Hive insulation using cork

There has been considerable interest over the years in the use of hive insulation to provide an environment for the bees that is closer to their natural habitat of the interior of a hollow tree trunk. Reported benefits include reduced expenditure of energy by bees in cold weather and fewer condensation problems, leading to lower colony losses over the winter in insulated hives.

Last year, I was given a colony of bees in a top bar hive and I decided over the winter to buy a second hive. When it arrived at the end of January, it was packed in a large cardboard box, and surrounded by masses of shredded cardboard to protect it in transit. Looking for a use for the cardboard, I packed it in large polythene bags which I fastened around my occupied hive, underneath, on top and on all four sides. When this hive was the first OxNatBees colony to swarm this year, I attributed the health and strength of the colony at least in part to the added insulation. I therefore decided that I should insulate the hive properly for the coming winter – with the expectation that the hive would also benefit from insulation should next summer be as hot as this one has been. My initial thoughts were to use extruded polystyrene, but when Gareth said that he’d bought cork boards from this convenient supplier, this struck me as a much more attractive material.

The thermal conductivity of western red cedar, generally recognised as the best material for hive construction, is 0.11W/(mK). Other woods typically have values in the range 0.14 – 0.3 W/(mK). The thermal conductivity of cork is in the range 0.037 to 0.04 W/(mK) – ie this lower value demonstrates that a given thickness of cork insulates as well as almost three times the thickness of western red cedar and more than four times the thickness of many other woods. Cork also has the advantage of being light, so that the insulation can be improved without a huge increase in weight.

The literature is not consistent in its description of the permeability of cork to gases such as air. Cork is made up of air-filled closed cells, so there is not an obvious route for air to flow through and it is frequently described as ‘impermeable’ to gas. However, gases are able to move through cork by diffusion, so it can reasonably be described as ‘breathable’. Adding it to the outside of a hive shouldn’t cause the problems that might occur (eg mildew on the hive walls) if it was really impermeable.

I therefore ordered 30mm cork board (this was the only size offered by Mike Wye in packs of less than 10) but 20mm would have been fine too. I cut the cork to size using a panel saw and glued on a backing of 6mm plywood (also hand sawn with a panel saw). I taped and stapled the edges with 73mm wide tape to improve resistance to rain of the cut edges of the ply and screwed these composite panels on to the sides and end of a top bar hive. Very little woodworking skill was needed to achieve the results – though someone more competent than I am would have produced a more elegant finish! I placed a sheet of cork on the underside of the hive roof and pieces of another sheet were packed underneath the hive, so each face of the hive was insulated. The photo shows the hive at an intermediate stage, with insulation in place on each of the long sides, but not yet on the ends. The dark cork can be seen behind clear tape, between the thin ply and the hive. The very substantial eke beneath the roof contains additional insulation, consisting of an old down duvet and a pillow case stuffed with washed sheep wool.

Warre with added internal cork lining

An alternative approach would be to place the cork inside the hive, as Gareth and Brian have done. The photo, taken by Paul, shows one of Gareth’s hives. This avoids the need to make the cork weatherproof, but does reduce the internal volume, and is really only feasible parallel to the bars – ie giving a reduction in one dimension. It would still be possible to insulate the roof and below the hive. Gareth has reported an advantage of internal cork in that the bees cover it in propolis, which promotes the health of the colony.

Do the bees appreciate the extra insulation? Who knows – but the pattern of temperature variation on top of the top bars changed following the addition of insulation. Temperature was recorded at both ends of the top bar hive, Temp 1 above the likely location of the brood and Temp 2 at the end furthest from the entrance. The external temperature was recorded beneath the hive and the results are shown in Fig. 1 below. The pattern for nearly four weeks in September was a variation in Temp 1 (blue) between 23°C and 35°C, while Temp 2 (red) was typically from 5°C to 15°C lower than Temp 1, and 0°C to 5°C higher than the outside temperature (green).

The solid vertical line marks the installation of the front and back panels and it can be
seen that Temp 2 jumped almost immediately to be close to Temp 1. A week later, Temp 2 dropped again for a day or two, and then increased at the time the final insulation was placed. It then remained high during a period of warm sunny weather, coinciding with a lot of activity, with some evidence that the bees had started drawing new comb at the far end of the hive. The same behaviour may be seen more easily in Fig. 2 below, a graph showing the difference in temperature Temp 1 minus Temp 2 against time, where the blue lines represent the condition before side insulation, green lines show the period of partial insulation and red lines represent complete insulation.

Detailed discussion

A further representation of the measurements is shown in Fig. 3, where Temp 2 is plotted against Temp 3 during the period before side insulation was in place. The solid line marks the limit of equality of these temperatures and it can be seen that Temp 1 is consistently higher than Temp 2 over the whole temperature range before insulation.

Fig. 4 shows the same correlation after insulation, with green diamonds representing the period of partial insulation and red diamonds the subsequent period of full insulation. It can be seen that Temp 2 is now significantly closer to Temp 1 for higher temperature values than it had been previously.Fig. 5 shows Temp 2 plotted against the external temperature before insulation. The solid line represents equal temperatures, and the values of Temp 2 below the line occurred when the outside temperature was increasing, and the insulation led to a delay in the response of Temp 2.

Fig. 6 shows the results after insulation. Again, a clear difference in pattern between Fig. 5 and Fig. 6 can be seen at higher external temperatures, with Temp 2 reaching values well above the external temperature in Fig. 6.There is therefore some evidence that the bees reacted to the added insulation, at least in the short term and in warmer weather. Although, having said that, it’s also possible that the change in Temp 2 was due to the unusually warm autumnal weather which allowed the bees to start drawing comb at the far end of the hive, underneath the sensor, and that it was a coincidence that the insulation was added at this time. Nevertheless, I’m pleased with the process and am currently making an eke and composite panels of cork and ply for the new hive which I hope will be occupied next year.

More information about the advantages of insulation can be found online. Paul reported a talk in 2015 given by Elaine and Derek Mitchell and the research he described was eventually published in 2016. William Hesbach has a long and interesting discussion on the same topic, posted in 2016.

With thanks to Paul, Brian and Will for comments and suggestions.

This entry was posted in Experimentation, Hives, Research and tagged , . Bookmark the permalink.

Leave a comment ...

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s