Since getting my first digital camera, and later, my first video camera, I have wanted to push the envelope of what these cameras can do, and where they can go. I have taken them on roller coasters, mounted them on my motorcycle, pulled out the IR filter and taken infrared photos. In short, anything is worth a try (thanks to Instructables for inspiration!).
Putting a camera underwater was another realm to conquer, and having two young children who love the pool gave me a great reason to work on this.
The purpose of this Instructable (my first) is to detail how to build an underwater video camera housing. The end result will be the ability to take underwater video on the cheap - less than $50, significantly less if you have parts on hand.
It took about 3 days to complete this, but most of that was testing for leaks or waiting for the sealant to set. Actual workbench time was less than 2 hours.
My inspiration for this project comes from these Instructables: http://www.instructables.com/id/ESAN0HCF2FRVIUV/ http://www.instructables.com/id/EKU3DUKR8WEPA8KXRG/
The first showed me how easy and inexpensive this could be, and the second set the bar for aesthetics. Thanks to both Kipkay and Bobby_M! I want to give credit where credit is due.
Step 1: Parts & Tools
1. Igloo (or similar) insulated water jug - $4
2. Sheet of Lexan or Plexiglass - $5 (enough to make several lenses)
3. PVC fitting, threaded at one end, flared at the other, about 2 inches across - $3
4. Fast Cure 5200 marine-grade adhesive/sealant - $5 (way more than needed)
5. Teflon plumbers tape - $??? (way more than needed, had it on-hand)
6. Industrial-strength Velcro - $10 (way more than needed)
7. Pair of metal door handles - $5
8. Two metal bands (type used to connect rubber hoses), 5" to 7" fit - about $5
9. OPTIONAL - water sensor - $10
I purchased all the parts except for the water jug at Home Depot. The water jug I bought at K-Mart, but I have seen the same or similar just about everywhere, from Target to CVS. We even have one at home that is easily 25 years old, so it should be fairly easy to find one that works for you.
Video camera that fits the water jug
1. Electric drill and bits
2. Hole saw of approximately the same size as the flared opening of the PVC fitting
3. Hole saw of same size as threaded end of PVC fitting (if different)
5. Cap from a can spray paint (an old one, as this cap will be destroyed)
6. A place to test for watertightness (I used an old 5-gallon paint bucket with a lid)
7. Heavy stuff to hold the lens in place while it sets and, if needed, to hold the lid on your test chamber
Step 2: If it doesnt fit, you have to quit - testing for fit
Before you go any further, take your video camera with you to the store where you intend to buy your water jug and make sure the camera will fit inside with a little room to maneuver. My video camera is a Sony DCR-DVD92 Mini DVD camera with a wide-angle lens. Depending on your camera, this Instructable may or may not work for you, as a larger camera will not fit properly into the water jug housing.
Step 3: Keep your powder dry - testing for watertightness
At several points, I tested for watertightness to ensure that everything was safe and secure. Not having a pool in the backyard, and not wishing to fill a bathtub every time I needed to check, I used an old 5-gallon paint bucket with a tight-fitting lid. I filled this in a utility sink, and then put the sealed jug in to displace the excess water. After the first use, I just kept the bucket full of water until I completed the project.
First, I checked the water jug for watertightness by filling it with water, putting on the lid, and turning the jug upside-down, sideways, etc. I figured if it didnt hold water in, it probably wouldnt hold water out. It passed.
Next, I filled the water jug with newspaper, put it into the paint bucket open-end down, then put the bucket lid on and put something heavy on top (my power washer) to keep the lid from popping off. I left things this way for several hours, figuring a slow leak might take awhile to show up. Because this method didn't put the entire water jug underwater, I then repeated the process with the jug oriented with the opposite end down.
I found that it leaked, but very, very little. I solved this with Teflon plumbers tape. I also found that starting the threads on the lid at exactly the right spot made a difference in how well the lid fit, so I made a mark on the lid, then marked where the threads started (with an S) and where the jug's lid fit the tightest without overtorquing it (with an E for End).
Step 4: Measure twice, cut once - or just get lucky
Once I was satisfied that the uncut jug was watertight, I started measuring.
This was probably the hardest part of the project, because the inside of the jug tapers down toward the bottom, making it hard to measure properly. I ended up winging it with an educated guess, so take your time and do your best. Remember, if you mess it up, you can always buy another jug for $4 or so.
I placed the camera, with the wide-angle lens attached, lens-down into the jug. Using a long pencil, I did my best to mark a circle around the lens. I then placed the jug on a level surface and eyeballed about where the circle was relative to the outer edge of the jug, and marked a dot with a Sharpie on the outside bottom of the jug. I hoped the dot marked the center of the circle. I was about to find out.
Step 5: Wreck the jug
I used a small drill bit and drilled a pilot hole through the bottom of the jug from the outside where I'd marked the "center" of the circle I'd made inside. I was off by a little, but that helped me correct where I needed to drill with the hole saw. WARNING: the jug is no longer watertight. :-)
Carefully re-marking the bottom of the jug, I used my hole saw (with the pilot bit in) and drilled a 2-inch hole. You'll need to hold the jug securely while doing this, and be prepared to go through two layers of plastic with a layer of foam insulation between them. Not only will the jug want to spin, but as the hole saw cut through the layers, it caused my drill to jump slightly, which is why I'm noting it. Be careful.
Once I drilled the 2-inch hole, I put my camera inside the jug and tested fit. I got lucky and the hole was exactly where I wanted it.
Step 6: Screw in the PVC fitting
In hindsight, I should probably have done this AFTER I cut and sealed in the Lexan lens, as it might be easier to assemble the lens that way, but since it worked this way, I wrote it up the way I did it. Should I get around to building the Mark 2, I'll try making the lens assembly first.
I took the PVC fitting, and using a good bit of force, screwed it into the hole in the jug. It was a tight fit, which is what I wanted, but once I got it started it threaded easily because the jug's plastic shell was fairly thin. I was surprised, but not distressed, to find that the threaded end of the PVC fitting did not thread through both the outer and inner plastic. Instead, the PVC fitting cleared the outer shell and most of the insulation, but left a small gap between it and the inner shell. I figured the sealant on the outside would take care of that, and if needed I could also seal the inside where the PVC fitting didn't meet up with the inner plastic shell. This ended up not being necessary.
Step 7: Cut the lens
This was the second-most difficult part of this project. I took my piece of Lexan, and was immediately stuck with the problem of how to cut the Lexan with my hole saw WITHOUT using the pilot bit. After all, I didnt want to end up having a lens with a big hole right in the middle. Having no experience with this particular problem, I improvised.
First, I clamped down the Lexan sandwiched between two boards. I was careful to leave the plastic sheeting on the Lexan in place on both sides, to avoid scratching.
I then took my hole saw and the cap from an old spray paint can, and (on a different board!) drilled through the cap with the hole saw to cut out the top. The cap and my hole saw happened to be a perfect fit, so again I got lucky. Using the spray paint cap to guide the hole saw, I placed the cap on the Lexan, held it with one hand, and used the drill and hole saw (without the pilot bit) to cut the Lexan lens. This took a couple of tries, but I had plenty to play with. If anyone has suggestions for how to do this step some other way, please let me know!
Once I had my lens, I cleaned the edge with my thumbnail to remove the bits of melted plastic caused by the hole saw. Could I have done this with some fancier tool, like my Dremel or some kind of buffer? Sure, but I also knew this would all get covered by the sealant, so I wasn't fussy. I just wanted a relatively clean edge.
I tested the fit of the lens in the flared end of the PVC fitting, and it was a good match. I was ready to seal it up.
Step 8: Sealant time
Before starting this step, a few words of caution about the sealant. First, the marine sealant is VERY sticky and VERY hard to clean up. Make sure you follow the instructions on the tube carefully. Second, you will have vastly more sealant than you need for this project, so if you have other projects in mind where you could use a waterproof sealant/caulk, you might want to line them up and do them all at the same time. I speak from experience: I wasted 90% of the sealant by not using it immediately, as it sets up within 48 hours, even with the cap on the tube.
First, I put a bead of sealant around the inner edge of the PVC fitting. I carefully placed the lens into the opening, and then put a second bead of sealant around the outer edge of the Lexan. This process is the part I would do before putting the PVC fitting into the jug, as it would be easier to seal and place the lens. I'm not the best at caulking, and I made a bit of a mess.
I then sealed around the edge where the PVC met the outer shell of the jug. This didn't have to be cosmetically perfect - just sealed. I did a fair job on both counts.
Lastly, I sealed the spigot in the lid by filling the spigot with sealant. While this didnt leak when I tested it the first time, I figured it was safer to fill this in, and besides, I had more than enough sealant to work with.
In order to ensure that the lens seal was tight, I put weight on it and let it set overnight. This required more improvising, and I ended up putting an old prescription bottle on the lens, and balancing a light sledgehammer on top. I was careful not to put anything on the lens that would touch the sealant, as this stuff is sticky and hardens to the point that it's impossible to get off.
Step 9: Back to the bucket - Test for watertightness (again)
As noted above, once the sealant was set, I put the housing in my "test chamber" following my earlier methodology, and waited to see what happened. Nothing leaked - not even a drop. Feeling very pleased with myself, I continued with the last steps.
Step 10: Put in Velcro
I put the hook side of the industrial-strength Velcro in what was now the bottom of the housing, figuring that if anything ever did get wet, I'd want the fuzzy side to be the part that came out (icky wet fuzzy stuff = bad). Also, as my wife pointed out, the fuzzy side would feel better on my hand when I was using the camera outside the housing. I then put the fuzzy side of the Velcro on the bottom of my camera and tested for fit.
I also put some of the hook side at the front of the housing, just under the lens, and a bit of the fuzzy side on the front of the camera. This helps to position the camera all the way in the housing before pushing it down into place on the bottom.
Everything lined up, and my underwater housing was ready for a field test.
Step 11: Field test
I took the housing with me on vacation to the beach, and played with the kids in the water for a bit. We all thought it was a hoot, and the beach gave us some good targets (aside from ourselves): shells and rocks along the bottom, seaweed and sea grass, and small fish. I took a few short videos first, checking carefully each time I took the camera out to see if there were any leaks. The only water to get inside the housing appeared to be from opening the lid itself, so I was careful to dry the outside of the housing and my hands before opening it.
Step 12: Add ballast and handles
After field-testing, I found that the whole rig was much too buoyant to handle, and it was hard to hold onto and keep oriented properly. While using it in shallow water at the beach, I needed to lean heavily on it to keep it underwater. I needed ballast and handles.
For ballast, I used metal rods cut to length, but just about anything really heavy but shapeable will do. Beanbags filled with lead shot would be good, or an old set of ankle-weights. I chose to put the ballast inside the housing, tucking it alongside the camera. This kept it dry (no worries about rusting) and it helped keep the camera securely positioned. I wrapped the metal rods in paper towels (to cover the sharp, rough ends) and ziplock bags (to keep dirt off the camera and water, should it get in, off the rods).
I suppose you could wrap the ballast around the outside of the housing if you wanted instead. There is certainly a very limited space inside the housing, and I cant get it to be neutrally buoyant as I have it. I just figured I would keep everything inside if I could.
For the handles, I used two painted steel door handles, held in place by two metal bands. As an added bonus, I lined up the screws on the bands as a sight to help me correctly aim the camera. This isnt the look I was going for, but for now its the most functional alternative Ive come up with. I intend to spraypaint the handles white to match the rest of the housing.
Step 13: (Optional) Peace of mind - adding a water alarm
If you trust your skills enough to risk your camera, this is unnecessary. I prefer "better safe than sorry" to "Oh *&%$#@!" so I decided to add a water alarm, just in case something leaked.
I bought a water alarm at Home Depot for about $10, and removed the plastic case, keeping only the electronics. Sorry I have no photos for this step, but this is very easy - remove every screw in the plastic case and gently remove the electronics (keeping the long wire and sensor pads attached). It comes out all in one piece.
I attached the water alarm to the lid of the jug with Velcro, and I put Velcro on the sensor so it would stay at the bottom of the housing.
The water alarm was a worthwhile addition, as I believe it saved the camera a couple of times. As noted, my first field tests were at the beach, and I believe some sand got into the threads of the lid and caused a very small amount of leakage. The alarm sounded, and I was able to immediately get the gear out of the water, being careful to hold the housing lens-up, so any water would stay away from the camera. While no significant amount of water got inside the housing, even a small amount of water in the right (or wrong) spot could fry the camera.
Step 14: Your mileage may vary....a few addenda, some quid pro quos....
If you haven't guessed by now, I'm a stickler for testing, especially when something expensive like my video camera is on the line.
After completing the housing and playing with it, I read up on the fancy new $500 waterproof video cameras (no housing needed), and saw that they are rated based on depth and time at depth, e.g. 1 hour at 5 feet.
I took my housing (minus the camera) into the pool and kept my foot on it for 30 minutes at a depth of 5 feet. I chose 30 minutes because that is the maximum length of time to record on a mini-DVD at best quality, so I know the camera wouldnt be actively used at that depth longer than that. I also know I can't hold my breath for 30 minutes at any depth. :-)
When I brought the housing up and checked it, there was some slight leakage - enough for me to know that I wouldn't want to keep the camera underwater at 5 feet for half an hour.
So, while I would never take this scuba-diving, I would take it to the beach, in the pool, and snorkeling. Perhaps you are a better craftsman than I am, perhaps not. Weigh the risks versus rewards for yourself and have at it.
Step 15: Enjoy!
Since completing the housing and testing it, I took it to the pool a few times, and the kids and I have had a lot of fun playing with it. It can be a bit hard for the kids to keep the camera steady underwater, but they do a good job of it and have a blast. The housing is easy enough for me to handle that I can swim underwater while holding it, and you can turn it toward yourself as well: that is, if you want to see silly videos of yourself with bubbles blowing out of your nose.
I was concerned about how the colors would look, as there are lots of amateur underwater photos and videos where the colors are washed out. However, with my camera set to bright sunlight, the colors came out fairly well, although the above-water colors are very washed out. I havent yet tried to let the camera auto-adjust it, but as my video camera has a built-in Easy button, I will try that on a future outing.
Posting videos is new to me, so hopefully this gives an idea of how it looks. The squeaking sound is my wet hands against the plastic housing - not sure what to do about that yet, but maybe some friction tape would fix it. I also have some vid captures below.
Step 16: Plans for improvements
Should I make the Mark 2, I plan on doing the following:
1. Use a water jug with a handle molded into the outer shell. It just so happens that Igloo makes one, which sells for $8 and is otherwise identical to the one I used except for the handle and the fact that the new one is blue, not red. I already have one ready to go.
2. Find a way to turn the camera on/off and record/stop without having to open the housing. My camera has an optional wired remote: I would have preferred a wireless remote, which I could seal in a plastic bag and aim in through the lens, but I am stuck with the wired one. My thought is to use a coax-cable grommet to make and seal a hole in the side of the housing, and put the remote and its cord in some watertight tube. I havent figured this all out yet, but when I do, I will post an update.
3. Complete the lens assembly before putting the PVC fitting into the jug, as noted above.
4. Improve the ballast arrangement to try and get the housing to neutral buoyancy.