From (by) RocketSlinger@SBCGlobal.net (email me there please)… This is a sub-site to main site at www.rocketslinger.com …

This web page last updated 03 Oct 2021

 

Harvesting and Managing Energy While Re-entering an Atmosphere Using a Shuttlecock Design

 

 

Abstract / Pre-Summary

This sub-page to www.rocketslinger.com is a companion or supplementary document to another document found at https://www.researchgate.net/publication/354612024_Methods_of_Decelerating_a_Spacecraft_Through_Atmospheric_Re-entry_Using_a_Shuttlecock-like_Design and at http://www.rocketslinger.com/BadMinton/ (same document, 2 locations).  In BOMs (Bills Of Materials) lingo, the “parent” BOM calls a “child” BOM, where the child is a sub-assembly of the parent.  This document deals with sub-assemblies of the designs described in the “parent document”, and so, it will henceforward be referred to as the parent document for brevity.

Temperatures endured during re-entry (both on the windward and lee sides) are briefly examined here, as are temperature tolerances of materials and subassemblies.  Then a “heat wall” design is described, which can be used to protect subassemblies from heat damage.  Much of this is focused on hydraulic machinery, which will need to be used to control the “petals” (or grid fins) as were described in the parent document.

Many variations of a powered hinge are described here.  One of the best versions consists of a curved (arc-shaped) friction plate attached to each petal-becoming-a-grid-fin (AKA, fairing segment).  The friction plate is gripped between two powered, slowly rolling funnel shapes, to create a torque-limiting device, to absorb sharp peak loads without damage.  This hinge system is called a “Bifrost Hinge”, and the name given to it, is explained.

To harvest energy (or power used to perform work) in a spacecraft entering an atmosphere, we could use roller chains similar to (but larger and stronger than) the chains used in bicycles and chain saws.  In a chain saw, some of the plates (in the chains) are extended outwards with teeth, for cutting wood or other materials.  Here, instead, some of the plates are extended outwards to “catch the breeze” during re-entry, the same as windmill vanes catch energy from the wind.  The overall device, then, here, will be called a chain windmill.

The vanes in such a device could possibly take many forms, but two are described here.  One is a single plate per each set of chain-plates that is so equipped, which is stopped (by a hard-stop) from folding past 90 degrees, with the right angle formed to “catch wind”, and not fold further towards the lee side.  However, this vane is allowed (during the chain segment’s return journey towards the windward side) to fold out of the way of the prevailing wind.

The alternate (here-described) vane is a “butterfly vane set”, with twin segments that “flap” to catch wind in one direction, and fold out of the way in another direction.  They can each perform a partial rotation around pins protruding outwards from each pair of roller-chain plates that is so equipped.  Each of the two twin vanes (“butterfly wings”) provides the hard-stop for the other, in this design.  In either form, a chain windmill is not at all an optimal solution here, for us.

An energy-harvesting device could also resemble a paddle wheel on a riverboat.  This approach will be far less disadvantaged by lack of lubrication than a chain windmill would be.  And as will be shown, lubrication WILL be a major problem for the chain windmill (unlike the paddle wheel).

Also described here is an airflow spoiler that could be placed to the windward side of the energy-harvesting device, possibly sized and located in a compromise between partially protecting the energy-harvesting device (best choice being a paddle wheel) from too much heat and strong airflow, and not getting enough airflow to harvest enough power.  The spoiler is composed of a hinged heatshield-covered plate that can be thrust out into (or retracted from) the airstream.  It could be activated by a (heat-walled) “Bifrost hinge”, which could also double up as a “hydraulic battery” used to store extra power.  These are sometimes called “hydropneumatic accumulators”.  In the design context here, the variably pressurized nitrogen in such a device could be partially or entirely replaced by the variable pressure of passing air in the airflow spoiler (which pushes on the plate).  Such a design choice may have to be balanced with the other possible purpose of the spoiler, which is to partially protect the paddle wheel.

            The body of this document contain far more details, and sometimes-implausible variations.  However, the above covers the most important basics.

            As with other sub-pages of www.rocketslinger.com, the intent here is to “defensively publish” propulsion-related (and “misc.”) ideas, to make them available to everyone “for free” (sometimes called “throwing it into the public domain”), and to prevent “patent trolling” of (mostly) simple, basic ideas.  Accordingly, currently-highly-implausible design ideas (usually marked as such) are sometimes included, just in case they ever become plausible, through radical new technology developments (often in materials sciences).

 

 

Introduction / Basics

 

            Please read the abstract above… Some of those basics may not be thoroughly (completely) repeated here below.

            Dear Reader, excuse me as I will often slip out of stilted formal modes of writing here.  I have no boss or bosses to please with these writings, so I’ll do it my way!  I’ll often use a more informal style from here on in, using “I”, “we”, “you”, etc.  “We” is you and me.  “You” are an engineer, manager, or other party interested in what’s described here.

            The parent document describes what I now briefly summarize:  Payload fairings can be built with embedded grid fins buried inside other materials (plastics, etc.) that will burn, ablate, or fall away in the heat of re-entry.  The fairings segments (4 of them being a good number here) can also be arranged on hinges that join them to the base of the payload section.  The only sensible option here is for those hinges to reside outside of the circular profile of the cylindrical rocket body, and of the base of the fairing.  This, then, I call a “partial hammerhead design”, where, in fairings lingo, the “hammerhead” design refers to the girth of the fairings (at the base of the payload cavity) exceeding the girth of the rocket body.

            The hinged fairings splay out from their base like the petals of a flower open up, so here they are called “petals”.  The payload is deployed, and the de-orbit burn is performed.  The petals stay deployed or partially deployed till the plastic or other “sacrificial materials” ablate or fall away; now the petals become “grid fins”.  Via “powered hinges”, the grid fins (which were formerly petals) can now be adjusted more-open or more-closed.  When this is done in a rolling-around-the-circle pattern, we can induce a heat-spreading “rotisserie roll” in the body of the rocket, as we descend.

            No details were given (in the parent document) about how the “powered hinges” should be designed and built.  Here, I will present my moderately-educated ideas about how to do so.  OK, more honestly, here come my best SWAGs at doing this, and associated things!  Further below, of course.

            The parent document also briefly describes what were there called “screw-propellers” or “screw turbines” as energy-harvesting devices.  There are a few mentions here of associated ideas.  But, presumably at the base of these screw turbines, we would want to add hydraulic pumps (far more likely) or electrical generators (far less likely) to harvest energy.  Either way, the pumps or generators would have to be shielded from ambient heat during re-entry.  Heat-shielding will be addressed in some detail, further below.  Heat-shielding discussions are new here, compared to the parent document.  Anyway, the screw-propellers or screw-turbines are dismissed here as not being very suitable for our use, especially as compared to a paddle-wheel design.

            I’ll describe “chain windmills” here (see a brief summary in the abstract above), as an alternative to “screw turbines”.  Their best energy-harvesting modes would also be to drive hydraulic pumps.  The other choice is electrical generators.  Why do we chose hydraulic pumps?  See https://sciencing.com/differences-hydraulic-motors-electric-motors-7351549.html  “Differences Between Hydraulic Motors & Electric Motors” and http://info.texasfinaldrive.com/shop-talk-blog/hydraulic-motors-vs-electrical-motors-why-hydraulic-wins “Hydraulic Motors vs Electrical Motors: Why Hydraulic Wins”.

            Motors v/s generators considerations are VERY highly similar here!  Also, the more times we convert one form of energy to another (oil pressure to electricity or vice versa), the more energy we lose to inefficiencies.  So we’ll stay mostly hydraulic here, per my preferences, in these design notes.

            Airflow spoilers and “hydraulic batteries” (“hydropneumatic accumulators”) are also mentioned in the abstract.  They are also discussed in much more detail below.  Let’s move on!

 

Internet Research on Relevant Facts About Re-Entry Heat and High-Temperature Materials and Subassemblies

 

            Here, we assume re-entry is from low Earth orbit.  Any higher-speed re-entry seems to be out of reach with current materials and technologies, for any kind of “shuttlecock” re-entry style, as described here.

            The below is some fairly cursory internet research on associated matters, to be used to guide our design work here.  My general intent here is to conduct “amateur public-domain rocket science” (AKA “open source”), so, as usual, your comments questions, and contributions are welcome at RocketSlinger@SBCGlobal.net.  So, in accordance with my intent, research matters (links) that are NOT used here (but are highly relevant, and could be used by others) are included here.  We generally start with the most-relevant, and conclude with the least-relevant.

            Temperature Tolerances Needed…  Assuming re-entry from low=Earth orbit, of course.  https://en.wikipedia.org/wiki/Atmospheric_entry is of general interest.  https://www.airspacemag.com/how-things-work/shuttle-tiles-12580671/  “Shuttle Tiles” is subtitled “Why the space shuttle can withstand reentry temperatures up to 2,300 degrees.”  https://www.nasa.gov/centers/ames/research/2007/faq-shuttleglass.html  says (about the lee as opposed to windward side)…”This side experiences lower high heat transfer compared with the windward side, and so it does not reach as high a temperature. This is because the pressure is much lower, at least two orders of magnitude lower (1/100 or less pressure) on the back side. The hot gas on the windward side expands to the leeward side, which means the pressure drops quickly, and so does the gas temperature.  (Deletions by me here).  So windows must still be able to withstand high temperatures, say about 1000 C.  So Shuttle windows are made from a high-temperature quartz glass that can withstand heating and cooling without cracking.”

https://www.daviddarling.info/encyclopedia/R/reentry_thermal_protection.html says almost the exact same thing as above.

Note that 1,000 C = 1,832 F.  There’s a handy C to F converter at https://www.google.com/search?q=Centigrade+convert+to+fahrenheit&oq=Centigrade+convert+to+fahrenheit&aqs=chrome.0.69i59.225990j0j4&sourceid=chrome&ie=UTF-8.  Also (importantly!) note that the “lee” side is about 100 times lower in pressure than the “windward” side.  https://www.nasa.gov/mission_pages/constellation/orion/orionheatshield.html  says “The shuttle enters the atmosphere at lesser speeds, 4.7 miles (7.5 kilometers) per second, generating a lower maximum temperature of 2,900 degrees Fahrenheit”.

So we can conclude that our temperature-tolerance targets are about 2,300 F to 3,000 F or so for the windward side then, and 1,832 F for the lee side.  These are our temperature-tolerance targets.  Also importantly, note that windward side pressures can be about 100 times as much as lee side pressures.

Metals to be used?  General knowledge (trivial to look up) is that the melting point of titanium  3,034°F.  Alloys of stainless steel also work.  https://www.businessinsider.in/bird-poop-and-dust-could-seriously-complicate-elon-musk-and-spacexs-latest-plan-to-reach-mars/articleshow/68085712.cms is relevant…  From there…

“Musk has said the nose of Starship may be exposed to temperatures of about 2,700 degrees Fahrenheit. The type of steel alloy SpaceX may use on Starship's outer skin, called 310S, melts at about 2,400 degrees.”  (My comment; so they need heat shielding there at the nose).

In passing, let me add, ceramics are of interest as well.  They’re generally good in “compression” modes, but not in “tension” modes.  Bending and torsional stress modes are therefore dicey for ceramics!  I still think metals are better for most of what high-temperature and high-stress structural uses are described here.  But you might want to look into the materials used here: https://www.imperial.ac.uk/news/176628/new-record-worlds-most-heat-resistant/ “New record set for world's most heat resistant material”, which says that “Tantalum carbide (TaC) and hafnium carbide (HfC) are refractory ceramics…”, and a ceramic made of them is rated at up to nearly 4,000 C, or 7,232 F!

            We want to use hydraulics a LOT here!  An obvious place to start is the hydraulic fluid itself.  The following web site discusses new high-temperature TARGETS for the fluids, so this is best-case!  https://onlinelibrary.wiley.com/doi/abs/10.1002/jsl.3000090203  “The rationale is presented for the development of 315 °C bulk oil high-temperature hydraulic fluids…”  315 C = 599 F, so we must conclude that any hydraulic gear (to include pipes or hoses) MUST be heat-shielded!

            I am NOT (here, much) recommending the use of hydraulic cylinders, but they’re certainly an option, if properly heat-shielded.  Keep in mind that, if you use these, the fluid isn’t the only heat concern.  https://www.hydraulicspneumatics.com/technologies/cylinders-actuators/article/21885384/hydraulic-cylinders-keep-their-cool says… “…premature seal degradation can usually be avoided by specifying a seal material that (is) designed to operate in higher temperatures. For example, ethylene propylene and Viton seals usually accommodate temperatures exceeding 350° F.”

            Lubrication is a concern!  How about high-temperature lubes?  https://www.machinerylubrication.com/Read/30674/high-temperature-lubricants  says 220 C = 428 F is perhaps the best that we can do!  If we can heat-shield a “well” of oil or grease (lube), and have our otherwise-exposed moving parts dip into there now and then, we are doing extremely well!  Enclosing the lubricated bearings in heat-shielded areas is even better.  In the worst cases, we’ll have to make do with un-lubricated parts.

            Mechanical or structural cables aren’t used in the here-described designs, but heat-resistant cables are strong candidates for use in design variations here.  https://lapplimited.lappgroup.com/products/high-temperature-cable.html provides cables tolerating up to 1,565 C = 2,849 F, and the lee side max is about 1,832 F, so we could use these, very cautiously, if we wanted to.

            Now we transition to parts that we can’t or don’t use, that are of less interest.  They’re included here (more or less) on an “FYI” basis.  High-temperature belts (think V-belts or similar)…  https://www.durabelt.com/hightempresistantbelts.php  says “Works up to 230oF (110oC).”  Belts are out!  High-temperature hoses…  https://www.ducting.com/high-temperature-hoses says…  “There is one specific variation of the stainless steel hose that uses a titanium alloy. It can be used with temperatures up to 1650° F…”    Hoses are HIGHLY questionable, would have to be used VERY cautiously!

            Between belts and hoses, we can make two very important conclusions:

‘1)  ANY use of flexible or pliable materials for heat-exposed covers or shields, is a VERY bad idea!  Hydraulic parts (with support hoses), then, can’t move around through heat-exposed areas.  In a heat-exposed area, their positions MUST remain fixed, and heat-shielded!

‘2)  Any ideas about using “heat walls” that use a cooling fluid (like water or liquid nitrogen), that therefor need cooling-fluid-supplying flexible hoses, as the heat-wall-protected part moves through heated areas, is off limits as well.  Any such-cooled devices will need to remain in fixed positions as well.  They will need to be supplied with cooling fluids by insulated high-temperature (inflexible) metal pipes or ducts.

            High-temperature cloth or fabric used for insulating hoses (or other devices) is off limits as well.  Fiberglass can withstand 1,000 F, and we need 1,832 F or higher.  Asbestos is ALMOST good enough, at a melting point of 1,600 F (or so), but is also hazardous as well.

            If you’re really set on trying to use moving, flexible hoses, and insulating them with fabrics, then see https://nutec.com/products/ceramic-fiber-blanket/ “MAXWOOL CERAMIC FIBER BLANKET”, rated 2,600°F (1425°C).  However, notice that the recommended uses here are as “blankets”, not wrappings on hoses, and so, this fabric might not be suited for uses involving being flexed a lot.  And as you can see, ceramics can sometimes be turned into flexible fibers.  You might, then, look into the materials used here: https://www.imperial.ac.uk/news/176628/new-record-worlds-most-heat-resistant/ “New record set for world's most heat resistant material”, which says that “Tantalum carbide (TaC) and hafnium carbide (HfC) are refractory ceramics, meaning they are extraordinarily resistant to heat. Their ability to withstand extremely harsh environments means that refractory ceramics could be used in thermal protection systems on high-speed vehicles…”, and is rated at up to nearly 4,000 C, or 7,232 F!

            As previously mention, we’re primarily interested in hydraulics here.  However, electrical generators and motors are still of some interest.  Here, https://netl.doe.gov/node/3754 says (in an oil-well context) “None of the commercially available generator systems developed by major service companies are capable of operation at 250°C. “  Their target (at this web site) is 250 C or 482 F.  Electrical motors self-heat and so they would be even worse.  We will have to robustly cool ALL parts of power management here, in our design context!

            Not used here at all, but of general interest, are thermoelectric device and Peltier “heat pump” devices.  For thermoelectric devices, https://www.energy.gov/sites/prod/files/2014/03/f13/acep_04_elsner.pdf says “Large increases in thermoelectric conversion efficiency (>3 times) appear feasible from T H’s of 150°C to 1000°C.”…   TH meaning temp-hot and 1832 F = 1000 C…  This is perhaps the upper limit of what may happen some sunny day…  Which is the same as our lee side…  Probably, this is not of much practical interest for now.

Now on a Peltier device, we’d want the cold side in our cooled-down-devices boxes, and the hot side out in the ambient hot air…  Ambient is best-case 1832 F; can we beat that?  A high temperature Peltier device here:  https://tetech.com/peltier-thermoelectric-cooler-modules/high-temperature/  says “Very High Temperature (VT) modules are rated for intermittent use to temperatures up to 200°C whereas all other modules we sell are rated for use up to 80°C.”  200°C  = 392 F, and so we write this off, except possibly for uses entirely within the avionics circuits, which is outside of our scope here.

            On a practical (non-academic, non-developmental) level, Digikey is an excellent place to see what is available “off the shelf”.  For Peltier devices, there, https://www.digikey.com/en/products/filter/thermal-thermoelectric-peltier-modules/222 lists maximum –Th  (Temperature-Hot) of 50 C, which is even worse than above…  For practical “off the shelf” coolers that is.

            Moving on to some other assorted matters, “hydraulic batteries” are of much interest here.  See https://www.hydraulicspneumatics.com/technologies/accumulators/article/21882959/hydraulic-batteries-save-fuel “Hydraulic ‘Batteries’ Save Fuel”.  From there, “Hydropneumatic accumulators are widely used in hydraulic systems because they provide auxiliary power during peak periods. This lets designers select smaller pumps, motors, and reservoirs in the main system.”  We’ll refer to this again.

            Of general interest are the following: https://scholarworks.iupui.edu/bitstream/handle/1805/7396/Vaezi_2014_energy.pdf;jsessionid=B7BC2AB03D619A47177D51FF43D852BF?sequence=1 is “Energy Storage Techniques for Hydraulic Wind Power Systems”.  Also see https://www.sandia.gov/ess-ssl/EESAT/2005_papers/Lemofouet.pdf “Principle of Hybrid Energy Storage Systems Based on Hydro-Pneumatics and Supercapacitors for Distributed Generation and Renewable Energy Sources Support”.  Then https://www.mobilehydraulictips.com/energy-efficient-hydro-accumulators-for-energy-storage-or-conversion/ is a commercial.  At my casual glance, it seems to involve mostly dampeners to dampen out hydraulic pressure oscillations.

            Let me briefly add that hydraulic accumulators need not be heavy and thick-walled, which we wouldn’t want for aerospace applications.  The can have COPV (Composite Overwrapped Pressure Vessel)-style lightweight yet strong walls.  For samples of providers of aerospace-type accumulators, https://ph.parker.com/us/en/aerospace-accumulators and https://www.valcor.com/missiles-and-aerospace/aerospace-accumulators/.

            Also of general interest for actuators, see https://en.wikipedia.org/wiki/Worm_drive for possible use (not used here).  A linear actuator may also be of interest: See https://en.wikipedia.org/wiki/Linear_actuator for actuators.  A sub-type being a “screw actuator” is what we may sometimes want, along with a hydraulic motor.

            OK, now I think that it’s high time to move on!

 

 

Building a “Heat Wall” (or Ice-Box) for Subassemblies

 

            High-temperature shielding tiles are well enough known from the Space Shuttle days, and now from Space X’s “Starship”.  These are easily enough researched, and I have no special comments to add here, about those.  Under the tiles, one can locate a blanket.  Search here in this document for “Maxwool” search-string for a good candidate material.  One or both of these could be on the outer layers of a “heat wall” for components that need to be kept cool.  This is also true for fluid-carrying pipes (don’t forget, fixed locations, not moving hoses!) that carry hydraulic fluids, or water, or liquid or gaseous nitrogen, for cooling.  Liquid nitrogen (or helium or other inert coolant) may gasify during its journey, but that will be OK, as long as it is kept as cold as possible.

            Now please see a document that is kept at duplicated locations at https://www.researchgate.net/publication/331556573_Designs_for_Passively_Thermally_Gated_Fluid_Flow_Switches and at http://www.rocketslinger.com/Psv_Tgt_Fsw/, titled “Designs for Passively, Thermally Gated Fluid Flow Switches”.  This was written when Space X was still considering the use of “transpiration cooling” for “Starship”.  So it is written assuming that the cooling gas (liquid nitrogen being an excellent choice) would be kept in a “sandwich layer” between two metal outer rocket-walls, and that tiny simple weep holes and-or passively thermally gated switches would allow cooling fluids to flow OUTWARDS, through the outer skin of the rocket body.  Here, for this application, we could let the cooling fluids weep INWARDS, into the “ice boxes” that keep our most sensitive gear (some of which self-heat when used) cold.  Passive, simple, tiny weep holes, thermally gated passive switches, and active electrically gated, temperature-sensors-equipped fluid switches could be used, any one alone, or in any desired mix.  In the below drawing, let’s just label them all as “pores”.  The coolant flows to the “ice-box”, and is injected (or flows) inwards as is needed.  At some “quiet spot” where there’s not much external airflow (ambient air flows undesirably, excessively pulling out our cooling gasses via the “chimney effect”), we place a small outlet hole to allow cooling-gasses outflows to ambient pressure, preventing our “ice-box” from becoming a balloon, and blowing up!

            See the below drawing of an “ice-box”, and keep in mind that this is the “deluxe” version.  Not all ice-boxes will need all of the layers shown.

 

 

Figure #1

 

            A good-quality “ice-box” will be needed for all devices that are located in hot ambient air, in the (during ascent, fairings-enclosed) payload bay, during descent (after the “flower petals open” and the payload is deployed, and the de-orbit burn is performed).  For many devices, we will have no sensible choices, other than to deploy them there, in the otherwise-now-emptied payload bay.

However, let’s assume we’ll want to keep things modular.  The payload bay will often contain a satellite (more often satellites plural) that need clean-room conditions, unlike the rocket itself.  The payload bay will be positively pressurized with clean air, on the way up.  This isn’t true of the inside of the rocket body, where ambient air pressure (outside of fuel and oxidizer tanks of course) is acceptable.  So there will be a payload bay “floor” (that needs to be airtight) and a rocket-body “roof” (not airtight), both of which will need to be bolted (or otherwise fastened) together.  We could call this a “deck” of the payload bay.

We can leave voids (holes) or indentations (pockets) in the roof of the rocket body.  Now we can put SOME of our heat-sensitive components “below the deck” instead of “above the deck”.  Below deck, using the shelter of a colder, LARGE rocket body (a large heat sink), we can use less-bulky and less-expensive “heat walls” to shelter some of our heat-sensitive components.  These might include avionics, batteries, hydraulic-fluid tanks, hydraulic accumulators, liquid nitrogen tanks, and (perhaps) a hydraulic motor driving an electrical generator, to keep our electrical batteries topped off.

To keep things modular and not too messy, we’ll want to keep the connections and disconnections between the rocket body and the payload compartment minimized.  Since many components MUST be above deck, that means we’ll want to cluster most of the associated parts together, associated with the payload bay.  On the other hand, that makes the payload bay (which must move from a clean room where the payload is assembled) heavier (harder to move).  For that reason, heavy (and dirty?) items might want to stay with the rocket body.  Some balance (of conflicting objectives) must be struck here.

 

 

Fleshing Out the Details of the “Powered Hinge”

 

            The parent document mentions a “powered hinge”, but never describes it in detail.  The time is now to describe a good design for such a device, in this application.  The “powered hinge” opens the “petals” of the bud (closed payload bay) to become petals instead of fairings-segments.  Petals later become “grid fins” as sacrificial materials fall, burn, or ablate away.

            Many of us are familiar with “gas shocks” for partially lifting or counter-balancing a (vertically opening) heavy door, or for softening the opening and closing of many kinds of doors.  A “gas shock” (pneumatic cylinder) here is highly similar to a hydraulic cylinder.  Envision it spanning from the payload bay “floor” (deck) towards a central strong spar on the “petal”, in a right triangle (in a petal-closed condition), with a triangle side on the petal, and a triangle side on the “deck”.  The hypotenuse is the hydraulic cylinder.  At way-wide-opened positions, the door-opening force exerted is WAY sub-optimal!  Far worse, the hydraulic cylinder would sweep through different positions for us, and we can’t use hoses (hydraulic or otherwise), or flexible “heat walls”, here.  I won’t bother to provide a special drawing here (unless asked, at RocketSlinger@SBCGlobal.net), so here are some illustrations of such a device:  Amazon Lifter.

            So a good solution that I could devise, is to add (rigidly, with no deliberately added ability to flex) an almost-1/4th of a circle (an arc, like a pie crust on a pie slice) of a strong heat-resistant metal, firmly bonded to the central, strong structural spar on a petal.  The arc can be called a “pusher arc”.  It will need to VERY strongly push on the petals, during descent, against prevailing winds.  If the “pusher arc” is ever “asked to” pull, instead, it will be only very weakly pulling inwards.  The “pusher arc” will be cogged on both sides, phased cog-to-cog at 180 degrees from one side to another, so that hydraulic-motor-driven cog wheels (qty 2) can grip it firmly at all times.  From a side profile, it will look like this:

 

Figure #2

 

            The metallic pusher arc will need to be built large and strong, to handle all of the forces involved.  The petals-becoming-grid-fins will try to twist, flutter, shake, and bend.  More on that later.

            The heat wall encloses a curved metal sheath (a slot; think of a scabbard).  The scabbard can dispense lubrication (grease) onto the heat-walled cogwheels and (cogged) pusher arc.  When the pusher arc becomes exposed to hot ambient air (as it must, when extended), the grease will burn or be blown away, to be replenished when it re-enters the “scabbard”.

            The cog wheels (driven by hydraulic motors) MUST be VERY firmly held into place, to handle the tremendous forces involved!  Structurally, they can be held up on a strong “post”, reinforced by “braces”.  These structures may or may not be encased by heat walls (or partially enclosed), as is convenient.  The below drawing should clarify and emphasize this point.  Note that these drawings can become quite cluttered, so they will sometimes be repeated, adding and subtracting different elements.

 

Figure #3

 

            The floor here (and the roof of the rocket body, together perhaps best called the “deck”) will need to be built stronger than it would otherwise have to be, with all of the forces exerted by the petals-becoming-grid-fins.  The petals will act like a manual, simple can opener with a prying handle (not the kind that you turn).  We don’t want to “open our can”, here, we want to recycle it without excessive damage!  Another way to think about it is to compare the “deck” here with what Space X calls their “thrust puck”.  We’ll likewise need to “beef up” this structure at the top lip of the “can” of the rocket body, similar to a “thrust puck”.  But this is outside of our scope to say more.  I have no detailed mechanical or structural expertise here!

            Another thing that can be done is to double up the pusher arcs.  Think of another one behind the one shown, in the above drawings.  One can be located at each end of the hinges.  If the hinges are long enough, perhaps we can line up 3 or 4 of these structures (pusher arcs, scabbards, etc.) on each hinge.  If we do so, adjustments will need to be made on the petals.  A strong central spar becomes 2 or more such spars, perhaps with holes in the grid fins in between multiple spars.  We don’t want to go “solid planar” out there, or we’ll collect too much force!

            Then yet another thing that we can do to “beef up” the powered hinge is to nest larger and smaller pusher arcs together.  Below, the arrangement is shown, conceptually, with the cogs stripped off of the pusher arcs, and the cogwheels not show, to simplify, and to make room.

 

Figure #4

 

            The above could have 4 separate “scabbards” for the 4 separate pusher arcs.  Or, we could go further yet, and smear all 4 arcs into a single solid plate (like a pie shape with the innermost tip nibbled off), and place arcs of tracks of holes in the plate, for cogwheels to run through these tracks (cog wheels rotate 90 degrees from what is shown in above drawings).  Now we would have just one scabbard, not 4, of course.

            In Norse mythology, the “Bifrost Bridge” is a rainbow bridge between otherworldly realms.  Our “otherworldly realms” being bridged by our rainbow bridge here are space, the exosphere, thermosphere, mesosphere, etc., so, for several reasons, we will call this entire design of a powered hinge (with variations) the “Bifrost Hinge” (please don’t think of me as being unhinged, for that).

            Now I have yet another idea that I consider worthy of describing, in terms of the “Bifrost Hinge”, then.  My mechanical intuition tells me that this is one of the best versions, actually, for whatever it is worth.  Picture in your mind’s eye, each of these cogged-on-both-edges pusher-arcs being replaced by TWO solid arc-plates, with rungs (like ladder-rungs) placed between the two (identically sized and shaped) pusher-arcs.  This should be easier to manufacture than the cogged-on-both-edges pusher-arcs.  The cogwheels then grab the “ladder” rungs (be sure to phase the cogs-pushing in an interleaved pattern, from one cogwheel to the next, for evenly modulated force).  It is NOT mandatory to place the cogwheels on both sides of the “ladder”.  If the outermost edges of the ladder-sides are VERY firmly supported and lubricated by the “scabbard”, then all cogwheels could reside on the inner edge.  The above-described solid “pusher plate” with tracks of holes, option, is OUT, for this design variation, because such a plate resides where the cogwheels want to go, here, and in the exact same plane.  Other options (nested rainbow style and one behind the other, on the same hinge) do remain viable, for the “ladder” approach.

            An illustration is easy enough to provide, so here it is:

 

Figure #5

 

            There’s one more mechanical aspect of “Bifrost Hinges” that I’d like to discus, but first, let’s briefly detour, and then return.  You’ll see why (assuming you’re a linear reader) in a moment.

 

 

Operational (Flight) Modes

 

            The “flower bud” will remain tightly closed, enclosing clean, pressurized air (for most applications) surrounding the payload, all of the way up to orbit.  The “petals” will then open fully, to get well out of the way for payload deployment.  Petal-opening forces there will be minimal.  There may be several stages of payload (satellite) deployments, sometimes even with orbital-adjustment burns in between.  If the user desires this, one could close the flower buds back up for intermediate journeys between different orbits.  If nothing else, closing the “bud” back up would protect the payload from being impacted by orbiting trash, during these journeys.  Opening and closing the petals will require minimal forces in a vacuum.

Next, the de-orbit burn is performed.  I’m assuming that we’d leave the flower fully opened up for this, or open it back up immediately after the burn.  Now, in the outermost, thinnest whispers of air, we can kill a maximum magnitude of speed, spreading out the heat and speed-wasting efforts, in the “calm before the storm”.  More slight stress now means less peak stress later.  So fully opening up the petals during this time would be best.  But stresses on the petal-spars and Bifrost hinges will remain minimal, during this time.

As the heat-storm builds up, the forces will amplify on the petals.  The petals will degrade into grid fins.  During this time, the petals-becoming-grid-fins should retract more and more, and not be allowed to be subjected to dangerously excessive forces.  Imbalanced forces from one petal to another (as some degrade into grid fins faster than others) can be corrected for, by deploying the most-degraded (less force-gathering) grid fins out further than those who are less degraded.  Measuring the tilt of the vehicle may be “good enough” to inform the decisions of the avionics package, with regard to this.  One might wish to equip the support structures internal to the Bifrost hinge system with “strain gauges”, and hydraulic pressures within the hydraulic circuits, to help inform the avionics control circuits, is an added parenthetical thought, though.  During this time, there will be tremendous forces pushing IN on the arc-pushers, making them hard to push OUT, but ridiculously easy to pull IN.

At some time towards the end of the process of petals degrading into grid fins, or shortly thereafter, we should enter a slow “rotisserie roll” mode (to spread the heat on the rocket body), which is caused by a slow process of rolling around the rim of the deck, a pattern of the grid fins being deployed in an imbalanced fashion.  The rotisserie roll will be accompanied by a corkscrew descent pattern for the entire vehicle.  Once again, the grid fins may not be deployed very far outwards, and the Bifrost hinges will be hard to push out, but easy to retract.  For more details about the rotisserie roll, and a drawing, see the parent document.  Figure #3 there is of interest.

Now, during this (rotisserie roll) mode, will the grid fins ever be deployed to their maximum reach?  It’s hard to tell…  How much mass are we willing to spend to make the grid-fins and grid-fins spars, the Bifrost hinges and supports, and the payload deck strong?  How much of the grid fins will be planar (as in the spars for interfacing to the Bifrost hinge’s pusher arcs), v/s how perforated will the grid fins be?  There are many variables at play here.

            Once we’ve reduced our speed (to sub-sonic levels) in the lower (much thicker) atmosphere, where we might want to deploy parachutes, the forces on the grid fins will be much reduced.  Here, for maximum added “parachute effect” of the grid fins, they’d best be deployed to their maximum possible reach.  But forces at this stage, on the grid fins, will be minimal.  If parachute fabric can survive here, grid fins (and their supports) will survive with colors flying!

 

Back to the Bifrost Hinges Designs

 

            So the most critical remaining open question from above is, during “rotisserie roll” mode, will we, or will we not, want to maximally extend our pusher arcs outwards, in any maximal-stress mode?  (I think this is the only time we’d maximally extend them in a high-stress mode, but I could be wrong).  ***IF*** we NEVER extend maximally, in a high-stress mode, then a question arises, which is, “Can we attain any safe mass reductions in our Bifrost hinges design, at the maximal reaches of the highest degrees of extension”?

            We can’t play games with the strength of the deck, or of the post or braces that fix (in space) the hydraulic motors and cogwheels (“drivetrain”), or the true hinge, or the drivetrain itself, or the spars on the petals-becoming-grid-fins.  All that is left is pusher-arc itself.  There, if our conditions are met, we can weaken the points on the pusher-arcs, at the points furthest-removed from where these arcs meet the petals.  This assumes that pushing hydraulic motors and cogwheels are clustered at the ends of the pusher-arcs which are closest to the petals (grid fins), as they should be.  There, at the far-removed tips of the pusher-arcs, we can add voids (holes), or internal inclusions of lower-density (weaker) metals, or other materials.  It’s not much, but there it is!  We can weaken the remote tips of the pusher-arcs, if the design (and operational) parameters allow.  Mass reductions are precious here, in ANY space-travel context!  Enough said.

            Now, all of that being said, let’s describe one more variation (or set of variations) on the Bifrost Hinge.  This is another top contender, and perhaps even the best of them all.

            We have already briefly discussed a variation of what is shown above, in Figure #4, where the multiple pusher-arcs can be smeared together, and tracks of holes can be placed into the smeared-together plate (in the shape of a truncated pie-slice).  Cogwheels could push on these arc-shaped tracks of holes.

            Let’s take this in a different direction.  Place no holes in the truncated-pie-slice-shaped plate, and turn it into a dry friction plate.  This means that we don’t intend to lubricate it.  The dry friction plate mates to a strong spar on the petal, of course.  On both sides of the friction plate, there are truncated-funnel shapes, which strongly squeeze the friction plate together.  The truncated funnel shapes have multiple grooves cut into the ODs (Outer Diameters) where V-belts can couple to multiple hydraulic motors.  The V-belt grooves are cut deep enough, so that the V-belts don’t touch the friction plate.  Parenthetically, V-belts could here be replaced by chains, with cogs on the funnels.  I prefer the idea of using V-belts, because the funnels are easier to manufacture, and there’s no cogs to be broken off.  V-belts would need to be strong and durable, and inspected (or replaced) between uses.  A torn V-belt caught up between a funnel and the friction plate could easily (and quickly) spell disaster!

            In either case, the funnels are free to spin on bearings at the small-tip area of the funnel (which has been truncated), and at or near the open end of the funnel as well.  Probably best, a spin axis (axle) would run clear through the entire funnel, and would be lubricated.  All of this (the entire drivetrain and support structures) would need to be ruggedly built.  Possibly, some areas (of the funnel) could include voids (or inclusions of less-dense, weaker metals or other materials), AKA be “honeycombed”, for saving mass.

            The funnels, motors, and V-belts can all be heat-walled, with only one significant area of heat exposure, and that’s where the (much-squeezed) friction plate travels in and out of the slot between the two rolling funnels.  Advantages here, then, are that no lubrication is exposed to high temperatures, and that we allow “slippage”.  If there are sudden jolts (“peak loads”) of high forces imparted to the petals (grid fins) and their spars and the friction plates, we’ll not get broken cogs…  We’ll instead get slippage between the friction plate and the funnels, and possibly also between the funnels and the V-Belts.

            The above is an introductory summary of where we’re going here, and why.  Keep in mind that the linear distance travelled on the surface of the funnel (for a given degree of rotation) is much greater at the OD (Outer Diameter) than the ID.  We want to optimally lay out a “grip line” where the friction plate meets the 2 “squeezer funnels”, and the linear speed rates of all elements agree with each other, all along the “grip line”.  Else our design elements conflict with one another, and we get “binding” instead of relatively free motions.

            Now before we start in on some drawings, it might be best to look at some basics of the design here.  I like to go intuitive and experimental on these things, with a smattering of theory.  The truncated pie-slice friction plate here, if we ignore the thickness of it, can be modelled by a piece of paper, or the walls of a cardboard box.  Get yourself a funnel, where the funnel walls come together at 90 degrees at the tip of the funnel (if you can find one, go “close enough” and imagine it being 90 degrees, or play with some paper and scissors, maybe even some tape too).  Imagine the friction plate being “rolled out” by your funnel.  Or place paint all around the outer surfaces of at least half of a funnel, and roll it on a piece of paper, with the tip of the funnel held in one spot.  The paint laid out on the paper will look like our “friction plate”.  A ½ (180 degrees) of a roll of the funnel is needed to “lay out” the 1/4th of a circle that is “laid down” in our 1/4th slice of pie, that represents the friction plate.  As we imagine the spinning funnel on the far side of the friction plate, we also see that the “grip line” MUST reside (no choice here) at 45 degrees, halfway between the petal wall and the deck.  This assumes we put the funnel in “square” with the petal and deck.  We can make SOME adjustments to that, as we’ll see further below.

            The whole scheme here can be envisioned as a sphere, with the center of the sphere being in the middle of the center of the true hinge (where petal meets deck).  The two squeezer funnels are tapered, cylindrical cones, whose tips “want to” reside at the exact center of our sphere, if we want precise agreement along the length of the “grip line”.  The friction plate (ignoring thickness) can be envisioned as being “laid out” by the same kind of rolling cylindrical cone (right angles of the sides, don’t forget) as we’ve described.

            It is NOT a good idea to have the “grip line” between the two rolling funnels meet at any kind of offset (of center of rotation) from the center of rotation of the friction plate, whose center of rotation MUST reside in the middle of the hinge!  “Geometry testing by drawing software” (Visio in this case, in the below drawing) should prove this point.  “H” is the hinge (the true standard) and H1, H2, and H3 are sample alternate centers of rotation, for the funnels, whose “grip lines” shift with them.  “Grip lines mismatched” means troubles, of course!  This means cheating is dangerous!  The centers of rotations for the funnels need to be at the center of the hinge, and, if offset, offset VERY little!  The point of the below drawing might be a little bit subtle, but a moment or two of consideration should help to clarify matters.

            More descriptions of this:  We are building a “slip clutch” or “torque limiter”.  The two spinning-top-style funnels can squeeze the friction plate very hard, and still remain fairly free to spin, while gripping (and moving) the plate firmly.  This is true IF the above rules are followed, and IF the funnels have 90-degree tips.  (This last part is true because the friction plate meets the petal at 90 degrees).  Otherwise, due to mismatched “grip lines”, your device will bind badly!

            It’s probably high time for more drawings. 

 

Figure #6

 

            The below drawing (of a funnel) assumes that we use V-belts.  We’ll assign this idea as being less than optimal shortly, but here, let’s cover it anyway.

 

Figure #7

 

            Several things quickly become apparent.  The funnel bangs into our petal, and the deck!  Both petal and deck can be “bulged” or moved out of the way.  We can’t put holes into the petal (which needs to stay airtight on the way up), but we can put airtight pockets into the deck, for the funnels.  For the friction plate, we can put pockets (“scabbards”), or, perhaps rubbery plastic air-seals (gaskets) attached to the slot-lips to seal the minimal clearance-spaces between the deck-slot and the friction plate.  These gaskets would be free to burn or abrade away later, during descent.  If we want to push the petals out more than 45 degrees, we’ll have to grow the friction plate below deck, with a slot cut into the deck.  I’m not sure how far the petals would have to be extended, to balance out their usefulness v/s the costs of growing the friction plates.  This is beyond my expertise!  PS, note, the axle (along with the entire funnel) will tilt towards the viewer at the tops of the axle and funnel.  The drawing is busy enough, without trying to show that.

            More than one set of V-belts could be used (I showed only one, to cut clutter).  The “grip” of the V-belts on the funnels would be good and long, and gently curved, for low wear and tear on the V-belts.  However, one would have to find room to squeeze a motor and pulley (per each V-belt) in there as well.  If they are kept small, we can find some room (between the funnel and the deck, away from the friction plate, and away from where the deck will have to be bulged down to make room for the lowest parts of the funnel).  Any other space, especially if we want a large pulley, eats up valuable room.  And a small pulley sharply bends the V-belt, causing rapid wear.  Parenthetically, to envision all this in 3D, it helps to place 2 funnels in 2 cardboard boxes, with the boxes wall-to-wall, and the tips of the funnels close together, in the bottom corners of the 2 boxes.  The walls between the 2 boxes is the friction plate.

            Probably yet worse, to protect the V-belts from the heat, we’ll have to shield (“heat wall”) most of the funnel (except we can’t do that where the funnel touches the friction plate, of course).  That means we eat up more space, and have to “invade” the petal and the deck some more.  Note that we ARE free to rotate the funnel clockwise a tad if we want to (in the above views), to avoid bulging the petal as much, but that, of course, invades the deck yet that much more.  Also note, the cut-outs for the V-belts (grooves in the funnels) will segment (void parts of) our grip line, reducing the grip of our slip-clutch.  So it’s probably best to totally drop the ideas about using V-belts.

            Now why did I like V-belts in the first place?  For “defense in depth” against “peak loading” on the slip-clutch function here.  V-belts can slip, without breaking cogs.  But if we think about it, looking at figures 2-3-4-5, the slip-clutches could be embedded into the couplings between hydraulic motors, and the driving cogwheels.  The driving cogwheels will still have inertia, so EXTREME (and sudden) peak loads could still break cogs (the slip-clutch function is further removed from the true hinge, with the use of driving cogwheels).  All in all, though…  V-belts have too many disadvantages!

            More minor parenthetical notes:  The bearings at both ends of the axle through the funnel will need to be lubricated, so therefore, heat-shielded as well (regardless of V-belts v/s cogwheels being used).  The bearings stay in fixed locations, so we are fine (I just don’t show the obvious things in the drawings, to reduce clutter).  Also, I make no comments here as to dimensions or to scale, in a lot of cases.  How large will the “Bifrost Hinges” system need to be, with respect to the sizes of the petals, and the entire deck?  I don’t know!  I don’t have the expertise required, here, to hazard a good guess, even.  “Not tiny” is clear, though, because they will need to be STRONG!

            Also note, the above drawing over-simplifies structural supports, above and beyond not showing braces.  The shorter post close to the true hinge can be just one post.  The taller post should be 3 or 4 posts, since the support as shown, if duplicated on the side where the funnel meets the friction plate at the “grip line”, should be moved to become two posts, straddling the grip line, with an overhead structure to mount the bearing.  Alternately, the axle can be longer, and the post further away from the petal (and longer as well).  I think that 2 shorter posts is best.  For symmetry and overall compactness, we might want to do the same thing on the near side (close to the viewer).  Re-stated, the drawing overlaps the far-side post (hiding behind the front post) and the grip line a bit, which is clearly not possible.  Almost the entire oversimplified “support structure” drawing above is a “glib fib”, because the funnel is tilted more complexly, in 3D, than what is shown.  We need firm supports, that fit correctly, and aren’t excessively massive, is really about “enough said”, I think.

            If we combine the funnel with cogwheels, we could do something like the below:

 

Figure #8

 

            The number of cogwheels could be 1 or more…  And the bulges for accommodating the funnels can be placed into the petals or the deck, or both.  I show (above) budging the petals out of the way.  Upon thinking about it some more, bulging the deck is probably far better.  Why?  Because tilting the funnels (towards the petals) and where the friction plate becomes squeezed between them may invite FOD (Foreign Object Damage), since materials from the petal (as it degrades in high winds) may get caught in there.  Tilting the funnels away from the petals reduces this risk.

            Tilting the funnels towards the deck, instead, means that we have to invade the rocket body some more, to extend the friction plate, while still retaining our ability to maximally extend our petals.  In comments immediately below Figure #7, we discussed adding rubbery gaskets to air-seal the narrow gaps between the frictions plates and slots in the deck.  This approach, too, would risk the gaskets (as they degrade) creating FOD between the funnels and the friction plate.  So we might as well sheath (think “scabbards”) the entire potential travel distances of the friction plates, keeping an air seal.  The furthest-away-from-the-petals points of the scabbards can meet (or be braced against) the rocket body wall, below deck.  This helps build structural strength.  Search for a “can opener” search-string further above in this document, to see comments about structural strength here.  In summary, to prevent FOD and to maintain structural strength, the friction plates should best be sheathed (not gaskets-in-slots-sealed), the funnels should be tilted into bulges in the deck, and the sheaths (“scabbards”) should be buttressed against (or otherwise merged into) the walls of the rocket body.  Rocket-body designers will have to accommodate this invasion, in the name of recycling most of our uppermost stage, if this version of the Bifrost hinge is used.

Also note that (not clearly shown) where the friction plate meets the petal wall, there’s plenty of room between the funnels, for putting a “V” shaped structural element thickening the friction plate (bottom, sharp point of the “V” shape pointing towards the grip line), for strengthening the joint where the petal meets the friction plate.  This is where a strong structural spar resides in the middle of the petal-becomes-a-grid-fin.  We’d have to look “downwards” onto the friction plate (edge-on, onto the friction plate), between 2 funnels, to show that.  Ask for more drawings, if needed, at RocketSlinger@SBCGlobal.net.  Small parenthetical note:  Placing more than one SET of funnels (plus one friction plate per set) on one true hinge seems to me to be a crazy idea, unless the funnels (etc.) are very small, or the hinge is very long.  I wouldn’t recommend the idea.  Other than that, I think that the “Bifrost Hinge”, in many variations, has now been thoroughly described.

 

Routing Airflow to an Energy-Harvesting Device or Devices

 

            As we’ll see later (further below), I believe that a river-boat-style “paddle wheel” energy-harvesting device will be most sensible for our uses here.  So a crude drawing of such a wheel is shown in the drawings below.

            The parent document mentions the idea of placing the energy-harvesting devices on twin rails or a grid-fins-style perforated plate, to thrust the harvesting devices “out into the breezes” after payload deployment.  Due to lack of high-temperature lubricants (thus making transfer of mechanical energy via roller chains be not such a good idea), and high-temperature hydraulic pumps, hoses, and fluids likewise off bounds, the entire idea of mobile-location energy-harvesting devices should be written off.  And trying to embed the harvesting devices into the petal walls (with good air seals) is similarly complex and therefore off limits.

            I have grown wiser now!  As was apparently retold by Francis Bacon, “If the mountain will not come to Muhammad, then Muhammad must go to the mountain”.  If we can’t move our energy-harvesting devices into the airstreams, then we must bring our airstreams to the (fixed locations) energy-harvesting devices.  Where one petal meets another (where they are latched together by electromechanical latches during ascent), we place bulges in the petals.  These bulges will turn our entire design into a yet more-so “partial hammerhead” design, but it is worth the price.  I envision only 2 (not 4) of these bulges, to preserve symmetry, while keeping complexity down to a dull roar.  Below, we repeat elements of Figure #4 from the parent document to show what I mean.

 

Figure #9

 

            The purple-colored sliding airflow-damper plates (above) should be located in line with, or slightly above or below the deck.  When closed (shutting off all airflow), they might mate directly to slots in the bottoms of the petals, where the petals bulge out, there.  I can see no reason why we can’t do it that way.  If, however, we have to leave an arc (half-circle) of structural material out there, for some reason, after the petals open, that might work as well.  If these structural arcs get blown away during descent, that might be acceptable as well.

            The sliding plates can be driven by hydraulic cylinders, with the cylinder bodies themselves heat-shielded, and the long skinny metal rods left exposed to the heat.  Hydraulic cylinders are fairly simple, which is why I favor their use here.  Another alternate choice is a “screw actuator” type of linear actuator, driven by a hydraulic motor.  Close the damper until after the payload is ready to deploy.  Afterwards, withdraw the damper to allow airflow (unobstructed by the rocket body) to power one edge of an energy-harvester (I favor a paddle-wheel design).  The sliding damper can double up to adjust how much or how little airflow is fed to the energy-harvester, which should be located right above the damper.

 

Figure #10

 

            The wind-collecting sides of the paddle wheels are now exposed to high winds (and high pressure), while the sheltered sides are not.  The wheel axles (and the hydraulic pumps that go with them) are now also fairly well sheltered from the wind and heat, while staying in a fixed location.  That’s what this whole scheme is about!

            I don’t know how much or how little wind we want to collect.  The above shows no “air ram-scoops” tacked onto the flanks of the rocket body.  We could put amplifying scoops or reducing scoops there as may be needed.  The below drawing is easy enough to provide, so here it is:

 

Figure #11

 

            The reducing scoop may not make sense if one regards it as being redundant with the airflow-reducing function of choking off the damper a bit.  However, if the adjustable damper is delicate compared to the strength of the wind, a strongly-built reducing scoop may make more sense than beefing up the adjustable damper.

            It is entirely possible, too, that we could put air-scoops down on the flanks of the rocket body, and invade the body of the rocket with angled air-pipes, and bring the airflow fountains (geysers?) up into the deck, further away from the edges of the deck.  We’d still need our sliding dampers to choke off airflow during ascent, to keep our clean air (around the payload) protected.  This method would eliminate our need to put special bulges in the petals (fairings), at the twin expenses of making the rocket body more complex, and of impeding straight-line airflow.  This idea seems simple enough to me, that I provide no drawing.

            Another idea is to put one single straight-line air-pipe right up the centerline of the rocket body (in the middle of the rocket engines and everything).  Fuel and oxidizer tanks might turn into toroids or “donut shapes” to accommodate this pipe.  Now we could have just one “air fountain” in the middle of the deck.  I doubt that rocket-body designers would be too happy to make all of the changes required (I could be wrong).  Once again, this idea is simple enough, to skip the drawing.

 

Energy-Harvesting Devices

 

            I favor the paddle-wheel, for its simplicity.  I don’t think I need to elaborate on that much at all.  If, as described above, one side of the wheel is in the high winds, and the other side (as well as the axle and hydraulic pump) is sheltered (and heat-walled), it should work just fine.

            The other is a wind chainsaw, or chain windmill.  See the abstract way up above for a quick summary.  See https://en.wikipedia.org/wiki/Bicycle_chain to understand the terms used to describe the parts of a “roller chain”.  Here, for a chain windmill, we could just add (to the “outer plates” in the chain) rigid wind-catching wind-vanes, similar to the vanes or paddles on a paddle wheel.  We’d need them, not on each and every outer plate of the chain, but, say, on every 5th or 6th or 7th one, or some such.

            If one side of the chain is in the stiff breeze (two cogged wheels carry the chain, with the wheels being of a sufficiently large diameter), and the other side is sheltered, we are doing fine…  Except now, we have TWO wheels that need lubricated, and their axles protected from heat, AND we have NO reliable method of keeping the chain itself lubricated.  This is extra trouble and complexity, for no good reason, compared to the paddle wheel, in my opinion.

            However, I did have 2 ideas that might be useful for a “chain windmill” in some other context, so I’ll briefly describe them “just for grins”.  ***IF*** we did NOT have a clear difference between a stiff breeze on one side, and shelter on the other (and we could ideally also lubricate the chain, of course), then “folding the vanes out of the way” on the chain’s return journey would make sense.  Here is what that would look like, in incarnation #1:

 

Figure #12

 

            The hardstop angle could be adjusted if the angle of the wind chainsaw (to the wind) is desired to be oblique rather than 90 degrees.  The hardstop and the vane should span from one outer plate to the other, on both sides of the roller chain.  Other than that, the drawing should make the idea fairly obvious.

            The “butterfly vanes” approach is to have two pins poking upwards out of the chain (from both outer plates on both sides of the chain), with the pins serving as pivot axles for the “butterfly wings” (vanes), and each vane containing the built-in hardstop for the other.

 

 

Figure #13

 

            The parent document mentions a “screw turbine” as yet another energy-harvesting device.  Now that I know that the windward side of a re-entry vehicle has pressures roughly 100 times that of the lee side, and have researched the state of the art for heat-tolerant materials, I believe that the “screw turbines” idea (whether ducted or not) is a vastly inferior design, for this application, compared to the champion, which is the paddle wheel.  A screw turbine would require transferring power away from the center-line of the device, which presents all sorts of troubles, with obstructing airflow and with heat-sheltering components.  I can speculate and provide drawings (contact me as usual at RocketSlinger@SBCGlobal.net if desired), but I think that the screw turbine deserves no more discussions or drawings here.

 

Hydraulics and Hydraulic Power Management

 

            I’m an amateur here, as in so much else.  Internet research (and the Mighty Google, Which Knows All) empowers me here.  Bidirectional hydraulic motors are clearly a “thing”.  At a casual glance, see https://www.globalspec.com/industrial-directory/bidirectional_hydraulic_motors.  So, in the Bifrost hinge, we can do what we want, there.  One motor can be used to open and shut the hinges.

            For bidirectional hydraulic oil pumps, see https://www.rg-group.com/resources/blog/bi-directional-pumps, which says… “NOTE: Don’t confuse bi-directional pumps with bi-rotational pumps. Bi-rotational pumps can flow out of either port, but only when rotation reverses. A bi-rotational pump has one port hooked to tank and the other port piped to the circuit.”  Also it says “Normally, bi-directional pumps do not have a port piped to tank.”  So, in the one and only place where I recommend the use of a hydraulic cylinder (in the adjustable airflow dampener feeding the air to the paddle-wheel), we might wish to go “closed hydraulic circuit” and use a bi-directional pump, to power the cylinder, here.

            As far as the “tank” is concerned, to my knowledge, this is a general-purpose reservoir of hydraulic fluid, which, unlike a “hydraulic battery”, doesn’t place the fluid under high pressure.  In our application, though, we’ll have to deal with zero gravity, as in, for example, when we first activate the Bifrost hinges (to open the petals), before deploying the payload.  We can’t deal with “sucking vacuum” from the tank, because our oil is vacuum-cavitating and floating around in blobs.  We’ll have to create a “tank” (which can have weaker walls, unlike a “hydraulic battery”) which at least weakly pressurizes the hydraulic fluid, behind a sliding wall, just like in a hydraulic cylinder, where the wall is called a “seal” (or piston), usually, it seems.  This will keep us from spilling the contents of our tank, in zero gravity!  On the empty side of our seal, we can place (sealed) weak nitrogen pressure, or mechanical springs.  Use nitrogen here, not air; we don’t want to risk burning our hydraulic fluid.

            One of the things that I am wondering about is, during the “rotisserie roll”, there will be tremendous amount of air pressure pressing in against one petal-becoming-a-grid-fin, as we retract it, and extend its neighbor.  Instead of bleeding high pressure fluid back to the (low-pressure) “tank”, and having to re-spend much energy fighting air pressure on the neighboring petal as we extend it, can we switch high-pressure fluid from the being-retracted petal’s motor(s), to the being-extended petal’s motor(s)?  Solenoid-switched fluid-flow valves (including “check valves”) and switches are a given.  In electrical power management, we can combine a motor and a generator and create an “alternator”.  Can we do the same thing in hydraulics, and create a combined hydraulic motor-and-pump?  This would allow us to efficiently do the “rotisserie roll”, by having one combined pump-motor supply the next, with high-pressure fluid, I think.

The answer isn’t immediately clear.  https://en.wikipedia.org/wiki/Hydraulic_motor seems to imply that it’s not easy to do that.  https://www.plantautomation-technology.com/products/kazel-hydraulic/hydraulic-motor-pump-combination isn’t very clear (probably involves electrical power).  https://www.hydraulicspneumatics.com/technologies/accumulators/article/21883829/accumulators is good background information on accumulators.  https://www.graco.com/us/en/in-plant-manufacturing/products/lubrication/automatic-lubrication/hydraulic-accumulator-supply-pumps.html describes air-powered pumps to feed power to accumulators (that one’s not very directly related to our application).  So in any case, in our “Bifrost Hinges” design, whether we need to combine a (birotational or bidirectional) hydraulic motor with a pump (as 2 discrete units or as one unified unit), we CAN convert mechanical motion (from relieving high fluid pressures as the petal retracts), to pump-power the extension of the next petal.  It CAN be done!  If “Graco” can do it (pump hydraulic fluid) with air pressure, we can do it with hydraulic fluid pressure.

Partially repeated from a few paragraphs up, see https://www.rg-group.com/resources/blog/bi-directional-pumps, which says… “NOTE: Don’t confuse bi-directional pumps with bi-rotational pumps.  (Deletions here).  A bi-rotational pump has one port hooked to tank and the other port piped to the circuit.”  So then, my question would be, when the petal is pushing in strongly on us (and we want to allow it to retract), can we use a bi-rotational pump, and pipe high-pressure fluid, not to the tank, but to an accumulator?  Or over to the next petal that needs to be extended?  If the answer is “yes”, then we are “home for free”!  We get to extract power as we retract!

If all else fails, we can STILL do this job of recovering some power, I will bet!  See some of the above drawings of the Bifrost hinges (especially figures #2, 3, 5, and 8) and we can see that there are multiple locations available, in the drivetrain, for pumps and motors.  If need be, we could place a weak unidirectional motor for powered retraction (as in a vacuum, for post-payload-deployment petal-retraction, if desired, when the wind isn’t pushing us), plus a strong unidirectional motor for powered deployment and place-holding (braking), PLUS a pump, to extract power to be fed back to an accumulator, or to a neighboring petal that needs extended, as we are retracting the strongly-being-pushed-on petal on our given instance.

As I understand it (I could be wrong), our strong unidirectional motor (in this scenario) will be fed high pressure fluid-feed to push outward on the petal.  It will maintain that high fluid pressure (from a pump and-or accumulator) in “braking” mode.  Now we could suddenly relax that motor-brake and feed excess fluid back to “tank” (don’t let the being-fed high-pressure fluids short-circuit to tank), at roughly the same time as we engage the pump, to turn strong mechanical energy (from the petal being pushed inwards by winds) back into fluid pressure (fed to the accumulator and-or the neighboring petal and Bifrost hinge, of course).  That should do it!  (Parenthetically, in electrical switching, especially with electromechanical relays, we talk of make-before-break, v/s break-before-make connections.  I don’t know what we’d do here with hydraulic switching).

Certainty concerning the above (or more details) are beyond me!  No hydraulic circuits drawings will be provided, at this time, from Yours Truly!

            A “hydraulic battery” will need to be thicker-walled (compared to a “tank”), to contain higher pressures.  See https://www.hydraulicspneumatics.com/technologies/accumulators/article/21882959/hydraulic-batteries-save-fuel “Hydraulic ‘Batteries’ Save Fuel”.  I’m no hydraulics expert, but I’m willing to bet that we need one (or more) in the systems envisioned here.  What I can offer up, is an idea:  The air (or mechanical springs, etc.) pressure in an accumulator could be indirectly replaced, in our application, with the airflow pressure exerted by our vehicle’s re-entry.  After our high-speed re-entry is done, of course, our high-pressure source will be fading away, and our Bifrost-hinges-stored hydraulic-fluid pressure (more importantly, also our stored potential mechanical energy) will largely fade away with it also.  But during the maximum-stresses parts of re-entry, this scheme may work just fine.

            That is, our special accumulator is an “airflow spoiler” down on the flank of the rocket body.  That does means that we have to move hydraulic fluids down there and back, in a heat-walled manner.  I think (and hope) that this can be done without invading the body of the rocket all that terribly much.  We also assume that we can find a way to feed hydraulic fluid pressures back and forth through a Bifrost hinge (see slightly above) to substitute for air pressure in a hydro-pneumatic accumulator.  The “accumulation” function is stored potential mechanical energy in the form of the extended friction plates (or pusher arcs) in the Bifrost hinges, which can pump fluid as they are pushed back in by the strong prevailing winds, is another way to put it.  Such a combined airflow spoiler and accumulator could be located wherever best makes sense.

 

Figure #14

 

            The hinged plate should roughly conform to the rocket body, and be heat-shielded in the same manner as the surface of the rocket body is heat-shielded, or better.  It is assumed to be solid planar here, not a “grid fin”, although I suppose it could possibly be grid-fin-style, as another choice.  If it is solid planar, it will gather a lot of force, and so, should be built thick and rugged.

            The airflow spoiler could possibly be located near the bulge in the deck and the petals, and the optional reducing air scoop; see figures #9, 10, and 11.  The spoiler would perhaps protect the sliding plate (airflow damper) and paddle wheel from too-strong airflows.  If this is going to be a purpose of the combined airflow spoiler and accumulator, the two sometimes-conflicting purposes must somehow be balanced, of course.

            The spoilers might best be added in balanced pairs, to maintain overall symmetry.  Note also that the “power storage” function here of a spoiler plate’s extension on a Bifrost hinge is really the same as what we have in a petal-becomes-a-grid-fin and a Bifrost hinge.  The difference is that the former isn’t envisioned as being part of the “rotisserie roll” function, and the latter is.  So the former is more purely an “auxiliary power accumulator”.  Once again, all of this assumes that we can pull fluid-pumping power out of extended-against-the-wind grid fins and spoilers, which I do firmly believe is a safe assumption.

 

Avionics and Flight Control

 

            I’ve previously commented (both here and in the parent document) about the “rotisserie roll” and an associated corkscrew descent pattern for the entire vehicle.  The avionics circuits (controlling the hydraulics of course) may, from time to time, shift from overall flight-path corrections as being the primary concern, to maintaining the rotisserie roll.  If all is working well, most of the time, both objectives can be met at the same time.  And as previously remarked, during the time that the petals degrade and become grid fins, imbalances here can be corrected by adjusting which petal is further deployed than another.  I have nothing more to add to that, other than the below…

            Immediately below Figure #8, I added comments about FOD (Foreign Object Damage), as materials from degrading petals may get caught between elements of the Bifrost hinge (of any design variation, really).  We should design to prevent this, both in the Bifrost hinge, and in the materials that will ablate, burn, or fall away, as petals degrade.

            However, we should probably design into the avionics, methods of getting ingested FOD materials out of the Bifrost hinges.  If sensors (motion sensors, hydraulic fluid pressure sensors, strain gauges, etc.) detect a “stuck hinge”, the hinge should be extended a bit, to see if that fixes it (disgorge the FOD materials).  Make several attempts before giving up.  So along with maintaining the flight path and the rotisserie roll, this might best be a third duty of the avionics circuits, with the duties interleaved as need be.

 

Conclusions

 

            “Bifrost Hinges” (in any of several design variations) would make good powered hinges for a shuttlecock-style atmospheric re-entry.  Paddlewheels could be used to collect power.  Airflow spoilers could serve combined purposes of slowing the descent (spoiling airflow of course) and as “hydraulic accumulators”, which store energy in the form of potential mechanical energy, as Bifrost hinges are extended against strong winds.

            Open questions remain.  Can a port from a bi-rotational pump be routed to an accumulator instead of a tank, thereby conserving power?  Can a hydraulic pump be combined with a hydraulic motor, without electrical power being involved?  Are the above design notes optimal, especially concerning hydraulics?  What would a detailed “hydraulic circuit” diagram look like, for a Bifrost hinge?  I plan to see if I can get a hydraulics expert to comment.  Maybe I’ll write or co-author another paper to flesh this out some more.  Please contact me if you are (or know such) an expert who might be willing to comment, contribute, or co-author.  See my email address immediately below.

 

I have no special expertise or any more plausible ideas concerning any associated matters here, so I will sign off at this time.  This concludes my ideas as of this time.  Comments or questions (or idea contributions) are welcomed at RocketSlinger@SBCGlobal.net …

 

Stay tuned…  Talk to me!  RocketSlinger@SBCGlobal.net

 

References

 

Stauffer, Titus. (2021). Methods of Decelerating a Spacecraft Through Atmospheric Re-entry Using a Shuttlecock-like Design.

 

Stauffer, Titus. (2019). Designs for Passively, Thermally Gated Fluid Flow Switches.

 

Back to main site at  www.rocketslinger.com