Air Seal Retention Means

The following relates to systems and mechanisms for maintaining coolant-gas pressure in a closure. More particularly it relates to such systems and mechanisms as adapted for replacing a printed circuit board in a card rack.

Workers in the art of designing and using computer modules and like arrangements are aware that often a plurality of printed circuit cards must be mounted, side-by-side, in a card rack such as rack R in FIG. 2, and cooled there. FIG. 1 shows, in very schematic fragmentary side-view, a computer unit 1, characterized by three card racks R-1, R-2, R-3 each including upper input-plenum means (respectively P.sub.1, P.sub.2, P.sub.3) adapted to receive input coolant air, e.g., pumped-in from outer inlet (cf. "AIR IN"), urge it down through its set of printed circuit cards (arrows, FIG. 1) to exit the rack and the machine (cf. "AIR OUT").

FIG. 1A shows, in schematic fragmentary perspective, one of the card racks R with three plenum means P (arrows show coolant path), it being understood that an array of printed circuit cards (e.g., as card pc in FIG. 1D) are to be stacked in parallel alignment, each engaged between respective upper and lower channels ch. Two exemplary lower channels ch-1, ch-2 are depicted in part in FIG. 1B, each comprising a spaced pair of guides or walls (g.sub.1, g.sub.1 ' and g.sub.2, g.sub.2 ') joined by an intermediate base (B.sub.1, B.sub.2 joined by side segment SD, FIG. 1B). For instance, guides g may be formed by cutting and bending a single sheet metal piece (SD, FIG. 1B) so that, for instance, g.sub.1 ', g.sub.2 are cut and bent-up out of a slot SLL (FIG. 1B) through which the coolant air is to pass (as known in the art). Thus, the bases (e.g., B.sub.1, B.sub.2) and connecting side segments SD extend along a relatively flat common plane (pl, FIG. 1C) out of which the like guides g are raised. For various reasons, the guides are formed to diverge slightly in the "up-direction", e.g., in FIG. 1C, guides g, g', of height hg diverge a few degrees, being spaced about 100-120 mils apart at their base (plane pl, base-width ch-w). So, a pc board wedged-in between guides g, g' is apt to engage them for only a portion of their height hg (as with PC, cf. width ch-w, in FIG. 1B--here assume a like upper channel ch').

FIG. 1D shows, in schematic perspective, a typical printed circuit card PC, comprising a circuit-board Bd with connector pins cc at one end and enlarged head or handle means h at the other. The head will span its respective channel slot when fully-inserted in a slot, and so block egress of coolant-air from within rack R. Each of the relatively planar circuit boards PC is inserted from one side of the rack (note "IN" side of R) in respective "slots" defined between receiving upper and lower channel means Ch, Ch', being thrust between the channel guides g, g' defining each channel (e.g., see PC in FIG. 1C) snugly to be relatively firmly secured at the opposite (back) end of the rack where they are typically lodged in a connector pin array, etc. (cf. FIG. 1D, pc connectors cc engaged by back panel connector B-cc and held in channels ch, ch'). FIG. 1B schematically illustrates (fragmentarily) a part of a typical channel Ch defined between a pair of like spaced guides g, g', projected from a common base plane pl to be projected a prescribed height hg thereabove.

As workers know, heat dissipation from such a closely-packed array of boards (PC) can be a problem unless supplemental cooling means is supplied, such as forced air, or a like gaseous draft, as indicated by the arrows in FIGS. 1, 1A--the coolant gas being entrained, for instance, down along both sides of all the cards (from top to bottom, here). Typically, rack R will be enclosed within some cabinet means, such as a computer chassis (as FIG. 1) wherein the entrained air may be kept at a positive pressure (above ambient; e.g., with an intake and fan means above each rack, as workers in the art appreciate). A typical card PC is shown in FIG. 1D and discussed above.

It will be apparent that, unless all the "slots" in a rack are filled with cards, and the cards thrust-in fully to block egress of coolant-air, such air will escape. Thus, if heads h of a PC array filling all slots in rack R (FIG. 1A) are disposed in contiguous abutment across the entry plane "IN" of rack R, the pressurized coolant air within is kept from escaping--otherwise an air-leak may radically degrade, if not destroy, the inner over-pressure and interfere with necessary cooling. (For instance, see adjacent heads h.sub.1, h.sub.2, h.sub.3 in FIG. 1E and discussion thereof below).

Thus, when any given card PC is withdrawn from a full set in rack R, the body of pressurized air should still be kept up to pressure within the rack by substitution of what workers call an "air seal" (e.g., see AS in FIG. 1A). An "air seal" typically comprises a replica of enlarged head means h along with associated attachment means for gripping the guides g, g' of both channels Ch. As workers well know, it is problematic to manufacture such "air seals" inexpensively and efficiently and so they exhibit precise dimensional and shape tolerances; e.g., so they are configured with sufficient precision to assure insertion and retention in any slot, willy-nilly, without some "play" that allows pressurized air to escape. This invention is directed toward improved "air seals".

This problem is exacerbated, of course, by attempts to standardize the design and manufacture of a card rack R from sheet metal stock (cf. using inexpensive material and methods, e.g., to form the channel pairs ch, ch'). For instance, the guides g, g' making up each channel can obviously vary in size, shape, spacing, etc.--though each channel must snugly receive any card and its associated head, as well as any and all "air seals", however fabricated.

This problem is even further aggravated when the channel guides g, g' are made relatively shallow (cf. reduced height hg thereof, FIG. 1C).

This invention is intended to develop methods and means for addressing such problems and features, e.g., for improving such air seals so they are better adapted for such sheet metal card racks where the depth of the card guides is minimal, and to particular methods and arrangements for better retaining air seals in such guides.

This invention is further intended to provide improved means for retaining air seals in a sheet metal card rack adapted for a computer processor and/or an associated I/O chassis, particularly where the air seal includes molded plastic spring means thrust against the bases from which the card guides are projected, wherein such spring means are adapted to so engage such bases as to provide superior frictional retention of the air seal. This invention is thus adapted to provide a more positive, more reliable engagement and retention of air seals in shallow card guides. More particularly this invention is adapted to provide engagement-forces normal to such a guide-base, rather than relying only on lateral (side-to-side) forces, directed parallel to the plane of the guide-base.

These and other features and advantages of the present invention will be appreciated by workers as they become better understood by reference to the following detailed description of the present preferred embodiments which should be considered in conjunction with the accompanying drawings, wherein like reference symbols denote like elements:

FIG. 1 very schematic fragmentary side view of a computer unit card racks apt for using the invention, while FIG. 1A is a fragmentary perspective of one such rack; FIGS. 1B, 1C an enlarged partial perspective and side view of guides therein; FIG. 1D a side elevation of a printed circuit board therefor; and FIG. 1E a very schematic fragmentary front view of a set of three such boards inserted into a rack.

FIG. 2 is a highly schematic fragmentary section of such guides with a first "sinous" air seal means, shown in elevation in FIGS. 3, 3A in end-profile in FIG. 3B, in enlarged se in FIG. 3C and seated between guides in FIG. 3D;

FIG. 4 is a view after FIG. 2 of an air seal according to a first embodiment, this being shown in elevation in FIGS. 4A, 4B and FIG. 6, and in plan view in FIG. 7 and in schematic side-view, as installed, in FIG. 5;

FIG. 8 shows a view like FIG. 5 of a second modified embodiment; and

FIGS. 9, 10, 11A show enlarged very schematic fragments of a seal portion of third, fourth and fifth embodiments, with FIG. 11B showing the fifth embodiment stressed and installed.

The invention will be better appreciated by workers upon consideration of the following detailed description of some preferred embodiments.

General description, background

The following illustrates an air seal S-1 by way of example and background.

This, and other means discussed herein, will generally be understood as selected, formulated, and operating as presently known in the art, except where otherwise specified. And, except as otherwise specified, all materials, methods, and devices and apparatus herein will be understood as implemented by known best expedients according to present good practice.

PRELIMINARY CONSIDERATIONS (FIGS. 2, 3)

Preliminary consideration is given to a non-preferred "side-engaging" air seal S-1 shown in FIGS. 2, 3, 3A, 3B, 3C, 3D. Seal S-1 comprises a relatively conventional handle or head portion h (like the head of an ordinary pc board) with pull-slots ts, ts' for engagement of a well known puller-tool (to remove from rack). From the ends of the head h are orthogonally projected a pair of like tongue portions T-1, T-1' adapted to be thrust along respective upper and lower channels ch, ch' and configured to frictionally engage the sides (guides) of those channels, being given a sinuous or serpentine segment (cross-sectional configuration) adapted for "side-engagement" of the guides (portions to be thrust vs. opposing inner sides of guides). This serpentine configuration will be noted as adapted to engage the inner sides of the channel guides g, g' (rather than the base from which they are mounted), generating "sidewise" (or lateral) engagement forces (parallel to base B--note direction of engagement force arrow F.sub.r in FIG. 2).

FIGS. 3B, 3C, enlarged end views of one of these tongues T-1, best suggest this sinuosity, showing the tongue extension t-d projected from head h and from which is projected an "offset" segment t-c projected along a prescribed reference plane P--P. A somewhat reentrant arcuate segment t-b is projected from segment t-c, extending at its maximum a prescribed channel-spanning distance dc away from plane P--P, sufficient to frictionally engage the opposing guide g' (and, preferably, be resiliently distorted thereby--this illustrated in exaggerated fashion in FIG. 3C). Arcuate segment t-b, in turn, projects another planar, terminal segment t-a coplanar with t-c along reference plane P--P. Workers will appreciate (for instance, from the exaggerated schematic views in FIGS. 2, 3C) how the coplanar segments t-c, t-a may be inserted to engage (be thrust against) the inside of one guide g, while the intermediate arcuate segment t-b is curved away therefrom sufficient to engage the opposite guide, and, ideally, be distorted somewhat by it, for a tight frictional "sidewise" engagement. Tongues T-1, T-1' of seal S-1 may be molded nylon or like suitable semi-rigid material as known in the art.

FIG. 1E may be taken to depict a fragmentary illustrative portion of the card rack R of FIG. 1A where, for instance, the heads h.sub.1, h.sub.3 of two printed circuit boards will be understood as inserted fully and engaged in their channels ch (see guides in phantom), while the intermediate channel is sealed by the head h.sub.2 of some suitable "air seal" device. Here, it will be understood that, ideally, all heads h are abutted contiguously and fill the entry side "IN" of the rack R so that no pressurized coolant air can escape (through that "entry side").

PROBLEMS

Workers in this art will be rather familiar with the problem of making "side-gripping seals" (like S-1, discussed above) so they are sufficiently accurate in dimensions and configuration as to be reliably, positively and efficiently engaged in slot channels, with their head placed, and kept, firmly in position to prevent air leakage--particularly in view of the difficulty in making card rack structures which exhibit tight tolerances and sufficient rigidity to reliably receive and hold any air seal. And even when such a seal is positively, firmly engaged in its channel (which is not all the time), it is too commonly upset and released by ambient shock and vibration, or by the pressure of the coolant air within, etc., as workers well know.

And such common regrettable shortcomings are badly exacerbated when the channel guides (e.g., g, g') are foreshortened (that is, their height reduced, see FIGS. 1B, 3D), since it will be evident that a sinuous, "side-gripping" tongue design (like S-1 above) will generate engagement forces which are proportional to the guide-area contacted by the tongue. Thus, if guide height is radically reduced (for instance, as suggested in FIG. 3D where channel guides g, g' are shown in phantom extending from their base to reach only about half way up the width of tongue T-1')--the gripping forces are likewise radically reduced. The invention is directed toward alleviating such problems (e.g., in embodiments to follow).

PREFERRED SOLUTIONS

FIRST EMBODIMENT S-1: FIGS. 4, 5, 6, 7

Thus, workers in the art will readily concede that, in many instances, air seals with "side-gripping" tongues leave much to be desired; e.g., too often being not so positively and firmly retained in their card rack as to adequately maintain coolant over-pressure. Workers will certainly concede that this becomes an ever greater problem when the height of the channel guides is reduced. Thus, one salient feature of the present invention is to provide air seals with "base-engaging", rather than "side-engaging", tongues, as exemplified, for instance, in a preferred embodiment S-2 (see FIGS. 4, 4A, 4B, 5, 6 and 7) described as follows.

FIGS. 4, 4A and 4B show a modified improved "air seal" S-2 improved according to the subject invention to be "base-engaging" rather than "side-engaging" ("guide-engaging"). It may be compared, for instance, with a comparable view in FIG. 3 of "side-engaging" seal S-1. FIGS. 4A, 4B are xerox reproductions, to actual size, while FIG. 6 is a similar schematic line drawing, FIG. 7 is an associated plan view and FIG. 5 is a side view of improved seal S-2 as inserted in exemplary upper and lower channels of a card rack, with spring detent portions d, d' shown bearing against respective channel bases GN, GN'. (Compare base B in FIGS. 1B, 4, etc.).

It will be seen that embodiment seal S-2 is an example of a structure that is frictionally engaged by thrusting opposed tongue-spring means against the channel bases, this thrust directed elongately along its elongate axis (see elongate axis H--H of the head h of S-2 in FIGS. 4A and 7 and see arrow F.sub.r ' in FIG. 4)--rather than thrusting laterally (from side to side) against the guide means (as for instance, with prior described seal S-1; see FIG. 2 for comparison).

Seal S-2 may be looked upon as relatively conventional (e.g., in construction of head h, etc.) except as otherwise described; in particular except as regards the base-engaging structure of its tongues T-2, T-2'. Seal S-2 includes tool-puller slots ts, ts' and further includes a pair of projecting "stops" tp, tp' to limit the thrust into rack. Tongue structures T-2, T-2' (best seen schematized in the fragmentary schematic of FIG. 5) each comprise a relatively flat, thin planar body projecting orthogonal to head axis H--H (to extend along the elongate axis of the channels ch), having an inner portion thereof relieved to form an elongate slot (SL, SL') extending along a good part of the tongue length and adapted to form a thin, resilient outer "spring bar" portion (SP, SP') closing the slot. This spring bar is characterized by a relatively central detent or projection (d, d') adapted to bear against a respective base (GN, GN') of its intended channel when seal S-2 is inserted in its rack. Thus, for instance as seen in FIGS. 4A, 5A (note in FIG. 5A phantom location of unstressed detent d', and compare for instance FIG. 3 for seal S-1 above), when seal S-2 is inserted into channels (in slot between bases GN, GN') where the guides (gg in phantom) are relatively shallow, it will matter relatively little since the detents d, d' will project resilient engagement forces thrusting outwardly against bases GN, GN' (note arrows denoting retaining forces F.sub.r ' FIG. 4A).

TYPICAL DIMENSIONS

Thus, for instance, for a seal S-2 about 10" long (nominal length Ln between outside of unstressed springs sp, sp', FIG. 5) by about 0.07" wide (width of head h, which is about 3/8" high), tongues T-1, T-2' may be about 1.75" long (extension away from head h) by about 1/2" wide and about 3/32" thick, and present slots SL, SL' about 1" long by about 1/8" high, to form spring bars sp, sp' about 5 mils thick surmounted medially by respective detents d, d' projected about 35 mils therebeyond and having a base-engaging "flat" therealong about 0.2" long. More particularly, each detent is projected 0.035" beyond its spring bar (sp, sp'). Thus, the two detents add 70 mils to the nominal length Ln of seal S-2 [ignoring stops tp, tp' each projected well beyond Ln, sufficient to overlap the slot (on bases GN, GN', as depicted in FIG. 5, for instance; e.g., each tip may extend about 0.1" beyond the engagement plane of its nearby detent, or more if necessary]. These stops are shown as separate pieces but may be made integral with head h.

Thus, for instance, a seal S-2 can be mass-fabricated reliably and inexpensively with an overall engagement length Le (between detents d, d') of a nominal 10.070" plus/minus 0.005. This may be compared with known card racks where the receiving-slot nominal length Ls (reliably constructed in mass production techniques) is a nominal 10.015" plus/minus 0.015", varying Ls between 10.030 and 10.000". Thus, workers in the art can see how, even under worst case conditions, detents d, d' may still be positively engaged into a receiving slot, and be displaced enough to invoke reasonable slot-retaining forces, despite expected variations in the size, shape of the seal parts or of the rack. For instance, consider the following worst cases.

A. Worst case "shortest" seal S-2 (in longest slot):

Minimum effective seal length Le (between detents) is 10.065", vs. maximum slot length of 10.030" (i.e., the longest receiving slot). This means a difference of about 35 mils, or about 17.5 l mils deflection per spring bar--something quite adequate to firmly seat seal S-2 and keep it so.

B. Worst case "longest" seal S-2 in "shortest" slot:

Maximum length seal is about 10.075" vs. minimum length slot of 10.000"; the difference would be about 75 mils, or about 37.5 mils deflection per spring bar--againt not over-stressing it and quite adequate for good results and engagement.

Workers will appreciate how seal S-2 acts to invoke base-engagement rather than side-engagement; how it tolerates a good deal of variation in part dimensions and configuration, how it can isolate the design from concern over guide-height (less problem with "shallow guides") and, in general, provide an air seal exhibiting good engagement forces and good retention in a card rack.

EMBODIMENT S-3, FIG. 8

FIG. 8 shows a further modified, more-preferred embodiment S-3 which may, in general, be taken as the same as the above-described embodiment S-2 except as otherwise specified.

Base-engaging air seal S-2 is replicated here except that a relatively rigid brace segment B is provided between tongues T-2, T-2' to further rigidify them and help assure that essentially all (preferably 100%) of the seal-deflection between detents d, d' will be confined to spring bars sp, sp'--that is, so the body of tongues T-2, T-2' will be effectively prevented from deflecting or distorting under the insertion-retention forces between bases GN, GN' as described above. Without brace segment B, the tongues may tend to bow toward one another, and/or to be twisted, etc. under engagement forces as the seal is inserted into a rack slot.

EMBODIMENT S-4, FIG. 9

Embodiment S-4, FIG. 9, is the same as S-3 except for being further modified by, in essence, varying the configuration of the slots in both tongues (see SL-4 in FIG. 8), opening each adjacent the proximal end of spring bar spp to form a slot-mouth SL-g. Thus, each spring bar spp becomes a re-entrantly cantilevered flexure spring (having a projecting detent protrusion dd adjacent the free end of bar spp; the other opposite tongue T-3' of seal S-4 is, of course, symmetrical and similar, though not shown here). Thus, insertion of base-engaging seal S-4 into a rack-slot (channels) will thrust respective detents dd, dd' (dd', spp', etc. not shown) against respective channel bases GN, GN', thus deflecting the end of the so-formed cantilevered spring bars (spp, spp', as in phantom in FIG. 9), to render the prescribed base-engagement.

Detent dd on the end of spring bar spp will be understood as preferably formed with a smooth gradual slope gradient on each side so that, as the seal is first thrust into its receiving channels, each base-engaging spring spp will slide smoothly along its receiving-base, with this base eventually bridging the slot-gap SL-g adjacent the end of spp, as workers in the art will understand.

This should give a desired gradually-increasing, resilient engagement force as the seal is inserted further and further. This design will minimize the chance that the free end of such a spring bar (spp) may be caught, bent or broken-away by rack parts or other external forces (e.g., vs. S-5 in FIG. 10). Similarly, as seal S-4 is withdrawn, the base-engaging forces will be gradually reduced, as desired. This is quite advantageous as those skilled in the art will see, especially as compared, for instance, with companion modified seal S-5 in FIG. 10 (described below).

Here, as with the embodiment of FIGS. 8, 10 a rigidifying brace B preferably extends between tongues T-3, T-3' of S-4 for greater rigidity, of the main body of the tongues, allowing the spring bar spp to undergo substantially all of the distortion of engagement forces, with essentially no distortion of the rest of either tongue body.

EMBODIMENT S-5, FIG. 10

FIG. 10 shows a further modified embodiment seal S-5 understood as essentially the same as S-4 except that its spring strips ssp, ssp' face in the opposite direction relative to their tongues (and relative to insertion and withdrawal motion) and except for the fact that its slots SL-5, SL-5' are opened not upwardly toward the guide base as with SL-4, but outwardly along the direction of insertion (cf. mouth SL-m). As seal S-5 is inserted, initial engagement forces will be relatively high (vs. S-4). Also, base-engaging air seal S-5 presents greater risk of catching and breaking or distorting the free end of its springs ssp, ssp'. By comparison, it will usually be preferable to form the spring cantilever in the fashion of embodiment S-4 (FIG. 9) to experience a more gradual increase in engaging spring forces as the seal is inserted (and gradual decrease as it is withdrawn). Also, as mentioned there is less risk that bar spp will be caught and distorted or broken by the channel or other external forces in edges, etc.

EMBODIMENT S-6, FIGS. 11A, 11B

FIG. llA presents a somewhat different "base-engaging" air seal S-6 where the body of its related tongues T-5, T-5' are sufficiently long, thin and resilient to, themselves, exhibit bending engagement flexibility, so that, when seal S-6 is inserted into respective channels, against the bases thereof (bases not shown), seal S-6 will stay firmly in place. (Tongue T-5 cooperates as like tongue T-5', not shown, as understood in the art). The tongues will be bent somewhat between channel bases GN, as indicated in exaggerated fashion in FIG. 11B. As before, each tongue presents an associated detent or projection doo projected outward for engagement with a base GN (doo' for GN'), these detents being sloped and disposed somewhat inward of the tongue end for purposes like those described with respect to S-4 in FIG. 9 (vs. FIG. 10) above.

Also, according to a related feature, detents doo, doo' are preferably provided with a down-stream slope portion d-s (see plane S-S in FIG. 11A) which is relatively flat and adapted to be thrust into conforming engagement with the surface of the receiving channel base when T-5 is inserted; e.g., d-s aligns with the surface of base GN in FIG. 11B as understood by workers in the art. Of course, in this case the rigidifying brace B depicted above in FIGS. 8, 9 and 10 will be counter-productive and not called-for, as workers in the art will understand.

Conclusion

Though air seals S-2, etc. have been described exemplarily in terms of a certain kind of circuit board and card rack workers will perceive that it may readily be adapted to other uses, such as different boards and/or racks. And the handling means h thereof may evidently be adapted to other equivalent configurations.

It will be understood that the preferred embodiments described herein are only exemplary, and that the invention is capable of many modifications and variations in construction, arrangement and use without departing from the spirit of the invention.

Further modifications of the invention are also possible. For example, the means and methods disclosed herein are also applicable to other analogous air seal situations. Also the present invention is applicable for providing better seal retention and resistance to other analogous adverse influences.

The above examples of possible variations of the present invention are merely illustrative. Accordingly, the present invention is to be considered as including all possible modifications and variations coming within the scope of the invention as defined by the appended claims.