GB2427627A

GB2427627A - Internally supported vacuum panel

Info

Publication number: GB2427627A
Application number: GB0616788A
Authority: GB
Inventors: Michael John Rickards
Original assignee: Individual
Current assignee: Individual
Priority date: 2005-05-24
Filing date: 2006-05-22
Publication date: 2007-01-03
Also published as: GB0616788D0; GB0510518D0

Abstract

A vacuum panel comprises first and second boards 11, 12, spaced apart by a peripheral wall 13 and by an internal wall 14 or central rod (not shown). The space within the panel is evacuated to less than 100 Pa. The height of the internal wall 14 or the central rod may be greater or equal to that of the peripheral wall. The vacuum panel my be square, triangular or hexagonal and the internal wall 14 may be the same shape as the panel. The internal wall 14 may be one third of the length of the peripheral wall 13.

Description

* 2427627 An internally supported vacuum panel This invention relates to

the construction of high efficiency sound barriers, the principles described can also be applied to heat barriers.

In the examples and descriptions that follow reference will be made to the accompanying drawings: Figure land lb show the essential components of a vacuum panel.

Figure 2 shows the basic problem of inward flexing.

Figure 3 illustrates the problem of panel warping.

Figure 4 shows how an internal support affects inward flexing Figure 5a and 5b show the reduction in height of the component walls.

Figure 6 shows the effect of using an internal wall higher than the perimeter wall.

Figure 7a and lb show the effect of a centrally placed supporting rod.

Figure 8 shows the effect of using a central rod higher than the perimeter wall.

Figures 9 and 10 illustrate panels shaped as triangles and hexagons.

That sound cannot travel through a perfect vacuum is generally well known and an effective sound barrier would be a panel, as shown in Figla and ib, comprising upper (1) and lower (2) boards separated by a perimeter wall (3) enclosing such a perfect vacuum. The perimeter wall, however, must be strong enough to support the compression due to atmospheric pressure acting on the upper and lower boards.

In practice a perfect vacuum cannot be attained and the air enclosed in the panel transmits some sound from the upper to the lower board. The attenuation of sound caused by the reduced pressure may be expressed as Attenuation = 2OLog Pint In this expression P represents atmospheric pressure and 1 represents the air pressure inside the panel.

S.. a *.

S S S S S * S * :: ::: * :: Reducing the pressure inside the panel to less than lOOPa offers little advantage since the perimeter wall itself transmits some of the sound through the panel. The fraction of the sound acting on the surface of the upper board and transmitted through the perimeter wall is spread over the area of the lower board. Consequently the overall sound attenuation due to transmission through the perimeter wall may be expressed by the equation Attenuation = 20Log- Aper In this expression 4,d represents the external area of the upper board and A,jer represents the contact area that the perimeter wall makes with the upper board. This equation is used to calculate the theoretical sound attenuation in the examples that follow.

Taking account of the inward flexing, Fig. 2, of the upper (4) and lower (5) boards due to atmospheric pressure it is evident that the perimeter wall (6) must be at least high enough to prevent central contact between the boards. This imposed minimum height in turn demands a minimum thickness to enable the perimeter wall to support the compression due to atmospheric pressure and this minimum thickness will in turn influence the transmission of sound getting through the panel.

Vacuum panels have a strong tendency to warp across a diagonal, Fig 3, and this can lead to rupture of the seal between the boards and the perimeter wall.

As an example, to make a square vacuum panel with 400mm sides, Fig. 2, and upper (4) and lower (5) boards made from 0.8mm thick mild steel it was found necessary to make the perimeter wall (6) 35nun high to prevent the upper (4) and lower (5) boards from touching at the centre when the internal pressure was reduced to less than lOOPa. In order to withstand the atmospheric compression the mild steel perimeter wall (6) had to be made 1mm thick. The sound attenuation measured for this panel, over a frequency range of 100 to 800Hz, was 30 to 35dB, the theoretical attenuation is 40dB. Vacuum panels constructed in this manner have a strong tendency to warp.

The inward flexing of the boards, indicated in Fig.2, can be reduced Fig. 4, by placing a square internal wall (10) centrally within the panel to separate the upper board (7) from the lower board (9), the side of the internal wall being about one third that of the panel. Although this arrangement adds the contact area of the internal wall (10) to that of the perimeter wall (8), nevertheless, as the atmospheric compression is now spread over a greater wall area, the thickness of both the perimeter (8) and the internal wall (10) can be reduced thus bringing the overall contact area closer to that of just the perimeter wall (8) on its own. The internal wall also has the benefit of stopping the panel from warping.

*. *. : *. * * :: ::: :: Furthermore, Fig. 5a and 5b, the introduction of an internal supporting wall (14) allows the height of both the perimeter wall (13) and the internal wall (14) to be reduced because the inward flexing of the upper and lower boards is so much less. The walls, at their reduced height, can now be made even thinner and yet retain sufficient strength to withstand atmospheric compression.

To illustrate this idea, Fig. 5a, a square vacuum panel having 400mm sides was constructed with upper (11) and lower (12) boards made from 0. 8mm thick mild steel separated by a perimeter wall (13) enclosing a square internal wall (14) having 133mm sides made of mild steel and placed centrally in the panel, Fig. 5b. When evacuated to lOOPa this panel required 12mm high walls to prevent contact between the upper and lower boards and a thickness of 0.25mm to withstand atmospheric compression. The sound attenuation measured for this panel, over a frequency range of 100 to 800Hz, was to 45dB, the theoretical attenuation is 50dB.

A further development, Fig.6, is to increase the height of the internal wall (17) and reduce the height of the perimeter wall (16) correspondingly but still preventing the upper (15) and lower (18) boards from making contact. The disadvantage of having to make the internal wall (17) thicker, to strengthen it, and thus increasing its contact area with the boards, is offset by making the height of the perimeter wall (16) much smaller and thinner giving an overall reduction in the total contact area. Vacuum panels made in this way do not warp.

To illustrate this development, Fig.6, a square vacuum panel with upper (15) and lower (16) boards made from 0.8mm thick mild steel and 400mm sides was constructed with a square internal supporting wall (17) having sides 130mm long, 15mm high and 0.3mm thick and placed centrally in the panel. The perimeter wall (16) was 5mm high and 0.1mm thick. The sound attenuation measured for this panel, over a frequency range of 100 to 800Hz, was 45 to 50dB, the theoretical attenuation was 54dB.

Replacing the square internal supporting wall, Fig. 7a and Th, with a centrally placed rod also permits a reduction in the height of the perimeter wall and consequently its thickness and contact area. The crosssectional area of a rod capable of supporting the atmospheric compression in a vacuum panel is generally much smaller than the contact area of the perimeter wall. This type of panel, however, has a strong tendency to warp.

The above approach is illustrated by a square vacuum panel, Fig. 7a, having 400mm sides with upper (19) and lower (22) boards made from 0.8mm thick mild steel and separated by a mild steel perimeter wall (20) 20mm high and 0.5mm thick. The panel was supported by a mild steel rod 20mm high and 4mm in diameter placed in the centre of the panel. The sound attenuation measured for this panel, over a frequency range of 100 to 800Hz, was 35 to 40dB, the theoretical attenuation was 46dB.

* . *** S 55 5** S * * S * :: :: ** :: By increasing the height, Fig. 8, of the central support (24) in a vacuum panel the height of the perimeter wall (25) can be reduced and so made even thinner thereby gaining the advantage of a much reduced contact area. Vacuum panels with a central support having a greater height than the perimeter wall do not warp.

As an example of this development, Fig. 8, a 400mm wide square vacuum panel was made with 0.8mm thick mild steel boards (23, 26) separated by a mild steel perimeter wall (25) 5mm high and 0.1mm thick. The central support (24) was a 30mm high mild steel rod 6mm in diameter. The sound attenuation measured for this panel, over a frequency range of 100 to 800Hz, was 45 to 50dB, the theoretical attenuation is 58dB.

The essence of the invention claimed herein is the construction of vacuum panels with an internal pressure of less than 1 OOPa and with the upper and lower boards separated by a perimeter wall and including an internal support with the object of minimizing the total contact area that the upper and lower boards make with the perimeter wall and the internal support together thus providing the highest level of sound attenuation.

The idea of using an internal support in a vacuum panel is not restricted to square panels. Triangular, Fig.9 and hexagonal, Fig. 10, vacuum panels can all have greater sound attenuation when an internal support is used to reduce the total contact area of their perimeter walls and the internal supports thereby conferring on the panel a higher level of attenuation than could otherwise be attained.

Whereas the descriptions given here refer to the attenuation of sound the ideas disclosed are also relevant to vacuum panels intended for thermal insulation.

Claims

* . *** . S. a.. * S * * U * :: ::: * :: :: CLAIMS

1. A square vacuum panel comprising upper (11) and lower (12) boards, as described herein with reference to Fig.5a and Fig. 5b, spaced apart by a peripheral wall (13) and by an internal wall (14), the internal wall having the same height as the peripheral wall and one third of its length and being placed centrally within the panel with the enclosed space evacuated to less than lOOPa.

2. A vacuum panel as described in Claim 1 but with reference to Fig.6, in which the internal wall (17) is greater in height than the peripheral wall (16).

3. A vacuum panel as described in Claim 1 but with reference to Fig. 7a and Fig. Th, in which the internal support is a central rod (21) of equal height to the peripheral wall (20).

4. A vacuum panel as described in Claim 3 but with reference to Fig 8, in which the central rod (24) has a greater height than the peripheral wall (25).

5. A vacuum panel as described in Claim I but with reference to Fig. 9, having a triangular shape with a triangular internal wall (28) one third the length of the peripheral wall (27) and of the same height.

6. A vacuum panel as described in Claim 5 but in which the height of the internal triangular wall (28) is greater than that of the peripheral wall (27).

7. A vacuum panel as described in Claim I but with reference to Fig.10, which is shaped as a regular hexagon with an internal hexagonal wall (30) one third the length of the peripheral wall (29) and of equal height.

8. A vacuum panel as described in Claim 7 but in which the height of the internal hexagonal wall (30) is greater than that of the peripheral wall (29).

9. A vacuum panel as described in Claim 5 but with the internal triangular wall replaced by a central rod of equal height to the peripheral wall.

10. A vacuum panel as described in Claim I and Claim 5 but with a central rod of greater height than that of the peripheral wall.

* * *** * ** * * * * I * I S * S S SI S S SI :. ::. : *: ::::

11. A vacuum panel as described in Claim 7 but with the internal hexagonal wall replaced by a central rod of equal height to the peripheral wall.

12. A vacuum panel as described in Claim 11 but with the central rod of greater height than that of the peripheral wall.