Water Tight, machining for o-rings, sealing lenses, ect..

[VanIsleDSM]

I'm getting to the point where I want to start making some lights now on my lathe. I whipped up an easy little direct drive 16340 light with a CREE quickly before I went camping last weekend, and just discharged the batteries down to 3.8V before use. Worked well.

Now I want to take things further. In doing so I need to know how to properly seal up these little lights. Does anybody have any information or first hand tips on how to properly machine grooves for 0-rings? How much should they squish? Is an 18mm o-ring the OD, ID, imaginary line in the middle?

What is the difference between a dive light's seal and a regular light? I think I've seen some with up to 3 o-rings before. Say you wanted to make a little 16340 water tight enough for diving, what would you have to do to the lens? I see on some dive lights instead of the cap screwing on there are fasteners all around the perimeter, to hold the lens tighter I suppose, but how does it seal exactly? just between 2 pieces of rubber?

Just learning how to use google sketchup right now so I can make some models to go from when machining. I'd like to figure out how o-rings work exactly so I can incorporate the o-ring grooves into the design.

By the way.. anybody have any idea to make screw threads in google sketchup?

 

[dom]

Hi Van
Here's a handy little chart with some basic type static and moving seal types and clearances.
http://www.norseal.com.au/tolerances/ormach.htm

Alot of good info here on what o-ring material type to use
http://www.marcorubber.com/materialguide.htm
http://www.lutzsales.com/chemical_resistance_guide.asp

The sizing is usually the ID and material diameter
18 X 1.5mm -so 21mm OD

Cheers
Dom

 

[mudman cj]

I like to use a free design program offered by Parker Hannifin called inPhorm.

 

[KC2IXE]

Besides the sorces mentioned above (which make it easy), I'm going to also recommend the "Usual sources" - The Machinery Handbook, and then a copy of the Marks Manual and a copy of the CRC. Between those 3 books, you probably have every piece of information you need to do just about anything technical - the trick is learning to read them

 

[gadget_lover]

When building a dive light, you have to take several things into consideration beyond simply keeping the inside dry.

You also want to think about how the pressure will be distributed, and how to work the switch. Neoprene gloves (can you tell I'm from the northern part of california?) make swithing things on/off difficult. Then theres the pressure. At 30 feet there's about 15 PSI pushing on your switch. It might not want to turn on (if you have a reverse clickie) or will not turn off (with a forward clickie).

Then there's corrosion and electrolysis. Rust. Sand. Lot's of little things.

It's not surprising that lots of dive lights have twist switches or levers. A rocker is also usable at depth, since the volume does not change as you press it.

Daniel

 

[VanIsleDSM]

I was thinking of machining a little twist ring with a magnet in it to trigger a reed switch inside.

Won't really be for diving though, snorkeling at best, Seems it would be neat to take a light into the water.. For this I imagine I could probably still get away will a well sealed clicky to keep things easy? I dunno.. we'll see.

Starting to get the hang of google sketchup now.. takes a while to get used to. Almost done my first model, the single CREE 16340. I'll post it up when I'm done.. it has the "threads" but they aren't helical, just for size reference.

 

[precisionworks]

I was thinking of machining a little twist ring with a magnet in it to trigger a reed switch inside.

Surefire did a nice job with a Hall Effect sensor in the U2. Their sensor & processor allow for multiple lighting levels - yours would be easier since you want only on or off. I would think that the multi-level concept would be pretty easy to do ... except that few companies offer it.

 

[VanIsleDSM]

So the hall effect sensor sends a signal to the uC to control the brightness depending on the proximity of the magnet? That's brilliant.

But I think I'll need to use an array of reed switches around the ring to control brightness for the other drivers I have made... Since I need to be able to adjust resistance between 0.05ohm to 3ohm or so to control brightness.. I know no other way than a bunch of precision resistors in steps. Or tapping sections of nichrome wire with a rotary switch. I wish I knew a way to seamlessly adjust resistance at that low of a level. There are no pots in that range, other than using the first part of a 25turn 10ohm pot. That works reasonably.

I'm capable of programming simple programs into uCs, so I wish I could think of a way to utilize the hall effect sensor.. I know how I read the signal and all... but there's just no device I can control digitally or by analogue means that will adjust resistance like that.. nothing I know of anyway.

 

[precisionworks]

... there's just no device I can control digitally or by analogue means that will adjust resistance like that ...

I believe the U2 uses a multi-level driver, sort of like the driver in the 120P. Wayne @ Sandwiche Shoppe might have something that would work. And both Allied & Newark have a zillion LED drivers, but not sure if any are multilevel.

 

[VanIsleDSM]

That's the thing. Any driver can be multi-level if you can adjust the current sense resistor. It's just always such a low value to keep the efficiency up.. awkward to adjust. The drivers I've made so far can also be adjusted by a PWM pulse.. but this makes less sense. PWM dimming preserves a better white point throughout the range, but dimming by just lowering the current to the LED is so much more efficient.

 

[Tekno_Cowboy]

If you capped the pushbutton seal with a metal plug, you could use a secondary swapable spring to offset pressure when diving. I'm working on something similar to that myself. Also, you could use orings set in machined grooves on the face, rather than the sides on a thicker light to give a set repeatable torque pressure seal.(hope that makes sense to everyone, I don't exactly know machining terms. )

 

[VanIsleDSM]

I'm working on a design right now that I would like to be able to be used as a diving light. Even though I've never been diving, I want it to be that robust. How are the lenses on diving lights usually sealed in? just squished between 2 o-rings?

I found an old mag sitting around, and I see the way they do it, but I don't think that would be good enough for diving.

I've noticed that often dive lights have many fasteners around the bezel, I imagine to apply more force to seal the lens than standard threads. But I'm still unsure exactly what I need to do to seal the lens for diving.

If anybody has any commercial dive lights and they could take a pic of how things seal up, I'd be forever grateful

 

[Tekno_Cowboy]

I'm not sure how the pros do it, but I've found permatex ultrablack automotive gasket maker works very well for enhancing the water resistance in applications like that. (It made my G2 last for 24 hours at 40ft under, applied everywhere and allowed to set. Changing the batteries took some work though )