Thread Tools
Jul 27, 2018, 01:37 AM
Go small or go home
ruzam's Avatar
Thread OP
Discussion

SkyCam build


OK, I think I've hijacked enough threads, it's time to start one of my own.

This is my newest blimp "SkyCam". It's not complete yet, but it's getting really close now. This is the result of a steady progression of bigger, more sophisticated blimps. Working off my earlier successes of "Buzzbomb Bob", and the "One eyed, One horned, Flying Purple People Eater", Skycam will be just big enough to carry a gopro hero4 session cam. In fact, the whole design was based around carrying the gopro. It's still very much a micro blimp. I have to be able to load it in and out of the car

It's a fairly basic shape. Round nose, tube body and a cone tail. The shape lends itself to easy building and, more importantly, easy calculation of envelope weight, gas volume and center of lift. Stability will come from a ring (skirt) at the rear. Power will come from power pods in the front and power pods in the rear. The front power pods will contain forward facing (steering) geared props and vertical facing direct drive props (both sourced of micro quad spares). The rear power pods will contain just vertical facing drive props. Six independent motors in total. The vertical pods are channel switch selectable to either provide full lift up/down or full tilt forward/back. It will be my most manoeuvrable blimp yet and for the size it should be able move like a dog chasing a cat.

In the past I've used directed the thrust with servos (two motors, two servos). But I really hate working with micro servos in this scenario. Not only are the mechanical "paddles" delicate, but it couples the forward/steering controls with the elevator/aileron controls. When you're fighting large cooling fans in an arena it's easy to find yourself not quite going where you want when the sticks are all over the place. Using independent motors for horizontal and vertical movement ensures the control sticks always send you in the direction you intended.

I'm sticking with my favourite build method of using internal spars to support the motors and internal wires to run from the motor pods to the control circuits and batteries on the belly of the blimp. That does result in more holes in the envelope, but sealing those haven't proven to be a problem yet. The motor pods will be removable (plug sockets on the envelope). I've design a custom controller capable of mixing and driving 4 brushed motors. This blimp will have two controllers (one for the front pods, one for the rear pods). I've programmed the rear controller to act as a slave to the front controller. Each controller will have it's own battery to put the power source closest to the controller and motors. Wires will link the batteries in parallel. In theory the blimp could run on a single battery with this setup, but it might get sketchy with all 6 motors drawing full power. I'm expecting a really good run time. Maybe 30 minutes or more of hard constant running. Hours of just hanging out with minor motion.

Here's my design specs:

Diameter: 55cm
Tube length: 110cm
Tail cone angle: 60 degrees
Tip of the Nose to tip of the tail length: 185cm

Calculated lift: 343 grams

Envelope material weight: 83g
Stability ring: 35g
Internal pod spars: 11g
Electronics, wires, mounts: 19g
misc mounts: 4g
Front motor pods: 27g
Rear motor pods: 11g
Batteries: 31g

GoPro: 72g
Camera Mount: 25g

Total weight: 319g (give or take some rounding)

If you're paying attention that allows approx 24g for dead weight ballast "after" the blimp is fully loaded. I've never had that much free payload on any of my blimps. I might even be able to slap a logo on the sides!

My weight design is a mix of estimated weight and measured constructed weight. Nearly everything is constructed and weighed now so I'm pretty confident I'm going to come in or at the design weight.
Sign up now
to remove ads between posts
Jul 27, 2018, 02:07 AM
Go small or go home
ruzam's Avatar
Thread OP

Power system


Power and control will come from two controller boards I designed, and 4 motor pods (two front, two back). The front motor pods have two motors each. One larger geared prop for forward/turning, and one smaller direct drive prop for altitude/attitude control.

Each controller board provides four reversible brushed motor drivers and weighs approx 1g (before wires). Mostly it's just an ATTiny24 microprocessor wired to two DRV8833 dual H-Bridge motor drivers. It takes a receiver ppm signal, splits the channels, mixes the throttle/rudder channels into the horizontal drive motors (tank steer) and mixes the elevator/aileron into the vertical drive motors. With the mixing built into the board, the transmitter doesn't have to do anything special. It flies like an airplane. The only real trick to the transmitter is that the throttle stick must be spring centered. On good transmitter the gimbal can usually be changed to spring centered with an adjustment screw. On RTF transmitters it's a little more involved. I want as many blimps in the air as I can find pilots for, so for me it's important to have individual transmitters to go with each blimp.

The controller board also contains a battery voltage monitor. When the battery voltage drops too low, the motor drivers will start pulsing. I can detect the surging even from the other end of an arena. At this level there's still plenty of battery power for a slow fly back to base. Usually it's enough to just ease off the power for a bit to let the battery recover. If the battery drops to critical, the motor drivers are simply disabled.

The controller initializes into a stick calibration mode when powered on. The throttle stick has to be moved full forward, full backward and finally returned to center to set the channel end/mid points. Centering and limits are crucial to getting the most out of the motor drivers.

Each motor pod attaches to the spar by way of a nylon bolt which has hole drilled through it. Combined with an O-ring for a little friction and padding, the motor pod is easily attached/removed and adjusted for angle.
Jul 27, 2018, 02:19 AM
Go small or go home
ruzam's Avatar
Thread OP

GoPro


The point of the blimp is to carry this GoPro Hero Session4. It's not GoPro's best camera, but it is their lightest one at 72g. Still, that's like trying to fly a brick with my blimps. I'll probably replace the GoPro with a lighter better quality FPV camera I have eventually. But for now it's easy to use and when I talk to other photographers about blimp video, they can relate. Plus I can actually monitor the camera view through my phone with it's built in wifi. The range isn't great and it starts dropping out half way across the arena, but it's enough to frame up a shot. And it has it's own self contained battery, so it really is a convenient all in one solution.

I've designed the mount point of the camera to balance the center of gravity at the center of lift. Hopefully this will make the blimp easy to ballast level without much fuss. I haven't quite worked out all the details on how it will be attached yet, but I've allowed lots of extra weight for the mounting bracket so I'm sure it will work out.

I'll run threads internally from the motor pod mount points to the camera mount point to distribute the load around the envelope so it doesn't rip straight through the blimp.

That's about 5 batteries worth of weight right there!
Jul 27, 2018, 11:49 AM
Go small or go home
ruzam's Avatar
Thread OP

Internal spars


More about the internal spars...

Rather than attach motors directly to the envelope, I build an internal spar into the envelope and attach the motor pods to the spar. The disadvantage is that the spar adds weight and creates more penetration points in the envelope (leak potential).

But the advantage is, you've got a solid structure to attach a motor to that doesn't sag or fail as the envelope deflates. That also means you don't have to rely on internal pressure to keep things in place. And for my high speed aerobatic blimps, that also means you can position the thrust along the center of drag for straight and true flight control.

I use a CF hollow tube for the spar. The strength to weight ration of a tube vs solid rod is much better but a big concern is that the CF tube is not a 100% helium barrier. A crack in the tube, or imperfections in the manufacture can ruin your day. To account for this, I rub the portion of the tube that sits inside the envelope with hi-float (I swear by hi-float). But hi-float by itself doesn't bond well with CF. After it dries it's really easy to rub off and break the barrier. So to seal the deal I wrap the tube with a strip of tissue soaked in hi-float. Now you've got a permanent barrier, and unfortunately more weight.

Initially I built blimps with a single CF tube (no balsa). But now I only use the CF tube for the ends of the spar and construct the internal spar from a triangle box made of balsa. For starters, now I can make the spar any size without having to find long length CF tubes in my stash (I have various smaller sections of cut tube). The cheaper CF comes in smaller lengths and is easier to find. For seconders, now I only need to seal the short section of CF that attaches to the balsa spar. Typically, that's only about 3cm . Plus, the balsa box construction is stiffer than a similar length CF tube. One thing I haven't tried yet is just a solid square balsa spar. Without the glue, that may be even lighter, with a small hassle of accurately drilling holes in the end for the CF stubs. As it is, each spar only weighs 5.5g

Oh, one more thing about the spar construction. I used 1/16 balsa because that's what I had on hand. 1/32 would probably work just as well. The spare doesn't have to be super strong. Once it's in the envelope all that's exposed are the stub mounts on the outside and they don't have much leverage. Cutting an edge angle to line up a triangle box is tricky. Instead of trying to angle and line up three edges at once, I cut the sides wider and angle only one edge (60 degrees). The box is made by gluing the angle cut edge to the flat of the next side with a basic wood glue. First two sides, until it dries and finally the third side (locking in the CF tubes at the end). I'll bet I could make some kind of jig to simplify this. After the box is finished and the glue has dried, I cut the excess balsa off each corner.

The spars exit through the envelope and mount to (are glued to) plates that are attached to the envelope material. Initially I made mounting plates of thin 1/64 ply cut into circles. That's still a good way to go. But they are time consuming to make, and align. Since then I've started designing my own circuit boards. It's amazing how cheap you can have circuit boards produced! So now my motor mount plates (or any mounting point for that matter) are made from specially designed circuit boards. The board material is very thin and additional holes are designed in reduce material. The boards have markings that make aligning holes easy, pads for mounting motor plug sockets, and they're double sided and reversible. Between the socket pads and aligned holes, I've really cut down the time it takes to attach the plates and wires.

The plates are attached to the envelope material with "pan glue". I think this glue is my new favourite. Pan glue is essentially hot melt glue. Only instead of coming in a stick that you can put in a glue gun, it comes as pellets meant to be melted in a pot (pan). I understand that florists use it to dip flowers in to attach them to things. I'm sure it has a million uses. Anyway, pan glue is exactly like hot gun glue with one notable exception. It's not affected by moisture. In case you don't know, hot glue can be easily removed from anything by swabbing it in alcohol. The alcohol dries it out and it just falls off. It's a neat trick when you need it. Unfortunately, hot glue's biggest downside is eventually it just lets go. If you use it for blimp construction, after some time, things will just start falling off. But not pan glue. Pan glue is not affected by alcohol, it won't dry up and lose stick. When you use it, it's on for life (or at least as much life as we should expect out of a blimp). It bonds well to the envelope foil. Too well in fact. When you use it on the 'colour' side, you can rip the colour and metalic coating off the foil if you're not careful. The foil is sandwiched between two plates. The holes in the plates are aligned using a couple of sewing pins while it gets ironed in place.

So how do you use pan glue? Simple. Sandwich some glue beads between parchment paper and iron it out until it's a thin flat slick. When it cools you'll have a thin disk of glue. I put the mounting plates on the disk I just made, and with the help of more parchment paper and the iron, melt the plates into the glue. When that cools just cut the plates from the slick and you have tidy mounting plates ready to attach with a hot melt side. When you iron these plates to the envelope material, they are permanent.

When it comes to finally attaching the spars to the plates, I line up the spar with the markings on the plate (so the plates aren't twisted side to side) and lock them in place with a little gorilla glue (the white formula). It foams up to lock things in place so the spar doesn't break free. Then finish sealing up the envelope and enjoy having good solid posts to mount motors to
Jul 28, 2018, 01:58 PM
Go small or go home
ruzam's Avatar
Thread OP

The Envelope


There's a reason why the shape and dimensions were chosen as they were. The cylinder length is based on the foil width of Fluffy's foil, 110cm (after trimming the edges). There's no easier shape to create than a cylinder. One seam (at the top where it will be less noticable). So from there it was just a matter of punching in different values for the cylinder diameter to come up with the size I needed to lift the weight required. For example, if I increase the diameter by just 2cm, that results in over 23g of additional lift (after the increased weight of the additional foil).

The dome nose section is to give it a nice blimp shape. Being a perfect sphere, the formulas to calculate volume and surface area are straight forward. Likewise the gore tool and gore patterns are also straight forward. I've got a spread sheet to automatically calculate the gore pattern dimensions, and I can plug in different values for the number of gores.

Can't get much simpler than a cone for the tail. I've never put together a cone pattern, so well have to see what problems that presents. But since it's just a single straight seam, it couldn't be that hard right?

The three sections are joined with seams. Made easier by fact that the nose gore tool just happens to be the same radius curve as the reset of the seams.

The gore tool is cut from some scrap chip board I had lying around. The thickness of the wood determines the width of the seam overlap. 7mm in this case (1/4"?). There's no advantage to making wider seams and I'd even prefer them a little thinner. After the tool is cut, and power sanded smooth, I glue two layers of thin maple veneer on the seaming edge. By the time that's done, the edge is super smooth, flat and true. My previous gore tool I added one final layer of woven cloth ribbon to create a mesh pattern in the seams. The ribbon gives a pattern of pressure points and channels to allow trapped air to escape while it's being ironed. But the foil colour appears to be heat sensitive and easily melts leaving a silver patterned seam that I'm hoping to avoid here. So this time I left it at the flat wood surface. The colour still melts and smears as the seam is ironed, but the silver underlayer doesn't 'pop' as bad so IMO the seam is more subtle. The only real problem I'm having with the bare wood seaming surfaces that it's too flat. It's tricky to run the iron completely parallel with the gore tool edge. So the outer edges of the seam get too much pressure and the center of the seam doesn't get enough. Next blimp I'm going to try to round off the gore tool edge just a little to even out the process.

Once the gore tool is made and sanded smooth I use a spray contact adhesive to put a little stick on the edge. Don't spray the gore tool directly. You'll end up with adhesive everywhere and the foil will be forever sticking to the sides of the tool when you don't want it. Instead, spray the adhesive in a disposable container, then use a gloved finger to dip and rub it on the gore tool edge. Use sparingly! You want just tiny bit of tack to hold the foil. The heat of the iron will activate the glue even after it's dry. If you've used too much glue you may find it peels off with the foil leaving the inside of your envelope full of sticky surfaces just waiting to bond where you don't want it. Or worse you could find your foil permanently glued to the gore tool. The tack doesn't last for ever. Some times I use a little iron heat to stick down foil prier to heat sealing. Sometimes I freshen up the glue with a rub down of acetone to get some stick back. Depending on the glue you'e used, there are probably different things you can do. The whole point is to use the glue as a third hand to hold foil in place until you can put the iron pressure on it to seal it permanently.

The gore pattern was calculated with a spreadsheet, and carefully marked out on paper. Then the paper was folded in half along the length and cut out. If you were careful, both sides should match exactly what you drew out. If you're not happy that it's perfect, make a new one. A piece of paper is a makes a lousy pattern for working with so I glue it directly to a thin sheet of cardboard (cookie/cereal boxes). Spread a wood glue, use a pressure roller, make sure the pattern if fully bonded to the cardboard. Follow the paper pattern and cut the cardboard when it's dry. Between the glue and cardboard now you've got a pattern that you can trust to not crumple up or move on you when cutting gores. Doesn't hurt to soak the edge of the pattern with CA to make it even harder.

I cut the gores, two at a time, by holding the pattern down with one hand and running a roller cutter around the edge. It's pretty easy. I tape down each layer of foil to the work table first to keep it spread out and flat. You could cut more than two at a time. It just so happened that I was trying use a scrap that didn't lend itself to cutting more than two at a time.

In the past I would have cut the gores directly into the cylinder edge and avoided more seams. But those were smaller blimps and I couldn't do that here. One thing to be aware of. It's impossible to cut all 12 gores, and seam them to the cylinder edge with the accuracy needed to "exactly" end the gore edge with the cylinder seam. So I cut the cylinder sheet with one straight square edge on one end, and oversized the other end (the two edges of the cylinder seam). After attaching all the dome gores, I found the last gore edge didn't line up exactly with the cylinder circumference as I had measured it. But now that I had the 'exact' edge location I could trim the cylinder end to match. It's so much easier to do this cutting when the foil is still in a state where all the parts are flat. Finally I seamed each gore, then finished off the nose tip with a foil backing circle inside. Careful measuring, careful cutting, careful seaming and it all came together really well. This could be my best nose job yet ha ha! At least until the next one.

The tail cone (which I haven't started yet) will be cut as a single circle. I don't know the dimensions of the "pie slice" that will need to be cut from it to turn it into a cone. But I don't care. From my spread sheet I now the length of the side edge of the cone (which is the radius of the circle). I'll start by making one cut from the center to the edge, then seam the cone edge to the cylinder wrap, and finally cut the second side of the pie slice so it aligns exactly with the end of the cylinder roll.
Jul 29, 2018, 04:27 PM
Go small or go home
ruzam's Avatar
Thread OP

Battery pouches


I like my blimps clean. No wires to snag, no stuff hanging off it. It's 99% personal aesthetic, 1% practical. Nothing ruins your day like breaking a wire off because you snagged it on something getting it into the flying arena.

So what do I do with the battery then?

When it comes to batteries, I've settled on a 'pouch'. The basic principle is to use clear packing tape and scrap foil to create a 'pocket' so the battery can slide in and out. The pocket is sized just right so the battery is held in place with a friction fit. The battery is hidden behind the foil cover, and the envelope is protected with packing tape to avoid wear and tear.

My first try at this, I made the pouch long so the battery could be slid further forward or back for balance. But that's a pain. First of all I needed a special cardboard 'hook' to fish the battery out from deep in the pouch, secondly it seems like I got the balance wrong every time I inserted a new battery. It's very noticeable flying and annoying to keep adjusting.

My second try at this I made the pouch just big enough to fit the battery so it's just hidden behind the foil. Same battery location every time. I used similar pouches front and aft to slide ballast coins in, so I was guaranteed the correct balance every battery change. Much better. But packing tape, is not permanent. Sure it sticks like glue when you want to remove it, but when you want it to stay in place it slowly lets loose. As it lets loose the battery pouch opening gets bigger and bigger and simply lets go. I've tried everything I can think of to permanently stick down the packing tape. Unfortunately the paint/metal coating on the foil is not as strong as the inside heat seal layer. Eventually, you find something that will hold the tape down, and now the tape is ripping up point and metal coating, which is even worse. Maybe it just needs a "little" better tape to stick just a little stronger.

This my third try. Same tape/foil pocket, only this time I've reinforced the opening with threads and mount points located inside the envelope. The tape alone doesn't have to take the full force of wrangling a battery in and out. Well see how it works out this time. I also used the battery pouch as a controller cover.

It starts with a strip of foil the same width as the packing tape (which happens to be the ideal width). Tape goes on the silver side (the inside of the pocket) to protect it from the battery sliding in and out. I left the end of the tape hanging out over the edge to fold back over the thread later. Then flip it over and two strips of tape side by side (2 x wide) over the colours side of the foil. This will protect the outside of the foil and give you the stick sides that will be attached to the envelope.

The envelope "under" the pouch gets a strip of packing tape. Again, to protect the envelope from wear and tear. Finally, line the whole pocket up in the battery location over a spacer (I use two paint mixing sticks) and stick it down. Once the pocket is in place, I ran thread through the internal mounts out and tied them in place with the battery spacer. Simple right?

I find it really difficult to get the tape down on the foil without creating air bubbles. When taping the foil flat, I use a rubber roller to roll the tape down on the foil. If the stars are aligned this seems to work reasonably well. Fortunately you can pull the tape and lay it down again if you have to. Attaching the foil pocket to the envelope is even worse. I've lifted and realigned the tape many times getting it into position.
Jul 29, 2018, 04:59 PM
Go small or go home
ruzam's Avatar
Thread OP

Recap - the envelope tail cone


The tail cone proved to be un-eventful. The tail cone was designed and calculated as an equilateral triangle (all three sides the same length). Since the blimp design diameter is 55cm, that makes each cone side 55cm and in turn. When making a cone pattern you start by making a circle with the radius equal to the length of the side, then cut a pie out of it. The size of the pie you cut out determines the diameter of the base. So in this case, the radius of the cone is 55cm and the base I need to fit it to is 55cm in diameter. Now here's where Karma lends a helping hand. A cone of this dimension is exactly a half circle with a diameter of 110cm which is exactly the width of the foil. It's as if I tried to do this, but I swear it was entirely by lucky accident! I'll pay attention to this in future designs.

So that's about it. Cut out a half circle (with a little extra to trim), then seam it to the cylinder end, trim the second pie edge once it's all attached and the exact edge is known.

Closing the "pie" straight seam was un-eventful. I made a rookie mistake right at the very tip. I tried to finish the seam right to the tip with the gore tool inside the envelope. But at the tip, it became impossible to keep the joining strip lined up with the envelope sides. It was a tight location, foil was bunching up and some how I caught one edge in an overlap an botched it. By the time it occurred to me that I could just turn the tip inside out to finish it, the mistake had been made. I did my best to make sure it was sealed and it's a small boo boo, so I'm not going to worry about it. But next time...

I also used the cone as an opportunity to plan my final closing hole. The transition from cylinder to cone will be hidden behind the stabilizing ring. So I left a small section of the cone/cylinder seam open for access to the cone tip. Then I double folded the cone/cylinder seam with a sacrificial parchment paper between and closed the seam. You can see where the overlap caused a little heat marking. I actually caught a bit of wrinkle in the over lap which caused the foil to join and make a divot. But carefully stretching it out with a little heat from the tip of the iron released the divot.
Jul 29, 2018, 05:44 PM
Go small or go home
ruzam's Avatar
Thread OP

Envelope complete


And here we are! A finished envelope! That fateful moment when you fill it with air, inch up the pressure and hold your breath that it won't pop or slowly deflate. How do you repair a blimp envelope once it's sealed? Nervous times.

But I came out OK. I found only one small pin hole. So how do you find the leaks? Use your lips. I learned that from these forums. But there's a trick nobody mentioned. You have to wet your lips, constantly. Maybe it's just me, but I can't "feel" a pin hole leak with my lips. But I can sure feel the cold spot of evaporating spit from a pin hole leak. So that's my technique.

I found the leak rather easily. Checked all the through envelope mount points first and sure enough one of them was leaking (or so I though). Now for fixing leaks, it's back to the pot glue. Once you've located the leak, it's an easy matter of cutting out a small 'patch' of flattened pot glue. Place the patch over the leak, melt it down with the tip of the iron (through parchment paper of course) and Bob's your uncle.

The problem with the leak I found (felt) is I couldn't accurately find it. I assumed the mount pad was leaking through a wire hole under the edge. So I put a pan glue patch on the edge. But that didn't fix it, so I put more pan glue over the mount pad wire pads, but that didn't fix it. Eventually I discovered it wasn't the mount pad at all, but a small puncture near by. I must of nicked it during the construction. You can never be too careful with sharp tools around envelope foil. Once I found the real source of the leak, it was simple to just cover it with a dot of pot glue and iron it down.

The envelope has been sitting inflated for a few hours now. I "suspect" that there may be another smaller pin hole some where. But it's so big, that amount of deflation is trivial. I can live with just topping up the helium during an evening's run. It could just be a temperature differential. My compressor sits in a very hot room, I have no air conditioning and is super hot out today. Temperatures are all over the place right now. I'll give it one more college try to find another leak, otherwise I'll live with it as is.

I must say, I've very happy with how this turned out. There are things that aren't perfect. I still can't manufacture a long straight seam without making waves. But there are no crows feet and I think my technique is improving. I would very much like to smooth out the gore seams some how. I'm tempted to take a hot air gun to the nose to see if I can de-wrinkle it. But after 3 weeks of construction, I'm not about to risk it. I'll have to make a scrap blimp to try that some day. The camera mount point is just slightly indented, which tells me the support threads attached to the spars are doing their job of holding it in place as planned.

The envelope, in it's current state (spars, mounts, wires, controllers) came in a 127.1g. This represents all the things I can't change later (the rest of the blimp project parts can be re-built lighter if they have to). The design estimate for this stage was 117.6g. So I've porked up 9.5g. Regret time. If only I'd used one less drop of glue here, if only I'd cut the wires shorter there, etc, etc. Over use of glue and Hi-Float are probably to blame.

I have a design ballast of 23.8g so now my ballast (dead weight) is reduced to 14.3g. Hypothetically, I will probably have more lift than designed. As long as the stabilizer ring and camera mount don't go over budget I've still got room to spare (knock on wood).

Woot! We have a blimp!
Jul 31, 2018, 12:52 AM
Melbourne, Australia
Wow - fantastic effort Ruzam, and thanks so much for the really detailed build notes; I suspect I'll be coming back to these again and again..!

A couple of questions:
a: what's the deal with this 'stability ring'? Why do you need it, and how do you get such a large structure so light?

b: how do you find the brushed motors? I've been killing myself trying to use brushless ones because they're lighter, but then I need individual controllers and more wires and they're more expensive and I'm wondering if I should go back to the brushies. (They seem noisier as well which I don't like , but that may be more the props I guess... )

c: regarding pouches... I've had a bit of success using zip-lock sandwich bags; with a bit of grease-proof paper (which I think is the same as your 'parchment' paper) in the middle, they seal nicely to the foil and you can make a little 'pouch' into the main body of the airship. (Maybe it's because I'm Australian - the world's first marsupial airship!)

Anyway, brilliant work - I haven't quite got my head around how you join the seams - you use a thin strip of foil 'meltable side up' tacked to your gore tool, and then butt join the outer envelope over the top - it sounds really hard work, but I can't argue with the results! Can I ask why you don't simply heat seal the two envelope pieces together leaving a small double thickness 'flap'? Is it aesthetics, or does your method give a better seal (I can see that it might...). And also how do you get your thin strips of foil - that stuff drives me crazy with the way it curls up!!

- Chris
Jul 31, 2018, 02:38 AM
Go small or go home
ruzam's Avatar
Thread OP
I was beginning to think even the crickets have left the blimp forums! Thanks for the feed back!

a)
The stability ring substitutes for tail fins. My blimps fly fast. Those forward thrusting motors will really haul it around. Easily as fast as a fixed wing aircraft (ok, maybe a slow flyer fixed wing). The ring keeps the blimp flying straight and true. I've tried experimenting with removing the ring (no tail surfaces at all), but then it's all over the place and you're limited to very slow speeds. The idea started with my first blimp as I was trying to capture a flat tail "bullet" look. The ring was as much to hide the rounded blimp shape at the back as anything. But it's been a solid design so I'm sticking with it. Being a ring, it only needs to be attached at the blimp in a few locations to keep it in place (I use interlocking foam stand offs with magnets so it's removable). Over all I think it's a much lighter alternative to trying to support the largish fins that would otherwise be required. In the next couple of days I'll be posting the full build of the ring design for this blimp.

My first tail ring was made from hot wire sliced dollar store foam board. It was a solid band of foam built over a round form. To keep the round shape I doubled up the foam on the front and trailing edges with a 2cm strip. Built on a form and two layers glued, it kept a round shape no problem. But ya, way to heavy for the size of the blimp. So I cut more than 50% of the foam with a repeating circle pattern to lighten it into flying shape. Still does the job, holes and cut outs or not. I didn't like the look of the "holy ring" so I made my second tail ring over the same form but paid more attention to sparing use of glue. Then I cut a radiator like pattern out of the band to remove just under 50% of the foam. The beauty of mounting the ring magnetically is you can just take it off and try a new one. The third tail ring I made was the same as the second (I was getting good at this now). But the blimp required a more interesting tail ring, so I attempted to cover it again in foil with a polka dot pattern painted on. That was a disaster (too heavy). So the most recent tail ring I've constructed I switched to using 1/16 balsa strips laminated into rings for the front and back edge separated by balsa stringers. That ring I covered in a colour matching tissue, with polka dots painted on and it turned out great! (this is starting to sound like the Monty Python castle in the swamp story).

For this blimp, I'm planning on using balsa to constructed ring again. I haven't decided how it will be covered yet. If I can afford the weight, I'd like to cover it in a matching foil and apply the words "Kim & Monica - Twenty Five Years" round and round. This blimp was intended to be an show piece for a friend's anniversary, I just ran out of time.

b)
Mostly I get my motors off Banggood. I started out blimp building with salvaged micro quad parts, then started ordering quad spare parts, now motors when ever they go on sale. You can often find sets of four on sale. I just keep ordering them, so my parts drawers are pretty well stocked now. Here's a few examples: https://www.banggood.com/Eachine-E33...p-1152331.html https://www.banggood.com/Original-Up...p-1105973.html https://www.banggood.com/Chaoli-CL-7...p-1059072.html

Thank the gods for the popularity of micro quads! There's never been a better time to get cheap motors and parts that are perfect for blimp making. The SkyCam will be using this combination https://www.banggood.com/Eachine-E33...p-1145563.html and https://www.banggood.com/Eachine-E33...p-1141691.html for the main forward thrust. It will be a beast!

I would dearly like to go brushless. Geared drive systems are super noisy, multiplied by a blimp envelope acting like a drum skin. At least that's what it sounds like when I'm testing. I guess it's not so bad in reality. When the blimp is flying through a noisy arena with the music blasting, nobody can hear the motors grinding away. And I'm not looking forward to the day when the brushes start giving out. I have a large stock of micro brushless motors in my stash too, waiting for a future project. But I'm stuck with the lack of reversible micro brushless ESC. I've been following the re-programming threads, but I just don't have the expertise to dive into it yet. The ultimate goal would be to design a new version of my controller for brushless, but for now that's beyond my skills.

c)
That's a great idea! I like it! Parchment paper in this part of the woods is otherwise know as baking paper (not wax paper). It's treated with some kind of silicone coating that makes it oven proof. I use it on the cookie tray to keep them from sticking (hmmm... cookies). It's a dry, stiff paper that virtually nothing will stick to. Also handy as a barrier for holding things down when your using CA so you don't leave the tip of your finger attached. Building the pouch "into" the envelope would eliminate more than a few problems and look much neater. The evolution of my battery pouch is getting better, but still too much work for the purpose it serves. I have a future blimp in mind which is basically a "tube" with a strip of controllable leds running nose to tail (maybe a few rows). The blimp would spin on access with the help of the control motors at high speed and then the led strip would be come a video wall. I think it can be done. Key to this design would be getting as much weight as possible into the center of the blimp, including the battery.

I think you've got the right idea on the seams. Using the old surplus foil, every blimp had to be built heat seal side out to hide the party balloon painted patterns. Then it's the opposite, tack the gores on to the tool, then iron the thin joining strip on top. It's actually easier that way, plus its really easy to open the envelope, make repairs and patch it back together (the inside doesn't seal to itself). I could skip the strip and join the two gores directly (if I understand what you're saying), but then each gore would have to be reversed from it's neighbour. Maybe get away with that on silver foil, but not a one side colour foil. It would be much easier, and probably stronger as you'd actually have more seam coverage. I may not fully understand what you're describing.

I have a beautiful 3ft aluminium "sanding bar". It's got a nice flat surface, a strong tubular back and clean straight edges. It cost more than I though it should at the time. I believe it is intended to stick sanding paper down to the flat side then use it to make large flat sanded surfaces. Now I wish I would have paid more and got a longer one. Makes it real easy to flatten down foil, hold it down tight and cut a clean line with the roller cutter. Making seam strips is really easy. I cut a whole bunch of strips before I put the bulk of the foil away so they're ready when I need them. They twist and curl all over the place as you know, but it's no big deal to pull it out flat again on the gore tool. I'm getting better and just lightly tacking it down without stretching it too hard. I don't know how much that matters. Once you put the iron heat on it, I think it permanently stretches to follow the seam as it melts, so it's not like one side of the foil will curl up when you're done. Joining the seam strip is easy too. Just fold the end back and slide the new seam strip under it so you've always got a melt to melt surface.

I spent the night working on the camera mount. It took some doing by butchered the GoPro mount down to a fraction of the starting weight. Let's just say it won't be up to snow boarding abuse any more. There a still a few details to work out, but it's looking good to come in under the design weight. I'll post those as soon as it's complete.

Thanks again for the comments!
Jul 31, 2018, 06:41 AM
Melbourne, Australia
Thanks Ruzam - I'm fascinated by that stability ring, and don't for a moment pretend to understand the aerodynamics - I would have thought those holes would give you lots of air resistance, but maybe not at airship speeds? Anyway, I'm interested because I use rings for the interiors of my 'hollow' airships with their centre mounted motors... but I did the opposite of you, I started with carbon fibre, moved on to balsa, and ended up with polystyrene :-).

I've had a horrible time trying to get balsa rings to stay circular - I was steaming them in my wife's cake tins for a while; not very successfully and there was a certain amount of negative stakeholder feedback. Anyway, I could never get them to stay circular without a lot of cross bracing - whereas heat cut polystyrene seems pretty good, and you can get it pretty thin (I cut it into segments and use polyurethane glue to join the segments.)

Re the seams - most foil seams tend to be done by facing the meltable surfaces of the foil together and joining them, leaving a flap of double sided foil flapping around:

_________| |_________

wheras you seem to be doing something like this:
______
---------- -------------

(which feels like it might be stronger and better at gas retention, but also requires a bit more work - and apologies for the bad ascii art)

Anyway, all the best, and look forward to seeing the results!!


P.S. Sorry for the slow reply; the crickets took a certain amount of time to read through your detailed posts :-)!
Jul 31, 2018, 10:16 AM
Go small or go home
ruzam's Avatar
Thread OP
Haha "negative stakeholder feedback".

I don't understand the aerodynamics of the ring either, but it's a thing. Google "model rocketry ring fin" and you'll find plenty of examples like this https://www.rocketryforum.com/thread...-in-or.137773/. Aerodynamics aside, I look at it as one continuously changing fin angle. Even with the holes cut out, there's still plenty of flat surface to act in the air stream. There may even be some sort of laminar flow situation going on along the surface of the blimp (mine are cylinders at the ring point). The ring doesn't need to be much bigger than the envelope diameter. Another explanation might be that the ring simply acts as drag, like the tail on a kite. My thrust motors are up front pulling the rest of the blimp behind it. It might just need that little extra pull back from the tail to keep it flying straight.

For me, balsa and tissue is only slightly lighter than sliced foam. But when every gram counts, you do what you can The foam rings are soft and squishy, but they do hold their shape and squishy seems to work when it comes to blimp. The balsa/tissue ring is much stiffer, but I'm counting down the days to when it gets accidentally 'crunched'. So far, no sign of warping. I didn't steam the wood, just let it soak in the tub for an hour or two to soften up a bit. It's a large diameter circle, so not much of a bend for 1/6 balsa strip. The secret is to laminate two layers. The glue between the layers and the different length inner/outer layers all work together. I don't think you could make rings like this with just a single piece of steam curved balsa. Still, it's no fun trying to wrap wet wood around a form, and smear glue on it, and lay another strip of wood on top. It's a job that needs 5 hands and I swear a lot making it. I'd like to know more about your polystyrene method. ANYTHING that makes this process either easier or lighter will be extremely helpful. I anticipate I will be making many rings for SkyCam. Different colours, logos, etc.

Side note, I've constructed round boxes from foam board to contain the ring for transport. We call them the hat boxes.

I understand what you're saying about the seams now. The ascii art served it's purpose I don't know why I didn't go that path to making seams. Of course there's the aesthetic of the "sticky out" part, but I've seen many foil balloons where seams done like this are trimmed so close to the seam they're practically invisible. I started my blimp adventure following along with Alan Sherwood's excellent beginner's guide. So from the start I was of the mindset of gluing seams together over a gore tool. I was trying to make it work with emergency blanket foil (which is impossible to glue) and hot glue (which holds well, temporarily). There really was no other way to do it. So I was already using the hot iron to make seams. When I finally started getting my hands on real heat seal-able foil, I just continued the practice.

I've tried some experiments with edge sealing a pattern into a folded foil section of foil (the seam you demonstrated). I successfully melted deep gouges into my work surface and produced an wriggly shaped seam with more leaks than a kitchen strainer. So I guess that technique is not for me. Not without a different work surface and a hot roller seam tool. I do like the idea of a just tracing a line. Pyro makes the technique look so easy.

Speaking of hot gluing emergency blankets together...

Now that I've discovered pan glue, and more practical ways to use it, I think the emergency blanket foil material might have merit again. To hot glue seams, you melt out a long flat slick of glue (the parchment paper method), the slice the slick into thin strips maybe 2-3mm wide. Tack join the strips until you get one long ribbon of glue. Then when it comes time to seam, its one gore edge on the tool, one glue ribbon tacked on top, then the second gore edge lapped on top of that, then iron the foil/glue/foil sandwich to seal it. Emergency blank foil is much lighter than heat seal foil and more than makes up for the weight of the glue. And it's impervious to heat, doesn't heat or shrink, so you can iron the crap out of it. I did manage a couple micro blimps like this, but eventually the hot glue would just let go. First you're re-heating a pin hole leak, then you turn your back and a whole seam just falls apart. It's hard to keep pouring your blood and sweat into blimps that only have a life span of 1 or 2 weeks. But now with pan glue...
Last edited by ruzam; Jul 31, 2018 at 11:12 AM. Reason: typos
Aug 01, 2018, 01:21 AM
Go small or go home
ruzam's Avatar
Thread OP

The GoPro camera mount


It's getting down to the wire. This blimp has to be ready for an event on Friday.

The GoPro hero 4 session is a brick. And I mean a brick. The thing might as well be made of lead. It's GoPro's smallest camera. It's GoPro's lightest camera, but in the end it's still a brick at 72 grams. I'm pretty sure that once I get this blimp tested out, I'll be substituting it with another much lighter recording camera. At least it has a built in battery so it's ready to go at a push of a button without having to fiddle with plugs and setup.

One major problem is the form factor doesn't lend itself well to mounting. There are literally no mount points on this camera. The only sane way to use it is with the 'cage' that comes with it. The cage locks around the camera and provides the mounting point. Without the cage I can't think of any other way to attach to it. One side has the start/stop record button and display, one side has the hatch to access the micro sd card slot and charging. Short of permanently gluing a threaded nut or something on a side, which I'm not sure I could trust either. Who wants to watch a GoPro crash to the ground from altitude, maybe taking out a few spectators on the way? So the cage it is. And the cage is heavy. It's half the weight of the mount construction I estimated. Realistically, this camera weighs closer to 84g.

So the mount is built to accommodate the cage. When I bought the camera (refurbished) it came with a huge lot of extra locking swivel mounts. There just happened to be one that connected directly to the cage and had a threaded nut base. I could work with a threaded nut base. But these things are built for snow boarders. Thick plastic, stainless steel bolts. Grams, grams, grams. I took the dremel to the connector and whittled it down to a fraction of the weight (I wouldn't recommend using it for snow boarding anymore). I found nylon bolts to replace the metal ones. More weight loss. I drill out and chopped the twist knob to nearly half it's size (and weight). Another nylon bolt with a nylon wing nut finished off the "adjustable" part of the mount. Even the wing nut was ground down to half it's weight. I mounted the now lightened camera swivel to a bamboo skewer. I've got some experience with cameras on my micro blimps. Unless you're happy to see the top half of your videos consumed with the under belly of a blimp, you've got to get some separation from the camera. So this GoPro hangs below the blimp a distance on a bamboo skewer. The skewer is permanently attached to a cutoff nylon bolt with glue and a through hole wrapped in thread. This camera mount is not going to let go. Ever.

Once things have transitioned from plastic to bamboo, it's pretty straight forward. More bamboo, more thread, remember the triangles for support. The whole mount easily attaches to the blimp with a hook on a thread and a pin I bend once in place. I won't be transporting it with the mount. It'll probably snag on something causing who knows what kind of damage. The camera can be screwed on to the mount (or removed) after the fact. So it can be transported separately too.

And the best part is the whole camera mount came in at 24.4g! My weight budget was 25g. Small victory, but it's a happy day when your construction efforts are rewarded with an under weight result!

The only real trick will be making sure the blimp is tied down securely because without that brick, it's going to want to reach for the sky. I should also note that the camera is really bouncy. The bamboo mount appears stiff enough, but attached to the envelope the way it is, the camera gets shaking when ever the blimp moves. I don't know how bad this is going to be in action. Once the blimp is free floating, it will be a different dynamic, but this mount may need to be redesigned.

I've started soaking the nylon parts in black dye to knock the visibility down. They're a bit purple now. After a couple more days soaking in the dye we'll see how much blacker they get. The sharpie will no doubt make an appearance to completely blacken the camera mount.

The only thing left to finish now is the tail ring. I've got a bad feeling I may have been optimistic about the final weight of it.
Aug 06, 2018, 02:43 AM
Go small or go home
ruzam's Avatar
Thread OP

The tail ring


Finally, the last piece of the puzzle.

Instead of fins for stability, I use a tail ring. It's a really simple design. Make the lightest possible ring and attach it to the blimp with interlocking foam and magnets. For this blimp the ring will be covered in envelope foil to colour match, and an anniversary message lettered on to it

The ring starts with a foam board form to build it on. The completed ring is very delicate and easily broken with rough handling. Without the form I don't think I could ever attach the foil to it. As it turns out I'm using the original form from my previous blimps and making it larger with a slide on doughnut. Kind of looks like a car wheel now. The ring is constructed with a leading edge and a trailing edge separated by 12 evenly spaced stringers. I'm using 1/32 balsa for all construction. I took the time to buy several sheets and pick out the lightest wood. In case you're wondering the heaviest sheet of the bunch was easily twice as heavy as the lightest sheet. So it does pay to pick your wood. The rings are laminated from three layers. The layers are 5mm, 3mm and 5mm respectively. That leaves a notch in the middle for the stringers. The stringers are 5mm wide (lots of stripping down to 5mm wide). At each 90 degree I used a 1cm wide stringer. The 90 degree stringers will also be used as attaching points for the mounting brackets. I use Gorilla "white" polyurethane glue for this. It sets in less than 20 minutes and foams to grab the wood. It's not very "hard" when it cures, so the resulting ring is not as stiff as I'd like, but it doesn't really matter. You have to use it extremely sparingly. First because glue is extra weight and you don't need that. Second because of the foaming action. You can make a real mess of things if it starts oozing out the edges. Forming the rings is not complicated. A little tape to hold down the first layer, carefully glue and hold the second layer (only 20 minutes!), glue in the stringers, and finally the last layer of the ring. The entire ring frame, approx 60cm in diameter, came out at only 5.8g. I thought it might be a bit weak, so I soaked the laminated rings in CA. That brought the weight up to 10.2g. Glue is heavy. I still wonder if I shouldn't have gone with a hard wood glue instead.

The 90 degree stringers got another 5mm strip on the inside to strengthen the mount and level it with the inner ring. The designated "top" of the ring also got a 5mm strip to make it level with the outer ring. This is so the foil ends have something to attach to.

I've tried a few things with rings to attach a 'surface' to the frame. Tissue makes the best surface. It's light and you can shrink it smooth. Foil makes the worst surface. It's nearly impossible to glue, and hard to attach without wrinkles. This time I tried something new and used pot glue to iron the foil to the balsa. I ironed out a flat strip of pot glue, cut it into thin strips, then ironed the thin strips on to the balsa frame with low heat. After that it's just a matter of attaching the foil to the wood, again with low iron heat. The foil was cut in two strips 10cm wide (the width of the tail ring). They were joined with an ironed seam strip (just like every other seam). The join was positioned at the bottom of the ring, and the ends were joined at the top on the stringer. The letters were applied before the foil was attached.

Now, I could have done a better job. But it was 4 in the morning and I was trying a new technique. I think it turned out OK and I can always make a new ring. I'll use this technique in the future I'm sure. The ring form makes it possible to apply iron pressure. Without it, I'm sure this would have turned into a mess.

The mounting brackets are constructed of black foam tray foam. The idea is to mount a half round nub on the envelope, and attach a matching bracket on the ring. The nub slides into the bracket. I magnet on the nub and the bracket keep them locked in. The magnets are powerful 1mm by 2mm diameter. They weigh practically nothing. All four brackets, magnets and all, barely weigh more than a gram. I can't imagine transporting the blimp without the removable ring.

So that's about it. Final weight of the tail ring? 24.6g. I had designed for 35g so this was a clear win in the weight reduction. Without the foil and lettering this would have easily weight much less.

The envelope was pressurized one last time. Pouches were added for ballast. I use coins for ballast. Toonie, Loonie, quarter, nickle and dime. Each coin gets incrementally lighter than the previous. The coin pouches are nothing more than a strip of packing tape and a strip of foil. Unlike the battery pouches, these don't take much abuse (and are much simpler to stick on). I placed a strip under the nose and one near the tail so the ballast can be balanced out front and back.

So that's it! Blimp complete!
Aug 07, 2018, 01:14 AM
Go small or go home
ruzam's Avatar
Thread OP

The Receiver


OK, I should probably say something about the receiver.

I use postage stamp receivers like this one https://www.banggood.com/2_4G-8CH-0_...p-1102906.html There are only 2 requirements for the receiver. First it must operate on 1S (3.7-4.2v), and second it must have a PPM output. The motor controller board takes PPM input and splits the channels from that.

Three wires are needed to operate the receiver, V+, Gnd and PPM. Additionally, I extend the motor controller led out to the receiver, so the receiver gets a 4 pin micro socket attached. I'm kind of fond of these plugs for everything (receivers, motor connectors, etc) https://www.banggood.com/20Pcs-Mini-...p-1170261.html

The fourth wire connects to a smd LED. The other end of the LED is connected to ground. I position the LED next to the receiver LED so I can see both LEDS. When the motor controller is initialized the controller LED is solid. When both LEDs are lit, you're good to go. I wrap the receiver in coper foil, soldered to ground. I believe the shielding gives it a little bit of an edge with the reception. The receiver, with plug, weighs less than 0.75g. I don't even count it in the weight. Normally I hide the receiver under a cover made from foil and packing tape leaving just the antenna and the LEDs exposed. For SkyCam I just let the receiver hang from the belly.

So far, I've never run into a problem with radio reception, which is pretty good considering I'm flying in metal buildings with a huge metal foil blimp blocking most of the signal. Well, almost no problems. The GoPro contains it's own WiFi hot spot to link with a phone app. I learned very quickly that as long as the GoPro is transmitting WiFi, the receiver range goes down the hole. So once the camera is set up and the phone app view finder looks good, I shut off the app and WiFi. From then on it's manual start/stop recording.

Putting the receiver on a plug makes it easy to switch out. Most of my transmitters happen to be spektrum dsm2/dsmx compatible. But I've also experimenting with Futaba and Frsky transmitters.

About transmitters...

Because of the stick layout, the left throttle stick is spring centered. Center is stop. Up is forward, down is reverse. Without the spring center, you'd have no way to accurately "stop" any forward/back motion. My normal radio is a Devo 7e. If you open the case, you can change the throttle gimbal from no spring to spring center. It's a pretty easy (and reversible) mod. But more blimps means more transmitters and most inexpensive RTF transmitters do not have an option to spring center the throttle. It's built right into the gimbal construction. So I've been modifying extra transmitters for blimp use by replacing the throttle gimbal "cube" from one that doesn't self center to one that does.

It's an exercise in solder iron frustration. First the "cube" has to be removed from both the transmitter board, and a spring loaded donor board. Weirdly I can't seem to find a source of spring loaded gimbal cubes, or should I say "compatible" gimbal cubes. The ones used in transmitters have metal stick nubs (most comparable game controllers have plastic nubs that don't fit the stick). Anyway, once you find a compatible gimbal, it's a pain to get the "cube" pins released from the board. Sometimes they're bent in place (and soldered) and the solder is difficult to melt and remove. And you can't just use the replacement gimbal as is because the pot values are guaranteed to be some different value as well. But I've learned a trick. The pots on these gimbals are almost universally the same size and they snap into the "cube". The "cube" is almost universally the same size as well. So the pots can be snapped off the side of the cube (and don't need to be removed from the board), then only the "cube" needs to be removed and replaced. If you don't know what I'm talking about, just ask and I'll included more pictures. I probably should have just included pictures to begin with, but I haven't documented a transmitter mod yet. Next time.


Quick Reply
Message:

Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
New Product Ultra-lightweight skycam [email protected] Electric Heli Talk 1 Jan 09, 2010 09:24 AM
Question Skycam THOPE Sailplane Talk 3 Oct 09, 2009 07:46 PM