Visit us at

Return to

Building the Ultimate Expedition Kayak

Cutting the Hatches

Masking and tracing the perimeter of the day hatch.

Next on the agenda was to cut the day hatch. Since the day-hatch would be located on a very flat portion of the rear deck, I figured it wouldn't matter whether I cut it now or after the freshly-epoxied fiberglass had spent a week curing. Plus, I decided it was best to complete the entire day-hatch before working on the main hatches. This way, I would have an opportunity to test out my plan to use "rare earth" magnets to seal the hatch before deciding whether or not to use the same method to seal the main hatches. Since I have never had an opportunity to see a magnet-sealed hatch up close, I had my doubts about how well the magnets would work, but the aesthetic advantages of an undetectable sealing mechanism were much too tempting to ignore. If it worked, it would mean less hardware cluttering the deck or sticking up where it might bang my knuckles or catch on my PFD during rescues and re-entries.

To get a good, clear reference line for cutting the day-hatch, I masked off the approximate area where the hatch would be located with masking tape. Then I used a compass to trace the perimeter of the hatch. Since I would be using 1/2" wide magnets (1/8" thick), as well as 1/2" wide foam weatherstrip, I decided to make the hatch nine inches in diameter. The one-inch lip needed to hold the magnets and the gasket would subsequently reduce the effective diameter of the hatch hole to about seven inches, which is a fairly standard size for day hatches on commerically-built kayaks, and which seems sufficient to accommodate most of the items I would be likely to place in the hatch (flares, VHF radio, lunch, etc.). It would also be the perfect mini-cockpit size for my four-pound Chihuahua Missy, who sometimes joins me on day-trips, but craves a better place to sit.

The newly-cut day hatch opening.

Once the line was masked and marked clearly, I used my Dremel and its second-tiniest drill bit to drill a series of thin holes for starting the cut. It worked perfectly. After about four holes drilled in series, I was able to slide in the Bosch T101AO fine-cut jig saw blade and commence cutting. Instead of the Skil saw I used when cutting the stations, I borrowed Bryan's Bosch jig saw because it has a pressure-sensitive trigger, allowing me to saw at a much slower, more accurate rate. Whatever you use, be sure to go slow. Also, every few inches you should stop and put a piece of masking tape over the cut to hold the hatch even with the deck. If you don't, it may start to drop or lean, causing your cut-line to end up less-than-clean. Even though I went incredibly slow and followed my reference line as perfectly as my eyes would permit, I still ended up with a small (but tolerable) gap where the cut finally came around and met where I started. Nonetheless, the whole ordeal went more smoothly than I had expected. It seems nerve-wracking to cut into a nearly-finished deck, but cutting the day-hatch inspired confidence and prepared me for what was yet to come with the main hatches.

The nice thing about my day-hatch is that part of it is intersected by one of the deck stripes. As a result, it will be easy to ensure that the hatch cover is perfectly oriented before epoxying in the rare earth magnets. Once the magnets are in place (assuming they are perfectly aligned), they should always pull the hatch cover into perfect alignment with the rest of the boat. That way, the stripe will always look fair and flowing, even where it passes over the hatch.

Using "Rare Earth" Magnets to Seal Kayak Hatches

The magnets used to seal a kayak hatch must be much stronger than the typical refrigerator magnet. Specifically, Neodymium or "rare earth" magnets are required. I found an excellent discussion of "rare earth" magnets on the magnets page of the Lee Valley Tool website (the source from which most kayak-builders purchase their rare earth magnets). For the inside scoop on using magnets to seal kayak hatches, however, I had to scour the Internet and repeatedly pester the folks at the Kayak Forum bulletin board before I even began to fathom how to accomplish this simple-sounding task. Lucky for you, I'm going to spare you the trouble and spell out my findings in detail so that other first-time builders can get their bearings on this issue far more easily than I did.

First, as far as I can tell, Rob Macks of Laughing Loon was one of the first kayak-builders to use rare earth magnets for sealing hatches. The advantage of magnets, of course, is that when used properly, they require no additional hardware to hold the hatches down. Thus, you can preserve the clean lines of your hand-crafted kayak and obtain a very good, water-tight seal at the same time. In fact, in many cases, magnets produce a better seal than the traditional two-straps-over-the-hatch approach, which often doesn't produce sufficient downforce to seal hatches adequately—particularly on flat-decked kayaks. In the case of magnets, the attractive force between the magnets pulls straight down against the gasket, ensuring good downforce and sufficient compression of the seal to keep out water.

Rob has recommended using 1/2" wide (1/8" thick) rare earth magnets spaced three inches apart all the way around the perimeter of the hatch cover, with a matching set spaced identically around the rim of the hatch opening. A neoprene gasket or foam weatherstrip running all the way around the hatch provides the seal. Some builders use thin gasket material (about 1/8" thick) directly on the hatch rim, while other builders use thicker gaskets (1/4" to 1/2" thick) and recess half of the thickness of the gasket into the rim (producing a theoretically superior seal). Both methods seem to work fine, but the recessed gasket approach has the advantage of allowing the hatch to sit perfectly flush with the rest of the deck. Whichever method you choose, the magnets must be offset from the gasket material so that the closed hatch allows direct, magnet-to-magnet contact rather than sandwiching the gasket material between the magnets. This practice yields maximum attractive force from the magnets and also ensures the gasket is evenly compressed. Because magnets rust easily (and quickly), they must also be covered with one layer of 3 to 6 oz. fiberglass to seal them from the water. Merely coating them in epoxy will not be sufficient over the long term because epoxy alone will wear off much, much quicker.

A lot of builders using Rob's recommended magnet arrangement have estimated the approximate holding force of their magnetic hatches at 10 to 15 pounds—plenty of force to crush a soft gasket and effectively seal out water. However, a few builders recommend using 3/4" wide magnets (still 1/8" thick) on kayaks intended for extreme rough-water use. Naturally, the larger magnets produce a stronger bond, so if you consider "upgrading" to larger magnets, be aware that you may need to use a small pry bar or attach some kind of heavy-duty pull handle to get your hatches open. It's probably best not to use these larger magnets (or at least not too many of them) on day-hatches, as they could make it terribly difficult to open your day-hatch while out on the water, thereby undermining its purpose. In my case, I decided to use 1/2" magnets to seal the day-hatch and 3/4" magnets to seal the main hatches. Getting sufficient attractive force shouldn't be a problem. The holding force in rough conditions also shouldn't be an issue because the hatches will be flush with the deck, leaving virtually nothing for wind or water to "grab" and pry off. Moreover, because of the substantial attractive force exuded by the magnets, most magnetic hatch covers will jump right back into place again even if something does momentarily displace them.

If you're especially concerned about the attractive force of the magnets, you can further improve their strength by backing them with steel washers of equal or greater thickness than the magnets. This may not be practical in the hatch cover considering the already thin nature of the wood, but you could at least consider adding a steel backing to the magnets in the rim of the hatch opening because the rim can be built as thick as you like. Alternatively, if you want absolute protection against losing your hatches to the surf or to the wind on the Interstate, you could consider running traditional straps over the hatch cover to prevent it from ever falling off or shifting beyond the attractive range of the magnets. In my case, I'm considering running the bungee-cord deck-rigging over the edge of the hatches for precisely this reason. To open my hatches, I will simply stretch the rigging out of the way and lift the hatch.

To save money on magnets, some builders have contemplated using magnets in the hatch cover only, and using metal washers on the rim (or vice versa); however, as Rob has pointed out on various forums, this is a bad idea. The advantage of using pairs of magnets, instead of magnets and washers, is that the paired magnets will automatically align at the poles. Thus, if you align the magnets properly during installation, they will always automatically orient the hatch cover perfectly, thereby preserving the direction of the wood-grain or any portion of the deck design which happens to run over the hatch.

The underside of the rear hatch with 3/4-inch magnets evenly spaced around the perimeter. A layer of 3.8 oz. silk-weave fiberglass over the magnets prevents them from rusting in the water.

To recess the magnets into the wood, builders have come up with all sorts of different methods. But the most common techniques seem to be these:

  • using an appropriately-sized forstner bit in a drill, and using a custom-made collar or "stop" to limit the depth of the cut (my method of choice)

  • using forstner bit in a drill press, and setting the stop on the drill press to limit the depth of the cut

  • using a router with an accurate depth-control setting, and using care to keep the router from "grabbing" the wood or cutting awry

  • using a rotary tool (such as a Dremel) and a steady, steady hand

Personally, I used a drill with a forstner bit (1/2-inch bit for 1/2-inch magnets; 3/4-inch bit for 3/4-inch magnets) and controlled the depth by boring out little bits at a time. In other words, I let the bit spin a few revolutions, then test-fitted the magnet. If I needed to go deeper, I gave it a few more spins of the bit. Just two or three spins of the bit will remove a noticeable amount of wood when you're only trying to drill a 1/16-inch to 1/8-inch deep recess, so proceed slowly. While this method is mildly tedious and slightly time consuming, it makes it easy to achieve a fairly accurate, consistent level of depth without needing to build any sort of custom, depth-limiting collar or "stop" for your drill. Just be certain to hold the drill perfectly perpendicular to the hatch or you may end up with slanted, uneven recesses. Truth be told, a little imperfection shouldn't be a problem if you plan to mold the retaining lip out of fiberglass as I did because the lip should take on the exact shape of the hatch cover. In fact, the reason the magnets should be recessed into the hatch cover first is so that the lip can be molded for a perfect fit, and so that the magnets will align exactly as you want them to.

The underside of the day-hatch after being strapping-taped in position, covered in plastic wrap, and aligned with the temporary 1/8-inch gasket.

Since the hatch cover is fairly thin, I decided not to recess my magnets fully, but to leave them sitting slightly proud of the lip, by about 1/32 of an inch. The thinking here was not only to prevent accidentally drilling through the hatch cover, but also to ensure that mild, corresponding "dips" would be molded into the hatch retention lip. This way, the hatch cover magnets would sit down into these dips later, further ensuring that the hatch cover could not spin or twist out of place. Of course, the potential draw-back of not fully recessing the magnets is that it encourages air bubbles to collect around the edge of each magnet when you cover them with fiberglass later. Nonetheless, as long as the magnets only sit slightly higher than the rest of the hatch, the air bubbles should be minimal and chipping should not be a problem. Besides, if you do end up with large air bubbles, you can always inject them with epoxy later.

Since I had ordered twenty 1/2-inch magnets from Lee Valley Tool, I decided to use all of them on the day-hatch: ten in the hatch cover and ten in the hatch rim. This allowed me to position the magnets approximately 2.5 inches apart, all the way around the perimeter of the hatch cover. Once they were all positioned in place, I covered them with a single layer of 3.8 oz. silk-weave fiberglass to protect them from water penetration. Since the magnets are so strong, thicker fiberglass would probably also work fine, but in order to maximize the holding force of the magnets, I decided thinner was better in this case.

Laying the first layer of fiberglass for the hatch rim.

A day later, when the hatch cover finished curing, I strapping-taped it back into place on the deck, using care to keep it perfectly flush, and perfectly aligned with the rest of the deck pattern. Use heavy-duty, fiber-reinforced strapping tape or some other non-stretch tape (not duct tape or masking tape) so that the hatch covers remain perfectly flush with the deck even when sand is poured on top of them later; otherwise, the tape will stretch and the hatch covers will end up sitting slightly proud of the deck. Once I felt confident the hatch cover could not shift out of place, I turned the deck upside down and laid down Saran Wrap to serve as a mold release. I taped the Saran Wrap into place in order to minimize wrinkles and prevent it from shifting during the lay-up of the rim. Since I intended to use a finished gasket size of 1/2" wide by 1/4" thick, I next put down a disposable piece of weatherstrip measuring 1/2" wide by 1/8" thick. This would create an 1/8" recess around the rim, making it easier for the hatches to fully compress the final gasket.

After positioning the weatherstrip carefully to align with the perimeter of the hatch opening, I borrowed some of Bryan's 3.8 oz. silk-weave fiberglass to use as the first layer covering the rim magnets. I cut the fiberglass into small strips, about 3 inches wide and 4 inches long. I wetted out each strip first with epoxy, then laid them down one-by-one all the way around the rim, making sure to create a lip wide enough to cover the magnets and gasket, as well as to allow at least a two-inch wide overhang to bond under the deck. With this first layer in place, it was time to set the rim magnets in place. Because the corresponding magnets were already recessed in the hatch, the rim magnets jumped into perfect alignment. The remaining lay-up sequence went as follows: three layers of 5 oz. fiberglass, one layer of 5 oz. carbon fiber, and three more layers of 5 oz. fiberglass. With all of the layers in place, I completed the project by laying down one last layer of Saran Wrap and then pouring about a gallon of sand over the whole area to help force the layers of fiberglass to lay tight against the magnets and gasket. The low-tech version of vaccuum-bagging: sand poured over the newly-fiberglassed hatch rim to force everything into place.

Two days later, when the lip was sufficiently cured, I pried it off the mold and trimmed the outside edges and the inside opening to a smooth radius. Several large air bubbles were visibly trapped in the cured glass, so I used a tiny drill bit and an eye-dropper to carefully pierce and inject each bubble with epoxy. I also used a mixture of epoxy and sawdust (mixed to peanut-butter consistency) to fill in any gaps around the disposable weatherstrip to ensure a consistent, well-defined channel all the way around the hatch. A day later, when the epoxy had dried, I sanded the injection holes and the epoxy-and-sawdust filler smooth again, scraped out the disposable 1/8-inch weatherstrip, sanded the resulting channel smooth, and painted the whole rim black to give it a sleeker, more finished look. The paint will also serve another purpose in the long term by making it easier to spot signs of wear on the rim before water penetration to the magnets becomes a threat.

A few minutes of playing with the rim and hatch confirmed several things: First, the hatch bonds to the rim hard enough to give a painful pinch if you aren't paying attention to what you're doing. Second, the holding force is certainly more than sufficient to keep the hatch in place during calm conditions. Third, the bond is strong enough to crush a soft-compressing gasket like weatherstrip foam (which is what I intend to use), but not strong enough to compress anything as firm as high-density rubber, and probably not strong enough to trust to stay put while paddling in crashing waves or ocean surf. Even so, I think the hold is just right for a day-hatch: strong enough to seal, but not too strong for convenient access with one hand during regular paddling. If I were to do it over again, I might consider backing the magnets in the hatch rim with steel washers to help strengthen their attractive force, but since I don't know how much this would increase the bond, I wouldn't necessarily recommend it to other builders. It's possible even a slight increase in strength could make the hatch too difficult to open with one hand. At present, I would estimate that my finished day hatch has about 10 - 12 pounds of holding force.

Although I was satisfied with the results I obtained on my day-hatch, I was certain I wanted a much stronger bond for the main hatches. Since the main hatches are far out of reach while I'm paddling, I need them to stay put in even the roughest conditions. For this reason, I decided to use stronger, 3/4-inch wide magnets on the main hatches. The spacing for the magnets varied slightly from 2.75 to 3 inches all the way around the hatch, leaving me with 16 magnets in the front hatch and 18 magnets in the slightly longer rear hatch, plus a matching set of magnets in each rim. All in all, this added up to 70 magnets. Knowing that the bond these magnets produced would be noticeably stronger, I decided to use thicker gasket material. I used 1/2-inch wide, 1/4-inch thick weatherstrip to mold a 1/4-inch deep channel into each rim so that later I could use 1/2-inch wide by 1/2-inch thick weatherstrip to provide the finished seal. This meant that the magnets would only need to compress half the thickness of the gasket.

A view of the underside of the deck with the finished hatch stops epoxied in place.

Once again I used 3.8 oz. silk-weave fiberglass to cover the magnets in the hatch and also to provide the first layer of glass in the hatch rim. Afterward, I laid down the magnets and then used the same lay-up sequences as before: three layers of 5 oz. fiberglass, one layer of carbon fiber, and three more layers of 5 oz. fiberglass. As before, I covered the whole area with another layer of Saran Wrap and then poured sand over the entire area to compress everything in place while the epoxy cured. When the process was finished, I was delighted to find that the larger magnets had produced a powerful bond which I would estimate at approximately 20 - 25 pounds of holding force, perhaps more—enough to pick one end of the kayak up off the ground!

A view of the grab loop on the front hatch. The small loop is made using Joe Greenley's soft padeye technique, and provides an easy way to tug open the hatch.

With the magnet work finished, all that remained was to sand any rough edges, clean out the gasket channels, and epoxy each of the hatch stops into place. To make sure I had everything aligned correctly, I wrapped each hatch in Saran Wrap and taped all of the hatches back into their original position on the deck, taking care to ensure my deck pattern was perfectly aligned. Then I turned the deck over and glued the hatch stops into place with a mixture of epoxy and sawdust that was mixed to about the consistency of peanut butter (referred to as "dookie schmutz" by some builders). If your hatches are taped in place securely, the magnets will align and hold the hatch stops down securely during the bonding process. Nonethless, I laid another layer of Saran Wrap over the hatch stops and poured sand over them to ensure they would be pressed firmly against the deck until the dookie schmutz was fully cured. Be sure to use a liberal amount of dookie schmutz for this step. A thick layer of dookie schmutz ensures that every little crack and space between the hatch stops and the deck is filled in and sealed, effectively fusing them into a single, solid piece. If you're particularly concerned about durability, it may be advisable to lay fiberglass all the way around the hatch stops also, but I decided not to do this since the dookie schmutz seems to bond rather tenaciously.

Once the hatch stops were in place, I replaced the hatch covers and sanded any raised edges to make them appear more flush with the deck. This of course meant adding another layer of fiberglass wherever I happened to sand through the original glass, but I decided the extra effort was worth it to make everything look nice. One little tip while I'm on the subject: Be sure to put some kind of dependable handle on your hatch covers before you put them back in place or you'll have a hard time prying them off again. Trust me, the magnets hold very tightly, so you'll need a good handle. I cut a small slit near one edge of the hatch and routed a loop of webbing (melted the same way Joe Greenley makes his "soft padeyes" for the deck rigging) up through the slit and sealed the back with silicone sealant to waterproof it. This makes a nice soft grab loop that is easy to grab for opening the hatch, but that won't scratch, bang, or slice hands and flesh during rough-water rescues. It also lays perfectly flat and out of the way when you strap gear onto the deck, such as a spare paddle. If you use this approach, be sure to position the grab loop far enough in from the edge of the hatch so that the back of it does not get in the way of the hatch rim and interfere with its ability to shut tightly.

Previous Page       Next Page

Return to