sandwich airframe construction

[Chad]

I saw this youtube video of a guy using foam to sandwich a fiberglass/CF tube while working on a sailplane project. I wonder if this technique could be used to produce low density but very stiff rocket body tubes? I looked all over and couldn't find "7/10 of 1mm airex foam" (which is a strange dimension) so i was thinking about rolling balsa on a 4 inch mantral. The idea being create a fiberglass (or CF) tube, then spiral roll 1/32 inch balsa, sand, and then wrap with another layer of fiberglass or CF. You'd end up with a thicker body tube wall but I bet you could drop the density significantly. With a large rocket the low density would pay off in mass savings.

 

[cwbullet]

I can't see why not. There used to a company that sold a honeycomb Nomex or kevlar product call Aero glass (?). It worked quite well in the day.

 

[rharshberger]

You want to use a foam designed for composites or endgrain balsa panel material or nomex honeycomb. ACP Sales among others has various core materials.

 

[RocketFeller]

Interesting idea...

We built a large rocket using an end grain balsa and carbon fiber sandwich for fins and internal structure of the fin can. The body tube was carbon fiber and weighed 1.2 pounds per foot for a 12.75” tube with a 1/16” wall.

I would think that Nomex honeycomb would work best to bend around a tube. Both the foam and balsa flexible panels are small cubes of material attached to a fiberglass scrim on one side. It would make for a lumpy cylinder.

 

[kalsow]

John Coker has a nice how-to page describing a tube with a honeycomb core:
https://www.jcrocket.com/honeycomb.shtml
... and an actual project:
https://www.jcrocket.com/nike-asp.shtml

 

[MClark]

I use cardboard tube as the core material.

 

[Chad]

I use cardboard tube as the core material.

interesting, i was thinking the same. my latest idea is this:

1. make a very thin FG tube, let it cure but leave it on the mantrel
2. then wrap with layers of kraft paper to the correct thickness using a simple spray adhesive for the paper layers
3. finish with another layer of FG

This gives you a FG-Paper-FG sandwich. I like this idea because it's easy to control the width of the final tube so fitting existing nosecones is more straightforward which is my main issue with balsa or foam core. Eventually, I want to make my own nosecones but until then using paper as the core is an easy way to get the exact width i'm after.

 

[Steve Shannon]

interesting, i was thinking the same. my latest idea is this:

1. make a very thin FG tube, let it cure but leave it on the mantrel
2. then wrap with layers of kraft paper to the correct thickness using a simple spray adhesive for the paper layers
3. finish with another layer of FG

This gives you a FG-Paper-FG sandwich. I like this idea because it's easy to control the width of the final tube so fitting existing nosecones is more straightforward which is my main issue with balsa or foam core. Eventually, I want to make my own nosecones but until then using paper as the core is an easy way to get the exact width i'm after.

Instead of the spray adhesive, use epoxy and do it within 24 hours of mixing the epoxy used for the first layer. It will cross link with the first layer, forming a homogeneous structure.
When you build up a sandwich the shear strength of the inner core is critical to the bending strength of the structure.

 

[Chad]

Instead of the spray adhesive, use epoxy and do it within 24 hours of mixing the epoxy used for the first layer. It will cross link with the first layer, forming a homogeneous structure.
When you build up a sandwich the shear strength of the inner core is critical to the bending strength of the structure.

Word

 

[prfesser]

Instead of the spray adhesive, use epoxy and do it within 24 hours of mixing the epoxy used for the first layer. It will cross link with the first layer, forming a homogeneous structure.
When you build up a sandwich the shear strength of the inner core is critical to the bending strength of the structure.

Epoxy-soaked paper by itself can be a useful material. Photo is a 4"-3" transition made from 60 lb cardstock (obviously I haven't yet finished trimming the 4" opening). The tab on each of the five individual paper transitions was glued with ordinary white glue---carefully, as you want the five to be as identical as possible. When dry, each transition was rubbed inside and out with enough slow-cure laminating epoxy to wet it through (give it enough time to soak in well). The excess was wiped off and the layers stacked; don't overlap the tabs or the thickness will be very uneven. A spaghetti jar was covered with plastic wrap and transition was placed on the jar to keep it circular. A little weight on top to force the layers together. Cure with heat.

Wall thickness is about 0.045 - 0.050". Got two more of these to make for my semi-scale "260 Space Booster."
https://www.spacemodeling.org/jimz/eirp_36.htm

Best -- Terry

 

[JohnCoker]

Honeycomb core with composite skins inside and outside does work well. I used that to make custom tubes 5" ID for my Nike-Asp:
jcrocket.com/nike-asp.shtml#tubes

 

[Not Quite Nominal]

Instead of the spray adhesive, use epoxy and do it within 24 hours of mixing the epoxy used for the first layer. It will cross link with the first layer, forming a homogeneous structure.
When you build up a sandwich the shear strength of the inner core is critical to the bending strength of the structure.

I think it's the compression strength of the inner core that is the critical value, and the point of the structure is to avoid shear loading.

The purpose of an I-beam or sandwich is to take that shear load and turn it into tension on the outer skin, compression on the inner skin, and compression in the core. For a given length, the lower the aspect ratio the lower the stresses on the skins and the higher the stresses on the core.

Or so I remember. Engineering school was decades ago, and I dropped out.

 

[Steve Shannon]

I think it's the compression strength of the inner core that is the critical value, and the point of the structure is to avoid shear loading.

The purpose of an I-beam or sandwich is to take that shear load and turn it into tension on the outer skin, compression on the inner skin, and compression in the core. For a given length, the lower the aspect ratio the lower the stresses on the skins and the higher the stresses on the core.

Or so I remember. Engineering school was decades ago, and I dropped out.

Look at the second illustration on the first page of the following article to see why I said shear is critical in this instance.
https://wp.optics.arizona.edu/optomech/wp-content/uploads/sites/53/2016/10/OPTI_222_W10.pdf

That doesn’t mean shear is the only stress, but if a substandard adhesive is used between layers failure in shear weakens the structure significantly.

 

[Not Quite Nominal]

Look at the second illustration on the first page of the following article to see why I said shear is critical in this instance.
https://wp.optics.arizona.edu/optomech/wp-content/uploads/sites/53/2016/10/OPTI_222_W10.pdf

That doesn’t mean shear is the only stress, but if a substandard adhesive is used between layers failure in shear weakens the structure significantly.

I agree that shear is present, and interlaminar shear resistance contributes to beam stiffness. In a uniform beam, it may even contribute significantly.

In the case of an I beam or a stressed-skin beam/panel, the sides have an order of magnitude plus more resistance to tension/compression than the core. In these cases I think the main contribution of the core is compression resistance, especially in a high aspect ratio beam at low levels of bending (our case)

Consider the case of an I beam with the web replaced with a very strong, very flexible fabric in tension. This fabric has much more shear resistance than compression resistance (zero compression resistance, actually), and such a beam will easily bend.

On the other hand, consider an empty but capped soda bottle, or the inflated balloon tanks of Atlas. The inner medium (air) has approximately zero shear resistance, but high compression resistance. This structure is quite stiff.

 

[Steve Shannon]

I agree that shear is present, and interlaminar shear resistance contributes to beam stiffness. In a uniform beam, it may even contribute significantly.

In the case of an I beam or a stressed-skin beam/panel, the sides have an order of magnitude plus more resistance to tension/compression than the core. In these cases I think the main contribution of the core is compression resistance, especially in a high aspect ratio beam at low levels of bending (our case)

Consider the case of an I beam with the web replaced with a very strong, very flexible fabric in tension. This fabric has much more shear resistance than compression resistance (zero compression resistance, actually), and such a beam will easily bend.

On the other hand, consider an empty but capped soda bottle, or the inflated balloon tanks of Atlas. The inner medium (air) has approximately zero shear resistance, but high compression resistance. This structure is quite stiff.

I have no interest in arguing about it. I just made a suggestion aimed at preventing delamination in his project.

 

[Chad]

Honeycomb core with composite skins inside and outside does work well. I used that to make custom tubes 5" ID for my Nike-Asp:
jcrocket.com/nike-asp.shtml#tubes

from your site
"payload section weighs 565.5g, which is only 163 grams per foot! (Compare that to 258g/ft for unreinforced 6" phenolic tubing and 146g/ft for 3.9" tubing.)"

that confirms a lot of thinking and kicking-the-tires, did you happen to measure the tube's stiffness against other materials?

i bet if you made an 8" tube you'd come out even further ahead in terms of mass savings than you did with the 5" tube. You could keep the density very close to the same but you'd have a large increase in volume.

Madcow lists 4" G10 tubing at 12.8 oz/ft which equals 363 grams/ft. I'm positive the G10 tubing is way overbuilt for the job at hand, i can't back it up with numbers but I will eventually. So if your honeycomb core tube is strong enough, even with a 4" tube that's still a 50% savings in mass vs G10 on the airframe. The fiberglass 4" DX3 is listed at a weight of 6lbs, 48" of G10 weighs 3.2lbs so a big portion of that kit weight is the body tube, dropping the body tube weight by 50% may save around 25% on the total weight of a fiberglass rocket.

The wall thickness is what's stopping me because creating a custom nose cone is beyond me at the moment. That's what got me thinking about paper core, i could dial the wall thickness to fit OTC nose cones by varying the amount of paper.

 

[JohnCoker]

that confirms a lot of thinking and kicking-the-tires, did you happen to measure the tube's stiffness against other materials?

i bet if you made an 8" tube you'd come out even further ahead in terms of mass savings than you did with the 5" tube. You could keep the density very close to the same but you'd have a large increase in volume.

I don't believe I measured the stiffness, but given that the skin was C.F., I'm sure it was high.

Larger tubes are no problem as well. We fabricated an 8' long 12" diameter tube as well, which was easy for one person to carry (which would not have been the case with a normal fiberglass one).
jcrocket.com/waccorporal.shtml#tubing

 

[Nytrunner]

Someone (Australian I think) showed a sandwich tube with cork as the inner material. Light, permeable, flexible till locked in with epoxy.

 

[Wallace]

from your site
"payload section weighs 565.5g, which is only 163 grams per foot! (Compare that to 258g/ft for unreinforced 6" phenolic tubing and 146g/ft for 3.9" tubing.)"

that confirms a lot of thinking and kicking-the-tires, did you happen to measure the tube's stiffness against other materials?

i bet if you made an 8" tube you'd come out even further ahead in terms of mass savings than you did with the 5" tube. You could keep the density very close to the same but you'd have a large increase in volume.

Madcow lists 4" G10 tubing at 12.8 oz/ft which equals 363 grams/ft. I'm positive the G10 tubing is way overbuilt for the job at hand, i can't back it up with numbers but I will eventually. So if your honeycomb core tube is strong enough, even with a 4" tube that's still a 50% savings in mass vs G10 on the airframe. The fiberglass 4" DX3 is listed at a weight of 6lbs, 48" of G10 weighs 3.2lbs so a big portion of that kit weight is the body tube, dropping the body tube weight by 50% may save around 25% on the total weight of a fiberglass rocket.

The wall thickness is what's stopping me because creating a custom nose cone is beyond me at the moment. That's what got me thinking about paper core, i could dial the wall thickness to fit OTC nose cones by varying the amount of paper.

And if you were to add foam sandwich core fins to the build....

 

[Chad]

Someone (Australian I think) showed a sandwich tube with cork as the inner material. Light, permeable, flexible till locked in with epoxy.

Yeah I saw that! The cross section shot was awesome. I was at a craft store today waiting on my wife and spotted a roll of 3/32 inch cork for $5 or so and grabbed it. I had no idea it was available that thin.

All my supplies are in and I made the mandrel yesterday (full length 4" coupler filled with 2part foam and PVC handles). I need to make a couple jigs then I'll be ready to do some tests.

 

[Chad]

Quick update, the first part of my layup is done. I used part of the John Coker method and wetted out the FG while flat first but then I covered it with Dacron then paper towels, flipped it over, then another layer of Dacron and paper towel. Then used a rolling pin to squeeze out as much resin as I could while it was flat. It worked well because both layers of Dacron where saturated when I was done.

Once that was done, I rolled it onto my mantrel. It was tricky because I was getting air pockets at first. Tonight I'll do the paper and then Saturday the outside layer of FG. Goal is to have a 12" test piece Sunday. I have to go out of town next week for work.

Here it is curing

 

[Chad]

Finally finished this little POC. It turned out pretty well but keeping air pockets out of the paper while rolling was surprisingly hard. Also, I need to modify my technique for the fiberglass as the moment something doesn't go according to plan it gets pretty crazy. I'm going to do a 48" tube next and will post a standard build thread.

 

[Steve Shannon]

It is much easier to wrap a tube one layer at a time. When you stack flat layers and then try to wrap the stack around a small diameter mandrel you cause the layers to separate and wrinkle.

 

[JohnCoker]

It is much easier to wrap a tube one layer at a time.

+1

You won't loose strength as long as you put on successive layers before the earlier layers have fully cured.

 

[Chad]

+1

You won't loose strength as long as you put on successive layers before the earlier layers have fully cured.

Thanks, this simplifies constructing a larger tube.

Something I'm doing differently is placing the wetted fiberglass between two layers of Dacron and then using a rolling pin to remove excess resin prior to putting it on the mandrel. Any comment there? I know it def. removes resin because the Dacron ends up soaked but is it possible to remove too much? I doubt I'm using more pressure than a vacuum pump would

 

[Steve Shannon]

Thanks, this simplifies constructing a larger tube.

Something I'm doing differently is placing the wetted fiberglass between two layers of Dacron and then using a rolling pin to remove excess resin prior to putting it on the mandrel. Any comment there? I know it def. removes resin because the Dacron ends up soaked but is it possible to remove too much? I doubt I'm using more pressure than a vacuum pump would

You’re definitely not using as much pressure as vacuum bagging. More importantly you’re not applying pressure consistently across the area, but it’s an interesting technique that I’ve not heard of before. Does the fabric wrinkle ahead of the roller?
Yes, it is possible to starve the composite. Once material has been fully wetted, it is pretty hard to remove too much. You’re more likely to not distribute the epoxy everywhere in the first place, resulting in soft spots. With your rolling pin technique you’re removing the epoxy while the cloth is flat. I’d be concerned about not having complete coverage as you’re applying that layer to the previous layer, resulting in voids between the layers. As John said, be sure to apply the succeeding layers before the previous layer is fully cured; that will allow the epoxy in the different layers to cross-link, which is most desirable because it results in a homogeneous structure.
If you’re looking for alternatives to bagging there’s heat shrink tape which squeezes out excess epoxy.
At the other end of techniques is vacuum infusion which only sucks in the right amount of epoxy in the first place. I haven’t used it yet, but I’m very interested in it.

 

[JohnCoker]

Something I'm doing differently is placing the wetted fiberglass between two layers of Dacron and then using a rolling pin to remove excess resin prior to putting it on the mandrel.

Removing as much slack and excess epoxy as possible is ideal. Not only does it make a strong structure, but if you're going to vacuum bag excess fabric turns into wrinkles.

I like to "roll" the cloth on slowly, pulling and smoothing it out towards the end as I roll. That wipes excess epoxy out and helps stretch the cloth a bit so it is less likely to wrinkle.

Here's my video on reinforcing a tube, showing some techniques you mind find helpful.
jcrocket.com/tube-wrapping.shtml

 

[Chad]

Wife + kids went to the beach for spring break so yesterday I set out to make a 48" by 4" tube using this method. Man, working with large pieces of wetted out fiberglass is a lot harder than working with small pieces. I did two layers of FG then a layer of paper, as the attached photo shows, I just can't get the air out of the paper. I'm putting this idea (and build thread) on hold because it's getting in the way of actually having fun haha.

One good technique that came out of this whole thing is mandrel prep. I took packing tape and spiral wrapped my mantrel sticky side out overlapping by about 1/2 inch. Then I used some Goo Gone from Home Depot and took off the "sticky". It's a long tedious process but very forgiving. If you don't get a perfect wrap you back off the tape a bit and try again. Removing the adhesive with Goo Gone takes time but there's no "art" to it, just a lot of paper towels and mess. At the end you're left with a perfectly wrapped mantrel, you can see it through the paper in the picture.

Another good technique was using a something rigid to lift the wetted out FG onto the mantrel. After wetting out and compressing the FG I stuck an edge onto my trusty angle iron and used that to lift the FG onto the mantrel, trying to pick it up with only your hands is a nightmare because it sags and you'll never get it on straight.

Compressing the FG between two layers of Dacron is useful but it gets unwieldy with large pieces, I was working with 48" / ~ 13" pieces. I never had issues with wrinkles while compressing but when you're done, getting it on the mantrel, straight, with no air bubbles or wrinkles was very difficult. I didn't try John's method of rolling it on but I bet that's the best way to do it.

 

[Steve Shannon]

Wife + kids went to the beach for spring break so yesterday I set out to make a 48" by 4" tube using this method. Man, working with large pieces of wetted out fiberglass is a lot harder than working with small pieces. I did two layers of FG then a layer of paper, as the attached photo shows, I just can't get the air out of the paper. I'm putting this idea (and build thread) on hold because it's getting in the way of actually having fun haha.

One good technique that came out of this whole thing is mandrel prep. I took packing tape and spiral wrapped my mantrel sticky side out overlapping by about 1/2 inch. Then I used some Goo Gone from Home Depot and took off the "sticky". It's a long tedious process but very forgiving. If you don't get a perfect wrap you back off the tape a bit and try again. Removing the adhesive with Goo Gone takes time but there's no "art" to it, just a lot of paper towels and mess. At the end you're left with a perfectly wrapped mantrel, you can see it through the paper in the picture.

Another good technique was using a something rigid to lift the wetted out FG onto the mantrel. After wetting out and compressing the FG I stuck an edge onto my trusty angle iron and used that to lift the FG onto the mantrel, trying to pick it up with only your hands is a nightmare because it sags and you'll never get it on straight. View attachment 376658

You seem to be finding difficulties.
I never wet the fabric first. I draw a straight line along the tube (which could be a mandrel), then I apply epoxy along that line, lay the edge along that line into the epoxy. You need to set it up in a way that allows you to roll the fabric off the dry roll of fabric and onto the mandrel or body tube as you wet the fabric onto the mandrel or body tube. That way you’re not fighting the weight of the wetted fabric. I use a paintbrush with a sideways rolling movement, but there are special rollers for composite work that don’t froth the epoxy. I think a foam roller would add way too much air to the epoxy, which I call frothing.
I learned a long time ago by watching videos Dave Triano put out, but John Coker and Tony Fish (tfish38 on YouTube) have made some absolutely excellent videos since then. There are tons of other videos on YouTube that deal with surfboards and such.

 

[Chad]

You seem to be finding difficulties.
I never wet the fabric first. I draw a straight line along the tube (which could be a mandrel), then I apply epoxy along that line, lay the edge along that line into the epoxy. You need to set it up in a way that allows you to roll the fabric off the dry roll of fabric and onto the mandrel or body tube as you wet the fabric onto the mandrel or body tube. That way you’re not fighting the weight of the wetted fabric. I use a paintbrush with a sideways rolling movement, but there are special rollers for composite work that don’t froth the epoxy. I think a foam roller would add way too much air to the epoxy, which I call frothing.
I learned a long time ago by watching videos Dave Triano put out, but John Coker and Tony Fish (tfish38 on YouTube) have made some absolutely excellent videos since then. There are tons of other videos on YouTube that deal with surfboards and such.

I really like tfish38's setup and wetting out the fabric on the mantrel is def. the least aggravating way to go. Of course all of John's videos are amazing too, i've thought about doing a reinforced cardboard body tube using his method next but tip-to-tip glassing of fins will probably be my next project.