) floor of the unit to position them as required, on air bearings. We didn't see them in action - but the guide said that, with patience and great care, two people were enough to push the coach around and control it - though I don't believe that would have been routine. But the concept of a couple of people moving a railway coach around in that way was mind-blowing!

 

[lesaid]

What about a floating bearing? My wife has a set of large stainless steel mixing bowls sized to nest in one another. I'm picturing a large one set on the frame, with some light oil in it. Float a smaller bowl in the oil, and attach the pendulum to that smaller bowl. It would be self leveling, and I would think self-centering due to surface tension.

Your pendulum is 1.5 kg, so the bowls would need to be large enough to displace that weight, so ~ 1.5 Liters of water, a bit more volume with an oil that is lighter than water. Large, but maybe not unmanageable?

I recall reading that some old lighthouses floated a heavy lamp on a fluid mercury bearing, the density of mercury allowed it to float some large objects. Though mercury isn't so dangerous in that form, you probably still don't want to use mercury around students.

Interesting idea. Not sure how practical for this purpose though. Apart from anything else, I'd avoid having water anywhere near, as I hope to introduce an electrostatic driver eventually, probably running at something approaching 1 kV (very high impedance!) but I don't want water around! Oil also has the potential for an incredible mess !

I think such a bearing would be in principle similar to the ruby bearing - just that instead of resting on a ruby, the pivot would float. I would presumably still need a cage similar to that which holds the ruby pivot just now, only large enough to accommodate a bearing with something approaching a two-litre air volume on top of the volume of the oil? The floating bowl would, I think, have to be a pretty precise hemisphere (of constant radius), so that as it swings, there isn't a changing displacement profile (if that is the right term) to cause movement of the oil, and corresponding loss of energy. If the gap between the bowls is low enough for surface tension to maintain centering, does that imply a higher energy loss due to coupling between the bowls?

The ruby - or the carbide lathe tip idea, sound much more appealing at the moment ! But this is interesting to think about!

I was wondering about a permanent magnet levitation bearing if such is possible. I know one cannot simply levitate a magnet without active control, but I wonder if that principle applies to suspending something whose balance is maintained by a mass dropping below the magnets? Such as a disk magnet attached to the top of the pendulum, levitating above an appropriate arrangement of ring magnets, with the (non-magnetic) pendulum extending through the ring to the bob well below. Such a levitated magnet could not tip over, so intuitively, that just might work?

 

[lesaid]

Ahhh i FINALLY found your photos in original thread. What an impressive piece of work !

Thanks for your remarks! You might not think that if you were here and saw it close-up - it is my first attempt at this kind of construction and is very much a learning experience! But fun! And will be nice if I can prove wrong all those people who assert that such a small scale Foucault pendulum cannot be made to work!

Do i understand correctly that your ruby tip actually rolls a very small distance across the ceramic plate as opposed to sliding ?
And it's pock-marking the plate?
I have some carbide lathe tool tips , triangles about 1cm on a side but they've a hole dead center for mounting.
i just tried and it definitely scores glass(mason jar) and ceramic(old white porcelain mug)
old jim

Yes - it rolls - by my calculation, something like 30 microns each way as the bob swings with an amplitude of up to 3 cm.
It is pock marking the plate, though not easily visible without a magnifying glass. This is also after quite a few days of swinging. I think it is good that the ruby is harder than the plate - so if anything takes damage, it isn't the ruby. But I'd rather not have pockmarks either, as they imply that the roll might be becoming a slip, inside a depression. My work around is to move the pivot to a slightly different place on the platform each time I restart it.

 

[jim hardy]

Thanks for your kind words. Only one of these i ever saw was maybe twenty feet tall... in a museum somewhere

I think the air bearing would be impractically expensive unless one showed up surplus someplace. Compressed air should the easy part there's lots of home oxygen machines in a local junkyard each with a nice little compressor and filters inside.

Fascinating project.

old jim

 

[lesaid]

Probably a bit late in the day, I'll suggest zirconium dioxide ceramic as sold in the form of kitchen knives. I don't think as hard as WC, but probably harder than your glaze or steel. These knives are available quite cheaply now with blades up to about 4 cm wide..

Good idea! I had looked already at various blades only to find that the hard material was just along the cutting edge rather than the whole blade (e.g. carbide edged) - but I never thought of ceramic kitchen knives. Could be worth an experiment! Thank you.