How to design a "press fit" lens?

[lightime]

I am doing a project in which I am making a waterproof 316/316L stainless steel enclosure to install a small closed circuit camera which will be submerged under saltwater for extended periods of time. I need to install a glass lens for the camera to "see" through so I was simply going to make say a 40mm hole and then glue a 45mm piece of glass over the hole....but then I though there has to be a better way!

How are the press in lens setups designed? I have seen lenses that use a plastic ring/gasket and are pressed in from the front. The plastic ring fits snugly into the front of the housing and then the glass is pressed into it.

What kind of material is that ring/gasket made of? If I wanted to make something that could be submerged for many months or perhaps years at a time what material should I use for the gasket and for the lens?

Is it as simple as making a hole that is 40mm in diameter and then finding a lens that is say 39mm? How thick should the gasket be? Where could I source a gasket like this?

Thank you for any help and/or ideas

 

[qwertyydude]

Be sure you're using real certified 316 stainless steel some cheaper grades steel are being passed as "marine grade" and are not true 316, but if you really want saltwater corrosion resistance that'll be easier to machine and press fit you may want to look at copper nickel alloys. Stainless steels are a pain to machine.

 

[modamag]

The ingenious principle of the pressed lens first came about from Don (McGizmo).

It's basically using both axial and radial oring compression force to "float the lens".

a) using an undersize oring on one side to sustain the compression in the axial direction
b) an oversize oring on the other side of the lens to provide radial compression

example of setup
Reflector - #18 oring - glass lens - #19 oring

The interior groove that hold the oversize lens (eg #19) need to be deep enough to prevent the oring from popping out. 70% of nominal thickness might be enough. Easily calculated based on the oring's nominal diameter.

From your request to use 40mm lens. I would would slightly change the diameter if it's not too important to 41.8mm x 1.9mm lens because it's available at flashlightlens.com for cheap.

Here's what I would go with.
41.8mm OD x 1.9mm thick flashlight lens
#127 (undersize) oring
THRU HOLE = 1.440"
CBORE (shoulder where oring/lens sit) = 1.650"
#130 (oversize) oring
GROOVE OD (oversize oring sit) = 1.756"

 

[lightime]

Thanks for the replies.

qwertyydude, I have used the shop that is making me the enclosure in the past and they source good stuff.

modamag, I really appreciate your reply but I think we are talking about two different methods. I am interested in what you are describing but I am having a hard time visualizing it. is there a diagram I could look at to see how the system you are referring to works?

In my original post I am referring to a very simple method that is often used in diving watches (Rolex, etc..). Say the glass lens is 40mm diameter by 3mm thick....a plastic gasket that is 40mm inside diameter and 3mm tall x .5mm thick is used. this gasket is fit snugly to the edge of the glass and then the glass + gasket is pressed into the opening. I am looking for specs on how to do it this way. It works on diving watches that go 100's of meters underwater I just want to go about 5 feet underwater.

Thanks!!!

 

[wquiles]

The ingenious principle of the pressed lens first came about from Don (McGizmo).

It's basically using both axial and radial oring compression force to "float the lens".

a) using an undersize oring on one side to sustain the compression in the axial direction
b) an oversize oring on the other side of the lens to provide radial compression

example of setup
Reflector - #18 oring - glass lens - #19 oring

The interior groove that hold the oversize lens (eg #19) need to be deep enough to prevent the oring from popping out. 70% of nominal thickness might be enough. Easily calculated based on the oring's nominal diameter.

From your request to use 40mm lens. I would would slightly change the diameter if it's not too important to 41.8mm x 1.9mm lens because it's available at flashlightlens.com for cheap.

Here's what I would go with.
41.8mm OD x 1.9mm thick flashlight lens
#127 (undersize) oring
THRU HOLE = 1.440"
CBORE (shoulder where oring/lens sit) = 1.650"
#130 (oversize) oring
GROOVE OD (oversize oring sit) = 1.756"

Jonathan, I would "LOVE" to see a diagram of this - even a hand written sketch of it - I am way too visual, and although your description is awesome, my confused brain is having a little hard time visualizing it

Will

 

[PhotonFanatic]

I would look into either a polycarbonate or Lexan ring. If you make the ring just slightly thicker than the lens, i.e., 3mm thick lens and 4mm thick ring, then press fitting it into the body of the SS should be no problem, as you will be pressing on the plastic, not the lens. You might have to experiment to get the oversizing of the ring correct, but I'd start with at least .001" over the diameter of the lens bore. Once in place, you could machine down the plastic ring a bit.

You need to take into account temperature fluctuations, as some plastics can expand/shink quite a bit, depending on the temperature.

Hmm, this might be an interesting set-up for a flashlight, too.

 

[modamag]

Wil, you're making me work for my answer huh

 

[wquiles]

Wil, you're making me work for my answer huh

Jonathan - thanks. That represents well what I "though" it was - now I can "see" it

So the outer, oversize O-ring is applying some pressure on the lens, which applies some pressure on the undersize O-ring? At soon as the light is under water, the water will easily go pass the outer/oversize O-ring and through the side of the lens, and then will be temporarily stopped at the inner/undersize O-ring. What prevents the watter pressure from making the inside/undersize O-ring slide inwards and failing?

I have a diving light book from Steve Lindblom ("Divelight Companion"), that specifically warns against this same exact O-ring usage, unless the O-ring has a wall or groove on the inside. Maybe that is part of Don's solution that is not shown in your diagram?

And by the way, I am not attaching anyone's design, just sharing what I have learned about sealing water for diving use, since the OP talked about taking this under water.

Will

 

[modamag]

I do not think water will not pass the outer oring.
Well it all depends on the depth & pressure. You just need to make sure that the interference is more than the compression of the inner oring @ design pressured.

I've seen it work at shallow depth w/o problem. I don't know about 50-100m depth. At such depth I would think the traditional double sealed oring style is more appropriate.

PS: Notice how I left out the intricate dimensioning details.
Just like anyone can make BBQ sauce and only your mom knows the secret ingredients (Tobasco Sauce!)

 

[wquiles]

Gotcha - the devil is in the details

And by the way, thanks again for the drawing. What software package do you used?

Will

 

[greenLED]

What prevents the watter pressure from making the inside/undersize O-ring slide inwards and failing?

Hi, Will - in my HD45, there's a small "lip" that nests the undersized o-ring around the reflector. I think that would prevent the o-ring from doing what you describe.

 

[wquiles]

Hi, Will - in my HD45, there's a small "lip" that nests the undersized o-ring around the reflector. I think that would prevent the o-ring from doing what you describe.

Yup, that is "exactly" what I meant earlier. You need "something" to prevent the O-ring from sliding inwards due to water pressure. I have been studying more and more diving-rated equipment, and the "Divelight Companion" from Steve Lindblom has some specific recommendations/warnings about O-rings and lenses, since most don't have that O-ring backing/lip/groove/etc.. Quoting from Steve's book:

"The backing of the O-ring is not so obvious - I see a lot of DIY lights that omit it and rely on just the clamping force between the glass and the lighthead body. This is not a reliable solution!. The problem being that, as you do deeper, the water pressure pushes from the outside, and there is nothing to keep the ring from sliding inward except the clamping force of the glass. When the pressure gets strong enough to overcome the clamping force, the ring will distort and the seal fail. This is one of those things that works fine until it doesn't. It may work for a week or a year, and just when you start to feel smug about it, you will forget to tighten it enough, or lube the O-ring a little too much, or go a little deeper than usual, the O-ring will give a groan and a whimper, and you will have a flooded lighthead."


What I have learned so far is that making stuff water resistant is relatively easy. Making stuff diving-rated is entirely a much harder challenge, and as Jonathan said very well above, it is all about the little details

Will

 

[McGizmo]

I am no engineer but I have experienced some of the stuff in question. It is true that water pressure will want to push the iner O-ring in towards the center and off its "ledge". However that same water pressure is also pushing straight down on the window and all of that force is concentrated on the same O-ring. You have much greater force of the window pushing the O-ring against the ledge than the perimiter water force trying to push the O-ring inwards and off the ledge. In the design that Modamag has illustrated, if you took this down say 25', you could remove the outer retaining O-ring completely and be hard pressed to get the window out or break its seal on the inner O-ring.

I think you need to consider the surface areas subject to external pressure and how these forces are directed on the components of the design.

As Jonathan has drawn it, you could enhance the design further by having an inside lip that came forward producing an O-ring groove as opposed to just a flat face for the sealing O-ring. However due to the friction of the O-ring once under compression from the window. I doubt you could get it to shrink in diameter and drop down inside of the ledge.

I just looked at the thread and see that Will has brought up this point while I am composing my response. In fairness to a design I use without the benefit of a real engineer's understanding, I would also point to the retaining O-ring and what happens to it while subject to the external water pressure. In Jonathan's drawing. It is hard to imagine the retaining O-ring getting compressed sufficiently to even allow water pressure past it and allowing it to reach the sealing O-ring.

Looking at the window on a Nikonos Lens, I found that the thick glas window seated down on a metal shelf in the lens. It simply could not be pushed past this ledge unless subjected to forces much greater than those experienced in a dive. Its edge has a taper so that the outside diameter of the window increases towards the inside face of the window. There is a retaining O-ring that seats and seals around the perimeter of the window's edge. External pressure against the O-ring pushes it down towards the windows ledge and as it gets pushed down in encounters additional side compression by virtue of this taper.

I thought about using this type of seal but getting a tapered grind on the sapphire windows edge and ideally with a bit of concavity seemed likely costly.

Another point to consider is that once an O-ring does get compressed by the window being forced down on it, the surface area subject to the water trying to push it inward decreases and this is a factor of force reduction. I suspect this consideration is one that can justify the use of a flat gasket that is thin and presents even less edge surface area to the water pressure while relatively greater surface area subject to the window compression and friction against sliding past the window. :shrug:

 

[McGizmo]

......Quoting from Steve's book:

"The backing of the O-ring is not so obvious - I see a lot of DIY lights that omit it and rely on just the clamping force between the glass and the lighthead body. This is not a reliable solution!. The problem being that, as you do deeper, the water pressure pushes from the outside, and there is nothing to keep the ring from sliding inward except the clamping force of the glass. When the pressure gets strong enough to overcome the clamping force, the ring will distort and the seal fail. This is one of those things that works fine until it doesn't. It may work for a week or a year, and just when you start to feel smug about it, you will forget to tighten it enough, or lube the O-ring a little too much, or go a little deeper than usual, the O-ring will give a groan and a whimper, and you will have a flooded lighthead."


....

Will

Thinking more about this comment, I admit I am not clear on what the specific design the author has in mind is nor is it clear that he is allowing for the increase in water pressure to have a much greater magnitude in terms of the "clamping force" as compared to the water pressure on attempting to slide the O-ring out of place. He mentions forgetting to tighten it enough where the design shown by Jonathan relies on the water pressure itself to increase the clamping force. The other consideration is that of lubing the O-ring. This sounds more like a dynamic seal that is broken between dives and must be addressed as opposed to a static seal in which case it is rarely "broken" into. I have taken apart components involving static seals and the seals are often stuck and bonded to their sealing surfaces by virtue of long term surface contact. Surely this type of surface contact is different than a freshly lubed and slippery surface contact?!?

I have some designs for the window seal where the window would have to push past the O-ring and its forces by virtue of the external pressure are actually trying to push to O-ring out and away from it. These forces in combat with the water pressure on the outside of the O-ring trying to force it in towards the center. The greater force is that of the window and not the water.

Let the force be with you, and your seal. :nana:

 

[wquiles]

Looking at the window on a Nikonos Lens, I found that the thick glas window seated down on a metal shelf in the lens. It simply could not be pushed past this ledge unless subjected to forces much greater than those experienced in a dive. Its edge has a taper so that the outside diameter of the window increases towards the inside face of the window. There is a retaining O-ring that seats and seals around the perimeter of the window's edge. External pressure against the O-ring pushes it down towards the windows ledge and as it gets pushed down in encounters additional side compression by virtue of this taper.

I thought about using this type of seal but getting a tapered grind on the sapphire windows edge and ideally with a bit of concavity seemed likely costly.

That is almost verbatim how Barbolight implemented the seal on their diving-rated Boomproof hosts, except that he used a fairly thick plastic lens (4.86mm or about 0.191" thick) at about 2" in diameter, with a small 45degree cut that mates to a similar cut in the head. The lens is in compression by an O-ring on the perimeter of the lens, and this O-ring sits on its own groove, so it can't be pushed anywhere by water pressure. The retaining, larger diameter O-ring is press fit, pretty much the same way that Jonathan's design shown above. Javier (owner/founder of Barbolight) tested the head and at 200Meters there was only a 0.5mm inward deflection on the lens!

Another point to consider is that once an O-ring does get compressed by the window being forced down on it, the surface area subject to the water trying to push it inward decreases and this is a factor of force reduction. I suspect this consideration is one that can justify the use of a flat gasket that is thin and presents even less edge surface area to the water pressure while relatively greater surface area subject to the window compression and friction against sliding past the window. :shrug:

Thinking more about this comment, I admit I am not clear on what the specific design the author has in mind is nor is it clear that he is allowing for the increase in water pressure to have a much greater magnitude in terms of the "clamping force" as compared to the water pressure on attempting to slide the O-ring out of place. He mentions forgetting to tighten it enough where the design shown by Jonathan relies on the water pressure itself to increase the clamping force. The other consideration is that of lubing the O-ring. This sounds more like a dynamic seal that is broken between dives and must be addressed as opposed to a static seal in which case it is rarely "broken" into. I have taken apart components involving static seals and the seals are often stuck and bonded to their sealing surfaces by virtue of long term surface contact. Surely this type of surface contact is different than a freshly lubed and slippery surface contact?!?

Based on diagrams/sketches in Steve Lindblom's book, I think he is refering to simpler/DIY designs, which many times come with little though behind them - low cost and quick time to assemble/use seem to be a primary goal. The designs shown in Steve's book use a single O-ring, and are often built using 4 screws to clamp on the lens/O-ring sandwich to the head, which explain the lubing of the O-ring and Steve's coments regarding pressure applied while assembling the head. What you, Jonathan, and Barbolight have implemented is definitely head and shoulders above the stuff in Steve's book.


I guess if I were to design my own seal/system, I would do like you, Jonathan, and Barbolight have done and add a small lip/groove anyway, just to make the design more "bullet-proof"

And by the way Don, thanks much Don for chiming in - I always seem to learn something new on your posts here in the forum - much appreciated :bow:

Will

 

[lightime]

modamag, thanks very much posting the diagram...makes sense to me now after seeing it visually.

Keeping in mind that my project won't be going any deeper than 10-15 feet underwater.

The method I am referring to is different and may be easier for me to pull off. I made a drawing (please excuse my horrible drawing skills but I think it's simple enough of a design that it should make sense)...see below...

Below is a diving watch that I pulled apart which is using this same system. Many very expensive diving watches rated to 100+ meters use the same. It is a simple "plastic" (I say plastic because it is rigid NOT soft like an o-ring I am not really sure what it is made of). But basically this ring fits snuggly around the glass and then the glass+ring is pushed in to the opening with hand force...NO LEAKS...you can pop it right back out by pushing on the glass from inside the watch with your thumbs...Nothing else is holding/sealing the glass just the "plastic" ring....

 

[wquiles]

Thanks much for the pictures. Question: Does the lens have a bevel cut on the bottom (side facing inside the watch) and does the watch has a similar bevel as well?

 

[65535]

Not exactly a press fit, but Novatac uses this method for their bezels and I think it's a very elegant method, effective too.

 

[lightime]

The lens (as fas as I can tell) is just a straight cut with no bevel. Where the back of the lens meets the watch is also just straight. It's almost too simple but again a lot of watches use this design and go very deep under water.

I already have a 40mm x 3mm thick lens....If I could only figure out how to make this work for me and figure out what sizes everything needs to be. As well as where the heck I could get that little ring made??

Thanks much for the pictures. Question: Does the lens have a bevel cut on the bottom (side facing inside the watch) and does the watch has a similar bevel as well?

65535, that looks nice...I assume the bezel screws on/is threaded on? Is there actually a gap under the bezel?

Not exactly a press fit, but Novatac uses this method for their bezels and I think it's a very elegant method, effective too.

 

[dom]

Hi Lightime
If you have limited tools for the job,the best option is to go with the above drawing you provided.

Have , just say , 1mm total clearance around the glass.

Buy some mairine Sikaflex silicone,coat the surface sides and bottom that your glass will rest on.

Centralize your glass and put it in -but don't push it right down -so when it dries
you'll have approx 0.5mm silicone under the glass.

This will NEVER leak at the depth you'll be using it.

Cheers
Dom