Battle of the Epoxies: A Road to BALLS Experiment

The Rocketry Forum

Help Support The Rocketry Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

A5tr0 An0n

Well-Known Member
Joined
Sep 29, 2012
Messages
521
Reaction score
10
Intro:
I am getting ready to start a build that I will hopefully launch this year at BALLS 23 and I wanted to perform some simple experiments on different types of epoxies used for fillets. I am basically looking for the epoxy that offers the best combination of strength, adhesion, and high Tg. Now I have been wanting to do something like this for some time now, however this has recently been inspired by Jim Jarvis. In a conversation I had with him, he said that he had a weak adhesion with Cotronics 4525 when he tried to bond G10 FG to basswood. I had originally planned on using 4525 again but this has raised some concern. I thought that this was a good enough reason to go ahead and finally perform some experiments on the matter. I have read through the data sheets on a lot of epoxies and came to the conclusion that this in itself was not enough for me, I want to see practically how different fillets behave. I hope this information will be useful to others and hence the reason I am making a thread on the matter. Now if you have any ideas on how to make the testing of the fillet better then please share!

I am interested in a few different aspects of the fillet bonds. Those are the surface preparation, pull strength, flutter resistance, additives, and Tg (behavior at elevated temperatures). I believe that the fillet itself must be strong, but so should the bond and the Tg of the fillets. If the fillet is not properly bonded with the adherent and if the epoxy is not meant to work in higher temperatures then the properties can degrade and flutter might occur. As I slightly mentioned early these fillets are meant for a BALLS project that will experience ~M3.5 during its flight and this is the driven force for this experiment. Now I do not have much information on the actual temperatures at M3.5, however I do have data on a M2.5 rocket. This rocket's fin height is ~2.5'' and the diameter of the rocket is 54mm, during it's flight the airframe and fins experienced 275F and 350F, respectfully. We can thank James Donald for that data. I can only imagine that a 98mm airframe and a fin height of 5.8" will cause higher temperatures and thus is how I have arrived at wanting a higher Tg capable fillet. Also the aerodynamic forces will be greater at M3.5 and hence how I arrived at the desire for the best surface preparation method.


Details:
Below are the details of this experiment. This area is subject to be updated from time to time. Knowledge/understanding is ongoing and may differ as they progress, for that reason some things in the future may not be true as they are understood today.


Testing Method:
As far as the testing method goes, I am currently just planning on exposing the fillets to a static load by hanging weight off the fin to see how much force it takes to break the fillet (or the structure). In terms of testing the adhesion strength of the bond I plan on trying to pull the fin off the airframe by orienting the fin perpendicular to the ground while under a static load. This should act as a pull force, thus creating a shear strength test. I will perform these tests with and without applied heat. This is to simulate the loads and conditions that the fillets will experience. Mike Passaretti did this with his DDT vehicle by exposing the fillets to a static load; he also deserves much credit in this. Now I do not have any awesome sensors or testing equipment, so I will analyze the results in a very crude manner. Again if any body knows of some relatively inexpensive, more technical way of testing this, and gathering data then please let me know and I will do it!


Surface preparation:
Surface preparation is defined as one or a series of operations including cleaning, removal of loose material, and physical and/or chemical modification of a surface to which an adhesive is applied for the purpose of bonding. Surface preparation is intended to enhance the bonding strength to said surfaces.

Adhesion refers to the state in which two dissimilar bodies are held together by intimate interfacial contact such that mechanical force or work can be transferred across the interface. There is unifying theory of adhesion that relates the physical-chemical properties of materials to the actual physical strength of an adhesive bond. The interfacial forces holding the two phases together may arise from van der Waals forces, chemical bonding, or electrostatic attraction. The mechanical strength of the system is determined not only by the interfacial forces, but also by the mechanical properties of the interfacial zone and the two bulk phases.

The primary function of adhesives is to join parts together. Adhesives accomplish this goal by transmitting stresses from one member to another in a manner that distributes the stresses much more uniformly than can be achieved with mechanical fasteners. Adhesive bonding often provides structures that are mechanically equivalent to or stronger than conventional assemblies, at lower cost and weight per se. In mechanical fastening, the strength of the structure is limited to that of the areas in contact with the fasteners. It is not unusual to obtain adhesive bonds that are stronger than the strength of the adherends (surfaces that are bonded together).

So Why Do We Need Surface Preparation?
There is list of reasons for surface preparation here are some of the main reasons.
1. To remove or prevent the later formation of a weak layer on the surface of the substrate
2. To maximize the degree of molecular interaction between the adhesive or primer and the substrate surface
3. To optimize the adhesion forces that develop across the interfaces to ensure sufficient initial joint strength and during the service life of the bond
4. To create specific surface microstructure on the substrate

Normally, optimum surface energy and structure is achieved by chemical surface treatment. The chemical composition and the morphology of the surface are changed, so that the surface energy of the substrate is maximized for adhesion. Chemical treatments also increase the chances that hydrogen, dipole, van der Waals ionic and/or covalent bonding can form at the substrate/adhesive interface. A quick test for adequet chemical treatment is to place a bead of water on the surface of the part and note if the water spreads. If so, the contact angleit forms with the surface is very small.

Remember all surfaces no matter how clean they seem to be, have surface contaminates and these contaminates get in the way of reliable bonding. Advanced engineering polymers tend to be chemically inert – with few or no functional groups on which to bond. Some applications require specific functional properties like low friction, or adhesion promotion.

The 5 Methods of Surface Preparation:
There are 5 methods of surface preparation that I will test. These methods all receive the same surface treatment procedures and should therefore determine which preparation method yields the ideal conditions. These methods are briefly described below.

1. Sanding Method*. The material is sanded with a fine to medium grit emery paper, to an evenly abraded surface while not damaging the fibers. Since the strength of composites is mostly in their fibers, once the fibers are damaged then the structures overall strength is compromised. The main thought here, is that the bond being formed is a chemical bond due to the sharing of the outer electrons allowing the epoxy to fill the empty space. This empty space or the "contact area," between the substrate and the adhesive is what makes this bond work… If you are wondering yes, the contact area increasing when sanding.

13990649845_da8eff054e.jpg

Here is an example of the sanding method. Photo credit: flynfrog.

2. BTH Method* (Beat To Hell). Here you can take a saw, dremel, razor blade, etc. to the material and scuff it up pretty roughly. The main thought behind this is that it increases the surface area (more so than sanding) and thus allows the bond to have more to hold unto. The downside is your are in fact compromising the strength of the structure due to the damage caused to the fibers. This is a more extreme approach to sanding and may or may not prove itself.

14010646473_31829c548c.jpg

Here is an example of what the BTH method looks like.


*This methods are formally know as mechanical abrasion.


3. Nylon Abrasion Method. This method is an alternative to sanding and has extra-thick nylon bristles that are coated with an abrasive grit. I will be using a 538 Nylon Abrasive Brush at 120 Grit. This method is listed on Henkel's surface preparation methods guide and should prove possibly better than regular sanding.

14077769854_d1c60f3089_o.jpg

Here is an example of what the nylon abrasion tool looks like.

4. Abrasive Blasting Method. Some time back I read a paper from McDonnell Douglas (cannot find it anymore) that mentioned abrasive blasting as being the ideal preparation method. This method shoots a stream of abrasive material against a surface under high pressure. This can roughen a smooth surface, shape a surface, or remove surface contaminants. As of now I plan to use just regular media (maybe carbon) and I am also playing around with the idea of wet abrasive blasting (not to be confused with hydro-blasting). Thoughts?

14054199516_d7221619ab.jpg

Here is an example of the abrasive blasting method.

5. Atmospheric Pressure Plasma Method.
More research on this method is required before I can commit to it. I can say however that it seems very intriguing. It may however determine to be more of a surface treatment (method in addition to another).

14098195333_43f004aa41_o.jpg

Here is an example of the atmospheric pressure plasma method.

Note that all methods assume that all the dust, oil, grease, etc. has been cleaned off prior to bonding. Without the right pre/post care, the bonds will be less than optimal. I have used some of the methods aforementioned and all have worked. I am simply trying to determine which method yields first in my crude testing; I think it is valuable to know which method is ideal for a flight of this caliber.


Pull Strength:
More information coming soon.


Flutter Resistance:
More information coming soon.


Epoxy Additives:
[More information coming soon.] I will repeat the static load tests with the top performing epoxies and add various additives into them. In the past I have used graphite, milled glass, and West Systems 404. I now want to see how effective they are in strength and if they help with the Tg. I have never used fumed silica, but since it is more used to hold the shape of the fillet I am uncertain if I will give it a go this time around. I will do this test at an elevated temperature and at ambient temperature. If the price is right, I will also repeat this later with CNTs and graphene.


Elevated Temperature:
[More information coming soon.] I will only expose the top performing epoxies to elevated temperatures and monitor their performance under load. I will start the structure out at 250F and increase in 100F intervals until I reach 550F. The reason for this is to determine the general strength properties of the fillets when operating at higher temperatures similar to those temperatures one might expect to find in high Mach flights.


Epoxies Used:
[Will update soon] As of now I plan on picking from Aeropoxy ES6265, Cotronics 4525, HTR-212, ACP High temp epoxy system, Aeropoxy PT2876, Fibre Glast High Temp Epoxy Resin, and AdTech 336. Some of these mentioned epoxies are laminating systems so obviously they will have to be filled. I am also contemplating maybe using an epiglue like Nic used on his Mad Max rocket. Laminating systems once thickened can be used as adhesives, for example Aeropoxy makes a PT2876 that has some good properties and works well as an adhesive when cabosil is added. In fact an Aeropoxy engineer told me PT2876 is better than all of their other adhesives that they currently have for sale (they are converting this to an adhesive fyi ;). If anybody has any recommendations of any other epoxies then please mention it below!

I will edit this and the table below with the new epoxies, and the epoxy data once I have officially chosen all the epoxies.


Experimental Table​
14078601254_29fc4a28bb_o.png

I think this along with all the text, speaks for itself. It will be updated and more will be added as this experiment progresses.


Notes:
* I will add a voting pole on top. Basically I am looking to see which epoxies the collective thinks is best suited for the fillets of a high performance rocket.

*If there is an epoxy that you can think of that has not been mentioned then recommend it.

* As I mentioned, I do not have any thorough sensors or testing equipment, so I will analyze the results in a very crude manner. Again if any body knows of some relatively inexpensive, more technical ways of testing this, and gathering data then please let me know.


Conclusion:
Once all of the testing is complete I will update this part of the post with a summarized conclusion here. That should make it easier for some who wish to not read through the entire thread and for those who wish to download a PDF of the results.

***Disclaimer:***
Just to state the obvious I am not a chemist, material scientist, or anything related so this is not me stating the best or even right choice for anything mentioned above. It is just my current understanding on how things work and the best way they worked for me at this current moment. This may be subject to change in the future.

Cheers
Mat
 
Last edited:
I have just ordered the mock up fin materials. I am using 0.125" thick G10 FG. I plan on bonding this to a 54mm FWFG AF... however if it proves to hard to hold a cylinder stable during testing, then I will resort to bonding the fins to a G10 plate. Next up I will order the epoxies.... as soon as they are figured out.
 
Last edited:
I like your approach and how your going at this in detail.


Alexander Solis

Level 1 - Mariah 54 - CTI-I100 Red Lightning Longburn - 6,345 Feet
 
I don't remember who posted it, but there was some discussion in a thread about epoxy bonding.

If I remember right, one of the science types pointed out that the best bond is achieved by lightly wet-sanding (wet from epoxy) the surface immediately prior to bonding.

I'm just not sure which thread that was in....

-Kevin
 
I performed a lap shear test comparing epoxies many years ago, but not including Cotronics nor at elevated temperatures, so this test should be very interesting.
Epoxy Tests
 
I don't remember who posted it, but there was some discussion in a thread about epoxy bonding.

If I remember right, one of the science types pointed out that the best bond is achieved by lightly wet-sanding (wet from epoxy) the surface immediately prior to bonding.

I'm just not sure which thread that was in....

-Kevin

Yes I am aware of that. The photograph of that method (1st photo) is from flynfrog; he is the author of that article and I might add he did a brilliant job with that post. However he did not perform any test for a future flight (which is what I am trying to do). I am aware that people are down the line on the surface prep methods, so I thought I would do a couple simple test to figure out which one would work best for my application. As I mentioned I have used both methods and both methods "work," I want to know which method will yield first (at least to a static load). Some manufacturers say just sand it others say just rough it up and some say either or. Therefore it might depend on the epoxy brand; I do believe a very important part of the bonding is cleaning the surface and having no contaminates in the epoxy prior to application. McDonnell Douglas released a paper on sandblasting being the top choice. I may have to incorporate that soon enough.

Ya mean the sticky at the top of this forum?
https://www.rocketryforum.com/showthr...nt-you-to-know

Lol. Yes the sticky.... how appropriate for a epoxy thread.

I performed a lap shear test comparing epoxies many years ago, but not including Cotronics nor at elevated temperatures, so this test should be very interesting.
Epoxy Tests

Ahh so that is your website. Good job, I have looked at your site from time to time. As you mentioned I am more interested in testing the "professional," epoxies against one another. I thought about testing Aeropoxy ES6209 but chose not to because of its Tg (it is quite low). Nice to see that you have done that already, but I will try the Aeropoxy ES6525, which is a toughed up version of the ES6209 with a higher Tg. The price is comparable to Cotronics 4525 (expensive) and they do not advertise this stuff online.
 
Last edited:
Ahh so that is your website. Good job, I have looked at your site from time to time. As you mentioned I am more interested in testing the "professional," epoxies against one another. I thought about testing Aeropoxy ES6209 but chose not too because of its Tg (it is quite low). Nice to see that you have done that already, but I will try the Aeropoxy ES6525, which is a toughed up version of the ES6209 with a higher Tg. The price is comparable to Cotronics 4525 (expensive) and they do not advertise this stuff online.

I look forward to hearing about ES6525.

Have you considered adding Rocketpoxy to your test battery? It seems like it may have similar properties to ES6265.
 
I look forward to hearing about ES6525.

Have you considered adding Rocketpoxy to your test battery? It seems like it may have similar properties to ES6265.

If memory serves me right, Rocketpoxy G5000 has a Tg of 150F and a TS of 7500psi. So for that reason I opted out of it, I think it is more similar to ES6209. I am not sure how hot the fillet will get but 150F has me on edge. There will be a mixture of heating from the aerodynamics, motor casing, and sitting on the desert floor, so I would like to accommodate that. ES6265 has a Tg of 350F and a TS 8000psi-10000psi(?).
 
I don't remember who posted it, but there was some discussion in a thread about epoxy bonding.

If I remember right, one of the science types pointed out that the best bond is achieved by lightly wet-sanding (wet from epoxy) the surface immediately prior to bonding.

I'm just not sure which thread that was in....

-Kevin

I believe the wet sanding with epoxy was from a minimum diameter thread with aluminum fins mounted to a carbon airframe.
 
I believe the wet sanding with epoxy was from a minimum diameter thread with aluminum fins mounted to a carbon airframe.

They were the ones that ended up ripping the surface off of the tube.
 
The kicker would be to try to find something that is close to the temp specs of Cotronics but a little more economical with perhaps a better shelf life.
I've used it and boy is has worked out well. Can even do fillets with it but oh the price. Kurt
 
Henkel-Loctite makes a series of industrial epoxies that are widely used in industry; they used to be under the name "Hysol" or Loctite-Hysol", but I can't find them on the website anymore. I bought a small amount as part of a project in me senior year, but all of the leftover went back to the sponsor of the project. They make about a thousand different products, and I can't remember which one we had-but Bandman444 & Company (can't remember their user tags right now, sorry) were using a hysol brand epoxy when it delaminated their Performance Rocketry filament-wound tube. I'd be tempted to call up a US vendor for Henkel and ask to talk to an engineer who can help you pick the best epoxy, and see if you can buy a small sample (ours was 9430, I think? I don't remember, sorry.) They make many specialty epoxies for bonding composites at high temperatures.
 
I'd be much more interested if you were to test Proline 4500 instead of the Aeropoxy. The Proline does not have readily avalible specifications so I'd be interested in how it performs.

Alex
 
I'd be much more interested if you were to test Proline 4500 instead of the Aeropoxy. The Proline does not have readily avalible specifications so I'd be interested in how it performs.

Alex

I wouldn't say "instead" but rather "in addition".

None of the specifications apply specifically to fillets.
 
I'd be much more interested if you were to test Proline 4500 instead of the Aeropoxy. The Proline does not have readily avalible specifications so I'd be interested in how it performs.
I'd be very suspicious of an epoxy system that doesn't publish specifications. And, testing it wouldn't have much benefit because you'd have no reason to believe that such an epoxy was consistent from batch to batch. (The obvious reason not to publish specs is so that whatever is available at low cost can be sold under the brand.)
 
I'd be very suspicious of an epoxy system that doesn't publish specifications. And, testing it wouldn't have much benefit because you'd have no reason to believe that such an epoxy was consistent from batch to batch. (The obvious reason not to publish specs is so that whatever is available at low cost can be sold under the brand.)

Yes. This. Every word of this.
 
Hysol EA934NA

https://tds.us.henkel.com/NA/UT/HNAUTTDS.nsf/web/B8D891E40C3704B98525715C001BD4DD/$File/Hysol_EA_934NA-EN.pdf


Still has 1Ksi at 300F Lap Shear might be on the brittle side though.

Or its brother 9394
https://tds.us.henkel.com/NA/UT/HNAUTTDS.nsf/web/D61948C932B43DF38525715C001BD400/$File/Hysol_EA_9394-EN.pdf
gets you 1.6 Ksi at 300 1.2Ksi at 350


cant find the MD spec you are talking about but I know of it. Here is a DOD paper check out page 31 With bond prep methods comparisons.
https://www.niar.wichita.edu/niarworkshops/Portals/0/Jun17_0200_JimM.pdf
 
Last edited:
The 9394 is what I was referencing above (the one I used once). I don't know what the Project60K team used to strip filament-wound tubing.

Hysol EA934NA

https://tds.us.henkel.com/NA/UT/HNAUTTDS.nsf/web/B8D891E40C3704B98525715C001BD4DD/$File/Hysol_EA_934NA-EN.pdf


Still has 1Ksi at 300F Lap Shear might be on the brittle side though.

Or its brother 9394
https://tds.us.henkel.com/NA/UT/HNAUTTDS.nsf/web/D61948C932B43DF38525715C001BD400/$File/Hysol_EA_9394-EN.pdf
gets you 1.6 Ksi at 300 1.2Ksi at 350


cant find the MD spec you are talking about but I know of it. Here is a DOD paper check out page 31 With bond prep methods comparisons.
https://www.niar.wichita.edu/niarworkshops/Portals/0/Jun17_0200_JimM.pdf
 
The 9394 is what I was referencing above (the one I used once). I don't know what the Project60K team used to strip filament-wound tubing.

I checked my email discussions with you and it was 9430.
 
9430 is not a high temp epoxy. https://tds.us.henkel.com/NA/UT/HNAUTTDS.nsf/web/8459CF40F9F68AAD882571870000DB1D/$File/HysA9430-EN.pdf
 
9430 is not a high temp epoxy. https://tds.us.henkel.com/NA/UT/HNAUTTDS.nsf/web/8459CF40F9F68AAD882571870000DB1D/$File/HysA9430-EN.pdf

We never said it was. In addition to gluing the fins down with it, they coated the outside of their airframe with it and it melted smooth from aero heating.
 
9430 is not a high temp epoxy. https://tds.us.henkel.com/NA/UT/HNAUTTDS.nsf/web/8459CF40F9F68AAD882571870000DB1D/$File/HysA9430-EN.pdf

I implied that it was, my bad. The one I used was somewhat high-temp; i think it was the 350F 9394.
 
I'd be very suspicious of an epoxy system that doesn't publish specifications. And, testing it wouldn't have much benefit because you'd have no reason to believe that such an epoxy was consistent from batch to batch. (The obvious reason not to publish specs is so that whatever is available at low cost can be sold under the brand.)

Yes. This. Every word of this.

+1!
 
Hysol EA934NA

https://tds.us.henkel.com/NA/UT/HNAUTTDS.nsf/web/B8D891E40C3704B98525715C001BD4DD/$File/Hysol_EA_934NA-EN.pdf


Still has 1Ksi at 300F Lap Shear might be on the brittle side though.

Or its brother 9394
https://tds.us.henkel.com/NA/UT/HNAUTTDS.nsf/web/D61948C932B43DF38525715C001BD400/$File/Hysol_EA_9394-EN.pdf
gets you 1.6 Ksi at 300 1.2Ksi at 350


cant find the MD spec you are talking about but I know of it. Here is a DOD paper check out page 31 With bond prep methods comparisons.
https://www.niar.wichita.edu/niarworkshops/Portals/0/Jun17_0200_JimM.pdf

I am glad to see that you found your way over to this thread! If you have any input feel free to let it out. Thank you very much for the docs and links. I am reading through it now.

Mat


The 9394 is what I was referencing above (the one I used once). I don't know what the Project60K team used to strip filament-wound tubing.

I implied that it was, my bad. The one I used was somewhat high-temp; i think it was the 350F 9394.

I will still check it out and will also call them tomorrow to see if they have any advice or secret products like Aeropoxy!
 
Last edited:
Loctite Hysol 9430, wet sanded the fins with the epoxy to get a strong, non-oxide bond, left the aluminum chips in the matrix to give a small lift from the CF, and did normal prep (sanding, wipe with acetone) to the CF. Applied more epoxy to the airframe, laid the fins on, and cured it at ~180F.

We found that the tensile strength of the epoxy beat out everything else.

Bryce (Bandman444) did a test between different aluminum pieces using the hysol, some sanded, some not, some partial, and found that they were all incredibly well attached, regardless of the oxide layer. Feel free to message him with questions on that, as I didn't help him with that testing.
 
9430 is a beast as long as it doesn't have to deal with heat. I think I might have some in the freezer.

From the data sheet
Code:
Shear Strength, psi, ASTM D 1002 Etched Aluminum
Cure Schedule
 Test Temp oF   Typical Value
5 Days @ 77o
F -67                  5000
77                      4700
140                    1700
180                    750
 
EDIT:
I decided to approach this in a slightly more thorough way. I have just updated some of the methods, tests, pictures, literature, layouts, etc. It is different now… so yes reread it. Thanks.


UPDATE:
So I have the fins about ready for the different surface preparation methods. That translates into this thread getting close to starting experiment 1… Well there are still many things that I have to do before starting. For example I need to pick the epoxies, setup up the actual test platforms, and I am sure there are other things.

14075278041_0a83e1c49d_z.jpg




Mat
 
I would simplify your test. Instead of actual fin shape I would do something more repeatable. a 1"x1" tab makes the math easy. What are you hoping to learn that cant be found on data sheets?
 
Back
Top