Flight Club - Ultralight Build Log

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

Gregory Perkins

Active Member
Joined
May 25, 2019
Messages
39
1. .030 gussets, 1" x .058 tubes
2. 6061-T6
3. Steel pop rivets from Fastenal
Without doing any tests to destruction of your joints, my initial knee jerk reaction is that you should be using gussets that are at least as thick as your tubing and maybe get a stronger alloy (2024 maybe) for the gussets. The reason is that there is no doubt about a strength increase of the gusset for a negligible amount of extra weight. Also, if you dont use stainless steel rivets, you will regret it just a few years down the road because of rust.
 

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
4,565
Location
NJ
post 179 shows.....as regards your design problem you will need one gusset for the diagonals and one for the rest of the tubes

Analysis of a truss is not easy for the non expert....here is a nice little FEA I use.....to get the Moments at the joints with that you can get a handle on how much strength you need in the joint
this FEA will not model your gussets, just the tubes....you will have to know the constraints (where you grab the model)(where the rear fuselage attaches to the wing) and loads (use Glider Criteria Limit Loads from charts of the tail area)

 
Last edited:

radfordc

Well-Known Member
Joined
Feb 5, 2008
Messages
1,390
Without doing any tests to destruction of your joints, my initial knee jerk reaction is that you should be using gussets that are at least as thick as your tubing and maybe get a stronger alloy (2024 maybe) for the gussets. The reason is that there is no doubt about a strength increase of the gusset for a negligible amount of extra weight. Also, if you dont use stainless steel rivets, you will regret it just a few years down the road because of rust.
Yeah, I'm sure it's a lousy way to build an airplane. After all, Robert Baslee has only been doing this for 30 years....probably hasn't got it figured out yet.
 

Gregory Perkins

Active Member
Joined
May 25, 2019
Messages
39
So Radford, are you saying it is not worth the extra quarter pound to upgrade the gussets to be a little thicker and of stronger alloy ? Or that it is not worth it to keep your rivets from rusting out in a couple of years. Since you apparently are an expert on Baslee's methods, please tell us what thickness and alloy his gussets are as relates to the wall thickness of his tubes?. I'd like to know what models of UL that you own and for how long.
 

Bill-Higdon

Well-Known Member
Joined
Feb 6, 2011
Messages
631
Location
Salem, Oregon, USA
Howdy Home Built Airplane Enthusiasts,

First and foremost, thank you all so much for helping out with our project. At the moment, we have a few questions regarding how to best join the tubes on our plane. Here’s some context to our questions:

Based on all of your prior suggestions and some of our initial research, we opted to design a gusseted aluminum truss frame. Over the course of our gusset endeavor, however, we have run into several issues regarding how to come up with appropriate solutions for certain joints, especially the joints on our empennage. A couple months ago, we thought we solved all our gusset problems, but we’ve recently encountered some more, which we can’t seem to find solutions to.

We are struggling with several gussets on our empennage, and below is an example of our upper empennage gusset, it utilizes our double gusset technique, but the rivets are dangerously close to the edge of the gussets on the inside (violating the 2*diameter tearout rule). Since the rivets must be installed normal to the gusset and must be tangent to the circular beam, we are unable to move them further inward. We were originally planning to just use a single gusset on the outside of the tubes like the popular Nieuport replicas but some of our members who participate in FIRST robotics think a single gusset won’t be rigid enough.




Something else to mention is that we attempted analyzing our fuselage truss and were unable to figure out the complex math due to use having many indeterminate structures. Five members of our team watched and took notes on the first 50-60 statics videos from Jeff Hanson, which were super helpful but still weren’t enough to be able to fully solve our frame. We did, however, manage to determine the stress in the most critical beams (eg: empennage longerons, wing spars, etc…). Since we don’t know how to determine the exact stresses in every beam, we have been using many arbitrary dimensions on the gussets we are designing. We don’t know if we need single gussets, double gussets, or how much of each gusset needs to be in contact with the tubing. We also don’t know if filleting our gussets makes sense to save weight or if it will compromise structural integrity. Our gusset design decisions have been based on similar ultralight airplanes that use gussets and on the few stresses that we are certain we know. We want to be certain each of the gussets we are designing meets its necessary strength requirement.

An alternative to a gusseted circular aluminum truss frame would be a welded 4130 steel truss frame like the Legal Eagle uses. We’ve created a comparison between the two, but feel free to let us know if we left anything out:

Gusseted 6061 T6 AluminumWelded 4130 Steel
More home shop friendly in the sense that it doesn’t require a TIG welder or a level welding table. Some people also have an unusually strong aversion to welding for some reason...MUCH quicker to construct (probably about a full week based on what we’ve heard from similar designs)
Potentially lighter since 6061 T6 has a higher strength/weight ratio of 410,260 lbf*in/lb, while 4130 Steel is only 222,180 lbf*in/lbSignificantly cheaper.
If done properly, it will likely be stronger since the 1x0.035” aluminum will have a higher column buckling strength than the ⅝x0.035” 4130 steel. The gussets may also create a more rigid frame. Additionally, we eliminate the risk of all welding related issues, which may compromise the strength of any joint.Significantly simpler. By using a welded frame, we’d remove about a hundred gussets and thousands of rivets from the BOM.
Eliminates complex coping (requires expensive tools to be done precisely).If we had a welded frame, we could easily weld on more safety parts, like tangs, gussets, or other attachment material.
Removes the need to use arbitrary dimensions for the guests and rivets.

We have a professional welding mentor who has offered to help us weld the frame but the average home builder does not.

All that said, what are your thoughts on our current gusset solution? Can anyone think of a better one? Does anyone have thoughts on a single vs. double gusset design? Should we switch from circular aluminum tubing to square aluminum tubing to make putting gussets on easier? Or should we change our truss frame material to 4130 steel so we can weld it more easily?

Thanks,
Flight Club Design and Physics Teams
I'll try and get you a few pictures this afternoon of 1 solution to your gusset issue
 

Gregory Perkins

Active Member
Joined
May 25, 2019
Messages
39
Just a thought/suggestion regarding your gusset design... If you take both outside gussets and connect them together so in effect to be one solid piece that you wrap around the top tube and then rivet to the down tubes... You can then pull back/trim
the piece of the gusset that is most forward and most rearward on the top tube that is interfering with the inner gussets so that they can be designed to reach further around the top tube without interference and thus solve your problem of rivet to edge minimums not being achieved. The reason why you can trim back the top overlapping gusset is now you have captured the top longerone tube with the wrap around gusset and the extra area is no longer needed for the application of rivets for attachment. Plus, this reduces the total number of holes drilled into the top longeron tube which causes some weakening of the tube even though it may not be in the location of the weakest link to failure. I hope you can understand this. In other words... the two outermost gussets are now one piece and are bent with a same radius as is the top tube and they are placed on the top tube like a hat. After you do that, then trim the outside gusset as necessary to allow the inner gussets to be extended further onto the top tube for quality riveting.
 

Ollie Krause

Well-Known Member
Joined
Feb 26, 2020
Messages
96
Without doing any tests to destruction of your joints, my initial knee jerk reaction is that you should be using gussets that are at least as thick as your tubing and maybe get a stronger alloy (2024 maybe) for the gussets. The reason is that there is no doubt about a strength increase of the gusset for a negligible amount of extra weight. Also, if you dont use stainless steel rivets, you will regret it just a few years down the road because of rust.
We were actually thinking of using the CHERRYMAX CR3213-6-2 3/16 diameter rivet for our structural joints which is made of 5056 aluminum and has a rated shear strength of 1480lb. The issue with using stainless steel rivets is galvanic corrosion and we didn't want to have to go through the process of normalizing all our hardware or spraying it down with that light oil. 2024 T3 is a good idea for the gussets and we'll definitely look into it if our budget permits. All our rivet comparisons and calculations can be found in this document if anyone might find them useful.
 

maya.ayoub.32

Active Member
Joined
Apr 6, 2020
Messages
32
post 179 shows.....as regards your design problem you will need one gusset for the diagonals and one for the rest of the tubes

Analysis of a truss is not easy for the non expert....here is a nice little FEA I use.....to get the Moments at the joints with that you can get a handle on how much strength you need in the joint
this FEA will not model your gussets, just the tubes....you will have to know the constraints (where you grab the model)(where the rear fuselage attaches to the wing) and loads (use Glider Criteria Limit Loads from charts of the tail area)

This is incredible! We've been using Simscale (Fusion more recently) for simulation, but all our truss analysis work has been done by hand. This is will definitely help us out since, although our CFD member has poured hundreds of hours into doing these simulations, the softwares we've been using seem to be spotty at best for the accuracy we want it to produce. We really appreciate the website link!
 

Ollie Krause

Well-Known Member
Joined
Feb 26, 2020
Messages
96
Just a thought/suggestion regarding your gusset design... If you take both outside gussets and connect them together so in effect to be one solid piece that you wrap around the top tube and then rivet to the down tubes... You can then pull back/trim
the piece of the gusset that is most forward and most rearward on the top tube that is interfering with the inner gussets so that they can be designed to reach further around the top tube without interference and thus solve your problem of rivet to edge minimums not being achieved. The reason why you can trim back the top overlapping gusset is now you have captured the top longerone tube with the wrap around gusset and the extra area is no longer needed for the application of rivets for attachment. Plus, this reduces the total number of holes drilled into the top longeron tube which causes some weakening of the tube even though it may not be in the location of the weakest link to failure. I hope you can understand this. In other words... the two outermost gussets are now one piece and are bent with a same radius as is the top tube and they are placed on the top tube like a hat. After you do that, then trim the outside gusset as necessary to allow the inner gussets to be extended further onto the top tube for quality riveting.
Our initial gusset design actually includes the wrap around style gusset but we've since switched over to the separate gusset plate design due to potential bending and fracturing issues with a rigid alloy like 6061 T6. Gauging by the thickness of the gussets y'all seem to be using in your ultralights, this may not be much of an issue and we might be able to stick with the wrap around gusset. Something that's also nice about the separate gussets is that they are a little more forgiving in the assembly process because they can each be attached at different points in time without obstructing the installation of other beams. That being said, the single line of rivets for a bent gusset is definitely appealing structurally and economically.

I'm still a little confused on how wrapping the outer gusset around would make room for the inner gusset rivets. Do you have an airplane that uses this technique that you could refer us to? Thanks for all the help and advice!
 

maya.ayoub.32

Active Member
Joined
Apr 6, 2020
Messages
32
posted for educational purposes only.
spread sheet is mine please load test or have another "expert" check....or build your own spread sheet from the example....I have too many load test failures to be anything else but a poser....
Thank you! We've been trying to create our own bearing stress calculator which we've been using to decide thickness of our gussets, rivet diameter, etc., but yours seems a lot more advanced and thorough. I will definitely check it out and reevaluate our joints!
 

Gregory Perkins

Active Member
Joined
May 25, 2019
Messages
39
QUOTE " I'm still a little confused on how wrapping the outer gusset around would make room for the inner gusset rivets. "
REPLY- Just for rough explanation .... Assume your existing outside gussets at the top measure 7 inches front to back. I am suggesting that when you connect the two gussets and bend with a radius same as the top tube,..... you reduce the front to back dimension of the connected gussets to 3 inches. Do this by cutting away the material in the gusset that interferes with allowing the inside gussets to extend well past the top tube contact area where the rivets would be placed. RE the 6061 was generally considered to be very malleable in the UL era and many times gussets and connectors were bent with pretty sharp 90 degree angles and there was never a problem that I heard of with cracking. Your application with relatively large radius bend at the very top of the outside gussets should definitely not be a problem. Same with 2024. Have you had an opportunity to examine any of the older original tube based UL models up close to study assembly techniques and design? Which models ? There were some tube based models that had pretty accurate drawings of specific components.... Quicksilver of course, Weedhopper, Hummer, Buccaneer, B1rd, Kolb, Nieuport, Gypsy , DSK , Jensen VJ-24 and VJ-11, DeltaBird, Easy Riser, Invader, Lazaire, Whittaker MW-7, SR1 Hornet, etc.
 

Gregory Perkins

Active Member
Joined
May 25, 2019
Messages
39
We were actually thinking of using the CHERRYMAX CR3213-6-2 3/16 diameter rivet for our structural joints which is made of 5056 aluminum and has a rated shear strength of 1480lb. The issue with using stainless steel rivets is galvanic corrosion and we didn't want to have to go through the process of normalizing all our hardware

Yes of course aluminum rivets OK. I was just thinking about the horrible situation with regular steel rivets. Surprisingly in some of my 30 yr old planes with stainless steel rivets, galvanic corrosion has not been evident or a problem but I agree with you. A more humid environment or hanging around coastal salty air could be catastrophic.
 

maya.ayoub.32

Active Member
Joined
Apr 6, 2020
Messages
32
QUOTE " I'm still a little confused on how wrapping the outer gusset around would make room for the inner gusset rivets. "
REPLY- Just for rough explanation .... Assume your existing outside gussets at the top measure 7 inches front to back. I am suggesting that when you connect the two gussets and bend with a radius same as the top tube,..... you reduce the front to back dimension of the connected gussets to 3 inches. "
Gotcha! I get what you're saying! We're currently prototyping some joints, so we'll grab some 5056 and test this one out as well. Thank you!

QUOTE "Have you had an opportunity to examine any of the older original tube based UL models up close to study assembly techniques and design? Which models ? There were some tube based models that had pretty accurate drawings of specific components.... Quicksilver of course, Weedhopper, Hummer, Buccaneer, B1rd, Kolb, Nieuport, Gypsy , DSK , Jensen VJ-24 and VJ-11, DeltaBird, Easy Riser, Invader, Lazaire, Whittaker MW-7, SR1 Hornet, etc. "
Because we started this project almost exactly when corona forced places to shut down, we haven't been able to visit any ultralights in person. Fortunately, our local EAA Chapter 20 is allowing us to view their Sirius construction where we hope to gain valuable information on welding, rivets, and generally get an idea on the typical must-dos in an airplane construction project. We have been researching and examining build logs of some of these ultralights, especially the Nieuport, and we did some strength calculations on their gussets to give ourselves a reference frame. However, there are definitely some we haven't heard of before in your list- thank you!
 

Dana

Super Moderator
Staff member
Joined
Apr 3, 2007
Messages
9,259
Location
CT, USA
Unless a joint is very highly stressed, you probably don't need to go to the Cherrymax rivets, which are expensive and require their own special tool. You can get structural rivets, which are somewhere between cherries and ordinary pop rivets, I used "Klicklock" aluminum structural rivets, which have a retained mandrel but pull with a standard pop rivet tool, on the paramotor I built some years ago. IIRC the rivet was 5056 but the mandrel was 7075.
 

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
4,565
Location
NJ
Thank you! We've been trying to create our own bearing stress calculator which we've been using to decide thickness of our gussets, rivet diameter, etc., but yours seems a lot more advanced and thorough. I will definitely check it out and reevaluate our joints!
I model the rivet locations 2d in autocad convert holes to regions then union all the holes .
Do mass properties to find the centroid.....move the centroid to 0,0,0 and coordinate dimension the holes.

Looking at load each rivet and bearing strength each thickness and shear each rivet.
 
Last edited:

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
4,565
Location
NJ
the softwares we've been using seem to be spotty at best for the accuracy we want it to produce. We really appreciate the website link!
see if you can find worked problems in say a Schaums or other statics text...and practice with them.....I found it very good....however it does not do torque well except for round parts (or I can not figure out how to do it)
 

radfordc

Well-Known Member
Joined
Feb 5, 2008
Messages
1,390
So Radford, are you saying it is not worth the extra quarter pound to upgrade the gussets to be a little thicker and of stronger alloy ? Or that it is not worth it to keep your rivets from rusting out in a couple of years. Since you apparently are an expert on Baslee's methods, please tell us what thickness and alloy his gussets are as relates to the wall thickness of his tubes?. I'd like to know what models of UL that you own and for how long.
I'm saying that if a design is sufficient for the purpose adding weight or expense is absolutely the wrong thing to do.

My aircraft:
Quicksilver MX 3 years
Airbike 5 years
CGS Hawk 5 years
Sonex 8 years
Eindecker replica 4 years (no rusty rivets yet)
 
Top