LOHAN seeks failsafe for explosive climax As true patriots last weekend rolled out the bunting and unfurled a celebratory pint in honour of her Maj Queen Liz II's 60 years atop the throne of Merry Old Blighty™, we here at El Reg's Special Projects Bureau took a few moments from shouting "Gawd bless yer ma'am" to ponder matters of perhaps greater import, viz: just how to …
Why not have the failsafe line pull away a small piece of electrically isolating stuff that sits between the (normally closed) contacts? Or, as has been suggested several times before, use the pull on the suspension rope to keep one contact of a (normally-closed) switch away from the other.
...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.
I had the same thought since the spring pressure acting on it might make the frictional forces exceed the pull developed by the chute at altitude. I was thinking something that utilized a mechanical advantage so the pin doesn't need to hold the whole spring load like an old school railway knuckle coupler.
Two thoughts:
1) do you have to use the same parachute to pull the cord (I'm concerned that the cable might be just enough to prevent your parachute opening properly)
2) can you use the tension in the cord rather than something on the balloon itself ( two contacts head apart by the tension of the balloon with a weak spring that will pull them together once there is no tension on the string? )
You need a bit of cheese and a mouse. Sit the mouse on a balance with the piece of cheese on a string in front of it. As the mouse eats the cheese he will get heavier and the other end of the balance will rise. This can be used as a release mechanism. just make sure the mouse isn't too hungry otherwise the rocket will suffer from premature(chedar) ejaculation.
To avoid the fail-safe line tangling , couldn't you fit a hollow tube through the swivel fitting and feed the line through that ?
By the way , isn't there a break the balloon where helium is to be fed through ? I'm imagining it's tied into a knot like a childs balloon. Couldnt you tie it around some simple pressure drop detector instead? Unlike like the air during lift the balloons internal pressure will suddenly change upon bursting.
I wouldn't expect the parachute to surely open at that altitude (there's not much drag up there) while it's kept together by failsafe line/pin. I'd think it's rather safe that it will fail.
I believe (but what do I know) it would be much more failsafe to rely on the lack of pull from the suspension line - I know, you don't like the idea, Lester. Think about an auxiliary line, parallel to the main suspension line, which features a kind of spring at the bottom and some electric contact which is pulled down by the spring - maybe all within a tubular thing. The construction has to be long enough and the spring weak enough that it's not accidentally triggered in the jet stream storm.
Entering the road to balloon burst detection, it makes me wonder why not completely rely on that event. Of course, I also wouldn't rely on one ignition mechanism alone. An additional (or the main) electronic solution could be with contact-less sensors. I'd give it a try with optical (pulsed infrared) or ultrasonic proximity switches (both but especially the ultrasonic one should be tested with an inflated balloon) and using it to start the rocket motor when the sensors don't see the balloon anymore. One sensor alone will not do because during stormy weather the sensor may temporarily lose sight of the balloon. Those proximity switches usually wouldn't mind if covered by balloon debris because they cannot "see" objects that are too close (but make sure beforehand).
I agree, I think either the parachute will not open enough/swiftly enough, or you will have to have the line taut to begin with, so that the distance traveled isn't too large.
If you had to have a manual (not electric) system, you might be able to rely on the weight of the cage to provide friction that stops a weighted pin from falling, with the bursting of the balloon, and entry into freefall, the friction would be released allowing the pin to drop and the mechanism to fire ... but that is open to the pin sticking problem and in this case you cannot grease the pin.
OK, my idea as follows..
take 1 thin rubber sheet, 2 pieces of aluminum foil and 4 small plates of plastic/wood/insulating material of choice.
the stack of of the device is as follows;
1st plate with small hole in the center. Around the hole on bottom is aluminum foil and forms one part of the switch.
2nd plate is a insulating spacer and has a large hole in the center.
rubber membrane comes next with another piece of foil in the center, this is the second part of the switch
3rd plate comes next and has another hole same size as the 2nd plate. The rubber needs to form a seal against this hole
4th plate (with no hole), closes the air gap at sea level pressure. Again needs to seal.
The idea is that as the pressure drops outside the switch, the trapped gas inside the sealed compartment expands the rubber, eventually causing the two bits of foil to meet and close the contact. This sets off the BOOM time! Trigger pressure can be set by playing with the thickness of the rubber and the 2nd plate and testing in your nice vacuum chamber.
Simple, one one moving part, should be pretty reliable :-)
I don't like this solution there is not enough atmosphere for the chute to open fully, and you run the risk of fouling the chute, not so clever.
I like the load sensor technique the best but it would have to be wind-buffet-proof. There's a trick to making this work mechanically by using longer levers in a mechanism to absorb the buffeting effects.
I would suggest if you were to go this route that you find a helpful sky-diving expert as the mechanism you speak of is known as an RSL (Reserve Static Line) to said adrenaline junkies. To-wit, an RSL is a connection between the ripcord of the reserve chute and one of the straps of the main chute such that when you cutaway the main (uh oh!!) it self-deploys the reserve by triggering it as it is pulled away in the airflow (probably a lot more drag/'pull' than you'll get?).
Indeed, you may want to speak to them about the inner workings of an AAD (Automatic Activation Device) - the fail-fail-safe barometric (+ vertical velocity + clever computer gubbins) trigger that fires a squib to cut the retaining loop of the reserve chute in the event that you are falling too fast when too close to the ground (otherwise known as in "deep doodie").
While such a default setup is unsuitable for high altitiude detection/deployment, the basic bits - especially the AAD squib - is designed for utmost reliability, so you can be sure as possible that it will work when triggered. Hence a hybrid - an RSL setup triggering an AAD squib may be a good combo.
Why not have the pin held in place against a stop by the upward lift of the balloon? Run a fine cable from the base of the balloon to the firing pin in a loose hole. The pin is prevented from being pulled out the top by a stop on the underside that is larger than the hole in which it sits, and prevented from falling out the bottom by tension on the line to balloon. When the balloon goes pop, the tension on the restraining cable is released and the pin is pulled back to earth by gravity.
It may be necessary to springload the pin to ensure that it is ejected when the tension is released, but the principle is the same... Modern monofilament fishing line should be enough to take the load.
I like the simplicity, but the 'failsafe releaseline' is bound to become tangled and wrapped around the parachute cords during ascent which will stop the parachute from opening.... as mentioned by a previous commentator - you'll need a rigid guide tube of some sort to stop the entanglement.
At least scale-model tests should be easy to arrange!
One, if you're going to use the pull-pin release as pictured, the pin will only respond to pulling in a narrow range of angles (i.e., pulling sideways on the pin does nothing). This may be good, may be bad. If it's bad, an alternate release would be to machine a groove 'round the circumference of the actuating rod, and use a flat, forked piece of metal with a slot that fits into the groove in the rod. The forked bit does the work of the safety pin, holding the rod back against some solid surface through which it passes. Attach the pull cord to other end of forked bit, and pulling in any direction more or less perpendicular to the axis of the rod will release it. In effect, this gains you a wider acceptance angle in one direction.
Another thought: To reliably detect balloon-pop, put a smaller, mostly deflated balloon (call it, say, "Mini-me") inside the main balloon through the neck, with said smaller balloon being connected to a tube running out through the neck of the main balloon. The interior pressure of the main balloon will remain above ambient pressure until it bursts, and up to that point the pressure in the mini-me balloon will do so as well. At main-balloon-pop, mini-me will be exposed to low ambient air pressure, and will inflate (and possibly pop as well, which is OK). The interior pressure of mini-me, and/or its abrupt drop, should be usable for triggering purposes.
Dunno if these are goo ideas, mind, just different ones.
Just had a bit of a eureka moment... or possible a silly idea..
The balloon is the risk here because once it pops it may cover everything, but the reason it pops is because it'll reach max pressure from the altitude.. so....
Use two balloons, one as the main load bearer (ie filled to the correct stated capacity to provide lift for the entire rig) and have another balloon of the same standard but only fill with say 50% of the gas. Tie the rig a significant distance below both balloons but tie the small balloon above the larger of the two - with a safe distance between the two.
Then when the main balloon bursts first the entire LOHAN structure will start to descend but... and heres the key thing.. the burst main balloon stays well clear of the main rig as itll be kept above the rig by the smaller balloon (make sure the secondary balloon can take the weight of the burst primary).
Then we can use one of the many burst balloon scenario detections.
Cheers
Will
Not good; here's why:
- the line to the upper balloon will chafe the lower balloon causing it to fail prematurely. Very.
- one image I've seen shows such a balloon disintegrating into dozens of separate strips. It would only be the neck left hanging from the line to the upper balloon, and not much more.
Actually, I think this idea is likely to fail.
The problem with chutes is that they don't provide any drag until they are opened. And they don't open until they've opened a bit. Until this happens the chute just flaps about a bit like a piece of cloth on a string. With your rig, even the slightest resistance is going to keep the chute closed, meaning it can't open enough to catch air and pull the cord further. And then the whole rig just plummets back to earth under a streamer instead of a chute. And the scenario you thought up yourself, the line twisting around the risers, is also very likely to happen. I think you'd better give that plan a miss.
Now, how to solve it, I currently have no idea as I've been pondering some other mechanical problems all day.
So *voice of the Brain* I shall ponder this conundrum some more later, pinky *end voice of the brain*
LOHAN could use a zero pressure balloon, that does not burst, potentially has much larger lifting capacities, and can go much higher than latex, pressurized balloons. They can also be made in vrious shapes (e.g., a cube).
A drawback is the expense of the balloon. Another drawback is that they are a bit more fragile to handle during launch.
However, with the shed-centric approach taken here, it may be feasible to manufacture your own zpb. The main skills needed are some elementary geometry, to design the shape of the skin panels, and the ability to fuse the panels (plastic sheets) to form the balloon skin....
For the launch itself, laying out the balloon on a large enough piece of tarp so that the balloon is kept neat & untangled & undamaged does the job.
The biggest drawback: the balloon needs a self-destruct mechanism, to prevent it from traveling all over the world. Perhaps LOHAN can be perched on the top of a cubical zpb, such that ignition burns a hole thru the balloon skin?
* A purely mechanical system, with no electronics
Given the current state of electronics I'm not convinced this is a good requirement. It presupposes that electronics are in some way inherently less reliable than mechanical ones. I am far form convinced this is a reasonable assumption, and it may be unduly swaying the rest of the design choices. That the backup system be independent of the main system is good, but there is no reason it should be purely mechanical.
* Triggered at, or after, balloon burst
This seems to be a revision of the requirements, curiously you are on the way to making balloon burst the preferred primary launch time.
* Guaranteed to work, no matter what
Good requirement :-)
* Easy to test
The proposed design fails totally here. Unless you can test parachute deployment at the equivalent altitude you are not going to be able get a useful test.
My suggestion:
Get a three axis accelerometer, and a micro-controller. All you have to do is run a suitably damped sum of the accelerations. If they all sum to zero for more than a few seconds the thing is in freefall. You can test this easily. You can wave it about, drive around in a car, hang it from a flagpole in the wind, simulate just about every sort of buffeting and pathological scenario you can, and then chuck it off a roof to see if detects that it is falling. Power the device with its own battery. It will be vastly more reliable than any Heath Robinson contraption of springs and such that will most likely freeze solid or simply plummet.
Agreed, why not just make this the primary method of ignition, with the altimeter purely being used to prevent premature launch? You could include a mechanical solution as a "backup", but microchips come in a neat little package with a minimum operating temperature, which you won't get with a mechanical solution.
My preferred solution would be to detect the loss of lift from the balloon.
If you place a piston in a cylinder, that is then held to one end by the lift of the balloon. A spring works in opposition, so when the balloon bursts the piston is pushed to the other end and then triggers the launch. You damp the system (so as you don't get false detection during accent from turbulence) by sealing the cylinder and making some little holes in the piston for the trapped air to flow slowly through (although just a sloppy piston fit might work just as well).
how about a magnet at the midpoint of a stiff (carbon fibre etc)rod fitted inside the balloon. (top to bottom)
with a reed switch on the outside attached to the lower end of another length of rod..
with the top end attached to the top of the internal rod (or the plug/bung or whatever is there
positioned so that upon bursting the 2 rods will swing together allowing magnet and reed switch making the contact.
im sure you could improve on the basic idea with some thought.
Mark.
If you expect your electronics to fail, you're already toast. Besides which, the most likely failure would be the battery (if it freezes). Wire up the electronics carefully, check it works in low temp, and package it inside a styrofoam box (these are really quite good; you can even mail ice-cream and it will arrive frozen). Otherwise, your mechanical trigger just adds weight, and the significant risk of a false positive.
BTW, the easiest way to detect balloon burst is actually to measure the G force on the truss. Have a G-sensor, which is armed by the altimeter. Once a decent height is reached, you'll be going at approximately constant velocity, and the acceleration measured by the sensor will be 1.0 G (within 10%). At burst, it will suddenly become 0G as you go into free-fall. Use this to drive a MOSFET and the ignitor.
Consider a wishbone made from two pieces of material with good elastic properties, e.g. thin steel, mounted at ~90 degrees to each other. Attach a large-ish ring to the vertex with some nice lightweight string approximately the same length as the "arms". Then attach the payload to the vertex, and the balloon to the ends of the arms using more string.
As long as the balloon is suspending the payload, the tension in the "arm strings" will curve the wishbone arms towards each other, which effect will be magnified by the captive ring. But as soon as the balloon ceases to provide lift (i.e. bursts), the elasticity of the arms will cause them to spring back to (nearly) their original 90 degree orientation.
This motion can be used to trigger both an electrical and mechanical firing system. Because the arms will have to move from their "under load" tips together to something close to their relaxed 90 degree orientation before the system will fire, the device will smooth out a lot buffeting. The length of the arms and their elasticity can be tuned to satisfy whatever metrics for detecting the burst event, e.g. force on the lift line must be less than 30% nominal for 250msecs, or whatever.
So, your main release mechanism depends on absolute pressure. For the backup, use change in pressure; and variometers do this. In normal use these instruments measure rate of ascent/descent. You don't need that to be accurate, and it probably won't be at high altitude, you only need to measure rate of change. With proper filtering and smoothing to prevent premature release from turbulence, a change from ascent to descent should be straightforward to detect and used to trigger release.
The problem is that a variometer measures rate of the volume of air escaping. At altitude the amount of air leaving the thermos bottle is going to be absolutely tiny because pressure differences over altitude are not very high at all (Much less than in the lower atmosphere)
than any heath-robinson contraption you could reasonably build.
Two very simple and extremely reliable options:
1) A clock. After a given time, fire the rocket. Calculate the likely maximum flight time to altitude and add a bit, such that it will fire after balloon burst.
2a Strain gauge joining truss to balloon not via the parachute. When strain drops to near-zero for longer than 5 sec or so, fire the rocket.
2b) If you want to wait for the parachute to have opened and thus stabilised the truss, use two strain gauges - one on the parachute cord, one on the balloon cord.
When parachute cord has higher strain than balloon, fire.
2 a/b can be done entirely with discrete solid-state electronics (wheatstone bridge, op-amp, monostable, mosfet) no software or moving parts. (Avoid electrolytic caps)
Use two identical strain gauges in both cases to compensate for temp - both should respond the same.
Can easily test this in the freezer as pressure has no effect, only temperature.
Correct me if I'm wrong but afaik:
- Instrumentation Op-Amps are finicky little things who refuse to work at the slightest hint of mistreatment (ie, low temperature, buffeting, etc)
- Straingauges and their amplification electronics suck power like mad
Sorry, you're wrong :-)
OK, partially right, but depends on the circumstance and device you choose
if your source is relatively low impedance (e.g. wheatstone bridge device like a strain gauge), then they're pretty stable and will work down to -40 C (industrial grade devices). They will typically consume less than 50mW (an AA battery has around 3Wh of power, so that would last 60 hours.. plenty of time to launch)
older devices are less stable and use more juice, but modern devices are pretty good these days
Was using a pre-setup system. But with some very old instrumentation op-amps looking at a a TINY signal. (Barely registerable) So we had its sensitivity cranked up all the way. The system didn't like that though :-P Maybe it was just a bad design, but those measuring instruments were rather finicky in the most sensitive range.
try measuring the signal from a fuel cell (1mV full scale deflection), when you've got TETRA radio signals bouncing around (80V/m.. normal EMC test is 3V/m or 10V/m at the most) :-)
modern devices, properly laid out and implemented have incredible common mode noise rejection
Still.. ho hum, that's life
Sorry :-p well guess I should know this stuff seeing though I'm an Electronics Engineer... only mechanics I know is from A-level maths (I did Pure and Mechanics). I'm sure there's stuff you could teach me about fluidics, thermal flow, stress etc etc etc
Just goes to show, doesn't matter how much you know, there's always someone that knows more :-p
Having given it a bit of thought I still haven't really come up with a good "balloon burst detection" method.
What I did think up is an alternative and much smaller mechanical release (You have to admit the PARIS design was rather bulky). See http://i231.photobucket.com/albums/ee238/Kloppertjes/LOHANrelease.png (Is there a way to directly post images?)
Basic idea, a sealed chamber with a small diaphragm. Affixed on this diaphragm is a pushrod that in normal conditions keeps a microswitch pressed. Now suck the bejeezus out of the chamber and bring it down to the release altitude pressure. (With a good vacuum gauge this should be easy to calibrate) Once the device reaches altitude the diaphragm pops into the relaxed position and pushes the switch.
Alternative, fix an internal string to the back of the diaphragm pulling it back, and seal the chamber at sealevel pressure. Once the internal pressure becomes sufficient the diaphragm again moves forward and pushes the switch.
Bad plan, because it will be impossible to calibrate remotely accurately. By the time you get to 1/3 of our expected altitude, 3/4 of the pressure drop will already have occurred. Much better to use a simple timer.
OR, FFS, A ZERO G SENSOR, ARE YOU LISTENING LESTER?
Already been done in the Patriot Missile System.
Encase the Rocket in an insulated box with a bit of acceleration space with a window that will break by the passing of the rocket headed to the balloon.
The rocket will burst the balloon on contact and continue space ward. This Mechanism also allows for recovery of the rocket in case of failure for another attempt.
Two separate ignition systems.
Mechanical timer with power and timing device attached to the rocket box. This system operates on a best estimate of optimal launch time.
Radio command and control launch system contained in balloon basket/launch subunit with a power source separate from the mechanical timer unit. Fired by radio from the ground.
Both should be insulated and painted flat black to absorb as much solar energy as possible.
It's easy to see now why NASA didn't bother with helium balloons for the Apollo mission but just went with big MF Saturn rockets. Too much hassle otherwise. Maybe a chap with binoculars and a remote controlled detonator a la something out of a spy movie would be a more workable plan.
(Icon == this way up)
Attach the launch truss to the balloon via a simple spring scale.
Chose spring scale with a range of 1/2 to 1/4 (guesstimate) of the total weight of the truss (incl Vulture 2) so that once the balloon's inflated and the truss is suspended the spring scale will be fully extended and at the limit of its travel, so as to avoid any bouncing off the upper stop during the ascent.
Botch a new lower limit stop, at about 1/8 (another guesstimate) of the full range, so that the spring is still under some tension when there's no load upon it and it tries to fully retract. Then epoxy an insulated contact onto the scale, beside the new lower stop, and another to the moving spring scale indicator: when the balloon bursts, the load drops to zero, the spring scale retracts and makes a firing circuit when the scale indicator hits the lower stop.
So I just had a brainwave. Why not have a simple electret microphone, and analyse the signal for noises above a threshold? Because the balloon mostly moves with the wind there shouldn't be much wind-noise on the way up. Once the rig starts dropping, there should be a noticeable amount of rushing air going on. Simply detect this and trigger LOHAN. The bang of the balloon popping MIGHT be noticeable, but I can't remember clearly hearing that in any high altitude footage(And a simple mic can run for days off a single 9v block.)
And beer, because its friday.
Can that parachute open forcefully enough to pull the chord in the thin air at ~100,000ft?
Could I suggest an alternative which detects lift from the baloon: http://iamtheb.org/me/weightdetector.GIF
Aerodynamics have no part to play. No unreliable pin to pull. Spring still ensures electrical contact.
When the weight of the payload is zero, it is in freefall. Keep the tube long to account for turbulence.
Those of us who have flown video cameras on high-altitude latex balloons are quite familiar with a phenomenon that we've tagged "Post Bust Chaos"; in fact, it's so familiar to us that "PBC" is a broady-recognized acronym. In a nutshell, the behavior of the payload string becomes unpredictable within one second of the burst. This is due in part to the sudden loss of lift and partly due to the shock wave produced by the sudden expansion of now-unrestrained lift gas.
Just google for "balloon burst video" if you're not convinced. YouTube has almost as many of those as cute kitten vids.
So my recommendation, if you're taking the latex route, is to launch the rocket while the balloon is still intact, e.g., by using a GPS-based altitude measurement (far more precise than any reasonable absolute barometric pressure measurement at 1% of sea level pressure), and setting the firing altitude setpoint about 3000 m below the predicted burst altitude.
Alternately, you could use a pricey and fragile zero pressure floater or an even more pricey super-pressure floater. Or, you could pop open a vent in the neck of the latex to slow its ascent. But all of these alternatives violate the KISS principle. One could also try a minimal fill in the hopes that the rising tension in the latex envelope will halt fill gas expansion before the latex reaches its yield point; good luck computing THAT!
This post has been deleted by its author
The chute may become frozen in a closed position due to the cold, and the tether may become frozen to the rig, meaning that the whole thing would be in free fall for some time after the balloon bursts.
How about suspending the rig from the ballon by a spring which, when the balloon bursts contracts causing two electrical contacts to be brought together igniting the rocket motor? There would still be the issues of stopping premature ignition due to turbulence on the ascent and preventing the contacts and spring icing up of course, but these apply to all solutions.
Surely the simplest solution would be to have two electrical contacts glued to the balloon such that they close when the balloon is deflated but open as the rubber skin stretches apart during inflation. when the balloon pops the rubber will contract again, closing the contacts and firing the ignition circuit. Simple, light and rugged. You may want to cover the contacts to prevent ice forming over them but otherwise failsafe and fast.
The advantage of this system is that it is not affected by turbulence or altitude variations as it triggers only and exactly when the balloon bursts.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
