Re: Another option...
(a) The idea is to get the engine burning within a second; (b) the chute will not be effective until 25,000' or so. Non-issue.
327 publicly visible posts • joined 31 Oct 2009
...a mercury switch; when the rig goes into freefall it will break contact. If the closed switch pulls the signal down to ground then a pull-up resistor will raise the signal and then you use FETs or a relay or whatever to drop current into the igniter.
Of course, at altitude it will be cold enough to freeze the mercury, but (a) the insulation will help, (b) it's easy to add extra insulation or a minor heat source to the immediate vicinity of the mercury to keep it liquid.
This system meets all of your needs, it's tiny and lightweight and can be easily tested at ground level.
If you remove the requirement of the balloon's support from your dynamic equilibrium calculation of launch forces, then your launcher will work equally well before the balloon bursts as after. How do you make a stable platform in free-fall? (1) small stabilizing gyros; (2) design the launcher hardware such that the rocket's thrustline passes through the overall centre of gravity. At this location in the structure, of course, must be located an exhaust vent; but arranged around that location we have various masses (batteries, etc.) such that their collective c.g. lies in the thrust line.
Suppose the balloon has not burst. Then we will have a static or slowly ascending platform as the rocket fires. Since the launcher c.g. is in line with the force separating the objects, their inertia will not rotate the launcher (which is a really important consideration even here; a balloon string would do nothing to stop a sudden rotation of the platform). Also, note that in the static, pre-launch configuration the main masses are concentrated near one another; this does make the design prone to spinning due to a small moment of inertia. However, it also makes the platform easier to stabilize with gyros.
Now, suppose the balloon has just burst. Using some simple zero-G detection system such as a mercury switch, the ignition can be triggered rapidly (before much height is lost or too much downward speed built up). This should allow the engine to fire within, say, 100ms (plus igniter heating time and other delays which are TBD) so we can state with confidence we are as high as we're getting by balloon, and also that our vertical speed is still small. We can also state that the support line's gone slack and it's falling toward the launcher, so launch NOW please. :-)
Since we designed the launcher to be free of torque moments due to launch dynamics, and because the required gyro-stabilizers work in freefall, the platform orientation will be steady over the short term transient effects of the launch (although bets are off for, say, a couple of seconds after separation). The stability of the platform does not care about the balloon's effects for transient phenomena.
This also offers us the opportunity to capture a portion of the exhaust in a controlled manner, thereby offering a degree of control over the separation itself.
All of this can be supplied with low-cost, low-weight options. The gyros are the trickiest parts, mass-wise.
How much signal degradation do you expect from a wire that is at absolute most 1 metre long? With a digital system the parasitic capacitance will only reduce your maximum bit rate, which will NOT be an issue here, we’re just getting accelerometer data — not downloading Torrents.
If the interference is so enormous that a continuously shielded cable 18" long is unable to transmit a digital signal, then NOTHING will work. Funny thing, I see balloon flights with bog-standard digital cameras on them working just fine thank you. How’s THAT possible?
Non-starter.
(1) The side-by-side configuration would be unstable. getting the two balloons to rise in lockstep is essentially impossible. (2) The loss of one balloon would set the platform swinging wildly, it would take at least a minute to settle down, and long before that (3) the stress spike on the second balloon will trigger a loss there too.
The four-balloon idea is neat, if they do consider a cluster balloon again. It could be made workable by adding about 10% more helium to the top balloon.
One big problem with the two-balloon approach is that the tension waves along the support lines will destabilize the platform. I think it’s a bad idea in any case to depend on a flexible line for moment resistance — never try to push a rope! Stabilizing gyros on the platform are important in any case.
Who said anything about telemetry being used? That is far from a foregone conclusion, since transmitters take a lot of power which will be exponentially more costly as the temperature drops with altitude. The only really necessary telemetry is a phone to call in its GPS coordinates upon landing, or perhaps a short while before when it’s warmed up a bit and is in range of cell towers.
I think you underestimate the fragility of these balloons. As I understand it you can’t even handle them without gloves, because skin oil would make a weak spot that will cause premature rupture. A string being pulled taut would cut right through the latex as it got fairly high.
Good point about antenna pick-up effects of long wires. I agree that sending analog acceleration signals would be lousy at best. But I don’t see why digital signals couldn’t make it; if RF interference is an issue then you can make the ground line for the outboard electronics a brass tube, and run the signal and power wires inside. That would be one serious Faraday cage for almost the entire signal length.
It is probably a good idea to have shielded servo signal lines, while we’re at it. That would really only mean adding one wire to the above shielding tube (the PWM position signal). Install a wired female plug (GND/PWR/PWM) in the servo bay to make servicing or parts replacement easy.
Actually, the height difference between the wingtips would be an excellent approximation of the actual bank of the craft. There would be a small displacement due to the flexure of the wings, but (a) this disappears in differential measurement and (b) by symmetry the angle of bank at the fuselage is parallel to the angle between the wingtip sensors — certainly within the tolerance of the sensors. Furthermore, by separating the accelerometers by a metre or whatever, they should be able to achieve greater precision than a solid-state gyro in a compact enclosure.
As far as variations in GPS altitude, we should be able to refine the reading by averaging out all the values over the past few seconds. We will know (from the IMU) when we are in a steady gradual drop vs. a sudden gut-wrenching air pocket. Therefore we can use the data stream, integrating it with weighting more recent values more strongly, to get a pretty decent absolute altitude measurement. That can be used to update the IMU periodically.
For the word “inertial” I mean any guidance system based on input from accelerometers and/or gyros (optical or mechanical). It would not be able to fly the plane to a specific destination very well due to rounding error, but it will be useful first-level feedback for control inputs. It will be easy to distinguish various roll angles, for example (comparing g-sensors at the wingtips, differentially, is a great method) and so this is the system that would be used to fly the plane from moment to moment.
It would be a minor conceptual change to take a system that aims for a certain behaviour (say, uniform gliding in a spiral or straight line) and an autopilot capable of seeking some specified goal. “By the time the plane reaches 50,000 feet, it should be somewhere near location X.” Then, again supposing the plane can pick up a GPS reading at that time or earlier, the goal-specifying navigation software can update the current goals for the autopilot to aim for.
This is how I see the resolution of conflicts between systems; inertial measurements will attempt to keep track of the aircraft state, while periodic GPS data can be inserted to correct for rounding error or drift. If we have useable barometric indicators, the correction factor for altitude may be the best — especially if the plane can radio for ground conditions at location X: weather, elevation, stay-the-hell-out-of-here-if-possible warnings, etc.
Hmm, I have my doubts that a back-facing port would work; the opening would not detect laminar flow but turbulence from the tip itself, would it not?
What about multiple ram-air ports; if one or more (but not all of them) freeze up they will simply be sealed off and the remainder will continue to measure ram air pressure. If we had a row of ports along the wing outboard leading edge, with a common chamber connecting them, this would work, would it not? (It would respond more sluggishly than a small ram-air chamber, I admit.)
Lee, one point I wish to stress is that barometric measurement, GPS and accelerometers are NOT mutually exclusive; having a GPS on board does not preclude taking inertial measurement too, and allow that to be periodically refreshed by GPS readings; and, when the ground is near, taking aerodynamic measurements.
One way to avoid or mitigate tube icing problems is to have multiple ram and static ports; say a ram air port on the nose and somewhere on the wing leading edge of each wing. If the ports are built into the plane they would be really quite light.
Thanks for the information from an actual pilot’s mouth! You’re right, a glider pilot is exactly who we need in this conversation. One thing I would like to bring up: I believe the GPS cutoff is for a *combination* of great altitude and high speed; slow-moving objects over 18,000' can still use the signal, I believe. Is this accurate?
As far as the autopilot goes, I suppose we need system specs before we can decide if a pitot is optional. Keep in mind, we can build the static and ram-air ports into the actual fuselage design, so they could have quite different port locations: ram-air on the nose and static on the belly, for example, or multiple static ports. It may not add much weight at all — even if the information is “purely academic” and not used for flight control. A log in the black box could be quite interesting to analyze along with accelerometer and GPS data.
Finally, I can’t help but see how pitot-tube data can’t help the autopilot. Suppose the airplane flies through a front; as the wind direction switches, the inertial sensors would detect a sudden plunge in altitude but with the same horizontal motion (which would be hard to diagnose), but a pitot tube would measure the drastic airflow changes (which would make decisions easier).
That’s a good point about IAS, Lee. Fortunately I think it would be possible to apply a transfer function in the microcontroller to estimate the actual airspeed in a straightforward way, by encoding a hopefully accurate model atmosphere.
So really, with a Pitot tube, we can gain three measurements: static pressure, IAS and corrected actual airspeed. Combined with accelerometer and GPS data, one could build a nice system that uses primarily inertial guidance up high, GPS at moderate altitudes supplemented with barometric altimetry when the air thickens (as indicated by static pressure). The saved data could make for some interesting analysis after the fact as well.
I have been thinking about Lester’s comments about the engine; I think I may know what their engine idea is — if I am right, it will be brilliant.
That’s a really good point — thanks, Stuart.
I agree that the reaction experienced by the truss is the principal driving factor in the relationship of the truss and the plane. This is why I have been discussing variations of piston launchers — by controlling the reaction force of the launcher we can derive stabilization. Consider F=ma; by taking the mass of the launcher and knowing its acceleration, we have a recoil reaction force getting the plane going NOW, in a degree we can control.
So consider a plane held by its wings or fuselage, with quite short but sturdy and lightweight guide trusses. By enclosing its engine(s) in designed devices, we can control that force (we need tests, of course).
Now, by taking the moment of inertia of the launcher and the moment arm of the launch force into consideration, you can design a system to get the plane moving really fast off the line, which admittedly isn’t usually the way you launch a plane.
But here we have the unique experiment of being able to design anything we like. Now, I fully appreciate there are professionals working on this who KNOW WHAT THEY ARE DOING in a way we can only dream of doing it. So I'm just shooting the moon in thinking about this, because, well, why not? (The plane was designed only on the basis of looking cool, not much actual aeronautical thought was given to it. The high-aspect wing is a good idea, I agree; think of the U-2 for example. I am a bit concerned about structural stress in a long thin wing, it could be a limiting factor in how much thrust capture we can use. It’s a design tradeoff.)
I guess my central point is, if we design for dynamic equilibrium, the launch action will happen so quickly that it won’t really matter what the launcher forces are from other sources like the balloon — it would act the same in freefall, although it could go in any direction unless gyro-stablilized now that I think about it.
Let’s think about this. suggestions are welcome.
I figure accelerometers and gyros are pretty much a given for any autopilot. There will certainly be space: a solid state 6-axis accelerometer/gyro combination is a single integrated circuit. It seems to me that the best option, though, would be to put 3-axis accelerometers in the wingtips and perhaps one in the tail, and integrate their inputs to determine overall vehicle motion.
The effect you describe is called weathercocking (which is comedic gold, but I digress). It’s described in this handy NASA bulletin: http://www.grc.nasa.gov/WWW/k-12/rocket/rktcock.html
This will not be an issue for LOHAN. As you mention, balloon airspeed will be negligible. Also, weathercocking depends on extremely large tail fins and does not happen appreciably with airplanes whose major airfoils are the wings mounted amidships.
The thrust from the rocket motor MAY act solely on the plane or it MAY act partially on the launcher — do the words “piston launcher” ring a bell at all? The design of the enclosure for the motor(s) on the launcher, in addition to protecting the launcher structure from exhaust gas exposure, can be used to provide a DESIGNED amount of positive-separation force. It’s not an all-or-nothing choice.
Your arguments about ground launchers are totally specious and entirely dependent on launcher design, anchoring, launch angle, magnitude of the exhaust gas plume, bla bla bla bla bla. But the common factor of ALL ground launchers which you are dancing around is — ready for this? — THEY ARE ON THE GROUND.
The safety factor I was referring to is the MUCH larger aerostabilizing that is required at 80,000 feet than is required at the ground. With 20 times denser atmosphere, a successful ground launch using aerostabilizing is pretty meaningless. However, an UNSTABLE launch with 1 atmosphere ambient pressure tells us we haven’t a hope in hell at altitude. This question can only be settled with wind-tunnel tesing using a Reynolds number appropriate to the tenuous atmosphere.
And my use of the term “slender” is the engineering use — the rod will be really long compared to its diameter, which means the moments are intensely focused at the mount point. You point out correctly that suspending the rod at both ends quadruples the stiffness — while making a sticky problem of letting go and getting the support out of the way just in time. A launch RAIL, being continuously mounted to the launch structure, will be (a) many times stiffer; (b) may be made of thinner material so it’s lighter; (c) presents no tricky problems with clearance at launch.
I agree that’s a possibility — so I refer people once more to my sketch here:
http://autographic.ca/TheRegister/LOHAN_plan.png
What’s not really visible in plan view is the tail structure, which is in fact TWO opposed T-tails: one pointing up, the other pointing down. This will (a)provide the largest possible aerostabilization surface, (b) leave the sides of the fuselage clear for boosters/gyros, (c)provide redundancy if one of the tails snaps off, and (d) provides a nice automatic-landing feature: when it reaches the ground, even if it’s flying nearly horizontally, the tail will strike first and drop the nose to the ground, inducing a large negative AOA and firmly landing the craft with no guidance needed.
Two irrelevancies: (1) a ground-based launcher will transmit moments and forces to the ground in a way that a balloon-suspended launcher would be incapable of doing; (2) a launch at altitude will require a huge margin of safety w.r.t a low-altitude launch because of the 5% air pressure at altitude.
I like the way that NOW — after trolling for days — you act all innocent and explain what you have been keeping to yourself. Classy move.
There are two reasons why a launch rod, which works fine for rockets, will not for LOHAN. (1) A rocket has all of its mass concentrated in a linear structure close to the rod; LOHAN will have wings that will greatly increase the stabilization forces from the wire. Thus, a slender wire will be inadequate even for a lightweight plane. (2) Even with slender rockets, above a certain mass and power threshold launch rods are useless; that is why high-power launchers use RAILS. Try watching some YouTube videos of launches with real engines (as in J power or up). (3) The proposed launch rod mounting was a cantilever onto an aluminum plate; this provides roughly as much directional control as a used Kleenex.
A ground-level launch will *only* be relevant if the launcher is dangling on a thread — the enormous difference that you don’t seem to see is the ability of the ground to absorb reaction forces from the launch. Though you seem to think it’s minor, that is a HUGE difference, if you have a heavy craft (which our plane, unlike a rocket, is) which has to get moving fast — your claims these factors are irrelevant are meaningless handwaving.
How do you plan to compensate for the 20-times-thicker atmosphere in testing? What I’m saying is this: while a totally unstable test at lower altitude means we’re definitely in trouble, a stable test tells us nothing about a launch in the stratosphere.
Right, I thought so. You simply don’t have a clue what you’re on about, or you’re trolling.
You see, there are NO videos on YouTube of rocket-planes being launched from balloons, as we’re trying to design here — because as far as we know THIS HAS NEVER BEEN DONE BEFORE. So every launch video you can watch on YouTube is either (a) from a ground-based launcher and therefore irrelevant to LOHAN or (b) of a lightweight rocket, and therefore irrelevant to LOHAN.
Try paying a tiny bit of attention to what we are actually trying to do here, Lee.
Guess he doesn’t actually have any examples of rockets taking off without a launcher guiding them — I sure haven’t seen any on YouTube — and I DO have an example of a rocket taking off without initial guidance.
It was a (dis)proof of concept, with three engines at the nose of a finless rocket, intended to verify the Pendulum Fallacy (http://en.wikipedia.org/wiki/Pendulum_rocket_fallacy) after that was pointed out to me and others at the start of the LOHAN design. If the fallacy was indeed fallacious, the rocket — launched utterly without guidance other than tractor-configuration mini-rockets — would not fly straight.
Guess what happened? It went upward maybe a maximum of two inches, and flew essentially sideways in a totally uncontrolled tumble.
Conclusion: active stabilization IS MANDATORY.
NEXT!
How the fuck am I supposed to say “yes, it matches” or “no, it doesn’t” IF YOU DO NOT SUPPLY A URL, TROLL? I am not the one stalling.
Of course the physics are the same, but if you are too cowardly to post an example of what you refer to, then please go away and allow a useful discussion to occur.
If you don’t want it to burn like the Hindenburg, just don’t make the balloon out of thermite. The flames in the footage are brilliantly bright. Hydrogen burns with a dull red flame that does not appear on black-and-white film.
I also question the use of helium in welding, it would be useless for creating an inert atmosphere without complete enclosure because it’s less dense than air. Since argon isn’t, that’s presumably the 'IG' in MIG welding. Nor is argon in any way threatening to disappear as it’s a major component of the atmosphere.
Power consumption is the bugaboo. The temperature, about –60°C, will suck the life out of any batteries as the power requirements increase due to increasing range. To make this work you would need a helluva lot of batteries, even if some of them were used purely to heat the others.
This shouldn’t really be much of an issue. There are fundamentally two options for the motor when mounted in the launcher; one, an open-air firing, the other, an enclosed firing container that works to some extent as a piston launcher to ensure positive separation and dynamic stability through recoil action.
If the engine is not enclosed, then the firing of the motor itself will have little effect. The thrust reaction works directly on the aircraft’s fuselage. Essentially the only change as far as the balloon is concerned is a sudden drop in payload mass — which will cause it to rise, but not nearly as fast as LOHAN. The payload mass will swing, but that shouldn’t horizontally displace the balloon much — and at launch it will pull the balloon AWAY from the launch trajectory.
Using my supernatural powers of estimation and the geometry sketched by Lester, I figure the plane will be clear of the balloon between 1 and 1.5 seconds after launch, with a speed somewhere on the order of 100 km/h at that time, assuming no piston launcher.
If there is a piston launcher, then the effect of the engine firing will be the balloon rupturing if it hasn’t already (the launch pad will jerk hard on the tether and will surely push the tired balloon over the edge) so there really isn’t much to think about. The balloon will consist of latex ribbons at that point, and not be a substantial obstacle.
Hi Magnus. You should stop by the forum again — Lester read the riot act to the trolls and now they seem to be (a) behaving and (b) reading each other’s posts. Shocking, no? But don’t worry about the naysayers. I suspect these are people who never studied the principle of dynamic equilibrium.
About the freefall launch platform: it certainly IS possible to get a stable launch from a freefalling platform, and I think it is mandatory that we build that capability into the LOHAN launcher so our flight is not completely stuffed by a premature balloon rupture.
How would it be possible to stabilize a freefalling platform? Gyros of course. That would maintain the overall orientation of the platform. Next, you would have to make the launch happen REALLY FAST, which means a piston-launcher of some sort (probably detuned, as composite motors deliver lots of thrust promptly) which will apply a large impulse force to the launcher and the aircraft simultaneously to kick them apart.
So, with (a) small gyros to maintain orientation, (b) a design that places the launch forces through the CG of the aircraft AND the launcher, and (c) a detuned piston launcher, you can get reliable performance when it’s hanging by a thread OR in freefall.
It’s even easier than that: an accelerometer will tell you, since very shortly after the burst the gravitational forse sensed will have gone from ~1g to ~0. The problem with the idea is that you have to design the launch pad to provide directional control when it is itself in free-fall.
In order to work at all, the stabilization moment from the avionics package would have to be transmitted, around a huge balloon, to the launch pad. Otherwise the system will be statically unstable. Creating a rigid frame would dwarf the mass of the plane and the launcher, probably stop the balloon from getting aloft at all.