Case Design

[colinh]

As I mentioned in my intro, I've been a hobby goldsmith for quite some time, and something I've always wanted to do was make my own case. (I have a lathe and milling machine).

 

This isn't exactly watch repair related - but there are lots of knowledgable people here so... :-)

 

 

Why do some (all?) watches have a separate bezel? Is it so that if the glass-metal join gets damaged it can be replaced, without needing a completely new case?

 

 

What makes a watch waterproof to 1000 m as opposed to say 50 or 100 m? Do they all just have screwed down crowns, bezels and backs, with silicone-greased O-rings? My Sinn UX, which is rated down to 5000 m (and has withstood many showers and baths without harm), is oil-filled — partly to stop total internal reflection under water, but also because it's incompressible.

 

 

And here's an easy one I'm sure (although I can't figure it out): Why do watch manufacturers assume that while people will surely want to be able to read the time on their chronographs at night (although not the seconds, of course), they won't have any interest in the chronograph functions? I don't get it. I wan't to see everything! Even the **BLEEP** date!

 

 

Here's my idealised chronograph, based on the Valjoux 7750,

 

 

and modified :-)

 

[bobm12]

Collin, I love the watch design and I hope it becomes viable at some point...maybe I'll buy you one!

 

On the other hand, those are interesting questions. I'm not an experienced watchmaker, only a hobbyist and new at that.

 

So, I'll give it a try, -- maybe you already know what is coming --  but I believe a bezel is simply either to add finish/looks to the watch considering the simple opening for the crystal is sometimes "disrupting" to the the design or/and to add utilitarian functions like on divers. Also think of cheaper construction, like avoiding custom cut crystals, simple mass manufacturing/assembly, etc, 

 

By the way, a by-directional "diving" watch bezel is a contradiction since diving bezels should only go in one direction. When you mark how much air you have with the bezel and accidentally bump it against something underwater, it is supposed to move only in a direction that benefits the diver, i.e. more air available based on time left.

 

Also, pressures underwater affect the workings of the watch as it affects the diver. Divers decompress at certain depths to equalize the pressures. (Deep fish when fished, are pulled fast and so we see their eyes popping out because of the change in pressure). Some underwater work it done at pressures that the diver breathes a different air mixture to what we consider normal even. To work under those conditions, true diving watches are engineered in a way -- using decompression valves, helium, tough case/crystal, or other means -- to stay water tight up to their specs and resist the harsh environment. It is common that "diver" watches with extra crowns are not really diver watches as well as cyclope embeded crystals are not usually part of true divers'. The Rolex submariner that went attached to the submarine in its time was the simpler design (no cyclops, etc).

 

Of course, sport watches -- or wannabe diver watches  -- try to achieve their specs with different seals here and there (sometimes approaching the real divers) but some don't have screwed down backs and crowns. Even the materials they are made of may not be adecuate. They sometimes come with simple press on back/crown but some sort of gasket solution to seal the watch. Some are a one piece design, no separate bezel or other "main stream" features. As you can see, those solutions don't last or are not enough over time to keep the watch water tight -- there is always an opening where water, given time, will go in. Which is also true for the true divers but they are supposed to be more resilient to hard use and time.

 

Which eventually lead us to the manufacturers and their designs. It is my belief that only those watches build specifically for a purpose have all/most of what is required, including full night view if necessary/requested. All the rest, it is just like cars, they cut corners whenever they can get away with. Some one told me once: "manufacturers usually make the worst car the market will permit" and I believe that regarding to cost, the watch industry is not too far from that either.

 

There may be other issues and/or more technical explanations -- better ones -- that may answer your questions better. Maybe some of the more knowledgeable people here may want to give it a try.

Edited September 20, 2014 by bobm12

[WillFly]

Good stuff there, Bob. I'll just add the comment that the people I know who are serious divers don't actually use divers' watches these days - they use dive computers, which are much more serious things.

 

Some manufacturers made cases with a monocoque design, with only the bezel & glass unscrewing - front loaders with no opening at the back. I have a 1940s Hamilton like that, and my Mido Commander is also a solid back construction. But I wouldn't dream of getting any of my watches near water.

 

And Colin - I have a nice Longines movement for which I don't have a case. You're welcome to have a crack at making me one in stainless steel if you're raring to go... :thumbsu:

Edited September 20, 2014 by WillFly

[autowind]

Dive computers hey,  we live and learn.

[Geo]

Very smart design Colin, I rather like it. Are you using Autocad for designing?

[colinh]

I'm not even a wannabe diver, but I do like my watches to be rugged and waterproof. I was just wondering what makes one watch more pressure resistant than another (other than a sturdier case).

 

@Geo  :  No, I'm using a PostScript module I wrote for the Perl programming language a while back. So the file is actually a program which means it can be as parametric as I like. I don't know Autocad, but programming a tachymetre scale, for instance, is trivial in perl...

 

 

 

@Bob, Will  :   Well, I have some 45 mm ø stainless steel and bronze. If I ever get around to it, you'll be the first to know :-)

 

 

My goal is maximum clarity, under all conditions. Once I've got some money again I'd like to see if an idea I have is feasible: I'd embed some photodiodes in the face (did anyone notice the blue square above the day/date?). They'd charge a little supercapacitor and that would pulse some UV leds when it's dark - the idea being that the entire watch (ie all the white and orange markings including day/date) would glow as much as they do in the on screen image :-)

 

 

PS. These images are actually screen shots (!). The pdfs are much nicer, but I didn't bother to scale/convert them :-(

 

PPS. The styling is influenced by my Fortis and Sinn watches.

Edited September 21, 2014 by colinh

[bobm12]

Thank you for keeping me in mind colin, and I love the idea of the UV leds and what they do!

[Geo]

Hi Colin, you have some wonderful futuristic ideas for illuminating the watch dial, I do hope you manage to develop these. Regarding Perl, it is not something a that I have heard of, but having looked it up on the net, I am well impressed. It is way above my understanding when it comes to computer programming, but you have obviously done a wonderful job. I think you, svorkoetter and DJW should get together, you would make a great design team!

[bobm12]

I dabbled a little with Perl when I was in college, in my UNIX class, but never followed through. After that was mainly C++ interrupted by Java. Never liked the instructor on that last one so I didn't follow the programmers route and ended up in IT! On second thought I think I chose wrong! Those are beautiful projects.

[colinh]

Perl is very UNIXy. Mac OS X is also a UNIX, but tries not to show it :-)

You can write ununderstandable code in it, but you can also make it fairly clear, if you try...

 

foreach $num (60,65,70,75,80,85,90,95,100,110,120,130,140,150,160,170,180,190,200,220,240,260,280,300,350,400,450,500,600 ) {

    $secs_elapsed    = 3600 / $num;

    $angle           = 90 - $secs_elapsed * 6;

    $aa              = ($angle > 0) ? 0 : -$angle;

    $frame           = $bezel->frame2(240, 240)->place(240,240)->rotate($angle);

    $text            = $frame->txt($num, $font_helv18)->place(0.5, 0.3, 225, 0)->rotate($aa)->colour(0);

    $rect            = $frame->rectangle(5,4)->place(0, 0.5, 205, 0)->b_colour(0);

}

 

Which reads as "for each number in this list, find how many seconds it corresponds to ("60" is at 3600 / 60 = 60 secs, "120" is at 3600 / 120 = 30 secs, etc). Then calculate the corresponding angle, (one sec being 360 / 60 = 6 degrees and 0 degrees being at 3 o'clock). Ignore the $aa line for the moment.

 

Then make a frame, which is like a mylar sheet on which you can draw. You can move it around, and rotate or stretch it, if you like. This one's a certain size (say 240 by 240 mm) and has been placed with its origin (0, 0) or bottom-left corner at the centre of the the bezel (which happens to be at (240, 240) of a different sheet of mylar). We're going to rotate it by that angle later, when we take a photo of it all.

 

So, we have this sheet of mylar. On it we draw a bit of text, "60" for the first number, out at co-ordinate (225, 0) which is 225 mm along the bottom edge. So it'll be a bit to the left of the 3 o'clock position (ignore the 0.5, 0.3 for now). But it's going to get rotated 90 degrees to the 12 o'clock position (which would tip the number over onto its side). So we draw it rotated by minus 90 degrees, ie. on its other side, so that when the sheet is rotated the number is upright again.

 

Actually the angles are -270 and +270, not 90 and -90, which is convenient because, as it happens, I don't want to unrotate the numbers between 3 o'clock and 12 o'clock :-) That's what the $aa line does. For angles bigger than 0 (ie between 0 and 90 degrees) the anti-angle is 0, ie. it doesn't get unrotated. Otherwise (between 0 and -270), $aa is the angle to unrotate by.

 

Then we have a little 5 by 4 rectangle as the index mark, a little bit in from the number, at (205, 0). The text and rectangle are both colour 0, which is black.

 

Finally, the [0.5, 0.3] (or [0, 0.5] in the case of the little rectangle) says which bit of the text gets placed at exactly (225,0). Imagine a box around the text "60". Normally the bottom-left corner is what gets placed. [0.5, 0.3] means the centre of the number. It's not [0.5, 0.5] because the box actually goes down a bit further because of letters like g, p and y.

 

See? Easy isn't it? I mean, if you even only vaguely understand what's going on, the rest is just details. That's what programming looks like. It's really powerful and flexible, which is why computers have taken over our world. Unfortunately.

 

I'm going to get banned now, aren't I? :unsure: Promise I won't ever try to post code in a watch forum again. Honest.

[colinh]

Hi Colin, you have some wonderful futuristic ideas for illuminating the watch dial, I do hope you manage to develop these.

 

Ah—it's called electronics. Evil futuristic stuff, that doesn't belong in a decent watch. :)

 

But there's no battery, and it has nothing to do with the timekeeping, so maybe it's ok?

 

 

... unfortunately, I never get around to finishing anything :(

Edited September 22, 2014 by colinh

[bobm12]

Thank you for the code and illustration, Colin, very nice and compact! It reminded me of my old UNIX instructor's class (the good instructor, not the bad one)! I had a lot of fun there! And this is a wonderful idea! Please, finish it! I strongly encourage you to do so, there is beauty, practicality and even, maybe, profit in this!

[PadraicB]

I write PHP ;). I'll start running now so you'll never catch me! If you throw the forum name at Google, you'll find the PHP Security blog that I maintain. Nice to see a few fellow programmers out there.

[WillFly]

All this takes me back - I used to do the odd bit of programming using Foxpro+. Once wrote a programme to calculate and print 7 digits+checksum barcodes... But it's all gone these days!

[PadraicB]

What are the bets that everyone on this forum is a programmer? ;) I know that the challenge from one is part of my attraction to the other.

 

On the case design, I would love to see how the lighting solution holds up in practice. We all know that Quartz watches have been cheating for decades ;).

[Geo]

What are the bets that everyone on this forum is a programmer?

It takes me to program my sat nav, never mind the dark side of a computer!!

[colinh]

On the case design, I would love to see how the lighting solution holds up in practice.

 

Me too :)

 

It takes me to program my sat nav, never mind the dark side of a computer!!

 

Come to the dark side! We have cookies.

[Mark]

 

I'm going to get banned now, aren't I? :unsure: Promise I won't ever try to post code in a watch forum again. Honest.

 

 

Don't worry Colin - this is a code friendly zone. 

 

Perl was the first computer 'language' I learn't back in the late 90's and I was addicted to writing small scripts for my website. Then I got into PHP and this has been very useful over the years. In the past I have also enjoyed, sort of, using VB and VB.NET and QT. 

 

With the question of water resistance, you have to look at many factors when designing a case.

 

Obviously the material and the material thickness.

The back. For example, a snap-on back is almost useless for a divers watch. So a screw-on back is required.

The glass should be thick enough - 1mm for a basic swimming watch, 2mm or more for a divers fitted with a nylon gasket - or an armoured perspex.

The crown. A regular water resistant crown for swimming, a screw-down crown for divers

 

So all the weak points:

Glass

Crown

Pushers (if its a chronograph)

Back

 

And also remember - there is no such thing as water-proof when it comes to watches, just varying degrees of water resistance.