SMALL - espritmodel.com SMALL - Telemetry SMALL - Radio
Reply
Thread Tools
Old Aug 20, 2014, 10:02 PM
Registered User
Joined Feb 2009
400 Posts
Discussion
Metalized Film Envelopes

In this thread I want to examine the pros and cons of metalized film envelopes vs. traditional PU film used in the majority of commercial RC blimps. An 8 ft long test envelope made from eight gores is used for the evaluation. Two types of film are being evaluated: metalized nylon and metalized PET.

Metalized film Pros
-------------------------
Superior helium retention over PU
Very light weight compared to PU
Easy to thermally seal the seams with wider range of working temperature and little risk of damaging the film from over temping.
Film does not stick to the devise used to heat the seams.
Cheaper than PU

Metalized film Cons
--------------------------
Lower puncture resistance than PU (3lb for PET, 6lb for nylon on nail puncture test)
Rapid tear propagation that makes rips quickly become un-repairable, resulting in need for envelope replacement
Inelastic characteristic results in more wrinkles that can not be removed through over pressurizing unless seams are excruciatingly perfect when produced
The inability to stretch much also puts the seams under more stress during thermal expansion, with increased risk of rupture
Film tends to curl at the edges when laid flat on a surface, which can make working with it a pain
Film can only be sealed from one side
Thermally sealed seams do not actually melt the primary films together, but rather melt an applied adhesive layer applied to one side of the film, so seams are more prone to peeling.
Not available in a semi-transparent form, so internally illuminated envelopes are not an option.

So just looking at the list above, it becomes obvious why PU is the material of choice for commercial blimps. The only condition where metalized film becomes the more logical choice is at the small end where weight becomes the primary design limitation. It would be nice if PU could be metalized in order to take advantage of the superior helium retention a metal coating offers, but metallization is incompatible with stretchy type films. This is why you only see inflexible films such as PET and nylon offered with a metalized coating. If PU were metalized, the metal barrier would quickly fracture as the film stretches and the helium retention benefit would be lost, and it would also pick up a few negatives like only being able to seal it on one side and being too opaque for internal illumination. So it would seem to be a trade-off for which there is no solution, other than perhaps embedding an internal metalized envelope inside of an outer PU envelope.

Thus the primary application for metalized films would be in smaller sized blimps or blimps that use balloonets for pressure regulation.

The seam wrinkle problem
----------------------------------------
Because both types of metalized films being tested here are not nearly as elastic as PU, envelopes made from them tend to show wrinkles along the seams. Wrinkles will usually occur where a wrinkle was present in one or both sides of the seam at the time it was welded, but can also occur from uneven tension in one or both pieces of film at the time they were joined or also from a partial pealing of the seam after inflation. This latter defect will actually cause an outward bulge in addition to wrinkles.

In an attempt to produce the most precise seams possible, a vacuum table was used to hold the two films together for the purpose of holding the film against movement and insuring no wrinkles prior to starting the seams. A CNC machine driven from a CAD file is then used to lay down precise seams with a consistent amount of pressure, temperature and feed rate. Even with all this effort, currently there are still problems with the heat roller pushing wrinkles into the film which must be worked out, but the problem is most likely solvable. The images below are from a first attempt, so the seams are showing more wrinkles than what should be possible. These pictures also show an envelope made from PET, which is the least flexible of all films so it will be a worst case example for wrinkles. The nylon film is slightly more elastic than PET, so it should result in slightly less wrinkles.

Capping the nose and tail
----------------------------------------
The envelope shown here is constructed the same way beach balls are made, where the gores stop short of meeting at both ends and then a disk is used to patch the resulting holes. Creating the heat seal around these patches is quite challenging to produce without resulting in a bunch of wrinkles, making them also the most leak prone areas. Because the film is not actually melted the way PU is, wrinkles in a seam that result in tiny air passages can not be ironed out and melted shut the way then can be with PU. The size of the air passage can only be minimized once it exists, then some sort of glue must be used to patch it. Also since only one side of the film can be sealed, you have no choice but to have the dull grey heat sealing side on the outside, with the end caps showing the reflective metal side on the outside. Personally I think this looks better than the reverse anyway, but you really don’t have a choice in the matter unless you created some kind of re-sealable port that would allow turning the envelope inside out again after applying the end caps.

Temperature Stress
--------------------------------------
To see if the seams could hold up to a change in temperature, I filled the blimp with air at 78F and then shut off the AC and let the room go to about 95F. The wrinkles slowly started to get smaller, but eventually the seams did “herniate” in a few spots which resulted in tiny leaks. It wasn’t a full scale blown seam, but rather a partial peal that created ugly bulges, one of which had a tiny leak. This could be the result of a flaw in the sealer, since the majority of the seams were unaffected. I’ve had problems with partial seam pealing with my two piece balloons as well, which I suspect is the result of slop in the sealer wheel resulting in inadequate pressure at random spots while rolling. I’m in the process of redesigning the sealer wheel using better bearings. The sealer wheel used here was made from parts taken from the commercial heat sealing wheel you can buy for around $120 at many packaging supply websites, and that sealer has several flaws with it. So I will probably do a rebuild on the sealer before proceeding with the nylon film test.

Build time on the envelop was about 2.5 hours, with another hour added for patching the ends. I think the time sealing the gores together could be decreased to about 2 hours with some process changes. The envelope weighs 133 grams with a volume of .6 cubic meters (21 cubic feet).
pyronaught is offline Find More Posts by pyronaught
Reply With Quote
Sign up now
to remove ads between posts
Old Aug 21, 2014, 12:24 AM
Go small or go home
ruzam's Avatar
Canada, SK, Regina
Joined Mar 2008
1,469 Posts
That's some nice work there pyro.

A foil strip (tape) sealed across the seam will increase seam strength and prevent partial splitting under pressure. Then you can inflate to a pressure that should make the wrinkles disappear for all but the most extreme curves (nose and tail). Of course that would also add tremendously to the construction time as you would have to 'tape' the seams manually.

How do you manage to close it up without damaging the foil?

Given how hard it is to prevent 100% of seam leaks, I'm starting to wonder how important it is to have a metalized foil helium barrier anyway. Maybe it's just because even my most leak free envelope attempts still deflate some portion of air in a matter of hours, I can only imagine that the helium loss would be significantly more through the leaks than through the material permeability. Products like Hi-Float will easily put a coating on latex balloons boosting their helium retention from less than a day to weeks or more. This tells me that the right plastic barrier will work as well as any metalized coating. At least for small blimp envelopes. A material that bonds permanently with poor helium retention would be preferable to a material with tricky bonding qualities and great helium retention.
ruzam is offline Find More Posts by ruzam
Reply With Quote
Old Aug 21, 2014, 09:29 AM
Registered User
Joined Feb 2009
400 Posts
I kind of wish I did a slow leak test before doing the expansion test, since now I don't know how leak free the original seams were before popping them. I guess I didn't expect that the temp diff from turning off the AC would actually pop the seams in a few places. I guess I could try and patch the holes that I know were from expansion. I don't think a bulge can be eliminated without turning the blimp back inside out again and repairing it from the metalized side, which I don't have a way of doing without cutting the patch off and installing a larger one afterwards. The patches are already too large as it is though. I think there is actually no way to get one of those patches installed without wrinkles because you can't put a flat disk over a curved surface without getting wrinkles. I'm curious how other foil blimp builders have been doing this, since there are a number of them I have seen over the years. The metalized film blimps seem to be particularly popular at the RC Airship Regetta contest in Germany every year.

My only problem with the foil tape method of reinforcing the seams would be the appearance of it. One thing I like about the metalized film turned inside out is it really looks like the traditional grey color of old blimps from the Goodyear era, which I like. So for authentic modeling purposes it is a good choice of material IF you can get the wrinkles out. They do make a clear version of this film without the metalized coating though, which could be used for taping the seams. Unfortunately you'd have to buy far more than all blimp builders combined over a lifetime could ever use just to get your hands on some.

I think the majority of wrinkles in the seams were caused by wrinkles getting ironed into the seam during production. I wasn't taking a lot of time to try and prevent that, but rather using the minimal amount of taping to try and get the shortest production time. As the gores piled up in the stack, the vacuum became less effective at holding the top sheet flat. I have some ideas for how to do it better next time though.
pyronaught is offline Find More Posts by pyronaught
Last edited by pyronaught; Aug 21, 2014 at 09:39 AM.
Reply With Quote
Old Aug 22, 2014, 01:49 PM
Registered User
Joined Feb 2009
400 Posts
Here's my new heat wheel I made. This one has more thermal mass, better heat transmission from the element to the wheel and less play in the bearings than the original one I salvaged from a commercial unit. I'm going to try and have the thermocouple measure the temp directly from the wheel for more accurate temp feedback. The old one was measuring it from the stationary fork that holds the wheel, since that was easier to do but less accurate. The commercial wheels don't even use thermocouples nor can they be made to reliably maintain a certain temperature. They just use a variable voltage source to the element and you have to experiment to find the right temp, which is then subject to change depending on ambient temp and will not adjust to compensate for increased heat transfer out of the wheel during continuous use.
pyronaught is offline Find More Posts by pyronaught
Reply With Quote
Old Aug 22, 2014, 02:08 PM
Go small or go home
ruzam's Avatar
Canada, SK, Regina
Joined Mar 2008
1,469 Posts
How does the heat get transferred to the wheel?

It needs an infrared thermo sensor aimed right at the wheel surface as it leaves the foil. Not that it would be easy to construct or design.

I'd like to see a wheel on wheel setup (over under) on long thin extension arms. The kind of thing you could hand feed a long length of foil through. Maybe self powered wheels. Strictly for hand seaming (no computer controlled direction).
ruzam is offline Find More Posts by ruzam
Reply With Quote
Old Aug 22, 2014, 11:00 PM
Registered User
Joined Feb 2009
400 Posts
The wheel uses a graphite impregnated bronze sleeve bearing to spin around the canister heating element, so the heat is transferred through metal. I've tried those hand held IR temperature measuring units but they are pretty inaccurate at high temps. Even at low temps the reading fluctuates quite a bit on each read. I just made a spring loaded thermocouple plug that will thread into the fork that holds the wheel and push the tip into the side of the wheel near the edge.

I've seen ultrasonic seam welding machines that use the two wheels like what you mention. Actually one is a wheel and the other side is the ultrasonic horn. With heat that method might not work too well though, since you have to pause as you feed the material through. With ultrasonic you only activate it when moving the material and then easily stop it while you reposition to pull more material through. With heat wheels, there is no way to turn off the heat so you would over temp the seam at the points where you paused.

Sonobond sent me a sample of some metalized film that was welded with one of their ultrasonic sewing machines. The seams were very clean looking and very strong. Those machines cost about $20K though, and you still have to manually feed the material through them. I looked into the possibility of using an ultrasonic head on my CNC machine, but I don't think it is possible without a metalic surface on the other side of the ultrasonic horn. I could not get a straight answer from the sales guy though, it might be possible with a sheet metal table surface but I couldn't get verification on that. I know that the metalized PET and nylon can be bonded using ultrasonics as an alternative to heat though.
pyronaught is offline Find More Posts by pyronaught
Last edited by pyronaught; Aug 22, 2014 at 11:09 PM.
Reply With Quote
Old Aug 24, 2014, 01:29 AM
Registered User
Joined Feb 2009
400 Posts
Here’s the new heat sealer. Some interesting things happened with this new design. The most notable is that the temp at the wheel is only 300F when sealing the film, whereas the same seal required 400F with the old wheel. So on the old one the fork holding the wheel got a lot hotter than the wheel itself, and the perceived sealing temp was way off. The wheel really can’t heat up to much over 350F with the 75watt heating element, but that is enough for sealing the films I’m using. Another odd thing that happened on this new one is that the heat doesn’t transfer nearly as rapidly to the fork and piston as it did in the old version. The cable tie-down plate actually never even heats up beyond warm to the touch. The only real difference that could have caused this would be the wider wood block being used as an insulator and gripping the heating element all the way at the very ends, whereas the old one gripped the element about half inch in from the ends. The wheel itself heats up twice as fast as the old one though. I can be ready to go 10 minutes after turning on the power, whereas it took 20 minutes on the old one.



You can see the spring loaded thermocouple off on the right side. I took a piece of 3/8” threaded rod and bored it out to hold a small spring loaded piston that pushes the thermocouple to the side of the wheel. The plug is threaded into the side of the fork so that you can turn it to adjust the tension to an amount that doesn’t cause too much drag on the wheel.



I’m pretty happy with this new one so far. The only thing left that needs to be cleaned up is the amount of play in the vertical plunger. Some of the play is from the gears not being tight against each other and some is from the pin that sits in the guide channel having too much clearance. The gear one would be a major pain to fix since I’d have to rebuild the whole gear box and that thing takes a lot of work, so I’ll probably just leave it. The problem caused by excessive play is that the wheel will rotate slightly off it’s calculated tangent to the curve being sealed, thus the wheels axis of rotation won’t be exactly perpendicular to the path of motion and you get a slight amount of sideways drag. This unwanted drag has the potential to push some wrinkles into the seam if the film is not tight enough.
pyronaught is offline Find More Posts by pyronaught
Reply With Quote
Old Aug 24, 2014, 09:32 AM
Warpquad: practice, practice!
Côte d'Ivoire
Joined Jun 2012
1,391 Posts
Excellent R&D work, Pyronaught! I look forward to reading about your continued progress as well as the related discussion.
Brainstorm is offline Find More Posts by Brainstorm
Reply With Quote
Old Aug 24, 2014, 06:19 PM
Registered User
Melbourne, Australia
Joined May 2005
126 Posts
Thanks for sharing your experiences, Pyronaut; you must be up to professional standard in your production methods now. I would just like to add my experience on the subject of wrinkles. As I see it, there are 3 key factors to avoiding wrinkles in nylon or PET envelopes:

1. Large number of gores. I generally use at least 8. However, it also depends on the film thickness relative to bag diameter (which determines pressure ‘stiffness’) and operating pressure. Obviously to keep leakage rate and tear susceptibility low it is best to keep the pressure low; I generally use 1-2 inches water. With this pressure, using foil balloon film on a 1 meter diameter envelope with 8 gores, you should have no wrinkles except near the nose and tail.

2. High accuracy in calculating, cutting and joining gores. I talk about this in my thread ‘Making Mylar Envelopes’, where I do show an example of what happens if you don’t. For scale models (where appearance is important) I always use seam tapes rather than cusped joins, as the latter is not as accurate in controlling gore width, and is prone to peeling and bulging (as Pyronaut says) which again changes gore width. I have not tried to make a wrinkle-free cusped envelope, so I can’t say whether it can reliably be done.

3. Consideration of curvature. Even with perfectly accurate seams you will get wrinkles due to the compound curvature, most likely near the nose and tail. Pic 1 below shows an 8 gore envelope; it is free from wrinkles except at the nose they can be clearly seen on the seams. This can be avoided by more gores, overall or locally. Pic 2 shows an envelope with only 4 gores (no attempt to avoid wrinkles). Wrinkles are evident, but no worse at the tail, where the gores increase to 8 by splitting each gore, similar to ‘darts’ in dressmaking. From considering geometry, I arrived at a parameter which determines ‘wrinkleness’ at any location on the envelope:
R cos (theta) / ROC
where R = radius of envelope (from axis to surface at that location)
theta = angle of surface to the axis
ROC = radius of curvature of the profile of the envelope.
The greater this number the more likely to wrinkle, although it depends on number of gores, thickness and pressure, so no specific wrinkle threshold value can be given. It is clearly smallest at maximum diameter, where ROC is very large. If the value increases markedly near the nose or tail (as it does on the ZPG-3W hull, in Pic 2) that is an indication to double up on gores.

To get smooth nose and tail caps, make the hole at the end small (ie take the gores as close to the ends as you can) and make the hull profile so that it becomes flat at the ends within the gore runout. Then it can be laid over a flat surface and the cap stuck on (glue or weld) without wrinkles.

Incidentally, if you want transparent material for making seam strips, metallising can be removed by vigorous rubbing with an MEK soaked cloth.
Alan Sherwood is offline Find More Posts by Alan Sherwood
Reply With Quote
Old Aug 25, 2014, 10:13 AM
Registered User
Joined Feb 2009
400 Posts
That ZPG hull looks great Alan! Do you have other closeup pictures of it? That's the best metalized film envelope I've seen to date.

I like your tip for the end caps, I didn't think to change the gore profile to flatten out at the ends like that. I'm going to try this on my next envelope attempt today.

I came up with another idea I'm going to try too which should save some time applying and removing tape: since I'm using the cusp type seams, I can use the faster technique of sealing the gores together one on top the other, which requires folding each previous gore in half before adding the new one on top. One of the problems I had on the last build was keeping tension on the gore that was folded in half, since it sits on top of other gores and can not be taped to the table. It can only be taped to the film of the bottom gore, which is not ideal. So I've added a thin steel cable stretched across the table with a quick release at one end, with the cable running down the center line where the gores are sealed. This cable is put in place prior to folding the previous gore, then it is tensioned with a lever so that it is very taught and can then be used to fold the gore around it while holding the fold exactly at the center line. Of course the longer the gore is the tighter the cable will have to be to keep it from flexing, but on an 8ft long gore this is not a problem. For a 20ft gore it would really have to be cranked down like piano wire though.
pyronaught is offline Find More Posts by pyronaught
Reply With Quote
Old Aug 25, 2014, 10:36 PM
Registered User
Melbourne, Australia
Joined May 2005
126 Posts
Look in the thread 'Ballonets for blimp' for more details of ZPG-3W model.
Alan
Alan Sherwood is offline Find More Posts by Alan Sherwood
Reply With Quote
Old Aug 26, 2014, 11:55 PM
Registered User
Joined Feb 2009
400 Posts
Still working on getting the perfect seam before building another blimp. I have found that the right amount of pressure at a temp of around 350F with a specific feed rate will produce cusp seams that are VERY strong. The challenge, however, is getting 100% contact between the wheel and the film with even pressure across the face of the wheel. If there is a gap on one side of the wheel even thinner than a sheet of paper it will cause a seam that peels easily at that point. The seam itself will still be gas tight because there will still be a strong seal on the side opposite the weak seal, but if the weak edge falls to the inside of the seam, then you will get a partial peel at that point when the pressure is high enough. Even a partial peel of 1/16" is enough to create very noticeable wrinkle spots.

The pictures below show a good seam along with flawed seams where the weak side is on the inside (to the right) and the outside (to the left). The seam has actually been cut on the left, so that's what the line is.

These weak spots are actually not caused by a flaw in the wheel or the wheel not being turned evenly on it's axis. Re-sealing the same seam over and over again results in the same weak spots in the same places, whereas any defect in the wheel would shift locations and also appear at regular short intervals equal to the circumference of the wheel. Because the defects consistently occur in the same spot, the cause must be unlevel spots in the table surface. Tempered hard board looks pretty flat, but apparently there must be very tiny waves in the surface. I may try a few other types of coated boards to see if I can get different results. The ideal surface would actually be tempered glass. I've also found that it is not necessary to follow the edge of the gore closely with the vacuum channel. A pair of straight parallel vacuum channels works just fine, seen in the picture below. This simpler design would actually make using glass more viable since no special shape would have to be cut.
pyronaught is offline Find More Posts by pyronaught
Reply With Quote
Old Aug 27, 2014, 12:09 AM
Go small or go home
ruzam's Avatar
Canada, SK, Regina
Joined Mar 2008
1,469 Posts
How is your surface supported? It looks pretty thin. Are you sure the boards aren't bending under the pressure of the wheel?

Even pressure seems to be critical to getting a full seam bond.
ruzam is offline Find More Posts by ruzam
Reply With Quote
Old Aug 27, 2014, 01:10 PM
Registered User
Joined Feb 2009
400 Posts
The surface is 3/16" thick and supported by wooden runners spaced at 4" apart. There is a runner at each edge of the vacuum channel to keep the edge from bending. I wasn't able to fully support the edge on the previous blimp and the vacuum was actually sucking the unsupported surface downward in some places, which caused defects in the seams.

Using 3/4" MDF would solve any flexing problem, although cost considerably more. I always dread trying to lift a sheet of that stuff too.
pyronaught is offline Find More Posts by pyronaught
Reply With Quote
Old Aug 27, 2014, 01:33 PM
Go small or go home
ruzam's Avatar
Canada, SK, Regina
Joined Mar 2008
1,469 Posts
If the surface material is what I think it is, 4" may not be enough with the combined vacuum and wheel pressure. You should be able to check that easily enough with a metal ruler. Run up the vacuum and trace a path slowly with the wheel (no heat). Follow it with the ruler and check for light gaps.

I use shelf boards for my 'long' construction. 5/8 thick, 11-3/4 wide, 6' long. It has a plastic coated surface that's very durable and smooth (and so far I haven't melted it). I'm not sure if they come in 8' lengths, but they're cheaper than a full size MDF board and easier to work with. You might have some ready made "shelf boards" available.
ruzam is offline Find More Posts by ruzam
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion Polyurethane 6 mil for Blimp Envelope lightandcolour Blimps 3 Feb 27, 2014 08:04 PM
Discussion ATLAS on Kickstarter - Pushing the envelope ;-) FazerFlipper Multirotor Talk 0 Feb 20, 2014 01:26 PM
Question Blimps with ready-made envelopes (balloons) ?? Gabe_BigApple Blimps 4 Nov 09, 2013 06:39 PM
Discussion Making model blimp envelopes. HarryWA Blimps 0 Nov 07, 2013 07:50 PM