Considering the large public financial contributions regarding thermostat valves for radiators given during the seventies and the beginning of the eighties, these valves were easy to bring into the market. In principle no one had to make any efforts developing the technology. The long-run function of the valves were not much contemplated. As a consequence many thermostat valves ceased to work after some years, which led to an uneven in house temperature in properties equipped with such valves. Uneven heating firstly causes complains among the living, secondly causes trouble with increased energy consumption. The reason for this was that the proprietors were forced to increase the feed-temperature of said heating media to those flats getting a feeding temperature which was too low.

Quality thermostat valves were equipped with a bellows containing gas, comprising a thermostat part. These valves, however, did not have much of an ageing resistance regarding the valve part. They, however, contained a weak spot, namely the stuffing box. This part is mainly un- changed today compared to said time period. The stuffing box unit is construed as a threaded cartridge, named the stuffing box housing, containing a stuffing box sealing material, often an 0-ring made of rubber, often integrated with the valve spindle. All or almost all manufactures have missed this weak spot. It was not contemplated that oxygen diffuse along the spindle and beyond the stuffing material

that is used, often an 0-ring made of rubber. Ions of iron in the water in a pipe-system containing steel walls, react very easily and very quickly with the smallest amounts of oxygen in the system. During these conditions magnetite or black ore is formed. Newly formed magnetite has a strong tendency to deposit on the nearest free metal surface. It is thereby contemplated that the oxygen that has diffused along the spindle of the valve forms a growing layer of magnetite on the metal surface of the spindle. After a long hot summer, maybe four to five years after the application of such a valve, the described process has propagated to such an extension, that the valve spindle, which by all manufactures known, has been chosen to be kept pressed to the bottom when the valve is closed, is no longer able to move outwardly when the thermostat body releases the spindle at the start of the heating season. By that time such a thick layer of magnetite has accumulated on the spindle that it gets stuck in the stuffing-box. The only way to free the spindle is to frenetically knock on the same so that it may regain ability to move freely again.

The alternative is to unscrew the old stuffing box unit and apply a new one. Valves for this purpose, which until today has been used in connection with thermostat valves for radiators, have not changed very much over the time. Some manufacturers have chosen to increase the force from the reaction spring in the valve spindle, while others have chosen to diminish the diameter of the spindle and also to diminish the stroke of the spindle. These variations have not changed the life of the spindle to any essential extent, since magnetite as previously mentioned is both resistant and has a significant adherability to metals.

There is today one larger producer, namely the company Danfoss, who has developed a version of a stuffing box unit containing a small cartridge of grease in order to lubricate the spindle. Fig. 1 shows a schematic view of

such a stuffing box unit, which may be considered as the state of the art. The idea is good but at each movement of the spindle the grease layer is scratched off when passing through the 0-ring in the stuffing box unit, so that after some years the risk for precipitation of magnetite is obvious, as the grease layer successively gets thinner. By experience many of these valves have got stuck after approximately five years. Regarding other kinds of valves, those used at engine valves in municipal heating networks sub centrals, contain the same kind of stuffing box unit as describe above, however, containing stuffing including Teflon@ as a spindle-sealing instead of an 0-ring sealing made of rubber.

For industrial applications valves having a bellows sealing both for ball valves and for valves having a pivoting/ascending spindle have been designed. The bellows in question is, however, a classical folded bellows and methods of more advanced kind is needed for attaching the bellows to the valve housing and the movable spindle. Some- times the spindle has been split up in valves having a pivotable/ascending spindle movement. It has not been possible to find valves, in which the bellows is integrated with the spindle-sealing, as according to the present invention.

The purpose with the present invention is to achieve an arrangement of the kind mentioned above, solving the problems with the present stuffing box units at a valve in a piping system put under internal pressure, having O-ring sealings or the previous kind of stuffing box sealing of another kind, and to eliminate expensive valves containing folded bellows.

This is achieved by giving the arrangement of the above mentioned kind the characteristics according to the characterizing portion of claim 1.

At present the preferred embodiments of the arrangement according to the invention are elucidated by depending claims. By providing thermostat valves with a stuffing box unit which after some years use do not loose its function completely or in part, these thermostat valves will be able to work as long as the thermostat part works in a correct way. A great and yearly re-appearing problem is, that most caretakers after a traditional heating inter- ruption during each summer are loaded with the task to dis- mount thermostat parts in order to enable the loosening of stuck spindles at a large amount of thermostat valves.

The present invention eliminates this problem by making the spindle permanently easy moveable always, and at this portion of the spindle, moving back and forth through the part of the sealing having contact with the oxygen rich surrounding air, does not come in contact with the water inside the piping system, which water contains dissolved iron ions. The reason is that magnetite is immediately constituted when oxygen diffuse along the spindle, as a consequence of the reaction between the oxygen and the iron ions. If infinitesimal amounts of oxygen would have dif- fused through the bellows in the bellows sealing according to the invention, magnetite will precipitate outside the bellows, whereby the magnetite will come off as a conse- quence of the movements of the spindle, which makes the outer diameter of the bellows to vary. A covering layer which theoretically would be able to stop its movement, will simply be blown off by the variation of an diameter of the bellows at normal use. Regarding protecting the sliding partition of the spindle in engine valves to municipal heating networks subcentrals another of the great advan- tages of this invention appears, namely that the bellows tightly fit against the spindle can withstand very high pressures.

The bellows, which can be shaped like a hose, can be compared to a folded bellows that has collapsed and has been pressed flat against the spindle, with the large difference that in this case it is not stiff, but elastic.

The valve will consequently work excellently at high pressures with the thin layer of lubricant, on which the bellows according to a further developed embodiment slides upon. The chosen design further enables the introduction of the bellows in most of presently known stuffing box units for such valves.

In other words, it will be possible to revive all the thousands of thermostat valves which were mounted in many houses during the days when large contributions were given for such investments. The solution is this new kind of stuffing box, which has a very long durability and which is very simple to mount. Stuffing box units today available as a spare part have a durability of approximately five years.

The stuffing box according to the invention has a consider- ably longer durability, which makes it possible to discount the expense in connection with the mount.

Below, the invention is more closely described with reference to accompanying drawings, in which: Fig. 1 shows a conventional stuffing box unit in a length section, Fig. 2 shows a stuffing box in a real embodiment for sale and for a purely axial spindle movement in a length section, Fig. 3 shows a complete stuffing box in a further developed embodiment in a length section, Fig. 4 shows an enlarged portion of Fig. 3,

Fig. 5 shows an end view of a stuffing box according to Fig. 3, Fig. 6 shows an enlarged portion of a stuffing box unit in a length section for a spindle movement of a longer axial kind, Fig. 7A and B show enlarged portions of a stuffing box unit in a length section of the same kind as according to Fig. 6, which beyond fluid pressure from the pressurised piping system can withstand an increased pressure against solid bellows.

Fig. 1 shows a conventional stuffing box unit P for thermostat valves on radiators, which initially when mounted has been loaded with lubricant 2 inside a plastic casing 3. At 1 the magnetite deposit arising is shown, which after approximately five years of use makes the spindle stick to an 0-ring 5.

In Fig. 2 the stuffing box unit P is shown in a basic embodiment, one end of which is fixed to a so called solid bellows 10 in turn fixed to an extended part 20 of the spindle 4 by jamming with a compressed clamp 21.

The stuffing box unit shown in Fig. 3 has a hollow spindle house made of metal, which has an exterior thread to be attached into a valve-cup (not shown). The stuffing box unit P is at usual thermostat valves for radiators combined with a valve-spindle 4. The housing 7 is on its inside designed having a seat 8 for an 0-ring part 9, which is constituted by an elastic specially designed solid bellows 10, which in one end is designed like an 0-ring sealing 9. Due to its over-size and due to the force 12 from the valve-spindle, which is generated by the over-size and permanent existence of the spindle, the sealing 9 is kept in the seat 8. The other end of the valve-spindle 4

has been designed to have a diameter enlargement 13 which fits in an inner cavity 14 in the thinner end of the mas- sive bellows 10. In order to force the end of the bellows 10 against the diameter enlargement 13 of a pivotable valve-spindle 4, a cap 15 has been assembled over the end of the massive bellows 10 by gluing thereto. The cap 15 has been designed having a collar with lugs sliding in groves 22 in the housing for the stuffing box or the spindle 7.

In order to be able to supply lubricant after assem- bly, a portion of a spindle 4 has been provided with a hollow channel 16 with a radially bored penetrating hole 17. The hole-channel 16 is after stuffing with lubricant closed with a plug 28 made of stainless steel. The design allows a friction reduction by the fact that a mid portion 18 of the solid bellows 10 may be expanded/lifted from the spindle 4 by a"pillow"19 consisting of lubricant.

According to Fig. 4 an inner end of the stuffing box P is shown as an enlargement with the clamp 15 keeping the end of the solid bellows 10 with its cavity 14 pressed over the diameter enlargement 13 of the pivotable spindle, as well as the grooves 22 that guides/stops the lugs 23 on the cap 15.

Fig. 5 shows an end view from the fluidum side of the stuffing box unit P. It is shown how the collar lugs 23 of the cap 15 are controlled by the grooves 22 and the spindle house 27, and how a co-rotation of the end of the solid bellows is avoided by said collar lugs 23 when a turning movement is translated from the spindle 4.

Fig. 6 shows a design of the stuffing box unit 3 that allows the spindle 4 to slide freely there-through. A locking ring 24 is fixed in a groove 25 designed for the purpose holding the collar lugs 23 of the cap 15 against the goods of the spindle house 7. This implies that the end 26 of the solid bellows 10 may not follow the actual move- ments of the spindle.

Fig. 7A and 7B show in a length section and a cross section respectively, the later from VIIB-VIIB, a design of the stuffing box unit P that by using an elastic clamp 27, allows an increased pressure on the lubricant pillow 19, which has been pressed in between the spindle 4 and the intermediate part 18 of the solid bellows 10. The housing 7 of the stuffing box P is made of metal, e. g. drawn copper.

The solid bellows 10 is manufactured from industrial rubber and the spindle 4 is manufactured from stainless steel. The clamp cap is made from elastic stainless steel, while the clamp 21 (the embodiment according to Fig. 2) or the cap/lugged flange (embodiment according to Fig. 3) is made from tempered copper.

The housing of the stuffing box P is manufactured having one end slit in a conventional fashion having four grooves and is there beyond threaded in a dimension fitting to a valve cup for valves of said kind, in which cup the stuffing box unit P is supposed to be assembled, preferably thermostat valves.

The outer end of the stuffing box unit in its unas- sembled condition is on its inside designed as an open seat for the 0-ring part 9 of the combined 0-ring solid bellows 10. When assembling the stuffing box unit P the lubricated spindle 4 is mounted inside the elastic solid bellows 10 by pressing it in. The spindle 4 is for this purpose pressed from one end of the solid bellows until it comes into con- tact with the diameter enlargement 13. The outer end of the bellows 10 is forced over the enlargement 13 on spindle 4 until the enlargement 13 gets in position in the cavity 14.

Thereafter a clamp ring 27, which with a suitable assembly tool is opened to its possible maximum over the mid portion of the bellows, after which the ring 15 with its collar lugs 23 is mounted over the small end of the solid bellows and is glued thereto. At this time the collar lugs are folded on to the side of the ring 15 in order to make the

outer diameter thereof smaller than the smallest inner diameter of the stuffing box housing. The assembled parts are with its thinner parts brought through the housing 7 of the stuffing box unit from the end with the open seat. When the solid bellows 10 has been brought into a correct posi- tion, the orifice of the stuffing box unit is shrunk around the 0-ring portion using machining. The folded collar lugs 23 are thereafter risen in such a way that they are con- trolled by the grooves in the other end of the stuffing box unit. Thereafter the stuffing box unit is threaded in a tool which puts the bellows under a sub atmospheric pres- sure from the outside. Simultaneously grease is pressed into the hole channel 16 of the spindle. When a predeter- mined amount of grease has been pressed in, a plug 28, made of stainless steel, is pressed into the open end of the channel 16. The sub-atmospheric pressure counteracts any grease to be squeezed out during the plugging.

The stuffing box unit now ready, is assembled with thread sealing compound or sealing tape in a seat threaded in a valve cup, e. g. on a radiator. The spindle in the stuffing box unit fits in a cavity in the cone of the stationary valve and is during axial movement controlled by the sliding movement of the collar lugs 23 sliding with almost no clearance in the grooves 22 at the end of the stuffing box housing, and that the outer end of the spindle is controlled by the 0-ring portion of the solid bellows 10 and its lateral compression force.

The intermediate portion of the solid bellows 10 be- tween the guides just mentioned enclose a lubricant pillow 19, which by its viscous character gives no mentionable resistance against movement in an axial direction. The function of the lubricant pillow during axial compression of the solid bellows is to simultaneously counteract its crumpling, by bulging the solid bellows 10. The durability of the solid bellows is thereby multiplied as is the re-

sistibility against pressure from a high system pressure, which effects the outside the bellows 10. The diameter enlargement 13 on the spindle 4 also counteracts the risk at the bellows is blown out due to a high system pressure.

The spindle is free to rotate inside the solid bellows when the design according to Fig. 3 is used, while with a design according to Fig. 2 the elasticity of the solid bellows will delimit the co-pivotation of the spindle 4.