US2961164A - Humidity transmitter or relay - Google Patents

Description

F; B. NEWELL 2,961,164 HUMIDITY TRANSMITTER 0R RELAY Nov. 22, 1960 Filed July 15, 1957 2 Sheets-Sheet 1 INVENTOR. FLOYD B. NEWELL BY W W Nov. 22, "19-60 F. B. NEWELL 2,961,164

HUMIDITY TRANSMITTER OR RELAY Filed July 15, 1957 2 Sheets-Sheet 2 FLOYD B. NEWELL BY Q United States Patent HUMIDITY TRANSNHTTER OR RELAY Floyd B. Newell, Rochester, N.Y., assignor to Taylor Instrument Companies, Rochester, N.Y., a corporation of New York Filed July 15, 1957, Ser. No. 672,011

14 Claims. (Cl. 236- 44) My invention relates to improvements in mechanism for transmitting or relaying the response of a humidity sensitive element to relative humidity in the form of energy more suitable for utilization, as in the control of an agent, agents or means having an eifect or eifects upon which depends, in part at least, the state of relative humidity sensed by the said humidity sensitive element.

More specifically, I have devised a novel humidity transmitter having a fluid pressure output characterized by higher degree of linearity, freedom from ambient temperature effects, hysteresis and change of zero, than hitherto known. In the course of doing this, I have developed a novel and improved form of humidity-sensing device or hygrostat.

In essence, a hygrostat may be considered as consisting of a functionally unitary element having a configuration dependent on the degree of relative humidity. For many years, hygrostats of various degrees of structural complexity have been known to the art.

One of the oldest, perhaps the oldest, is the so-called hair-hygrostat, which in its simplest form may consist of a single strand of animal hair so arranged that the elongation and shortening of the hair in response to the relative humidity of an atmosphere surrounding it performs some useful task such as operating a switch, or a valve, or an indicator, and so on.

This response of the human hair to relative humidity is a fact of nature, and is exhibited by many cellular or fibrous substances, both animal and vegetable, natural and artificial, in many shapes and forms. It is no concern, however, of this application to provide a comprehensive listing of materials that may be fashioned and/or obtained in filamentary form and exhibit the property of response to relative or absolute humidity, nor am I concerned with an explanation of the phenomenon involved. Suflice it to say that there are many such materials, and that it is a fact that the dimensions of a properly shaped piece thereof depend on the relative humidity. I find that nylon thread is quite satisfactory, but hair, strips of goldbeaters skin, and the like would also do.

The object of the invention will be clear from the following descriptions and the appended claims.

In Fig. l is shown a diagrammatic sketch of a humidity transmitter in correspondence with the invention.

Fig. 2 illustrates an elevation view partially in section of a transmitter in more realistic form.

Fig. 3 is a view of a humidity sensing element.

Fig. 4 shows in detailed perspective a modification of a temperature compensator and linearity correcting device used in the device of Fig. 2.

Fig. 5 shows a modified detail of the compensator of Fig. 2.

Ideally, filamentary types of humidity sensing elements, being relatively fragile, and limited in their ability to respond elastically to stress, should be permitted to respond to relative humidity without reaction from the various means supporting and operated by the filamentary device.

However, although such filamentary devices are fragile 2,961,164 Patented Nov. 22, 1960 and of poor elastic qualities, in order to realize reasonable response times, filament dimensions other than length must be quite small, and in order to prevent their position from being changed by effects other than relative humidity, they must be restrained and supported so as to permit insofar as possible only dimensional changes of the filaments due to relative humidity.

This is ordinarily done by supporting the filaments under slight tension. Only the minimum tension should be used, and it should remain constant, or nearly so. Hence, very little power is available from the filaments, and allowable power demand on the filament and allowable variations in tension thereof are too limited in the usual control situation to permit the filaments to act directly on the thing to be controlled, or even to control directly the magnitude of an independent control force of any but a modest order of magnitude. Therefore, the power output of the filaments must be reserved for the purpose of detection of the response of the filaments to relative humidity.

According to my invention, the filaments are supported under the minimum tension so that change in length thereof varies the tension, and means is provided which in response to such change in length varies the tension in the filaments by an amount equivalent in magnitude but opposite in sense to the tension change caused by the change in length.

The result is that the filaments are subject to stress commensurate with the limits set by their fragility, size and elastic characteristics, and no power and consistency of performance is lost to prior art expedients such as trains of mechanism, and spring devices, which may introduce undesirable stresses and friction in varying degrees, thus absorbing power and causing inconsistent behavior of the hygrostat due to hysteresis, static friction, etc.

In Fig. 1, moisture sensitive element 1, which may be a hair, a thread, natural or artificial, or equivalent fiber or filament, or a bundle of hairs, fibers, threads, etc., is tensioned between bellows 3 and spring 10, which in turn are fixedly mounted as indicated by hatching 5 in the drawing. The free end of spring 10 carries a baflle 6 which throttles a nozzle 7 to an extent depending on the state of extension of the various parts. Clean, dry air at constant pressure is supplied nozzle 7 via orifice 8a. A relay R may be provided to amplify nozzle back pressure changes due to changes in batlle and nozzle spacing. As shown, relay R is fed from one side of restriction 8a, but may of course be fed from an independent source of air, if desired. Where, as in the device of Fig. 1, the pressure must increase to counteract filament contraction, the relay R should be direct-acting, i.e., have increasing output for increasing nozzle back pressure, unless nozzle 7 were positioned to face the other side of bafile 6, in which case filament contraction would increase bafiie and nozzle spacing and a reverse-acting relay could be used. A suitable direct-acting relay is shown at 31 of U.S. Patent 2,638,874 to Woodhull, and a suitable reverseacting relay is shown at 3 132 in my U.S. Patent 2,487,266, both patents assigned to the assignee of the present application.

In any event, relay R transmits pressures to bellows 3 via branches 8 and 9, which pressures are proportional to back pressure in nozzle 7. The effect of the relay is to increase speed of response to nozzle back pressure, and output in branch 2. r If a relay is not used, the device will still be operative if nozzle back pressure is fed directly to bellows 3, but speed of response will drop, larger nozzle back pressure changes will occur before bellows 3 recould be used to. perform such translation.

' is translated by the bellows into a batfie and'nozzle spacing change such that the bleed from the nozzle is :just

enough to permit such pressure to exist. Obviously, therefore, any subsequent change in lengtbiin the element will allow the'spring to take up slack, if the element lengthens, or will stretch the spring a slight amount if the element contracts. With battle and nozzle arranged as shown, a pressure change in the bellows due to change in baflle and nozzle spacing. will cause the bellows to take up slack in element 1 if the .latter elongates andto slack up on the element, if the latter contracts. The net result is that .the magnitude of the pressure-within the bellows follows changes in element length, and tends to maintain the elementunder constant tension in spite of elongations and contractions thereof. The tension is not maintained exactly constant since 'baffle position must vary slightly to maintain nozzle back pressure at levels corresponding to element length.

Inexplainingthe operation of Fig. 1, it has been tacitly assumed that the elbows or whatever form -of pressureresponsive .expansible chamberdevice be used, i.e., springbiasedv piston, or a Bourdon tube, diaphragm, etc., selfbiased or with additional spring or other biasing means, is, and such would be-the case in practice, sutficiently stiffer, than spring '10, or is so biased or otherwise prevented from elongation save as a result of pressure changing therein or thereon, that the spring 10 contracts or is stretched by elongation or contraction-of the filament without affecting the dimensions of the bellows. Generally speaking, if the bellows 3 were too responsive to the force change in the element, it would extend "(or contract) .too much, the spring 10 'would contract (or extend) too little and the latter would not have sufficient control over the nozzle pressure. The stiflness of the bel lows 3 determines the magnitude'of the change in the pressure output at 2. The stiffer the bellows the greater thepressure change require to balance a given movement .of the filament, Hence it follows that if the bellows is stiff enough not to be afiected by filament length change .andif spring 10 is only strong enough to restrict bafile .movements to those resulting from filament length change,

sensitivity will be high, and the changein pressure due to a given change of filament length will be determined by the sensitivity of the bellows to pressure change. Therefore, the gain in power due' to translation of filament elongationor contraction into fluid pressure may be set at widely different orders 'of magnitude simply by substituting a bellows or other expansible chamber device of a different spring rate for the bellows shown, and changing the supply pressure level to suit the bellows and the range of expansion and contraction desired for the bellows. When I speak of spring rate, I refer to the overall rate which maybe due to the material of the bellows, alone, or of a spring or other means biasing an exipansible chamber or flexible wall device ,to vgive it a spring rate, or a combination of these two. 7

While a baflle and nozzle couple has been shown as themeans for translating the response of the filament Linto pressure changes, any of the numerous translating or detecting devices in use today for similar purposes, such as pilot valves, jet'pipes, and a host of similar devices Likewise,

be replaced by a contact and solenoid valve system controlling the level of pressure in the bellows 3, or, nozzle and baflie, and bellows and air supply could be replaced by a contact system acting to operate an electric motor in one direction or the other to slack-oil or tighten up filament 1 to an extent sutficient to counteract the change in tension due to change in length of the filament in response to moisture.

The prior art is replete with motor control systems capable of automatically following a movable object, and as I make reference thereto merely for the purpose of clearly indicating the scope of my invention, and not to inform those skilled inthe art, sincethey should recognize those possibilities, it is unnecessary to further describe or to illustrate such matters.

In general, for quick, stable and reliable response of a humidity relay of the sort concerned here, thickness and tension of the filament, and hysteresis and friction in moving parts must be kept' at a minimum. It will be seen that the device of Fig. 1 inherently accords with these standards.

In a typical case, with an air source to 20 p.s.i., a throttling range of 3-15 vp.s.i., covering a substantial portion of the humidity range 0l00%, is obtained with an element comprising some twenty parallel nylon threads on the order of 10' long and initially adjusted to have an overall tension of 20 grams or there about (generally the less tension the better, but the more filaments, the more leeway in choice of tension) at an output of 9 p.s.i., using a bellows of psi. per inch characteristic as a follow-up. While the performance of the device Fig. l is reliable, it is non-linear, and also subject to temperature errors if used at temperatures different from that of calibration (generally, the transmitter would be calibrated empirically for a range of humidities at one given temperature).

The foregoing data apply to a typical device along the relatively crude lines of Fig. 1. In 'Fig. 2, however, is shown a more elaborate embodiment of the invention of much improved characteristics, which include compensation for temperature and better linearity.

In Fig. 2, the parts having reference numerals beginning with the numeral 1 correspond to those parts of Fig. 1 indicated by numerals identical with the number following the number 1 in said reference numerals. Thus, Bourdon tube 1'3 corresponds to bellows 3. It will be noted that no separate spring is' utilized. Instead, baffie 16 is elastically mounted on block 41 as sort of cantilever spring applying the desired tensionto elements 11. Elements 11 may comprise a number of strands of nylon threads tensioned between crosspieces or bars 22. The strands may be separate, or parts of one continuous strand reeled a number of times about a pair of crossbars -22 as shown in Fig. 3, but, however constituted, they must be proportioned such that if tension is applied to the cross bars tending to elongate the strands, the tension will be proportionately shared by, each strand. Use of a number of turns of a single strand permits winding of the element 11 on a jig under constant tension, after which cement or equivalent fastening means (not shown) is applied to the turns 21 where they bear on the cross pieces '22 (Fig. 3). After the cement has dried, the element is removed 'fromthe jig, and where nylon has been used, the uniformity of material and tension results in an element of very good consistency of behavior.

The elements 11 comprise what is equivalent to element 1 of Fig. 1 folded back on itself, so as to include the given strand length in a smaller longitudinal space. Accordingly, the elements are connected attheir lower ends to a connector generally indicated by the numeral 23o by means of'hooks 24 and 24a in the appropriate holes (not shown) in the extremities of the connector 23.

other forms of energy than that of a fluid pressure supply 7 Another extremity of connector23 is anchored by a fiex- 'ible' strip, flexible band or other pivot 25 permitting angular movement in the plane of the drawing, and about an axis determined by the pivot of =theconnector with- V out permitting upward movement of the connector 23. A flexible type of pivot is preferred since such introduces no bearing friction and can be made sufiiciently flexible so as not to interfere with filament motion although nevertheless constraining the connector to substantially only angular motion. The pivot 25 is anchored to a foot 26 extending outward from a support 27. The support 27 is fixed to a horizontal portion 14a of a base 14 by means of nuts 28 or equivalent means permitting upward and downward adjustment of the support 27.

One element 11 is connected to the end of baffle 16 by a hook 29, and the other element 11 is connected by a link 30 to a hook 31, this last being part of the adjustment device 55. The adjustment device 55 is in turn supported by the horizontal upper portion 32a of an angle 32, said angle having a vertical portion 32b slidingly mounted in the slot 34 of an arm 33, a set screw 35 or other means being provided to lock the angle into adjusted position relative to the slot. The adjustment device 55 may be raised and lowered by advancing or backing ofii nuts 36 which retain the upper end of the hook in adjusted position with respect to the portion 32a of angle 32. Arm 33 is provided with a portion 37 having a curved slot 38 therein and held to plate 39 of tube 13 by means of set screws 40 passing through the slot 38, so that by unloosening screws 40 arm 33 can be adjusted about the approximate center of curvature of tube 13. The nozzle 17 and branches 12, 19 and 18, are incorporated in a block 41, which, as shown, serves to support baffle 16. A restrictor 18a is provided in branch 18.

Block 41 is secured to the base 14 and tube 13 is connected to the branch 19 in the block by a piece of tubing 19a. To permit bodily adjustment of tube 13 left and right, it is mounted by a slotted bracket 44 which is held to the block by set screws 45.

Nozzle 17 is threadable in and out of block 41 in the vertical directions for adjustment purposes and lock nut 17a is provided to lock it in adjusted position. Likewise, baffle 16 is made adjustable in the same directions as the nozzle by having the fixed end of the baffle formed with a slot 16a through which pass set screws 16b which secure the baflle to its place of support, in this case, block 41. As is evident from Fig. 2, block 41 must have a somewhat irregular shape to accommodate the various air passages, Bourdon tube, battle and nozzle in operative relation. Obviously, the Bourdon tube and bafile could be mounted directly on portion 14b of base 14 independently of the block incorporating the various air passages.

It will be noted that no booster relay' R is included in Fig. 2. Such may be added if desired, of course, and/ or such a device may be interposed between the final load and transmitter output. For example the output branch 12 would be closed, and a booster relay would be inserted between branch 19 and tube 13, tube 13 being actuated by booster output pressure. The load on the transmitter-the motor of a control valve, or the likewould also be connected to the booster output rather than to the branch 12.

Portion 14a of the base is perforated at 42 and 43 to permit the elongations and contractions of the elements 11 to be transmitted to the baffle and the motions of tube 13 to be transmitted via the elements to the baifie.

It will be seen from the foregoing, that the elements 11 correspond to element 1 of Fig. 1 folded back on itself, and that the essential elements of Fig. 1 are functionally practically identical to corresponding parts of Fig. 2. Fig. 2 however, exhibits a number of refinements and additional features not possessed by the device of Fig. l.

The base 14 in addition to supporting the relay structure also acts to prevent the mingling of air from nozzle 17 with the atmosphere to which the elements 11 are exposed, portion 14a thereof forming a barrier.

As shown, connector 23 forms a sort of double-armed lever or bell crank. The effective arms of the crank extend from the hook and eye connections to the pivot of the connector forming the angles 0 and 0 with the vertical, as indicated by the dashed lines in Fig. 2. If the sum of 0 and 0 were vertical displacement of hook 24a would cause a proportional displacement of hook 24. However, in the present case the sum of 0 and 6 is less than 180, from which it results that the ratio of a given increment of motion of hook '24 is not constant but is different for difierent angular positions of the connector 23 about the pivot axis defined by pivot 25. Thus the displacements of the one hook are nonlinearly related to the displacements of the other hook, and the amount of non-linearity increases as the sum of 9 and 0 is decreased. With the configuration of Fig. 2 the non-linearity of the bell crank action is in the opposite sense from, and hence has a compensating efiect on the non-linear relation of relative humidity to elongation and contraction of the elements 11. To vary the eifect, screws 49 and 50 are loosened and arms 46 and 47 shifted up or down to decrease or increase, respectively, the angle sum. As is evident in Fig. 2, screw 49 passes freely through a suitable hole in strip 47 into a tapped hole 49a in strip 46, and screw 50 follows the same scheme with respect to strip 46, and tapped hole 50a. Between the vertical portions of the strips and providing vertical adjustment leeway is a slot 48a in plate 48. Plate 48, in turn is connected to the foot 26 by pivot 25.

For a given vertical displacement of hook 24a say upward, angle 6 decreases, but at an increasing rate, while 0 increases, causing displacement of hook 24 downward at an increasing rate. That is, repetition of the said given displacement upward of hook 24a while angle 6 decreases and angle 0 increases, results in a progressive magnification as 0 decreases of the vertical displacement of hook 24, with the displacements of hook 24a being non-linearly related to the resultant displacements of hook 24.

Insofar as explained, the transmitter is subject at times to a considerable temperature effect. The temperature of the atmosphere to which the sensing element is exposed has a twofold effect on the response of the element. First, there is the thermal expansion of the element (and of the other parts of the transmitter, of course). This may vary considerably depending on the materials, and layout of the transmitter. For example, the thermal expansion of human hair is reputed to be negligible relative to the effect of humidity. However that may be, I find that the thermal expansion of a nylon element is a large fraction of the dimensional effect produced. by' change in relative humidity, and may require compensation.

Second, the manner in which the element responds to relative humidity depends on its temperature. This is apparently true for all materials in the nature of hair or nylon.

These temperature effects cause changes in zero and in span of response, or, alternately, for any given value of relative humidity, the change in element length between such value and zero relative humidity varies with the temperature.

The connector 23, as seen in Fig. 2 includes bimetallic strips 46 and 47 vertically adjustably mounted as explained before to plate 48. Eye plates or eyes 51 and 53 adjustably mounted on strips 47 and 46, as by set screws 52 and 54, respectively, are engaged by hooks 24a and 24, the direction of eye adjustment being along the length of the outer portions of respective strips 47 and 46, slots such as 47a allowing this.

Where the transmitter is to be used over any substantial range of temperature, the effect of temperature may be compensated for, if arms 46 and 47 are formed of bimetal. The change of zero, i.e., element length, requires a simple displacement lengthwise of the elements Correction. r

counter to the effect .of their thermal expansion, which displacement results from an angular displacement of the arms of 1 the. compensator, ;a nd1,the.,change, in spam i.e., the'variation in 'amount'stheithreadexpands, for a given change in humidity with the temperature varying, requires a change in the radial length .of theeffective arms of said crank; Hence, .compensation for the vtwo temperature effects is provided as the components of one compound motion,'e.g., that of the bimetallic strips 46 and'47,'whicl1- behave similarly and cumulatively. The more active of the metals in the strip will be on the bottoms'thereofso that both bend upwardly on increase of temperature; Hence,1the one strip ,will be a reversed version of theother;

It shouldbe pointed out here that no precise mlesfor compensation can 'be laid down such as would apply exactly to specificcases, since when it comes toparticulars, the general characteristics and compensation needs form a broadscheme theoutlines of which guide a program'of judicious cuttingand trying. Hence, I do not intend that the scopeof my invention is to be takenias limited to a particular form of compensator, since, while the exact application of the principles of temperature and non-linearity compensations may depend on the structural form of the compensator or compensators, the principles do not, yet are sufficient guidance to those: skilled in the art for working out particular cases.

In the species of Fig. 2 humidity elements and temperature compensator elements, are more or less duplicated. The primary consideration for so doing is to conserve space, i.e., the combined response of the elements 11 is equivalent to that of one element' twicethe length of one of elements 11. The bimetallic strips, are duplicated to get suflicient temperature compensating action from strips of the form shown. A second consideration for duplieating the elements 11, is to reap the advantage of having the detecting follow up elements (i.e., Bourdon tube, batfile, nozzle, etc.) located together, a convenience, for example, when the humidity sensing element is to be located in a given space and the remainder of the instrument outside such space. Obviously, this latter advantage would be conserved were one bi-metal strip and one element 11,. replaced by, respectively, attemperature insensitive arm or strip, and a humidity insensitive connecting element, or the like, if the response of one element 11 only were desired and only one bimetallic, strip were needed for temperature compensation.

In Fig. 4 is shown a connector 123 corresponding to the connector 23 in whichthe temperature compensationis introduced by way of a bimetallic coil 60. In this form, onlyone piece of bimetallic material is required to introdue the amount of temperature compensation neededby the two elements 11. Obviously, if an extreme amount .of

movement is necessarya plural-tum coil could be utilized, or a one or pluralturncoil could be duplicated analogous to connector 23.

Coil '60 is fixed to a bracket 61, which in turn istfastened by set screw 62, or equivalent, to one portion of a slotted lever,63. Screw ,62, or whatever fastening device may be used should permit angular adjustment of coil 65 about anaxis through the screwpand the point of connection of hook 24 and the lever, which point isa hole 64 in the free end of coil 60. ,The other end of ,thelever. 63 has a hole 64a for the receipt of the remaining hook 24a.

.Set screw.65 secures lever 63 to. block .66which is secured to, or is part of bracket .67. in turn-brackeL 67 is securedto slotted plate v70.by means of set screw. 69 and block 68. Slotted plate 70-is pivoted to plate 72 by flexure pivot7'1,.and.plate 72 is mounted o-nrfoot 26; The connector 123 is incorporated into the transmitter structure-exactly like the connector 23 of Fig. 2, and performs the equivalent functions of temperature and non-linearity Adjustment is carried out by :uhloosening set screws 65 and 69, thus permitting lever 63 to be bodily shifted up or down, left, or right, blocks 66 and 68 being keyed, as at {74 and 75 respectively, into respective slots 76 and 77 so as to restrict these adjustments to straight line pathand to keep plate 7 0 and lever :63 in mutually perpendicular relation. As mentioned earlier, coil 60 can also be adjusted angnlarly about pivot 62.

In Fig. 4, the adjustment of the compensating coil does not shift in appreciable amount theplace'where hook 24 would connect to the coil. Somewhat the same result may be reached with the compensatorof Fig. ;2, if the slight modification of Fig. 5 i applied. In Fig. 5, strip 147 corresponds to strip 47, and eye or eye-plate 151 to eye plate 51. However, in addition to a slot, 147a {in strip 147, eye .151 also has a slot 153, and thenut and bolt fastener 152 has a washer 154 separating strip and eye. In order to locate eye plate 151 roughly respect to strip 147, a hole 155 is provided in the strip loosely receiving as shown one endiof eye plate 151. -When.eye and strip are clamped together by fastener 15 2, there should be sufiicient clearance in hole 155'so that in:no position of fastener 152 will any bending of strip 147 be applied to the eye plate except at the fastener. In tbis'case, instead of moving the eye relative to the strip for adjustment, .and thus shifting the hook in the eye, the fastener 152 is shifted along the slots, to vary the point of connection of eye to strip, and hence the amount the eye is deflected by the strip. The strip, of course, changes curvature with temperature, and the farther the point of connection with the eye is from the place where the strip is secured to the remainder of the connector, the more motion for a given temperaturechange occurs.

The calibration and compensation adjustment of a device such as shown are necessarily somewhat complex and time consuming. One general and essential consideration which must always be honored, is that whenever the relative humidity, or the temperature, to which the instrument is exposed during calibration and adjustment, is changed, the instrument must be permitted to come to equilibrium at the new value of humidity and temperature.

Also care should be taken not to stress the humidity sensing elements more than is absolutely necessary, nor to expose them to extremely low humidities,since the characteristics of the elements may be altered, except that inthe case of nylon, it is found that alternate baking and parboiling, as it were, is a good technique for conditioning the elements prior to incorporation in the relay.

vThe first step in adjustmentis to set the desired. force on the elements by substituting a 20 gram weight for the pull of the elements on the bafile, and adjusting thebafile up or down and/ or screwing nozzle cap. 17 up or downuntil the output of the transmitter is 8-10 p.s.i.,with a 20 p.s.i. air supply, so that if nozzleand bailie-are working properly, removingthe weightwill send output to within a few pounds of supply pressure, whereaspulling the baffle awayfrom the nozzle will drop output pressure to 1 p.s.i. or less. 1

Now,-the weight having been removed, the righthand element .11 is connected between baffle and one end of the connector (23 or 123), and support 27, is adjusted vertically to remove slack fromthe element and straighten plate .48 (it is to .be understood, that, unless it is stated otherwise, the procedure applies tov both connector 23 and connector 123, although, for convenience, only connector 23 and parts thereof will be referred to ordinarily).

The other element is then installed and is adjusted about the Bourdon tube axis by meansof slotted portion 37 (coarse adjustmentland the height of .hook.31

'is adjusted by nuts 36 (fineadjustmen-t), until output is at a known low humidity, say about 20% at 80 F., and transmitter pressure output noted, order to determine if the pressure output at 20% is of suitable value. For

Xample, if the useful range of variations of output pressure were 10 p.s.i., and this were to correspond to a range of 100% relative humidity, then the output at 20% should be 2 psi. plus minimum output (and at 80%, same temperature, 8 psi. plus minimum output). In any event, should the output pressure at 20% be other than desired, adjustment of arm 33 and/or hook 36 as aforesaid will bring the pressure output at 20% to the proper value. The instrument is now exposed to a humidity of 80% or more at 80 F., and transmitter output noted. If now, it is desired that output at 80% be linearly related to output at 20%, angle 32 is adjusted along slot 34 toward the Bourdon tube to increase output, or away therefrom to increase output, until the output is at the desired value. The instrument is now brought back to the low point and the output checked, and adjusted by means of angle 32 if necessary. This procedure is repeated till the instrument can repeat the desired outputs at both humidity points. The Bourdon tube may have to be shifted by means of slotted bracket 44 if adjustment of angle 32 results in rubbingonthe 'sides of holes 42 and 43 or too great a departure of elements 11 from parallelism with support 27.

After the instrument has been adjusted for two points near the low and high ends, it is checked for linearity by checking a midpoint such as 50% RH. where the output pressure should be 9 psi. The linearity adjustment is made by raising or lowering the lever assembly relative to the fulcrum 25 using the slotted parts of strips 46 and 47 held by screws 49 and 50. Should an adjustment for linearity be necessary, the outputs at the low and high points must be checked, and may possibly require adjustment, after which the midpoint is checked for linearity, and so on.

The instrument is now checked for temperature compensation by checking the output of the transmitter at the two relative humidity values used before but, at a difierent temperature, say 50 F and, according to the results of the check, moving one or both of eyes or eye plates 51 and 53 in or out to vary the efiect of the leverage on the temperature effects. In the form of connector shown in Fig. 4, the lever 63 is slid left and right to vary the magnitude of the compensating effect and the coil is turned about the axis of screw 62 to change the relative amounts of the compensator action on the zero change and on the span change. The construction of Fig. 4 (and that of Fig. has the advantage that these temperature compensation effects are to some extent independently adjustable, in contrast to the eifect of eye adjustment of connector 23.

At this point it may be necessary to go through the entire adjustment scheme again (not the initial tension adjustment, however), since in any case none of the adjust ments are entirely independent in effect.

The above described calibration and adjusting proce- 'dure is not exact, but may suflice where range of temperature variation is not too large and repeatability is the essential criterion. I Actually, the transmitter may be adjusted to an accuracy of about 1% relative to linearity and temperature effects and with unnoticeable drift and hysteresis, provided the device is permitted a reasonable amount of time to come to equilibrium. The constructions of Figs. :2, 4 and 5 are designed to permit the adjustments necessary to provide this kind of accuracy.

To begin with, three humidity calibration points are chosen, say 50% and 90% at 80 F., and the instrument adjusted to repeat the correct outputs at the extreme limits. Then the output at 50% is checked and if it does not fall at the half way point (i.e., between the output for 10% and the output for 90% the verti- "cal height of connector 23 is increased by unloosening V 10 screws 49 and 50 and adjusting the bimetals 46 and 47 to such a height that output for 50% is, or nearly is halfway, this being an adjustment of linearity. If the linearity adjustment is made, it may be necessary to go back to the zero and span adjustment point to restore the outputs at the 10% and 90% points, which would of course be followed by a check on the relation of the output at the 50% point to the outputs at the extreme points, and so on, until the outputs at 10%, 50% and 90% are linearly related. This being done, the compensation for temperature effects may be checked by checking several or all of the 10%, 50%, 90% at a difierent temperature, say 50 F. Depending on the result of this check, eyes 51 and 53 are moved in or out to bring the outputs at 50% as close as possible to the correct values. It may be necessary to bend the bimetallic arms 46 and 47 to vary the proportion of horizontal adjustment to vertical adjustment brought about by moving the eye along the strips. If any adjustment for temperature is necessary, the instrument must also be checked again at 10%, 50% and 90% at F., and, if necessary, adjusted, and subsequently temperature compensation at 50 F. re-checked, and so on. The temperature adjustment of Fig. 4 depends on the same principles as Fig. 2, but is more convenient when making adjustments ofthe temperature compensation, since the angular adjustment of coil 60 is predominantly eifective on the relative proportions of the components of compensation for change of response of the elements with change in temperature. However, no adjustment of the instrument is sufficiently independent of the others to obviate a check starting at the beginning, and possible re-adjustment or touching up of previous adjustments. Some time may be saved if all or most of the scheme is run through making only approximate adjustments and without backtracking during the first round of adjustments. In any case, the principle of the calibrating and adjusting procedure is that each successive adjustment of a given eifect is of lesser magnitude, whereby repetition of the procedure or of parts thereof reduces error until a certain tolerable residue of error remains.

Considerable difference in detail notwithstanding the devices of Fig. 1 and Fig. 2 possess in common the characteristic of defining kinematic mechanisms of the greatest simplicity. In Fig. 1, for example, two simple motors, i.e., bellows 3, filament device 1 and battle 6 and spring 10, are directly coupled to form the only movable links in a closed chain. Further, by what amounts to folding the device of Fig. 1 on itself, which in effect requires a pivot at a fixed point for directing the motions involved through the folded configuration, the arrangement of Fig. 2 is obtained, and by forming such pivot as a flexibly pivoted bell crank including as a portion thereof a temperature sensitive motor means (i.e, coil 60, or strips 46 and 47), there is obtained not only the increase in compactness due to folding without introducing friction or other losses into the kinematic chain thus formed, but at the same time provides the three-fold benefits of linearization of the relative humidity response of the filament or filaments, compensation of thermal expansion, and compensation of the effect of temperature on the relative humidity response of the filament or filaments. It is evident that a considerable complexity of function is developed from the device of Fig. 2, by what is in essence a very simple modification of Fig. 1. Putting the temperature compensation in at the connector 23 or 123, besides enhancing the utility of the connector, has the advantage that the resulting arrangement of transmitter parts is ideal for cases where the elements 11 are immersed in an atmosphere considerably different from that surrounding the Bourdon tube and other structure on the base 14, since the arrangement insures that the compensator will be exposed to the same conditions as the filaments.

However, both Fig. 1 and Fig. 2 are to be contrasted with the prior art, in which invariably some motion amplifying linkage or analogous mechanism involving considerable friction, is .combined with a humidity responsive element, in the attempt to get a useful output therefrom. The defects of such prior art devices are generally compounded, if it is attempted to add linearizing and compensation features.

I claim:

1. A humidity relay comprising a pair of elements adapted to change in dimension in response to change in relative humidity, such change being elongation or contraction depending on whether the relative humidity increases or decreases, said elements being arranged side by side in position to expand and contract along substantially parallel directional lines, a connector pivotally mounted and connected to said elements so as to define a double-armed lever; a bafiie and a nozzle, one of said baflie and nozzle being connected to one of said elements remote from said connector and adapted to apply a tension force thereto tending to pivot said connector so as to tension the other of said elements, motor means re mote from said connector to which the said other of said elements is connected, said motor means being responsive to pressure to move said other of said elements along its line of expansion and contraction, the relative positions of said baflle and nozzle controlling the pressure to which said motor means is responsive, expansion of said elements causing said one of said bafiie and nozzle to change its position relative to the other of said baflle and nozzle, so as to control said pressure to cause said motor means to move the said other of said elements along its line of movement in the direction of change in dimension of said elements, said elements being connected to said connector at points thereon so spaced that the angle between a straight line from the pivot of said connector to one of said points and a straight :line from said pivot to the other of said points is less than 180, whereby the said pressure is so controlled .as to cause the tension in said elements to remain constant although the relative humidity varies, and said pressure linearly corresponds to the relative humidity.

2. A hygrostat comprising in combination a pair of elongated, spaced, efiectively parallel humidity expansible and contractible elements, a pivotal connector having one part thereof connected to an end of one of said elements, and a second part thereof connected to an end of the other of said elements, a support for said connector, said connector being pivotally mounted on said support to have a pivot axis spaced from and lying at one side of a line joining the actual points at which the said parts are connected to said elements, first means on said support stiffiy connected to the remaining end of one of said elements, and second means on said support elastically connected to the remaining end of the other of said elements, the said first and second means being arranged to apply a tension to said elements, and said first means having at least sufliciently great stiffness in its connection to a said remaining end, relative to the elasticity of the elastic connection of said second means to a said remaining end, that upon humidity-induced length-change of said elements, a change in tension is created in said elements such as to cause said connector :to pivot and to cause the said remaining end connected to the said second means to move along the direction of the length of the element having the last said remaining end.

3. The invention of claim 2 wherein the said connector is provided with means for adjusting the length of said connector between said pivot axis and said line.

4. The invention of claim 2, wherein the said connector comprises a transverse arm, the ends of'which include said parts and a member connecting an intermediate portion of said arm to a pivotal mounting on said support, and the said member is adjustable as to the length thereof between said portion and said mounting.

5. A humidity relay comprising a tensionable element, a support, and a connector, said element being connected in tension between said support and said connector, said connector being mounted for deflection about a given axis and said element being connected to said connector at a point spaced from said given axis; said element being non-linearly responsive to relative humidity to change its length between said support and said connector in such manner as to cause deflection of said connector about said axis in one, angular direction or the other as the said element contracts and expands with changing relative humidity and to vary the tension in said element; follow-up means adapted to move said support to positions representative of the magnitude of a control force applied to said follow-up means and along the direction of expansion and contraction of said element; means to apply said control force to said follow-up means, detecting means adapted to control the magnitude of said control force in response to deflection of said connector from a predetermined position; connecting means connecting said connector to said detecting means for transmitting said deflection to said detecting means, the arrangement being that said detecting means controls the magnitude of said control force in a manner such as to cause the said follow-up means to move in the direction of expansion of said element when the said element expands, and to move in the direction of contraction when the said element contracts; said connecting means being connected to said connector at a point spaced from said given axis and from the first-said point, said points being so located that the angle included between straight lines drawn from the said given axis to said points is less than degrees, said support, said detecting means and said angle being oriented so that deflection of said connector due to change in length of said element causes the movement of said detecting means to positions producing changes in said control force in linear accordance with changes in-relative humidity.

6. The invention of claim 5 wherein the said connecting means is a tensionable element non-linearly responsive to relative humidity and is connected in tension between said connector and said detecting means.

7. A humidity relay comprising a tensionable element, a support, and a connector, said element being connected in tension between said support and said connector, said connector being mounted for deflection about a given axis and said element being connected to said connector at a point spaced from said given axis; said element being representative of the magnitude of a control force applied to said follow-up means and along the direction of expansion and contraction of said element; means to apply said control force to said follow-up means, detecting means adapted to control the magnitude of said control force in response to deflection of said connector from a predetermined position; connecting means connecting said connector to said detecting means for transmitting said deflection to said detecting means, the arrangement being that said detecting means controls the magnitude of said control force in a manner such as to cause the said follow-up means to move in the direction of expansion of said element when the said element expands, and to move in the direction of contraction when the said element contracts; said connecting means being connected to said connector at a point spaced from said given axis and from the first-said point, said connector including temperature responsive means movable in response to temperature, said tensionable element being responsive to temperature to vary its non-linearity of response to rela- 13 tive humidity as a function of temperature, and said temperature responsive means being arranged so that movement thereof in response to change in temperature varies the angle formed between a straight line through said axis and one of said points and a straight line through said axis and the other of said points, in a sense opposing the effect of temperature-induced variations in nonlinearity of response to relative humidity of said tensionable element.

8. The invention of claim 7 wherein said tensionable element also responds to temperature to change its length as a function of temperature, and said temperature responsive means is arranged so that its movement as a function of temperature varies the leverage of said connector relative to said given axis of one of said tensionable element and said connecting means, in a sense opposing the eifect of temperature induced change in length of said tensionable element.

9. A humidity relay comprising a pair of elements adapted to change in dimension in response to change in relative humidity, such change being elongation or contraction depending on whether the relative humidity increases or decreases, said elements being arranged side by side in position to elongate and contract along substantially parallel directional lines, a connector pivotally mounted and connected to said elements so as to define a double-armed lever; a bafile and a nozzle, one of said bafiie and nozzle being connected to one of said elements remote from the connector and adapted to apply a tension force thereto tending to pivot said connector so as to tension the other of said elements, motor means remote from said connector and to which motor means the said other of said elements is connected, said motor means being responsive to pressure to move said other of said elements along its line of expansion and contraction; the relative positions of said bafile and nozzle controlling the pressure to which said motor means is responsive, and expansion of said elements causing said one of said baflle and nozzle to change its position relative to the other of said baflie and nozzle, so as to control said pressure to cause said motor means to move the said other of said elements along its line of movement in the direction of change in dimension of said elements, whereby the said pressure is so controlled as to cause the tension in said elements to remain constant although the relative humidity varies, and said pressure corresponds to the relative humidity; said elements being of the type wherein the linearity of their response in terms of relative humidity is a non-linear function of their temperature, and said connector including temperature sensitive means movable in response to temperature, the said temperature sensitive means having a component of movement in a direction tending to change the angle between said lever arms in a sense opposing temperature-induced non-linear variation in the response of said elements to relative humidity.

10. The invention of claim 9 wherein said elements are responsive to temperature so as to vary their lengths, and said temperature sensitive means also varies the leverage of said elements on said lever relative to the pivotal axis of said lever in a sense opposing the efiect of variation in said lengths.

11. A temperature compensated hygrostat comprising a deflectible member, an elongated element connected at one end thereof to a relatively fixed first point, and at another end thereof to a second point on said deflectible member, said deflectible member having a deflection axis fixed at a third point spaced from said first and second points, whereby the length of said element between the said ends defines a straight line transverse to a straight line defined by the second said end and the second said point; the length of said element between the said ends varying in accordance with relative humidity and temperature, and the change of length of said element in response to relative humidity varying non-linearly with 14 t t i its temperature; said deflectible member including temperature responsive means responsive to temperature to shift the said second said end relative to said third point and along a path having a component along the length of said element and a component normal to the length of said element, bias means applying a deflecting force to said deflectible member so as to tension said element whereby said deflectible member deflects in accrdance with change in length of said element, detecting means for detecting motion of a fourth point on said deflectible member, said fourth point being so located as to have a motion including a parallel component substantially parallel to the first-mentioned straight line length of said element and a normal component substantially normal to said first mentioned straight line; and follow-up means controlled by said detecting means to shift said relatively fixed first point in accordance with said parallel component and in the direction of change of length in said element.

12. The invention of claim 11 wherein the said third point is spaced from a straight line joining said second and fourth points.

13. In combination, a support, a double-armed lever pivoted on said support, the axis of pivoting of said lever lying intermediate the ends of the arms of said lever, bias means connected to the end of one of said arms, an elongated, tensionable element connected at one of its ends to the end of the other of said arms, said element having its other end connected to a relatively fixed point on said support, said element being responsive to relative humidity to vary its length and being responsive to temperature to change its length; said bias means being arranged to act on the said one of said arms along a straight line substantially parallel to the length of said element and to cause said element to be tensioned between said fixed point and said connector, whereby change in length of said element causes said lever to pivot so that the end of said one of said arms has a component of motion substantially parallel to the length of said element; detecting means responsive to said component of motion to produce a control efiect in accordance with said component of motion; temperature responsive means in one of said arms, said temperature responsive means being constructed so as to vary the leverage relative to said axis of one of said element and said bias means; and said temperature responsive means being arranged so that as said element changes in length in response to temperature, said leverage varies so as to counteract the effect of temperature-induced changes in element length on the said component of motion.

14. In combination, a support, a double-armed lever pivoted on said support, the axis of pivoting of said lever lying intermediate the ends of the arms of said lever; bias means connected to the end of one of said arms, an elongated tensionable element connected at one of its ends to the end of the other of said arms, said element being responsive to relative humidity to vary its length and being responsive to temperature to non-linearly vary its response to relative humidity; said bias means being arranged to act on the said one of said arms along a straight line substantially parallel to the length of said element and to cause said element to be tensioned between said fixed point and said connector, whereby change in length of said element causes said lever to pivot so that the end of the said one of said arms has a component of motion substantially parallel to the length of said element; detecting means responsive to said component of motion to produce a control effect in accordance with the said component of motion; temperature responsive means in one of said arms, said temperature responsive means being constructed so as to vary the angle defined by a straight line from the said axis to the point of connection of said element to the end of the said other of said arms and a straight line from the said axis to the point of connection of said bias means to the end of the said one of said arms, and said temperature responsive means being arranged so'that as the relative humidity response of said element changes nonlinearly with temperature, said angle van'es so a's'to counteract the effect of non-linear temperature-induced changes in the relative humidity response of said element.

References Cited in the file of this patent UNITED STATES PATENTS r 16 Bohnstedt June 16, .1942 'Pett Oct. 12, 1943 Askew 'Mar. 21, 1944 Johnson j Apr. 11, 1950 Eddisonet' a1. Aug. 15, 1950 Joesting Sept. 8, 1953 L-oepsinger Dec. 21, 1954 FOREIGN PATENTS Great Britain an. 22 1925 OTHER REFERENCES Heinz: pages 456-460 inclusive, of Instruments Magazine, for July 1945.

Images