John Wilding 8 day Weight Driven Wall Clock

[Russell EberhardtParticipant @russelleberhardt48058]

Thanks for that explanation Dick. Ambient temperature (indoors) here is 20 C in winter and up to 35 C in summer so like Alan I will continue to keep my Loctite adhesives in the fridge.

Russell

[Alan CharlestonParticipant @alancharleston78882]

Hi,

I had trouble finding anyone who stocked Loctite 638 locally so I got Bondloc B638 which I was assured was the equivalent, and as it was made in England (i.e. not China) I bought a bottle.

I've done a test by gluing a piece of 3/8" brass rod reamed out to 3/32" onto a piece of 3/32" silver steel. I degreased both parts using isopropyl alcohol.

The bottle said it took 1 hour to fully cure so after 2 hours I tried twisting the brass off the shaft. It wasn't all that difficult and I got a bit worried as to whether the joint was strong enough so I downloaded the Technical Data Sheet from Bondloc. It gives the full cure time as 24 hours. It also says that the surface keying action is greatly increased by a rough surface finish giving higher shear strength. They recommend a roughness of 30 – 100 micons whatever that means.

Does anyone else roughen the surfaces, and if so, what is a good grade of emery paper to use?

I've redone the trial and given the silver steel a bit of a rub with 600# emery paper and I'll give it 24 hours before I try twisting it off again.

Regards,

Alan C.

[Stephen BensonParticipant @stephenbenson75261]

I have been using Loctite adhesives most of my working life and it is wonderful stuff I have never had any luck with cheaper alternatives, also if you are asked to roughen up the surface then that would have me worried I do sometimes grind or file a small flat on the diameter to allow the air to escape

[Neil WyattModerator @neilwyatt]

Why not use Loctite 640 – it has an open time of an hour. Or 641 (20 mins, medium strength to allow dis-assembly)?

Neil

[Russell EberhardtParticipant @russelleberhardt48058]

Alan,

The Loctite data sheet for 638 gives about 90% strength after 24 hours at room temperature and it takes about a week to reach full strength. So, if Bondloc is similar, two hours isn't enough. Roughening the surface will increase the surface area so I would expect it to accelerate curing as well as increasing strength. Perhaps one of our chemists could comment? If the arbour is polished I rub it with 240 wet and dry to remove the glaze.

Neil

I would always use high strength retainer considering the damage that can occur if it gives way in a clock. I have seen the result of that in a spring driven clock!

Russell

[Michael GilliganParticipant @michaelgilligan61133]

Alan,

Like others here; I have never used Bondloc, so cannot comment on its performance. … I have, however, used Loctite and Permabond products very successfully on both large and small cylindrical fits.

The standard test method for anaerobic retainers is a pin and sleeve 'compressive shear' … Please see my post date-stamped 27/02/2017 on this recent thread: **LINK**

http://www.model-engineer.co.uk/forums/postings.asp?th=125269

The ISO standard is pricey, but the US equivalent is freely downloadable from the link that I provided.

…. I recommend reading it.

The test pin is much bigger than a typical domestic clock arbor, but the behaviour of the retaining compound can [at least to a first approximation] be scaled linearly. Plug-in your numbers and you can work out roughly what performance to expect. … If your test results don't compare satisfactorily then there is something wrong with either the product or your assembly method.

I could 'ramble on' but if you search for Loctite on this forum, you will probably find everything you need to know.

The principle is very sound: So find a product and an assembly technique that works for you, and then standardise on that.

MichaelG.

[SillyOldDufferModerator @sillyoldduffer]

Posted by Michael Gilligan on 21/03/2017 12:45:04:

… there is something wrong with either the product or your assembly method.

…

MichaelG.

True confessions time. This is how I mess up:

• With adhesive that's past its use-by-date.
• By optimistically using in-date adhesive that I opened last month.
• Storing the adhesive on a shelf that catches the sun.
• Using adhesives unsuited to the materials I want to join.
• Not cleaning the joint carefully enough. It has to be REALLY, REALLY clean and I can't be arsed with that.
• Never rehearse or use a jig. Instead, on assembly I waggle the joint about until it looks good. (Moving setting adhesives is an excellent way of damaging the bond. Even the slow setting types.)
• Only wimps read the instructions!
• Understanding the instructions? Maybe, but then I'm always in a hurry…

Cheers,

Dave

 

Edited By SillyOldDuffer on 21/03/2017 13:23:43

[Michael GilliganParticipant @michaelgilligan61133]

[Neil WyattModerator @neilwyatt]

Posted by Russell Eberhardt on 21/03/2017 09:37:56:

Neil

I would always use high strength retainer considering the damage that can occur if it gives way in a clock. I have seen the result of that in a spring driven clock!

Russell

The 640 slow cure I suggested first is a 'full strength' type with twice the shear strength of 641.

620 high temperature is probably the best choice for general model engineering, 1 hour fixture time + high strength, but the extra heat tolerance will make dis-assembly more difficult.

It is well worth downloading the "Loctite Retaining Compound Design Guide" especially as it details some relatively new compounds ideal for the likes of us.

Neil

[richardandtracyParticipant @richardandtracy]

Green Scotchbrite pads roughen the surface of aluminium & many plastics (so probably brass too) to almost the perfect roughness for most adhesives, including araldite, general epoxy, Loctite etc.. The 'adhesive ready' roughness is where more rubbing doesn't change the general look of the surface.

This I learnt playing with aircraft interiors for a living and maximising structural strength was considered to be of some slight interest… In the stress office we did tests on 12 samples each for 9 different easy workshop preparation methods. The green scotchbrite prepared samples were the strongest when doing lap shear tests between aluminium sheet using Redux 420 adhesive. The strength was just over 1900 psi B value (95% confidence that 90% of samples exceeded the strength). The weakest preparation method led to shear strengths down in the 600psi region. We also tried some samples using different adhesives and found a similar trend in strength, but did not do enough samples to be statistically useful.

Regards,

Richard.

[Alan CharlestonParticipant @alancharleston78882]

Hi,

Thanks for all your comments. Both the silver steel arbor and the reamed hole in the brass were shiny which could have been part of the problem as well as not leaving it to cure for long enough. As per the comments, I'll try roughening the surface of the silver steel with 240# wet and dry emery and try getting a bit of scotchbright through the hole in the brass.

The first joint I'll make will be the arbor going through the barrel which holds the weight string. I'll test the integrity of the joint by hanging the weight 1/2" off the ground and without connecting the rest of the train. I'll leave it for a couple of days. Even if it fails, I don't see it causing any damage.

Thanks for the help,

Alan C.

[Alan CharlestonParticipant @alancharleston78882]

Hi all,

I repeated the test I did using a reamed brass collet onto a 3/32" silver steel shaft with both the surfaces of the shaft and the collet roughened with 240# wet and dry emery paper. The emery paper was rubbed along the axes. It made a big difference to the strength of the joint.

I did the same to the arbor through the barrel and after waiting 24 hours I hung the weight off it. No problems. If it's still OK tomorrow, I'll continue with the rest of the train.

Thanks for the help.

Alan C.

[Dick HParticipant @dickh]

Hi all,

I am finally trying to get my version of this clock going. I followed the later (updated) book which puts the barrel off-centre and uses a guide roller to help the weight cord across the barrel. If I use a straight drop with the weight, it runs with a 5.lb weight, if I use the guide roller I need a bigger weight, if I put the weight pulley in nothing goes. Either I´ve no idea how to string this thing up or I´ve got some massive losses somewhere. How far do the pallets enter the escape wheel, is the geometry of the anchor wrong or have I friction losses?

It runs but should have more ummph.

Any suggestions,

Any help appreciated,

Dick H.

[Alan CharlestonParticipant @alancharleston78882]

Hi Dick,

I've got exactly the same problem. Without the pendulum in place the crutch goes backwards and forwards but there is not enough "ummph" to maintain the pendulum after it is set going.

I was hoping that I had got to the stage where either it would work OK and I could proceed with the bits on the front of the clock or it wouldn't work at all which would give me an excuse to give up on this project from hell. Sadly it's half working so I suppose I'll keep on going to see if I can get it going properly.

I'm going to disassemble it and open up the pivot holes, polish the wheel bearing surfaces and the pinions. I've gone for the lantern pinions which may have been a mistake but I'm not going to machine solid ones.

One thing I found made a difference was to put a bit of clock oil into the pivot holes – this sped up the crutch oscillations a lot.

If all else fails, I'll try your idea of hanging the weight directly from the main wheel spool which will double the force to the train – I don't mind winding it every 4 days rather than every 8.

I wonder what the ratio of finished clocks to books sold is.

Regards,

Alan C.

[Russell EberhardtParticipant @russelleberhardt48058]

Posted by Alan Charleston on 21/04/2017 07:17:55:

I've got exactly the same problem. Without the pendulum in place the crutch goes backwards and forwards but there is not enough "ummph" to maintain the pendulum after it is set going.

The greatest frictional loss in a well made pendulum clock is the air resistance on the pendulum bob so that is exactly what I would expect. Indeed I found the same when building my J.W. regulator.

"One thing I found made a difference was to put a bit of clock oil into the pivot holes – this sped up the crutch oscillations a lot."

Am I to take it that you didn't lubricate it before testing?

Each pivot hole needs a small drop of clock oil. The winding drum pivots could use something a bit heavier. You also need to apply a very small drop of clock oil to the pallet faces as there is a rubbing action there.

Russell.

[Stephen BensonParticipant @stephenbenson75261]

Sounds like you like me made the classic engineers mistake making everything too well H7 fits do not work on a clock the old clock adage is "if it rattles it will run" applies also lantern pinions are superior to cut pinions on all but the smallest of clocks

[Alan CharlestonParticipant @alancharleston78882]

Hi Russell,

Thanks for your comments. I didn't make the pendulum bob as laid out in the book which is a quite streamlined design. Instead I used a piece of 40mm dia. steel 100mm long which has the same weight as I calculated the lead filled bob described in the book to have. The air resistance of this is going to be more than the bob described in the book which is only 25mm at its thickest point and 90mm in diameter. Although at this stage that's not what my problem is (there is barely enough ummph to keep the crutch alone oscillating) I'll bear it in mind if I make some progress in increasing the force from the escape wheel.

Yes, I initially ran it without oil as everything seemed to be rotating freely and I could drive the train by blowing on the escape wheel. Without oil I could only get a few beats out of the crutch but after I lubricated the pivots it more or less ran continuously.

You say I could try a heavier oil for the winding drum pivots. I have some clock spring oil. Would this be a better bet?

From what you say about lubricating the pallet faces, (which I didn't do) I presume one of the jobs to be done when I get it apart is to polish the contact surfaces of the escape wheel which still have the machined surface.

Hi Steven,

All of the small pivot holes are reasonably loose but the large diameter winding wheel pivots are a reasonably close fit. I'm not sure of the best way to ease these out as one end is an odd size which I needed to make a special reamer for.

The other area I'm not sure about is the adjustable front pallet bush. If this is rotated much, it binds onto the escape wheel pivot because the holes in the front pallet bush and the back cock aren't in line. This is made worse because of the thickness of the front bush. I've tried to get around this by drilling the hole out except for the back 1/8" but the amount of adjustment achievable before the pivots start to bind still isn't much.

As far as the lantern versus cut pinions go, one of the problems with lantern pinions is that you can't look at them end on to see how they engage with the wheels when you are depthing them. If I have the depthing wrong, I'm not inclined to plug the holes in the frame and start again. At some stage enough will be enough and I'll move on to a more rewarding project.

Thanks for the comments,

Alan C.

[Stephen BensonParticipant @stephenbenson75261]

I always test a clock before lubrication if all the pivots, pallet faces and pivot holes are well polished the clock should run very well indeed without lubrication, if it does not then something is incorrect or it needs polishing some more,

Friction at top of the train will restrict the power more than earlier in the train and the pallets are right at the top of the train these should be mirror polished best if you harden them first however for testing a couple of minutes with a Dremel type polishing tool with work wonders.

Lantern pinions are much more tolerant of depthing errors 

Edited By Stephen Benson on 21/04/2017 09:38:13

[Alan CharlestonParticipant @alancharleston78882]

Thanks Stephen. I "polished" the ends of the pivots before and after I hardened them using 1000 grit wet and dry paper. When I've got it apart I'll give them a polish using a felt pad on the Dremel and Brasso.

Regards,

Alan C.

[Russell EberhardtParticipant @russelleberhardt48058]

Posted by Alan Charleston on 22/04/2017 07:23:51:

Thanks Stephen. I "polished" the ends of the pivots before and after I hardened them using 1000 grit wet and dry paper. When I've got it apart I'll give them a polish using a felt pad on the Dremel and Brasso.

Regards,

Alan C.

I've always gone through the grades to 2500 wet and dry with oil and then progressed to Micromesh down to 8000. The latter is expensive but long lasting.

The oil I use is M&P clock oil. They do two grades, one for pivots up to 6 mm diameter and one for pivots greater than 6 mm. Some people even use a fine grease for the drum pivots.

The eccentric bush for the pallets shouldn't need much adjustment but of course the pivot must be free where ever it is set. More clearance or shorter bearing surface will do that if needed. I prefer an eccentric at both ends .

The tips of the escape wheel should have a good finish but it shouldn't be necessary to polish them.

Russell

[Alan CharlestonParticipant @alancharleston78882]

Thanks Russell.

I agree that an eccentric at both ends of the pallets would be a good idea. As it is, the amount of adjustment is pretty small before the pivots start to bind.

Regards,

Alan C.

[Dick HParticipant @dickh]

Thanks to everyone for the suggestions and comments.

I think the problem is with my bit tight depthing of the hour/reverse hour wheel in the motion work. Anyway it runs, sort of now, on the smaller weight without the pulley. I took everything apart, polished it and put back together.

Thanks for the input.

Dick.

[roy entwistleParticipant @royentwistle24699]

Dick. When depthing , concentrate on the feel rather than the look, and remember that the wheel drives the pinion

Roy

[Stephen BensonParticipant @stephenbenson75261]

I know this a bit daft but this was the first clock I had made, I failed to understand and it was not made clear in the book that bowed tension spring must not touch the front plate it should act on the step on the centre arbor

[Russell EberhardtParticipant @russelleberhardt48058]

Posted by roy entwistle on 24/04/2017 23:05:44:

remember that the wheel drives the pinion

Although it's the other way round on the motion work!

Russell

[Author]

Posts