Filament winding

I backed the x-winder on kickstarter (actually, I backed it twice), and received mine about six months ago. Just wondering if anyone else on the forum purchased one.



[video=youtube_share;FrCHP4VYgqY]https://youtu.be/FrCHP4VYgqY[/video]
this is a 6 inch, 5 foot section of 4-layer carbon fiber on a cardboard mandrel.





[video=youtube_share;dDEt5hDr3oE]https://youtu.be/dDEt5hDr3oE[/video]
this is a 38mm 6 foot section of carbon fiber, minimum diameter on a copper mandrel. I like the metal mandrels more because I can wind much tighter, but they are a real pain to remove once cured.





[video=youtube_share;3V97_Pkh4K8]https://youtu.be/3V97_Pkh4K8[/video]
this is my redneck curing process. After winding I cover the airframe with a specialized heat shrink tape, then connect a heat gun to the winder and let it rotate for about six hours. Creates a beautiful finish and squeezes out a lot of excess resin

Alex,

Was not aware of that but would be interested in the long term performance of that material. I have some specific interest in the weight to impact performance of filament wound airframe. That is to say well constructed wound airframes are heavy realtive to convolute cloth airframes. And if wound aifrframes are not well constructed they unravel on impact.

Perhaps you could report back on your particular flight experiences?

Feckless

Feckless Counsel said:

Alex,

Was not aware of that but would be interested in the long term performance of that material. I have some specific interest in the weight to impact performance of filament wound airframe. That is to say well constructed wound airframes are heavy realtive to convolute cloth airframes. And if wound aifrframes are not well constructed they unravel on impact.

Perhaps you could report back on your particular flight experiences?

Feckless

Click to expand...

These seem to be indestructible. I had a six inch booster fall from ten thousand feet (spinning sheered my kevlar harness) without a scratch. Actually, I have it on video (onboard video, vomit inducing) and will try and post it. There are pro's and con's to filament vs. convolute, but the two reasons I've made the switch are cost, and automation - the increased density and flexibility I have in winding angles for strength are just a bonus. Much more work to hand lay up carbon fiber, and more prone to error, at least with my skills From a cost perspective, I'm at 1/3 the price of my old convolute method, and that's buying carbon tow off ebay.

If you have a metal lathe, could you filament wind your own tube that way?

taiwanluthiers said:

If you have a metal lathe, could you filament wind your own tube that way?

Click to expand...

If it were CNC I suppose you could, a geared manual would not work out so well.

One of my team mates bought one, but it is still in the box while his shed (man cave) is being finished.
Give us the scoop....what tow's should we buy? Any tips?

Why can't a geared manual lathe work? Are there some weird ways you are supposed to filament wind a tube?

taiwanluthiers said:

Why can't a geared manual lathe work? Are there some weird ways you are supposed to filament wind a tube?

Click to expand...

I would think the feed rates available on a geared machine would be too fine, and you typically need to reverse the spindle direction to reverse the feed direction.

The secret to filament winding is in the software, not the hardware. Chadrog is right, you could use a computer controlled lathe, but you would have to buy windcad, or cadfil winding software, both of which are very expensive. Each layer has between 30 and 120 steps, each step being a different speed, which changes the winding angle, and determines strength and pattern.

The Carbon Fiber tow in both those videos is 6k, which means it is a bundle comprised of 6000 strands. When wet, it flattens out to about .125 of an inch. I won't make an airframe larger then 6" with 6k because there is so much surface area to cover, if I do four layers, the laminating epoxy in the first layer likely to harden by the time I get to the cure phase. I use 80k tow for the large rockets, and it goes much faster. I buy most of the filament on Ebay, as it is likely industrial "ends" (ie, the ends of spools to short to be used in manufacturing ,but great for hobbyists), and reasonably priced.

I've also tried fiberglass, kevlar, and basalt.

Basalt is interesting because it is superior to fiberglass in strength, yet costs less - it's about 1/20th the cost of carbon fiber.

Basalt? Is that the volcanic rock that basically lava turns into when it cools down?

Its basalt rock thats been spun into fibers, its marginally heavier than E-glass's density (by .05 gr/m3), and has better tensile strength for the fibers. Somebody will come along and give some better info than that, Wikipedia has a short article about it with common comparisons to steel, fiberglass, carbon fiber and a few other common materials.

taiwanluthiers said:

Why can't a geared manual lathe work? Are there some weird ways you are supposed to filament wind a tube?

Click to expand...

The patterns for a tube are not really "weird", but they need to be laid down in a controlled manner to ensure a uniform result. I don't think the average guy has the motor skills or patience for this. This can be best seen at the beginning of the winding process, when you have a good visual contrast between the fiber and the mandrel.
[video=youtube;1A3vaJaNDLY]https://www.youtube.com/watch?v=1A3vaJaNDLY[/video]

Reinhard

What is the advantage of filament wound tube besides the fact that they are just extremely expensive and difficult to make?

Is this for minimum diameter/max altitude rockets? I realize filament wound would be stronger than say convolute winding of carbon fiber/fiberglass around a mandrel, but is it enough to matter? How important is it for a rocket to survive intact when it crashes into the ground at 500mph? Isn't that what we're trying to avoid altogether?

I'll take convolute over FW any day -- you just can't get it.

The advantage FW has is lower cost plus [affordable] machines like this to do the work...

taiwanluthiers said:

What is the advantage of filament wound tube besides the fact that they are just extremely expensive and difficult to make?

Is this for minimum diameter/max altitude rockets? I realize filament wound would be stronger than say convolute winding of carbon fiber/fiberglass around a mandrel, but is it enough to matter? How important is it for a rocket to survive intact when it crashes into the ground at 500mph? Isn't that what we're trying to avoid altogether?

Click to expand...

There are numerous advantages to fillament wound tubing, the biggest being precise fiber placement that you cant get with convolute tubing. It's easier to apply tension to the fillament and get a lower epoxy content. The tubes themselves are cheaper (a pound of fillament is cheaper then a pound of cloth.) Shrink tape is also easier to finish then peel ply.

The only real downside is winder itself. $$$

Alex

I though about buying one. I wonder if they start selling them again.

How do you control the OD?

Also, where does the excess resin go using shrink tape? Does it just squeeze out the overlapping edges of the tape?

ID is based on the mandrel. OD is based on the ID plus the thickness of the filamented added.

I often pondered that too, back in the day I got some Aeropoxy and heat shrink tape from Shadow Composite and they said in their video to just wrap the tape around the bare resin saturated cloth tube... I tried it on a 24mm BT but used the university's curing oven to cure it. It ended up producing a glass smooth tube that had tape spirals on it, but I couldn't tell if it was better no tape. I was using carbon fiber and the carbon is incredibly stiff, that it had difficulty wrapping around such a small BT so I basically forced it around with the tape. I did not see much excess resin outside of the tape either, it felt as though the tape simply compressed the resin and fiber but did not squeeze the resin out.

I watched a few How it's Made episodes where they were making carbon fiber products... it was very educational. They all work with molds. They would lay the saturated cloth in the mold and then followed by peel ply and breather. The bag is placed over the part and vacuum drawn. Once cured the part comes out of the mold formed. On the other hand I seen one where they made carbon fiber bike wheels and all they did was laid up prepregs into some mold covered in plastic, and then it was squeezed into the mold while heat was applied to cure it (I don't think a breather/peel ply was used). A little hazy, have to watch it again.

Alex,

Could I please ask you to post some detail photos of the resin tank and its roller system?

Feckless

cwbullet said:

ID is based on the mandrel. OD is based on the ID plus the thickness of the filamented added.

Click to expand...

Ok. Thickness in = thickness out. I was thinking more along the lines of achieving industry standard thickness (i.e. a grinder).

You can come close without a grinder. It is not as exact. M

Be pretty hard to achieve good tolerance with a standard grinder, you'd need a surface grinder and a lathe combo.

Feckless Counsel said:

Alex,

Could I please ask you to post some detail photos of the resin tank and its roller system?

Feckless

Click to expand...

Feckless,

do a YouTube search for "x-winder llc" on YouTube. You will find about 10 chapters on building the x-winder, and they go into detail on the tank and roller system. I don't have mine setup exactly like they do (I use a dual squeegee), but you'll get the idea.

as far as filament wound vs. convolute, in terms of quality and strength. If they are built right, I don't think one is better then the other at all. As I said earlier, for me it's about cost, and automation.

It would be cool if you could do nosecones with this!!

We (Auburn University) just bought one of these to do our rocket bodies / nose cones on. Have you tried a nose cone yet?

I've made a few nosecone's, and done a lot of testing.

which version of the filament winder did you buy? They have a new extra axis version which is much better suited for nosecones.

As it stands, Conical, Spherically blunted, ogive, and tangent/ogive are difficult since you can't wind around a point without the extra axis.

Haack, Von Karman, and Aerospike work very well. I use a 3d printer using PVA filament to design the nosecone andI wind around the 3d printed shape. I then dissolve the PVA to get a carbon-fiber only nosecone. I've copied most of the industry, and use a metal tip at the end for the last part of the nosecone I can't wind.

I've written some code to build the Haak seris and Von Karman windings. Send me a PM if you would like me to share.


Alex

alexzogh said:

I've made a few nosecone's, and done a lot of testing.

which version of the filament winder did you buy? They have a new extra axis version which is much better suited for nosecones.

As it stands, Conical, Spherically blunted, ogive, and tangent/ogive are difficult since you can't wind around a point without the extra axis.

Haack, Von Karman, and Aerospike work very well. I use a 3d printer using PVA filament to design the nosecone andI wind around the 3d printed shape. I then dissolve the PVA to get a carbon-fiber only nosecone. I've copied most of the industry, and use a metal tip at the end for the last part of the nosecone I can't wind.

I've written some code to build the Haak seris and Von Karman windings. Send me a PM if you would like me to share.


Alex

Click to expand...

Hey Alex,

I am sure I speak for everyone in saying we'd love to see some pics of your finished product!

And can you post some pictures of your tubes? I would love to see more.

alexzogh said:

I've made a few nosecone's, and done a lot of testing.

which version of the filament winder did you buy? They have a new extra axis version which is much better suited for nosecones.

As it stands, Conical, Spherically blunted, ogive, and tangent/ogive are difficult since you can't wind around a point without the extra axis.

Haack, Von Karman, and Aerospike work very well. I use a 3d printer using PVA filament to design the nosecone andI wind around the 3d printed shape. I then dissolve the PVA to get a carbon-fiber only nosecone. I've copied most of the industry, and use a metal tip at the end for the last part of the nosecone I can't wind.

I've written some code to build the Haak seris and Von Karman windings. Send me a PM if you would like me to share.


Alex

Click to expand...

We have the 2 axis right now, but I plan on getting the upgrade next fall as the department budget allows. We have 3D printed and flown nose cones and those work well for L1 and even L2 certs, but I would only use the metal tip for the L3 and experimental stuff we want to fly. The students usually fly their stuff in the late fall / early spring but I hope to get my L3 and other stuff done in August / September. We would only do the blunted section and either 3D print or lathe a nose tip.