back to article Reg readers ponder LOHAN's substantial globes

Our Low Orbit Helium Assisted Navigator (LOHAN) project is beginning to work up a good head of steam, as we begin to consider the practicalities of launching a rocket-powered spaceplane from under a whopping helium balloon at a not inconsiderable altitude. Click here for a bigger version of the LOHAN graphic Thanks to all of …


This topic is closed for new posts.
  1. Martin 66
    Big Brother

    mercury switch

    Have you considered using a mercury switch to determine and correct the orientation of the plane ? Its quite simple, a small capsule (think of your standard antibiotic size pill) filled with mercury. One contact on the bottom, where the mercury sits when flat, one contact on either side. This can then be used to move a flap or rudder whenever the mercury makes contact (ie. whenever the plan pitches or rolls, depending on which way you install the switch).

    Lightweight and cheap. Mercury freezes at -38 degrees C, so this may be an obstacle depending on how high you intend to go.

    Anyway, implementing such a switch could enable you to maintain a steady flight path of 45 deg, or even allow you to start at horizontal on release, can correct to vertical over a few metres, avoiding the ballon then moving to a vertical path.

    1. RayG

      Or an accelerometer

      Better yet, tiny accelerometers are quite cheap now and can detect what direction "down" is quite reliably (this is how phones/tablets determine orientation flips).

      Not so great if the plane is swinging, or once you light the engine, but neither is the mercury switch - and it is easier to detect swinging from an accelerometer than it is from the mercury. Plus, we can easily log a peak acceleration figure :-D

      1. Ian Michael Gumby

        How complicated do you want LOHAN to be?

        Its not just a question of putting in tiny accelerometers but building the flight control computer behind it and then writing the software that will change the control surfaces as needed.

        You then have to test it, retest it, and that takes time, effort, and of course money. Essentially you'll end up with an auto pilot capable of landing a plane.

        I'm not an RC modeler, but assuming that something exists that could be put in to a model plane, it would be interesting. Assuming that it doesn't exist but could be built, you'd have to shrink it to be small enough and light enough to fit in your plane. That too takes time, expertise and money.

        You want to go 'low tech' you could do the following:

        Create a detachable weight and make the tail of the aircraft heavy so that when released from the balloon it drops tail first. When ignition happens, the weight detaches and the plane is angled up.

        Assume that if you time the delay long enough, your probability of hitting your balloon decreases to a small enough number.

        I haven't followed LOHAN enough to know just how complicated of a flight system that El Reg wants to build. The launch scheme I suggested could be done with models (gliders) that contain no flight control systems dropped from a balloon at a controlled height. They key is working out how to get the timer to detach your weight and start the ignition. I'd suggest explosive bolts but I don't know if El Reg has access to aerospace tech like that, but the timer and electronics would be very small and simple.

        1. RayG

          Easier than you think

          Hang on a moment. We're talking about a trigger used before flight here, not something that will move control surfaces.

          Not that there are not loads of people doing the latter now - google ardupilot. One of the great miracles of the modern age is that these things are not only cheap, they are very easy to use.

        2. Argh


          I wonder how heavy the weight would need to be and if there would be a problem dropping it from a great height where there might be people walking.

        3. ridley

          Not a good idea....

          I suggest you do a test of your theory by building a rocket yourself. If you put the weight at the bottom it is unstable, usually you have to put weight at the top to make it stable. Doesnt sound right does it....

          So by putting a weight at the bottom you are ensuring that the Centre of Gravity will go below the Centre of pressure and if that happens it will try to fly tail first.

          Now of course you might quite like LOHAN tail first.

          1. Ian Michael Gumby

            The point is to make the plane tail heavy on launch

            Assume that the plane is perfectly balanced.

            All you need is a small weight large enough to force the tail to drop as the plane falls from the balloon.

            the idea is to make the tail just heavy enough to force the nose up in free fall and when the engines ignite the weight is lost.

            Depending on how much weight is required, you could use something like dry ice or something combustible that gets burned on ignition. (The later would be more difficult and more dangerous.)

            1. Richard 12 Silver badge

              Physics fail there

              Gravity alone cannot cause objects to rotate*, because it acts on the centre of gravity - thus there is no moment.

              From the point of view of the accelerometer, it will see the rocket thrust, then zero once that burns out.

              It'll stay at zero until atmospheric resistance starts to become noticeable (going fast enough in dense enough air), at which point it'll read the atmospheric drag acceleration only. That's not necessarily in the direction of 'up/down' though!

              Gyros are the only way to get an orientation.

              *I'm excluding very high spacial curvature here, we're not near an object that massive!

              1. Ian Michael Gumby

                @Richard 12

                Ok... you have a plane that is perfectly balanced. That is the center of the plane is also the center of gravity.

                You add a weight to the tail. This shifts your center of gravity.

                Add to're not doing free fall in a vacuum. ;-)

                (Do I have to spell that out to you? especially when you're talking about lifting bodies? :-)

    2. ridley


      I have used these

      in water rockets. It is harder than you think determining which way is up when you are accelerating, it is also a non trivial task to determine apogee (which is the ideal with water rockets to deploy chutes at that point) which may not be much of a problem with LOHAN


      1. RayG

        Apogee WOULD be impressive

        I have something very similar. It's easy when you're not accelerating, hard when you are... but when simply swinging gently under slow movement it's not so bad.

        1. ridley

          Detecting Apogee is very hard.

          When the rocket is firing you will have +ve Accn upwards.

          As soon as the motor burns out you will still be accelerating just at 1g downwards (drag would make it higher but at that altitude not by much), however you are still moving upwards.

          When you reach apogee, from an acceleration point of view, nothing changes. So using an accelerometer how do you detect apogee?

          You need to detect the point the rocket tips over, assuming LOHAN aint a tail slider.


    3. Anonymous Coward

      One contact on the bottom, where the mercury sits when flat

      except under acceleration ...

  2. Andus McCoatover

    Two balloons, of course

    After all, El Reg. coined the "Bulgarian Airbags" term, so I think it would be most fitting (ooer, missus!)

    Seriously, one. Worked last time.

  3. Chris Tierney

    put the plane inside the balloon.

    when it bursts it will launch

    1. Steve X

      when it bursts it will launch

      and when it launches it will burst. Win-Win.

      If the ballon were filled with hydrogen it would be an even more spectacular launch.

      (OK, OK, I know)

  4. Chris Partridge

    Three balloons better

    Three balloons could be attached round a launching tube for the rocket, which would automatically point upwards with the business end of the rocket pointing down, away from the balloons. Simple.

    1. Yet Another Commentard

      Re: Three balloons better

      There is an Occam-esque appeal in the simplicity of this.

      Not having any physics/rocketry/engineering training isn't going to stop me commenting though.

      How tough are the balloons? As they rub on the tube/each other will it make them burst prematurely?

      Should one burst at (say) 100 feet, how do we deal with the now crazy angle of the launch tube? Does that matter?

      What would the tube be made from?

      Do we want to launch vertically, I thought the optimum angle was 45degrees for some reason?

      As I recall the PARIS craft froze to its platform - a tube infers a lot of potential contact points, is it likely to freeze in there?

      Does one launch these rockets from inside a tube (like a mortar) or from a milk-bottle type thing (like a firework) or does it not matter?

      1. Stoneshop Silver badge


        "Do we want to launch vertically, I thought the optimum angle was 45degrees for some reason?"

        45 deg is when you want to lob a projectile the greatest horizontal distance. Which is not a design criterium for LOHAN (apart from that, it's not a projectile but a glider). Altitude is what we want, hence you want as much power as possible to be spent in a vertical fashion.

    2. John Sturdy
      Thumb Up

      Nice and simple

      It does sound like the simplest one, to me. However, you'd have to make sure each of the balloons provides the same amount of lift.

      Also, you could name the balloon(s) stage "Eccentrica".

  5. Magnus_Pym

    three balloons vertically

    just use different lengths of cable to tether the balloons. Find the video of that guy who launched himself skywards in patio chair to see the idea configuration.

    1. Stoneshop Silver badge

      Lawnchair Larry

      He went up to about 15 kfeet, IIRC. If PARIS' performance is anything to go by, LOHAN's jubs will lift her to well over five times that, meaning way more expansion and thus weaker skin.

  6. Marcus Aurelius

    In favour of hydrogen

    If hydrogen explodes, it explodes upwards. Unless you plan to be standing on top of the balloon when it explodes, you have no problem.

    Hydrogen is mucho cheaper

    In addition, if you were really clever, you could have the hydrogen gradually empty from the balloon into the rocket tank with rising altitude and use if extra flight time.

    1. Dave 64


      "In addition, if you were really clever, you could have the hydrogen gradually empty from the balloon into the rocket tank with rising altitude and use if extra flight time."

      But you'd maybe need to burden LOHAN with some tanked oxygen to burn that Hydrogen...would there be sufficuent oxygen at that height to combust the hydrogen

      1. Stoneshop Silver badge

        Hydrogen plus oxygen

        Plus a pump to get that hydrogen from the balloon into its tank, because pressure in the balloon is just over atmospheric. Which it has to do in a few seconds.

        Doing the math on the energy equations shows you're seriously way better off leaving that hydrogen to dissipate into the upper atmosphere unburned. Really seriously way better off.

  7. Johnny Canuck

    This is not a title

    How about a long spar suspended under the balloon, with one side counterweighted and perhaps vaned, and the other side extending far enough horizontally to clear the balloon on launch?

    1. Ian Michael Gumby

      Not a bad idea...

      On the one end you have the connection to the balloon. The connection is a spring loaded hinge so that upon release it launches the plane in an upward arc like a catapult.

      I'd say depending on the weight of the plane, look at some graphite tubing like from a golf club. Its light and durable. The only downside is that you've increased the size of your debris that could come falling down...

    2. graeme leggett

      Thinking of something like this myself

      Not necessarily a long spar but long enough to act as the launch rail for the aircraft.

      The aircraft having a couple of loops on the top, or a hollow spar running through which fits over the launch rail. Rail doesn't need to be very long since unlike a Black Cat Super-whizzer you will have some degree of aerodynamic control.

      Perhaps an additional "carrier" element with substantial fins could be used to get the aircraft free on a striaght path, thereafter falling away. More payload, but which is better - a guaranteed launch or maximum altitude.

      Counterbalance for the launch rail would come principally from the rest of the launcher ( camera, telemetry etc) but a single guideline running out from the balloon tether point to a second spar in the opposite side from the launch rail should fix it at the right angle.

  8. Chris Long

    Thrust vectoring

    Mount the rocket motor on gimbals and attach it to the same servos controlling pitch and yaw. Launch horizontally (or even down) but program the autopilot to pitch to vertical after launch.

    The autopilot will clearly need different goals during different phases of the 'flight' (remember that it won't in any real sense be 'flying' when above 80k feet, not unless its doing SR-71 speeds).

    1. Malcolm Boura 2

      Thrust vectoring 2

      Or put vanes into the rocket exhaust. If they are only needed whilst accelerating to a speed where more conventional control surfaces are effective they don't have to last very long.

  9. Graham Bartlett


    You could put the plane on the end of a long enough pole that the balloon doesn't get in the way, as suggested on the last thread. If you know what height you want to release at, Boyle's Law tells you how big the balloon will be. These balloons have huge lifting ability, so a reasonably strong bit of wood will not stop it going up.

  10. Thomas 4


    But hydrogen is fun. Hydrogen goes *boom*.


    Being serious for a moment, if you can't have a platform on the top of the balloon, you could launch from the side. From the top, it would look like a + sign, with strips bending downwards along the sides of the balloon. Three of the strips would have a weight on it equal to the plane, with the fourth holding LOHAN.

    1. Poor Coco
      Thumb Down

      Stability is the problem

      You cannot have a stable balloon configuration with anything but the payload UNDERNEATH. Furthermore, the skin of the balloon CANNOT repeat CANNOT support a load; it’s not a consistent shape during the flight, and it’s really fragile.

    2. phuzz Silver badge

      no title

      And now you only have 1/4 of the mass avaliable for the aircraft. C- must try harder.

    3. Neil Barnes Silver badge

      Ah, but when *helium* goes boom...

      there's enough energy around to get to orbit. And there won't be too much of that nasty beryllium byproduct lying around.

  11. Jaymax


    What /is/ it with you lot at SPB and hydrogen?

    Stuff ain't as scary as you think -- especially when it's in something as flimsy as a balloon.

    Hell, just get yourselves a little bit, a few party balloons, a pole with a lighter on the end, and make some bangs, you'll get over your namby-pamby woosiness soon enough.

    1. Robert E A Harvey

      soft whumping noise

      I used to work with a bu ch of lads who would inflate bin-bags from the oxy-acetelyne welder, and let them drift upwards. Lighted fag-ends flicked at them would eventually burn through and produce a soft, wet-sounding ball of flame and a lot of soot.

      1. John Gamble

        Re: soft whumping noise

        Er... no.

        Dunno about your wimpy acetylene, but a hydrogen balloon makes a much more impressive sound.

        Scene from my freshman chemistry class: Prof has graded exams, a floating balloon, and a lit torch (not slang for a battery-powered flashlight) on his desk. Prof says: "According to the exam, a depressing number of you did not know whether an exothermic reaction produces energy or not. For those of you who don't...", and waves the torch (which is on a long handle) to the balloon.

        LOUD BOOM. Also, the flash of light was impressive.

        This should not be considered an anti-hydrogen posting, but do be careful with lit cigarettes.

        1. ridley

          Re: soft whumping noise

          I suggest that the balloon was not only filled with Hydogen, large volumes of Hydrogen do not explode they burn from the outside in as the oxygen gets to it.

          Look at the video of the Hindenburg, that is not an explosion it is just burning fast, pretty damn fast I admit. However if the balloon has the right amount of Hydrogen AND Oxygen in it then the balloon will explode. (Technically it is still burning just very very fast, just like gunpowder burns very fast, C4 etc really do detonate)

          Before H&S stopped the practice in Schools we had to open the windows before doing this experiment...........

          1. John Gamble

            Re: soft whumping noise

            Perhaps -- I didn't ask the professor. But a toy balloon is not a large volume of hydrogen, and mixing of the hydrogen and the oxygen-containing outside atmosphere could happen very quickly, unlike the very large volume of the gasbags of the Hindenburg.

            And the larger point, of course, is that it doesn't take much hydrogen to create an impressive explosion.

  12. RayG

    Arc shaped launch rail

    How about

    One big giant balloon.

    Suspend a big hoop below it, hanging vertically, with a few threads of fishing line like bicycle wheel spokes to keep it circular.

    Fit launch rail from the bottom of the hoop curving up one side 'til it reaches the desired angle, then sticking off for a little bit (the last bit will need a rigid brace)

    Rocketplane sits at the bottom, no need for any kind of counterweight. Light the motors and it accelerates along a curved path along the rail to the right angle, missing the balloon, then flies off at a tangent to the circle on the intended path.

    Shouldn't weigh as much as a long rigid pole with a counterweight (oo-err?) as it's smaller and shouldn't need to be too hefty to be rigid enough (as it's efficiently braced)

    Also means you have a bit of speed before being totally untethered so aerodynamic stability will work better.

    1. Poor Coco
      Thumb Down

      No, no, no, no, no, no, no.

      The centripetal acceleration of the plane would throw the entire thing cockeyed. Zero chance of success.

      1. RayG

        I'd better do the sums again

        It will certainly drag the rail a bit... but unmanageably? As you're only using the bottom part of the arc, most of the force is in the vertical plane and pulls against the balloon (which can stand it); the remainder will try to turn the wheel about its axis but it's a comparatively small force and hasn't long to do it in.

        Hmm, I must test this! For science!

        1. Poor Coco

          While doing the sums...

          ...don’t forget dynamic forces. You know, Newton’s Third Law, the action/reaction one.

          We’re talking a multi-kilogram plane — the MAJORITY of the weight of the entire ballocket — suddenly moving along a curved path with nothing to anchor it! The rail with thrash about and the rocket will leave at an utterly random angle, it’s worse than useless.

          1. RayG

            Happy with the laws of Newton, it's the laws of Sod I'm worried about

            The rail exerts a force on the rocket to keep it on a curved path.

            Maybe I've got my sines and cosines the wrong way 'round, but it seems to me that the larger part (cos @ less than quarter pi) of the opposing force exerted on the rail (on an axis through the wheel hub) pulls down against the balloon, and the smaller part (sine @ less than quarter pi) will tend to generate a moment that will rotate the (top-anchored) wheel in the direction of launch (as if the rocket were "dragging the rail around with it", although this is a separate force from the friction term, which I'm hoping I can keep nice and low)... this moves the wheel centre away from a vertical line through the anchor point, so its weight (and the cos part of the force) partly counteracts this moment.

            The actual launch angle will be slightly higher than the set angle.

            On a balloon, laterally there's not a lot of force if the thrust is nice and symmetric; there is no crosswind (we're drifting at jolly close to wind speed). On a ground level test the wheel definitely needs anchoring against crosswinds rotating it about the vertical axis, but I don't think this would invalidate the test results.

            It's quite probably not right for LOHAN but I'm confident enough in my analysis to try it on a smaller scale (admittedly standing behind a wall)

  13. Geoff May

    Stupid question

    Have you decided on the exact aim of the project?

    I reckon there are 2 targets:

    1 - Go as high as possible

    2 - Go as far as possible

  14. ISYS
    Thumb Up

    Two balloons - One half inflated

    As the baloons rise and expand, the first one will eventually burst and the second one can 'take over' lifting duties until it has expanded to bursting point therefore acheiving more height before launch. I don't know if this is scientifically possible but I thought I would add my 2 pence worth.

    As to how to fire the rocket - Fire it sideways. As long as you have a piece of buttered toast attached to the top (butter side down) it will naturally point skywards. Now I know this is scientifically correct - I would suggest using a cat (same scientific principal) but the RSPCA may take a dim view of this.

    1. ISYS

      update on the cat

      If we don't use the cat for directional control we could always use it at ground level - it would boost team moral and provide useful pest control in the command centre. As to the pet cats name - it has to be LOHAN's Pussy....

      Go on moderator - I dare you!

  15. Richard Gadsden 1

    If you actually want a vertical launch

    And I suspect that the 45-degree angle is better, but one approach would be:

    Long pole.

    One (or two so it doesn't twist around too easily) balloons at each end

    Rocket attached to middle of pole.

    Shape something like an H but with the middle of the H being 20-30 metres long and the cross bars only a metre or two each. Then attach a balloon to each corner. Make sure you know how big the balloon will be at launch so they won't meet and you get a vertical launch platform.

    The problem is that it's not a stable launch platform, so you still need control surfaces, and the rocket backwash has a <b>much</b> higher chance of trashing the balloon than on a 45-degree launch.

    Incidentally, if you're really afraid of trashing payload on the balloon from the backwash, you might be better off with a glide launch to get some separation before firing the rocket.

  16. Anonymous Coward
    Anonymous Coward

    I'd be inclined to launch horizontally

    Then use the control surfaces to move upright. No fancy elevation rig required and if the control surfaces don't work from ignition then they aren't going to work anyway so you're not going to lose much.

    I would have thought igniting a rocket at altitude would be more of a problem than which way it's going to go.

  17. breakfast

    Swing til you're winning

    Could you use the release timing so that the rocket swings around the balloon and the balloon's inertia can be used to orient it to a more vertical angle before releasing it?

    I know nothing about rocketry or physics, but in my imagination this works perfectly and I believe hard enough that I think it just could work...

  18. Anonymous Coward


    an inverted triangular frame suspended way below the balloon. The frame is rigid so will not deform when the rocket motor ignites and send LOHAN off at an unintended direction, and because it's a long way below the balloon you can ensure that the flight trajectory is well away from it. You can have the balloon's box of tricks underneath that to act as a weight to stabilise the frame and ensure that the LOHAN is kept at the right attitude.

    One of the lower sides of the triangle would be the launch rail which the LOHAN is guided along when the engine fires.








  19. Robert E A Harvey

    Launch Structure

    Surely the launch structure would be two strings of different length, hanging the payload from the balloon at the correct angle to launch and avoid the canopy?

    Once the motor fires up it would slacken off the strings which would fall out of their hooks and robert's yer familial relation once removed.

  20. Reg J Mitchell

    A long piece of string

    To miss the large balloon, just dangle the launch platform a long way below the balloon.

    Use an asymmetric Y yoke to support the launch platform at the 45 deg launch angle - which can be steeper if the string is longer. The direction of launch won't be controlled, but you can wait (using the guidance gyro or a compass or something) until it is pointing in the required direction if you want, as it will almost certainly will be spinning around during the ascent.

    These aren't my ideas, but are easy to easy to find.

  21. Anonymous Coward
    Anonymous Coward

    Let the balloon float free

    How about releasing the balloon before it pops, wait a second or so before igniting the rocket, the rocket would be in freefall and the balloon would be accelerating away so the chances of hitting the balloon would be minimal (the balloon would rush upwards at a rate of knots once the weight of the rocket is removed, it would also move to the side due to any wind/current)

    You could make the balloon last longer by having some kind of pressure release valve allowing the He/H escape at a slow rate (obviously launching before there is insufficient gas left to raise the vehicle anymore)

    1. Steven Roper

      There's just one small problem...

      That wind you speak of that will blow the balloon away will also be knocking the rocket around. There will most likely also be some oscillation of the payload prior to launch, which will persist while the rocket is in freefall. All of this combines to mean that the rocket will be pointing in a pretty random direction when the motor fires. If that direction happens to be the ground, what you have is an incoming ballistic missile, which among other things could see the El Reg crew off for an extended holiday in Cuba. There needs to be some way of making sure the rocket is pointing generally upwards before firing the motor.

  22. Hodders

    Launch it downwards

    Assuming sufficient air density for aerodynamic flight and an autonomous flight control system. could you not hang it upside down under the balloon and launch vertically downwards. This makes for a simple release mechanism.

    Once in free flight and with sufficient velocity for the Aero surfaces to work, the control system could make a U turn, point the nose to the heavens and go for it.

  23. kryptonaut

    Hydrogen balloons

    In my more experimental days I used to fill party balloons with hydrogen, by reacting caustic soda (drain cleaner) with aluminium foil in a drinks bottle then fitting a balloon over the neck. Once the mix warmed up it would fill the balloon fairly quickly.

    Of course hydrogen being somewhat flammable, the obvious next step was to ignite it at altitude - so I tied some string to a balloon, soaked the string in methylated spirit, lit it and released it. Only rose about 20 ft before it made a nice noise, and showered burning rubber over my parents' driveway :-)

  24. bertino

    Idea for a pole.

    Use a carbon fibre fishing pole. My dad has some in his shed I am sure he would donate. They weigh about 1kg for a 13M long job, So assuming you need only about 4 metres to poke above the top of the ballons, should only come in at a couple of hundred grams. They are also hollow so you could route any wires down the middle if needs be.

  25. Poor Coco

    A major issue to consider

    At the stratospheric altitudes we’re considering here, we have about 1% of sea level air pressure. Aerodynamic controls are going to be minimally effective. Forget about casually flying around the balloon, the turning radius would be ludicrous.

    I have been toying around with the LOHAN-001 concept posted in the last article and I have a stability suggestion. If the rocket pod on the tower is mounted on a nice bearing and the engines thereon are canted, then the pod itself can be induced to spin at a tremendous rate for gyro-stabilization. By using a series of engines (perhaps even ignited by as low-tech a method as chemical fuses), we can choose a thrust pattern that will start with a largish spike of thrust (to overcome static friction) followed by a long and relatively constant burn at low thrust. This will result in the mass of the gyro staying high (conferring stability) for as long as possible.

    The main engine can then be relocated where it belongs, at the tail. With a fuselage like I sketched out — it has a 10cm (4") diameter — there is room for a BIG main engine. Like in the M impulse class or thereabouts, the kind of engines that can get rockets to 50,000 feet from ground level and that can get LOHAN to truly preposterous altitudes. (The set of 12 E engines I first suggested would add up to the lower end of the I class; this new approach will increase impulse almost 20 times.

    With an engine like that the speed of the plane will be stupendous, and strength will be the overarching concern. That means scratching the folding wing (which is okay, since the main thrust will be behind it now) and skinning the wing in plywood or carbon fibre.


    — Murray Pearson

    1. Richard 12 Silver badge

      Thrust bearings aren't that good.

      So it'll end up spinning the whole craft pretty quickly, unless the 'gyro' stabilisation component is mechanically driven by the craft. Built-in reaction wheels might be a better bet.

      Building the craft really strong isn't hard - carbon fibre monocoque is easy to crazily overbuild. The 150g class ROV combat vehicle that I built with carbon fibre has been thrown across the room several times during 'battles' and it was completely unharmed.

      Better to build it in one or two pieces though, rather than trying to join lots of parts later - the joints are usually much weaker.

    2. RayG

      More engine must be better

      as the title says.

      Just as long as it only lights that big M when pointed upwards...

  26. Mike Richards Silver badge

    We're going to need a bigger balloon

    Does the plane need to be directly beneath the balloon? Instead couldn't you have a horizontal boom suspended from the balloon with the plane on one end and a counterweight on the other. Fire the plane up at an angle before the balloon bursts and it should clear the rest of the vehicle.

    Either that or dump the balloon entirely and build a bigger rocket to create your very own Dynasoar. The rocket, of course, would have to fuelled with liquid hydrogen for maximum hilarity.

  27. Richard 12 Silver badge

    At the kind of launch height and speed we're talking about, control surfaces do little

    It's going to be horizontally stationary relative to the air, and moving relatively slowly vertically.

    So pretty much all control at launch will need be done by thrust vectoring.

    I doubt you can even expect aerodynamic drag to be able to stabilise the craft. That said, the platform looked pretty stable at drop last time so a launch rail might actually be worthwhile.

    Either way, multi-stage means you'll need the ability to direct the burn. Two axis servo control is simple and easy, and high-torque servos are pretty cheap. Maybe wiggle 'em around on the way up to ensure they don't freeze up.

    To control it, you're going to need some solid-state gyros. Accelerometers are only really going to be able to give you the burn acceleration, which is less useful - though interesting, and will help to know whether it's worth trying to correct a poor orientation.

    - Once the engines burn out, there's no point in trying to correct a tumble with them - keep the battery power for the elevons later. The accelerometer would also tell the craft when it's experiencing enough aerodynamic drag to try the elevons.

    You can get a triaxial gyro and accelerometer in the same SMT package these days, an "Inertial Measurement Unit". They weigh 5g or so!

  28. Miek

    Wow, there is a market for donut shaped balloons ...

    Admittedly, they are not weather balloons.

    1. John Gamble

      Re: Wow, there is a market for donut shaped balloons ...

      Considering where you would have to place the rocket, this is unnecessarily Freudian.

      1. bertino

        Even more so

        If you use a pile cushion.

  29. paulc

    ARGHHH it HAS to be TWO balloons...

    After all, the project acronym is LOHAN in honour of Lindsey... she of the enormous funbags that "escaped" when she was frolicking in the sea recently... google it... safesearch off and not in work...

  30. Dodgy Geezer Silver badge

    If you have rocket direction control....

    ...the answer is simple. Fire it off sideways and then let it climb up.

    It is easy to detect 'up', particularly at these heights. 'Up' is dark, and 'down' is light. An array of 4 photo-cells will provide attitude positioning very easily - a sort of 'artificial horizon'.

    I remember a model aircraft autopilot from the 1990s which used this technique, and worked very well. You can probably still buy the kit - it wasn't that expensive, and attached direct to standard radio control servos.

    1. ridley


      Do you mean these? The ACT (Anti Crash Technology) worked well and consisted of light sensors pointing up and down.

      With this one up was light down was dark, as soon as the system detected you were being too "enthuisiastic" it recovered stable flight for you. (The system could be turned off.)

      However it could make for interesting flights at dusk............


      1. Astarte

        Orientation sensors

        The idea is simple but you might find that at high altitude there's more light reflected from the clouds below than the darker sky above.

        1. M Gale

          The more reliable autoland systems are IR-based.

          Warm earth, cool sky. Should work at pretty much any altitude.

          Though there is something to be said for using webcams for optical horizon detection.

  31. silentcoder
    Thumb Up

    This is so easy

    If you let gravity help. Here's what I'd do. Attach to the plane's tail a hook, consisting of a straight bar at a 90 degree angle to the plane itself, and a bar attached to the end of it at a 45 degree angle to the straight bar - line it up with the direction of the plane so the second bar is just behind the tail. The put a simple switch on the second bar, which will trigger ignition. Now hang a nice weight from the first bar, and attach the plane to the balloon with the nose downward. The weight should be hanging from a pulley-wheel so it slides smoothly.

    So here's what happens, as soon as the plane is released, it starts falling down gaining velocity. The wait in the back along with the lift from the wings start tilting it back up - which will be automatically curved and correctly aligned, as soon as the plane is tilted far enough (at about 90 degrees - e.g. horizontally) the weight rolls on to the second bar, it now drags the tail down further but the wings are moving the plane forward, depending on setup you can easilly achieve a curve of several meters here already - and by adjusting the sizes you can make it much larger with ease. As the plane tilts more up - the weight starts rolling because the second bar is less and less steep - as the plane hits 45 Degrees, the weight rolls off, and as it does so it hits the ignition switch starting up the rocket - which is now several meters below and to the side of the balloon already, with the 45 degree angle at ignition - it's launching safely around the baloon and up.

    Sure you can do it all in software controls and such... but why use a complex solution when a simple one will do ? If you insist on writing code, then write some modeling software to work out the perfect weight and length of the bars so you won't need to determine it experimentally :P

  32. Ben Norris

    every action has an equal and opposite reaction

    clearly you need 2 rockets pointing in opposite directions and launched simultaneously :P

  33. Martijn Otto
    Thumb Up

    4 balloons

    I think the best way would be to use 4 balloons. Create a launch platform on it to keep LOHAN. If the balloons are far enough apart it would be quite stable.

    In addition, one could use some accelerometers to determine if the platform is horizontal and move some gas between the balloons to adjust it. Wouldn't be terribly hard to make.

    1. Jonathan Richards 1 Silver badge

      Would be terribly hard to make

      I Am Not An Aeronautical Engineer. But, from first principles: The lifting power of the ballon is proportional to its density relative to the air around it. The pressure in the balloon depends on the tension in the latex, so a less elastic balloon will contain the helium at a higher density than a more elastic one. The probability of matching four ballons in terms of their elasticity is indistinguishable from nil.

      Thus as it rises, the foursquare platform will tilt. A mechanism to sense the orientation and to pump helium between the balloons with sufficient speed and precision to maintain a horizontal attitude would indeed be terribly hard to make, in comparison with the simplicity of a single balloon. Edit: and it would have to work at pressure regimes from 1 bar to a small percentage of one bar...

      If a balloon bursts prematurely, the platform will instantly be a total loss. Even a single-balloon configuration will need some sort of mechanism to detect this condition, I expect, otherwise one risks dropping an unignited rocket on an unsuspecting country beneath.

      Who is Range Safety Officer? :)

  34. Anonymous Coward
    Thumb Up

    Simplest solution

    A really long bit of string.

    Parachute cord would be ideal. Type 1 has a breaking strain of 43kg and you can run about 64m with 100g of the stuff. Type 3 (the most common stuff) has a breaking strain of almost 250kg and you could run 75m for half a kg.

    Remember you need some spare weight carrying capacity for wind loadings, but it strikes me putting the launcher 50m below the balloon would almost guarantee no collision!

  35. bertino

    Simple solution

    Well, simple ish. 3 balloons, pole up the middle. LOHAN has another tube on the underside with a capped end at the top. You simply place LOHAN on top of the pole using this tube. As long as the pole points above horizontal, LOHAN should stay on top of the pole. No need for any release mech as it is not technically attached. Just fire the rockets when the altitude is high enough.


    1. Mr Grumblefish

      Place LOHAN on top of the pole using this tube

      Is this a "simple solution" or your personal fantasy?

  36. Paul 77

    Vectored Thrust

    I'm with Chris Long, except that I don't believe you need to gymbal the whole engine, I think it might be possible to have a steerable thrust guide, sort of like a kort nozzle on a ships prop. I think that would get rid of the worry about flight control surfaces during the first phase of the flight - just leave them centred.

    1. ridley

      V2 Steering

      So you mean a little like the method used by the V2 to steer the rocket at low speed and presumably when the fins became useless at really high altitude, though I would have thought by then the carbon would have burnt away.

      and strangely


  37. Skizz


    ...think of using three airbags has obviously been thinking too much about Eccentrica Gallumbits.

  38. SMK8

    Conventional Helium Balloon, only the Mylar type instead of Latex.

    Considering Bulgarian Airbags: Why Latex?

    Why not Mylar? -- The envelope is much larger, similar if not same capacity as the Latex Balloon, and much tougher to destroy, and will achieve greater altitudes? shows the effect of Several CF-18 Hornet fighters, RAF fighters, and Possible Incursion into Russian Airspace, and Still it don't come down...

    Single Envelope, Lower hanging framework for Launch, and no Human Powered ornithoptics required...

    Launch Framework for 68-degree launch orientation and Solid Rocket propellant from Cesaroni Technology Inc. (In Canada-land no less.)

    I think this a doable project. See : for consumer products.


  39. spegru

    Launch Tube

    I believe these are commonly used for launching rockets anyway. Attach the tube in the gap between 3 balloons. I don't reckon a longish tuby thing would harm the balloons at all. Just abit of perspex tube or drainpipe should do the trick. You could even perforate it or make it from mesh to save weight. Much more effective than the hoop you mention and since it has length it could also be used to suspend the rocket under the balloons for extra stability

  40. Johnny Canuck

    This is not a title

    Just had a thought about flight stability in thin atmosphere - top mounted rocket engines on a longish fuselage, perhaps with thrust vectored slightly away from the fuselage.

    Sorta like this.


    / | \





    1. Steven Roper

      As was mentioned in an earlier thread...

  41. Hawkmoth

    use a long string

    I like the idea, mentioned earlier, of using a long line to dangle the payload well below the balloon. This will allow a rocket launched at 45 degrees to clear the balloon easily without any additional silliness (or any latex scraps from a burst balloon) and also stabilize the balloon from swinging. As I recall, wasn't there quite a swing developed by PARIS? Also I believe that a swing was a problem for the kid who launched an iPhone powered stratospheric balloon in NY state a couple of years ago. If your payload is separated from the balloon by a long string, you wouldn't have to launch at some preset altitude that was conservatively low (thus costing you ultimate mission height). Just let it go up until the balloon bursts and let the burst itself or the tension coming off the tether trigger the rocket launch.

  42. E-Penguin

    Hanging a weight on the back of the plane won't do anything...

    ... to the initial attitude at release. Heavier things don't fall faster than light things. Have we learned nothing from Galileo?

    Gravity will act at the centre of mass and won't provide any turning force.

    Horizontal launch with actively-controlled rockets to bring it to vertical is the way to go, like

    You could knock up a simple control system using an Arduino nano, a couple of tilt switches and a pair of servos, and to hell with roll rate.

    1. IWishIHadAHandle

      I think that weight on the back might ...

      Gravity acts through the centre of mass/gravity, and if there is a pivot then the system will orientate itself. So that would be the difficulty, making a reliable release mechanism that allows free movement during ascension. Of course getting it to set to 45deg is a bit different, and the whole system would add weight. (If you were going this method though, I think you just build the plane with a large rear engine. Once it is burnt the c.o.m. would naturally shift forward allowing for more glider like properties.) Either way, unless you can release absolutely smoothly, there is a good chance that the orientation would get knocked out on release - so you need to power launch from the angle if no other mechanism is being used.

      I am not sure that the initial launch is actually the complex bit, I think that getting it to go where you want afterwards is. If the aim is height, then a vertical launch is preferable, so at that altitude would a simple small drogue 'chute work for vertical orientation or is there too little air? i.e. drop the rocket off with the chute on the nose, wait a second or two, then fire up, (if the chute has to be too large, then this obviously wouldn't work).

      However the issue is still how to keep it in the right direction while it is accelerating; once you start on internal gyroscopic stabilisation, you are beyond my pay grade. Although it will accelerate nicely in the direction it is facing any already existing attitude changes will continue to occur. c.f. ground launched rockets don't generally move before the engine fires, and if one does, or falls over while it is firing, then much Benny Hill like hilarity ensues, because noone has any idea where it will end up.

      With all of these ideas/engines, how will they work when they have ice on/in them? The advantage of a normal engine is that it is already lit by the time it gets up there - or is the air above Spain dry enough for that to be a minimal problem?

      Too much rambling, too many questions and not enough answers, so I'll leave it there.

      Beer because... well it's beer isn't it?

  43. Alan J. Wylie

    Barrage balloons

    My mother had tales to tell of sitting on top of a hydrogen cylinder in Sheffield inflating a barrage balloon during WWII.

    > We're are not, under any circumstances – except possibly the complete

    > exhaustion of the world's supplies of helium – going to touch hydrogen

    > with a long, flameproof stick. Let that be an end to the matter.


  44. Jim Smith

    Launch attitude suggestion.

    I suggest dropping the aircraft straight down nose first. Light the engine after it has had enough time to become aerodynamically stable, then pull up.

    Problem: how do you light the engine away from the main payload? Reel of wire like a TOW missile? Or, drop the aircraft with the ignition equipment (battery, etc.) and release on ignition?

    1. IWishIHadAHandle

      ... as I think about it ...

      Going on from this simple idea and my previous suggestion of dropping with a drogue, could the craft be built with a relatively large 'open backed cone' shaped front end ? (leave a small whole at the front of the cone, and the craft can exist inside it), this might allow for it to aerodynamically put itself into a vertical stance as soon as it gets up enough speed - a sort of fixed parachute. As long as the engine can overcome this deficit in height/speed then you will win even if you have to give up some height. Using for the approx. terminal velocity, and my l33t estimating skills, I would guess at about 100m/s so reachable in around 10 seconds losing 1500 ft of height. I assume that launch will occur ~80000 ft as last time the balloon popped at 90k and that is not wanted. Aerodynamic control must be established before then - if anyone can calculate exactly when then you can reduce the deficit - but 10 seconds is ample time.

      The next question is, can this be overcome easily by the rocketry?

    2. Jonathan Richards 1 Silver badge

      What we need is a README

      ...with all the fallacies that people have previously pointed out set forth so that we don't think of the same bad old ideas time and time again.

      From what I understand of previous posts, if LOHAN is built to be aerodynamic at low altitude (so that the autopilot can land) then the atmosphere will be too thin at launch altitude for control surfaces to have any effect. Clearly, we don't want to waste that lovely helium-gained altitude by dropping the craft into thicker air. So we have to launch in a regime without aerodynamics.

      This is also the killer for ideas to orient the craft with tail weights, etc. In the absence of significant drag, the whole assembly will drop in the same attitude as that which it was released.

      Comments and corrections welcome...

      1. IWishIHadAHandle

        Comments and corrections

        I whole-heartedly agree with the first part, and the first part of the second part.

        The second part of the second part forgets however that even at that altitude there is still air resistance and it needs to dealt with/utilised - that is why there is a terminal velocity even up there. The problem as I understand it, is that it is difficult to get to a decent proportion of that speed quickly enough and stably enough before the normal control surfaces can be used. Hence a lot of suggestions for using rockets themselves or gyroscopic effects for stability during at least the initial phase.

  45. Anonymous Coward
    Anonymous Coward

    Keep it Simple

    Like this comment.

  46. Anonymous Coward
    Anonymous Coward

    Launch attitude

    You could try attaching a v. light A frame - rt angle Triangle anyway - shape to the payload. Attach with a heavily dampened pivot at the rt angle and have the rocket sat at the bottom. Launch along the long edge, I would assume that it will be clear before any other motion occurred, but you will need fairly impressive acceleration to get to aerodynamic stability by the end of the run (I am assuming 7-10 mtrs would be the maximum feasible). What is the speed for aerodynamics to be effective ? 50m/s ?

  47. Anonymous Coward


    Everyone keeps talking about latex balloons. The weather balloons supplied by Totex and others are actually made from neoprene. This is due to the fact that latex needs to be doused in 'anti static solution' before filling with hydrogen, as the action of filling the balloon can introduce Potential Difference between the balloon and the ground.

    This is also why all good Met Office 'balloon sheds' have a braided copper earthing cable attached to the brass filler.

    Having spent a good deal of my working life filling such balloons, I can assure you that as long as you wear outer clothing made from 'natural' fibres (a dustcoat) and have adequate head protection (navy anti flash hoods) and non-conducting footwear you are pretty safe. This includes the standard tactic of 'poking it with a long stick 'til it bursts' if it slips off the filler whilst being inflated :)

    You end up looking a little like this

    Anon as I'm not supposed to post about work related subjects without getting it cleared first.

    Icon for what you don't want near a hydrogen balloon in an enclosed space!

  48. M Gale

    Keeping balloons (or just one balloon) under control on the ground

    Grab your mallets and hammer a nice long metal stake or three into the floor. Use them to keep your assembly attached to terra firma without any danger of SPB team members going on an unexpected stratospheric joy-ride. There, now you can use a triple-triplet config!

    I also don't think some people here appreciate just how huge a latex balloon can get at altitude. I'm pretty sure a lot of people know about the infamous expanding capabilities of condoms though? Well, this is the same stuff on a vastly larger scale. Wikipedia quotes a 100:1 expansion ratio. Think about a balloon so large that you could stick a couple of Jumbo Jets inside of it, and you're getting to the sort of size a latex weather balloon will expand to before bursting.

    Dunno how Lester thinks that three balloons will only go as high as one. You would only have to inflate each balloon a third as much, and that's not including any other triplets also attached to the launch platform. I know my idea was a triple-triplet set-up where each triplet can lift the entire platform on its own, for a little added redundancy. Inflate the upper set the least.

    Anyway, as for deciding when to launch, there are a number of ways. One is the obvious pressure switch. Use a solid state barometer and have it used to initiate launch sequence when air pressure readings drop below X, indicating you are Y feet above sea level. This could be backed up by some system to detect a sudden altitude drop of say, more than 1000 feet. Just in case a single-balloon set-up bursts early. The good thing about going solid state, is that it has no moving parts to freeze up and screw with the release like with PARIS.

    Another redundant system, which could also provide some other benefits: I know the Reg Guys were thinking of (but eventually never bothered with) tracking via telescope for PARIS. What would be the chances of not only tracking optically with a telescope, but pointing a radio dish in the same direction and keeping a live communications link with the thing? Everyone gets to see the telescopic view, plus with a singular balloon platform you can see yourself if your only lifting device bursts and can hit the big red button for a manual launch from ground control. Maybe creatively re-engineer a couple of TV satellite dishes for the purpose? What's the regulations in Spain with regards pointing a few watts of radio energy into the sky?

    Oh, I know I keep saying this, but CAMERAS EVERYWHERE. Two on the Vulture 2 for forward/rearward footage, preferably with the rearward camera able to see any rocket-pod-jettison action if indeed there be any to see. At least two on the launch platform. One pointing downward toward the Vulture 2, the other upward toward the balloons to capture the burst and parachute deployment. Maybe an extra one pointing sideways towards the expected direction of travel for the little spaceplane.

    I keep making long posts don't I?

  49. Anonymous Coward
    Anonymous Coward

    keeping it simple...

    The balloon will be white/shiny, won't it? So why not use the simple "light/dark" autopilot thing to fire the ignition of the rocket at the exact moment the balloon bursts? Point it straight at the balloon, and then when it goes pop the rocketplane will fly off at the absolute maximum height possible.

    Would it be possible to build a basic RCS using a travel sized deodorant can I wonder? I know the pressure differential would increase with altitude, but it should also drop with the reduction in temperature. It could be used to tip the nose up for the duration of the burn, and then would be a nice bit of dead weight in the front to keep the ship stable on the way down?

  50. Simon_Edwards

    Rail guide

    Isn't everyone overthinking this launch sequence. Military aircraft designers solved this long ago by using a sliding rail to avoid rockets hitting the wings that they are mounted on. Why not use one for LOHAN?

    I like KISS for this part of the design. Flying the thing later will be hard enough.

    To get a vertical attitude from a rising baloon you can hang a weight beneath the main payload container - that'll be Vulture2. But you don't connect it directly to the payload container.

    You have it slot onto a rail that is suspended at a 45 degree up angle and connected to the main payload container.

    Use either low powered RF, or IR or old fasioned contact points to have the main payload communicate with Vulture 2 and get it to trigger Vulture2 to fire and fly it straight off the rail.

    The rail will need to be sufficiently rigid to withstand the initial tug of the rocket firing, so its hanging on a rigid lightweight structure not on a rope. But we're talking a few grams of structure not kilos of the thing hanging there.

    As for the Vulture 2 itself: Some missiles used to have rail mounts designed into the external structure, others had sprung trips to either discard the mount or to have it retract inside the body as the mount cleared the rail end.

    Since it doesn't exist yet either approach could be used with Vulture 2.

    1. Anonymous Coward
      Anonymous Coward

      Agreed. Rail guide as A frame launch above.

      It was mentioned earlier, but I don't know if the rockets available can accelerate the craft quickly enough in the length of rail available. If it can - it looks like the easiest solution, as long as the rail is stable.

  51. GateKeepers

    iphone launched into space

    I've just re-viewed the youtube video of the guys who launched their iPhone up to 30km. I've also just watched the video of the guy who jumped from a balloon at the same height back in 1960. From the two videos I can say that the ascent through the thermal differential layers of the atmosphere is going to require a lot of serious hardening of any designs for this balloonery. If not taken into consideration, in all likelihood, any loose cables will become seriously tangled.

    The other thing you get is that there is NO SIGNIFICANT ATMOSPHERE in which control surfaces could possibly work. The guy who jumped out of the balloon was falling at 1,000 m/s and the air wasn't even ruffling his hair.

    It's probably a good idea to use a larger balloon and to leave it slack at the surface on launch. We learn from scuba diving that as a buoyant device ascends, it also expands and/or experiences increasing internal pressure and we don't want our balloon to explode before topping out. Also, some sort of heating device inside the balloon would provide some extra lift if held until the last.

This topic is closed for new posts.

Biting the hand that feeds IT © 1998–2021