Experimental Pendulum Clock

[Michael GilliganParticipant @michaelgilligan61133]

Posted by SillyOldDuffer on 09/12/2022 11:26:05:

.

[…] I'm going to let the clock carry on logging until the pressure rises. If the period slows down in line with it, I'll accept the evidence.

.

If you were planning to put it in a closed vessel anyway: Why not do that now

… and adjust the experimental pressure with a bicycle pump ?

Surely you only need to demonstrate ‘principle’ at this stage in the proceedings.

MichaelG.

[SillyOldDufferModerator @sillyoldduffer]

Posted by John Haine on 09/12/2022 11:48:42:

I know that CF is supposed to be sensitive to humidity but I doubt that this is particularly a problem in a clock. The polymer matrix may change its size, but with a decent weight bob the load will be taken by the fibres not the matrix and the fibres are not affected by humidity AFAIK. My first clock had a 7kg weight on a 10mm dia 1mm wall tube, the new one has 5kg on a 6mm tube. Water is taken up by the matrix, but the change in mass according to something on the NAWCC website if at most 0.3% w/w, and if you take into account the very light weight of CF compared to the bob the effect on timekeeping is minimal. However such measurements as I have done do not include humidity, the next system will have a humidity sensor as well…

Results from my Mk 1 clock showed humidity affected the timing. I think the effect is due to the matrix rather than the CF. The rod on my clock is lightly loaded because the bob is tiny, which might allow it to stretch and contract. More, I was using the rod as a spring, where I guess the matrix dominated. I suspect the springiness of the matrix varies with humidity.

I've no idea what humidity would do to a big pendulum with a heavy bob, but your assumptions seem very sensible. However, as it's so easy to log humidity, go for it!

Dave

[SillyOldDufferModerator @sillyoldduffer]

Posted by Michael Gilligan on 09/12/2022 11:53:40:

Posted by SillyOldDuffer on 09/12/2022 11:26:05:

.

[…] I'm going to let the clock carry on logging until the pressure rises. If the period slows down in line with it, I'll accept the evidence.

.

If you were planning to put it in a closed vessel anyway: Why not do that now

… and adjust the experimental pressure with a bicycle pump ?

Surely you only need to demonstrate ‘principle’ at this stage in the proceedings.

MichaelG.

Convenience rather than principle! It's just that fitting the pipe and top disturbs the clock. Doesn't help that the wiring has to be plugged into a home-made bulkhead penetrator and I usually get it wrong. Another problem it the pipe and top are held on by air-pressure (vacuum inside), and the structure can't take positive pressure.

Funnily enough I was wondering about making a second clock so that I could change mechanicals whilst the other was left ticking undisturbed.

Dave

[blowlampParticipant @blowlamp]

Do you think the elasticity of the suspension spring could be varying due to temperature or ageing in-use?

 

Martin.

Edited By blowlamp on 09/12/2022 15:57:31

[duncan webster 1Participant @duncanwebster1]

So having spent all day on grandad's taxi duty it's too cold to go in the workshop and not worth heating it for a couple of hours, so I decided to work out the effect of buoyancy due to varying air density on the period of a pendulum.

Using various equations from Wikipedia, all of which look sensible, I reckon that the period of a 1 metre long pendulum in a gravity field of 9.81 m/.sec^2 swinging in vacuum would be 2.0060666807 seconds, with hot, damp, low pressure air this increases by 0.0001291836 sec which is 64 ppm, and with cool, dry, high pressure air the vac time increases by 0.0001539039 seconds, 77 ppm.

If anyone is crazy enough to want a copy of the sums, send me your email address via pm, as converting an Smath Studio document to jpg to paste into this is too much like hard work, and would probably be illegible

[John HaineParticipant @johnhaine32865]

What Bob material did you assume Duncan?

[duncan webster 1Participant @duncanwebster1]

Brass, it's denser than steel. Not as heavy as lead, but I've been told that lead bobs can change shape over time. What you want is depleted uranium, but it's not readily available

[SillyOldDufferModerator @sillyoldduffer]

Posted by blowlamp on 09/12/2022 15:56:52:

Do you think the elasticity of the suspension spring could be varying due to temperature or ageing in-use?

Martin.

Yes, I'm sure it does, but I hope the effect is tiny. As the spring is only a couple of millimetres long, I wouldn't expect temperature to have much effect.

Ageing could well be a problem though – the spring is made from disposable razor-blade, properties unknown, but it's a Stainless Steel, some varieties of which work-harden. Also, the spring is mechanically clamped rather than firmly fixed, and it might be working loose.

Dave

[John HaineParticipant @johnhaine32865]

Posted by duncan webster on 10/12/2022 09:26:08:

Brass, it's denser than steel. Not as heavy as lead, but I've been told that lead bobs can change shape over time. What you want is depleted uranium, but it's not readily available

Osmium wound be good but at EU1850/gram a tad expensive…

[Michael GilliganParticipant @michaelgilligan61133]

Posted by John Haine on 10/12/2022 10:09:23:

Posted by duncan webster on 10/12/2022 09:26:08:

Brass, it's denser than steel. Not as heavy as lead, but I've been told that lead bobs can change shape over time. What you want is depleted uranium, but it's not readily available

Osmium wound be good but at EU1850/gram a tad expensive…

.

I think Tungsten Carbide [scrap tooling inserts, etc.] might be worth investigating … it’s heavier than Lead, and could possibly be used as a filler in a Lead casting.

MichaelG.

[duncan webster 1Participant @duncanwebster1]

You can get a machineable tungsten alloy which has SG 17,which is double that of brass, but I bet it ain't cheap. Not much use to Dave either, he needs it to be magnetic. I'll run the sums again with different densities and see what difference it makes.

[Michael GilliganParticipant @michaelgilligan61133]

Posted by duncan webster on 10/12/2022 11:39:56:

You can get a machineable tungsten alloy which has SG 17,which is double that of brass, but I bet it ain't cheap.

.

That’s why I was pondering the use of scrap carbide

… but of course, at the scale Dave is working,

‘Tungsten’ darts would be a good “make-from’ source of material.

MichaelG.

.

Edit: __ I have no connection with this Seller … but she is now on my watch-list

https://www.ebay.co.uk/itm/385277852864

Edited By Michael Gilligan on 10/12/2022 12:26:38

[duncan webster 1Participant @duncanwebster1]

I've run the calc again, following table shows ppm change for different materials and air conditions

material hot/damp/lowP dry/cold/HighP

brass 64 77

steel 69 82

tungst alloy 32 38

Anything that increases density of bob helps, I'd be careful of lead/tungsten mix because of long term physical stability, and Dave needs magnetic, so some steel needed. Nickel and cobalt are not heavier enough to be worthwhile.

[blowlampParticipant @blowlamp]

Posted by SillyOldDuffer on 10/12/2022 10:05:27:

Posted by blowlamp on 09/12/2022 15:56:52:

Do you think the elasticity of the suspension spring could be varying due to temperature or ageing in-use?

Martin.

Yes, I'm sure it does, but I hope the effect is tiny. As the spring is only a couple of millimetres long, I wouldn't expect temperature to have much effect.

Ageing could well be a problem though – the spring is made from disposable razor-blade, properties unknown, but it's a Stainless Steel, some varieties of which work-harden. Also, the spring is mechanically clamped rather than firmly fixed, and it might be working loose.

Dave

The shorter the section of spring, the 'stiffer' it will appear to the system. Any change though temperature or work hardening might well show in your measurements, I would think.

Definitely make sure the clamping is reliable.

Martin.

[Michael GilliganParticipant @michaelgilligan61133]

Posted by duncan webster on 10/12/2022 13:00:23:

[…] I'd be careful of lead/tungsten mix because of long term physical stability, and Dave needs magnetic, so some steel needed. Nickel and cobalt are not heavier enough to be worthwhile. […]

.

Fair comment, Duncan

I should have made clear that I was only thinking of the ‘filled Lead’ for an ordinary long-case pendulum, not for Dave’s little special.

MichaelG.

[John HaineParticipant @johnhaine32865]

Dave, what size is your bob?

[Michael GilliganParticipant @michaelgilligan61133]

Posted by John Haine on 10/12/2022 17:54:50:

Dave, what size is your bob?

.

I can’t be bothered scaling the picture, John … but it’s pretty small

.

**LINK**

https://www.model-engineer.co.uk/albums/member_photo.asp?a=58643&p=917157

MichaelG.

[SillyOldDufferModerator @sillyoldduffer]

Posted by John Haine on 10/12/2022 17:54:50:

Dave, what size is your bob?

Well, that's a personal question, especially as my bob is tiny! I'm not embarrassed, it's what you do with it that matters:

The bob is a 31mm length of 3/4" diameter mild-steel rod, weighing about 40g from memory. The rod is 220mm of 0.8mm diameter carbon fibre. The end of the rod is superglued into an brass suspense piece inset into the bob's base, and the through hole is 2mm diameter. The rod passes through a brass top cap drilled 0.8mm and intended to stop the rod flopping about. Not sure it's necessary.

A redesign incorporating Duncan's suggestion is in progress, replacing the brass end with Aluminium and deeper inset so the bob is supported at the centre of mass. As always suggestions and criticisms gratefully received!

Dave

[Michael GilliganParticipant @michaelgilligan61133]

Posted by duncan webster on 10/12/2022 11:39:56:

.

You can get a machineable tungsten alloy which has SG 17,which is double that of brass, but I bet it ain't cheap.

.

This gives an easily comprehensible point of reference: **LINK**

https://www.ebay.co.uk/itm/184137520904

MichaelG.

.

Edit: __ and a useful guide to machining:

https://www.edfagan.com/wp-content/uploads/2020/06/Machining_Guide_for_Tungsten_Alloys_06_11_20.pdf

Edited By Michael Gilligan on 10/12/2022 22:43:26

[duncan webster 1Participant @duncanwebster1]

I wouldn't jump ahead too far. I imagined the ally going between the end stop on the cf and the seating halfway up the bob. Actually just above half way as the hole in the bottom bit is bigger than the hole in the top bit so the cg is just over half way up. Not by much tho as the holes are small compared to the OD. Then the whole bob is lifted up. However, with it suspended as you have it from the bottom, the bob expands upwards relative to the end stop, and I'd expect that to be more than the expansion of the suspension spring. So I'm a bit unsure of what's going on. I'd start by moving the suspension point to the cg, get rid of the top brass disc ( I have a split lightweight collet) and then establish the time/temperature relationship. In an ideal world you'd do this by enclosing it in a controlled temperature environment, you might be able to extract it from all the other variables using statistics. Then fit an appropriate washer, which can be C shaped so it can just be slipped on after lifting the pendulum.

[duncan webster 1Participant @duncanwebster1]

So having slept on it, for 1 deg C rise in temperature the cf gets longer by 0.7*220e-6=154e-6 mm, and the cg of the bob goes up relative to the end stop by 13e-6*29/2=188.5e-6mm, so as long as the suspension grows by more than 34e-6 mm the clock still loses with increasing temperature. Having said all that, I'd ignore what I said above, leave it as it is until the relationship between temperature and rate is known, then calculate an ally washer to go under the existing bob. The 0.7 figure doesn't have to be out by much to change the situation. 

There will be latency, as the bob will change temperature more slowly than the suspension, which will be slower than the rod. This could make a statistical approach even more difficult. 

Edited By duncan webster on 11/12/2022 09:39:06

Edited By duncan webster on 11/12/2022 09:40:10

[SillyOldDufferModerator @sillyoldduffer]

Posted by duncan webster on 10/12/2022 23:59:11:

I wouldn't jump ahead too far. I imagined the ally going between the end stop on the cf and the seating halfway up the bob. Actually just above half way as the hole in the bottom bit is bigger than the hole in the top bit so the cg is just over half way up. Not by much tho as the holes are small compared to the OD. Then the whole bob is lifted up. However, with it suspended as you have it from the bottom, the bob expands upwards relative to the end stop, and I'd expect that to be more than the expansion of the suspension spring. So I'm a bit unsure of what's going on. I'd start by moving the suspension point to the cg, get rid of the top brass disc ( I have a split lightweight collet) and then establish the time/temperature relationship. In an ideal world you'd do this by enclosing it in a controlled temperature environment, you might be able to extract it from all the other variables using statistics. Then fit an appropriate washer, which can be C shaped so it can just be slipped on after lifting the pendulum.

Agreed! I like the collet idea.

On Friday I added the code needed to extract the correlations, with these results:

Spent yesterday in a muddle about how to apply the information. Pearson shows strongish positive correlations between period (ticks) and temperature and period and pressure. There's also a correlation between temperature and humidity, but I don't think it matters.

My first problem is how to combine two formula!

tick = 2066 * temperature + 1.315e+07 AND tick = 359.7 * pressure + 1.281e+07

My brain has seized. I've failed to define f(temp, pressure) giving tick.

Second problem is, having predicted what the correct tick should be by applying temperature and pressure, what maths is needed in the motion to apply the information. (The period was 'n', due to known temp and pressure, what should period be changed to before input to the counter.)

Yesterday morning I thought both questions were easy. Perhaps they are, but I'm baffled!

Dave

Edited By SillyOldDuffer on 11/12/2022 11:52:48

[John HaineParticipant @johnhaine32865]

Posted by Michael Gilligan on 28/11/2022 19:21:49:

Dave

You may be interested in this:

I have just opened the December issue of the Horological Journal … Starting on p507 is a 4.5 page article entitled ‘A precision Internet-Connected Clock Timer’

The authors are engineers at Chronova Engineering Ltd.

**LINK**

https://chronova-engineering.square.site

MichaelG.

Has anyone studied this? I'm rather puzzled…

[Michael GilliganParticipant @michaelgilligan61133]

Posted by John Haine on 13/12/2022 10:15:04:

Posted by Michael Gilligan on 28/11/2022 19:21:49:

Dave

You may be interested in this:

[…]

Has anyone studied this? I'm rather puzzled…

.

I only skimmed through it, John … finding it sufficiently promising to merit mentioning here. … I am hoping to read it properly this afternoon [various domestic matters having taken priority ]

MichaelG.

Edited By Michael Gilligan on 13/12/2022 10:26:33

[John HaineParticipant @johnhaine32865]

Thanks Michael, I'll be interested in your thoughts.

[Author]

Posts