Help with a Devon Sea Clock

I have come by a Devon Sea Clock and needless to say it does not run. It is a grasshopper escape mechanism with some aspects of Harrison's H1 design.

After a few hours of frustration I still cannot get my head around the setting up of the grasshopper pallets and associated springs. I realise the pallets do not follow the EW profile but simply 'peck' in and out but the mechanism is not obvious.

Thoughts appreciated ... doesn't have to be on a Dawlish postcard.

Alan

< deleted >

Edited By Michael Gilligan on 22/06/2017 21:08:05

The problem is specific to the Devon Sea Clock design and not generic grasshopper or Harrison related both of which are well documented on the net.

Sorry I spoke

Ask the supplier? Can you post some pictures and explain more about the problem?

Devon Sea Clocks who are the original kit supplier seem to have morphed from Devon to Monmouth and so far have not replied to my messages. Interestingly while the original Devon based company offered a number of kits, the new company has reduced these down to one (Congreve) with 'Sea Clock to follow'. I am interpreting this as a redesign being underway.

I think the copy in my hands has been built from a kit and is marked 98/500. It is a nice looking device complete with cover and what is more important, the wife likes the look of it and it has been sideboard approved.

The main clock movement seems to work fine and has plenty of power but the EW and pallets are not controlling it.

There are two quadrants at the rear that mesh and are rocked by and control two front mounted pendulum arms with balls on each end as per Harrison H1. These rock in anti-phase. The shaft holding these components has ball races in the front and back plates.

The pallets sit at the centre of each of the pendulums bosses and should 'flop' (for want of a better word) in and out of the EW as the pendulum arms rock. The pallets are made of what looks like SRBF and rock freely within their boss and there is a small bias rod to change their emphasis as to when they flop in the pendulum cycle. There is a pallet spring on each which clearly is meant to give a return but both these were floating free on receipt.

The combination of these components is the cause of frustration.

The springs appear too strong for their action and the routing of the springs inside the centre hub and its intended pressure point is not clear. It is also unclear whether the arms of the two pendulums should be symmetrical about the centre boss. When I set this equal the rear quadrants naturally bias to the end of their curve rather than central so have a restricted movement.

If I remove the EW the arms swing freely but not for an extended period. It maybe that the ball bearings on their shafts have not been de-greased but this is a complete strip down which I want to hold off doing for now unless it inter relates to the problem.

Ideally I need someone who has done a similar laying on of hands.

Alan

If e-mail doesn't work, telephone is less ignorable, so it'd probably be a good idea to try them.

Regards,

Richard.

Both landline and mobile numbers go to voice mail

[quote]

The pallets are made of what looks like SRBF

[/quote]

.

That would, in my opinion, be a poor choice of material.

MichaelG.

The glossy sales sheet from Devon Clocks that came with the clock claims lignum vitae but it looks like SRBF to me.

Alan, there's a US based guy I know who has made an H1 replica and published a description - I have a copy which I can email to you if you PM me your address? I know some people with practical experience with the escapement too who I could put you in touch with - where are you located? I'm told that actually the escapement is very tolerant and should not need much fine tuning, so I think you probably have a simple assembly error.

"It is also unclear whether the arms of the two pendulums should be symmetrical about the centre boss. When I set this equal the rear quadrants naturally bias to the end of their curve rather than central so have a restricted movement."

They should not be symmetrical! As there does not seem to be a balance spring (which H1 had) they are compound pendulums relying on imbalance for their restoring force, so the moment arm to the lower bob should be greater than to the upper. If I knew the dimensions I could probably calculate where the pivot point should be (if I knew what the design period is!).

"The pallets sit at the centre of each of the pendulums bosses and should 'flop' (for want of a better word) in and out of the EW as the pendulum arms rock."

They should not "flop". Before a pallet engages with the EW it is supposed to be kept in position relative to the pendulum by the spring against a stop. As the pendulum swings the pallet moves in to the EW and hooks on to a tooth, recoiling the wheel, so the other pallet "hops" out of engagement with its tooth. The pallet stays hooked on, recoiling the wheel a bit further, then the pendulum starts to swing the other way while the EW's torque imparts energy to it, before being released when the other pallet hooks on to a tooth the other side of the wheel.

The grasshopper is unlike other escapements in that there is no sliding friction between tooth and pallet. Harrison used oak in fact, and the several replicas of his RAS regulator being made now use oak too. Had SRBF been available to him, I suspect he might have used it rather than oak!

Excellent input John and very much appreciated. I am near Newbury.

I will experiment further and come back to you, Will also PM.

Alan

.

John,

Yes, I am aware of Harrison's choice of materials, but must disagree regarding SRBF.

I stand by my opinion... but will say no more in this thread.

MichaelG.

.

P.S. ... Have just seen Alan's response regarding Lignum Vitae ... which is, of course, much more appropriate.

Edited By Michael Gilligan on 23/06/2017 10:58:34

Geneically speaking, grasshopper escapement action should be that the engagement of one pallet releases the other which to use your term flops out of the escape wheel. The pressure of the wheel on each pallet in turn holds it in contact. If you start with a pallet engaged and the pendulums stationary it shout stay engaged whilst there is power on the train. If you take the power 'OFF' by rotating the escape wheel slightly backwards the pallet should spring out of engagement. The escapement obviously needs one pallet engaged as any one time otherwise the train will just run out of control so don't let go of the escape wheel when you do the above if you have power on the train.

The swing of the pendulum should bring each pallet into engagement directly after which the pendulum 'controller' will leave it behind engaged in its tooth on the escape wheel.

Probably best if you provide power on the train by finger initially untill you have got things something like near and have understood the escapement.

Take a look at

**LINK**

which is a GIF of the action, and I know it does not look like your arrangement but the basic action is common to both.

Feel free to google other examples untill you are sure you know what it is supposed to do.

I cannot tell easily from your pic how your clock works but it would seem that it uses springs rather than a gravity counterbalance to create the release (falling away) action.

Hope this helps.

regards Martin

Edited By Martin Kyte on 23/06/2017 11:01:33

Edited By Martin Kyte on 23/06/2017 11:02:16

It's a pity Michael that you don't want to say why you think SRBF is inappropriate. In my view it has the benefits of strength, dimensional stability, and actually having a little more friction. The GH operates needs the pallet and tooth to keep engaged until released by the tooth being pulled back, so the pressure between them vanishes and the pallet is released to hop back. A little friction is probably helpful to make sure they stay engaged when you want them to, given that the friction vanishes with the pressure at release. Oak or SRBF seem right, but LV, which is inherently greasy and has low friction against metal (which is why Harrison used it for his pivots) seems less satisfactory to me.

I fail to see how comments on choice of materials help the OP.

The design can be assumed to have been working at some stage so the clock needs fixing not redesigning however valid or otherwise the comments on choice of material.

and by the way, the preferred material for the RAS  regulator pallets is snakewood because it is close grained and hard.

regards Martin

Edited By Martin Kyte on 23/06/2017 12:24:46

Progress but it is remarkably fussy thing.

Following your collective contributions I realised that the spring has a double action to push the pallet arm in both directions.

Pressure #A in the drawing is obvious but I had missed it needs the Pressure #B to force the pallet to disengage.

By adjusting this pressure I have got it to engage and disengage and it to run .... for a few strokes at least. Some fine tuning needed.

"has a double action to push the pallet arm in both directions."

Or as drawn the tail of the spring acts as a stop (*B) and the end(*A) provides cusioning for engagement.

As it clearly has a counter weight this would seem the likely candidate for the disingagement force. The spring action I think should provide a cusion to the pendulum motion driven engaging action. I would suspect that the spring should not be in contact with both ends and that there should be a certain amont of free movement in the pallet. That would take *B out of the action which would make more sense to me. The escape wheel then limits the forwrd movement of the pallet and the spring at *A the kick back. When the spring moves forwards with the pendulum it gathers the pallet and move it into engagement with the escape wheel again.

I'm not sure where you started from on this clock was it fully assembled? Like are all the bits where they should be?

regards Martin

At the risk of being ignored, the pallets do not look like SRBF more like SRBP, there being a total lack of texture on the surfaces when seen under some magnification. I always thought that snakewood was not only a dark brown but was also with patches of darker areas.