3D Printed 50mm F-104 Starfighter

[telnar1236]

The canopy is molded. I ended up needing to split it into two parts to get it onto my vacuum forming machine, but it was an issue with me forgetting how small my machine is, not one inherent to the mold. I didn't do the best job molding the canopy but after cutting the mold in half with a hack saw to get it to fit, it's good enough for me.

 

[telnar1236]

The covering job is mostly done, and the plane is now able to stand on its landing gear. I don't plan on fully finishing the plane's scheme before doing a test flight, but since I am using aluminum foil tape (the thinnest I could find and then sanded to remove additional material) which adds about 3% to the final flying weight, I wanted to have it before the maiden. A fair bit of sanding remains to do, but currently the weight of the covering is 20 grams. The covering on the wings and tail also adds some strength, so I included it as well. In the picture of the gear, the patch where the fuselage is not covered in aluminum is visible, which, when combined with uncovered nose cone and canopy, should let in the radio signal (I spent a couple hours messing around with a test receiver to make sure I wasn't blocking out the signal). I already planned to put a skid plate in the open spot on the fuselage in case of a gear up landing, so it's a simple job to paint it to match the aluminum.

And a picture of my range test apparatus. Of everything I have done related to this hobby, slowly walking away from an aluminum foil coverd box that is waving a piece of foam board at me has to have gotten me some of the strangest looks.

 

[telnar1236]

The landing gear is now able to extend and retract. All three gear operate off one servo. Unfortunately, my projected AUW is now close to 750 grams, which should still let it fly but is heavier than I wanted (the good news is that the aluminum is less than 3% of the AUW). I've ordered a couple of higher C batteries for it to hopefully help squeeze out every bit of thrust I can, but I may end up stripping off the aluminum anyway depending on how the first flight goes.

 

[telnar1236]

I took the F-104 out with me to the flying field this morning to do some taxi tests on the runway. Unfortunately, one of the steering lines to the nose gear was not tied tightly enough and came loose resulting in the plane veering off the runway into the grass and snapping the landing gear. I'm going to try and print it in a better orientation for the replacements since the reason for the original orientation was bed adhesion with ABS and I have given up on that material and switched to PLA. The failure point was at the same place as the ABS gear, so if this happens again, I know where to beef them up.

The positive notes of the test were that I got the plane up to 21 mph briefly before the failure, and that until the failure the nose gear steering was very effective and controllable and the suspension on the main gear worked great.

 

[Inq]

I'm going to try and print it in a better orientation for the replacements since the reason for the original orientation was bed adhesion with ABS and I have given up on that material and switched to PLA.

I wouldn't expect PLA to be any better. But I should say I'm biased toward ABS as it is my main go-to... 81 kilometers of the stuff so far.

https://all3dp.com/2/pla-vs-abs-filament-3d-printing/
In light of this, MatterHackers performed a test where they printed the same hook in different materials and recorded how much weight each could handle. In their findings, ABS held a minimum of 209 pounds and a maximum of 284 pounds. This is pretty strong and much greater than PLA’s range of between 119 and 184 pounds. Overall, ABS is a good bit stronger than PLA, which somewhat justifies how difficult it is to print.

PLA is stiffer than ABS, but more brittle, so less able to absorb impact. Unless you're the best pilot in the world and grease every landing, you might consider Polycarbonate... or even Carbon Fiber / Polycarbonate. Ever try to break a CD or DVD? They're claiming near Aluminum properties. I've been experimenting with a roll of Carbon/PC and it prints very nicely, but I have not verified their strength and stiffness claims.

But for all materials all primary loads need to be in the filament extrusion bed plane. Z is nearly worthless in most materials except maybe Nylon and TPU.

 

[telnar1236]

I wouldn't expect PLA to be any better. But I should say I'm biased toward ABS as it is my main go-to... 81 kilometers of the stuff so far.

https://all3dp.com/2/pla-vs-abs-filament-3d-printing/
In light of this, MatterHackers performed a test where they printed the same hook in different materials and recorded how much weight each could handle. In their findings, ABS held a minimum of 209 pounds and a maximum of 284 pounds. This is pretty strong and much greater than PLA’s range of between 119 and 184 pounds. Overall, ABS is a good bit stronger than PLA, which somewhat justifies how difficult it is to print.

PLA is stiffer than ABS, but more brittle, so less able to absorb impact. Unless you're the best pilot in the world and grease every landing, you might consider Polycarbonate... or even Carbon Fiber / Polycarbonate. Ever try to break a CD or DVD? They're claiming near Aluminum properties. I've been experimenting with a roll of Carbon/PC and it prints very nicely, but I have not verified their strength and stiffness claims.

But for all materials all primary loads need to be in the filament extrusion bed plane. Z is nearly worthless in most materials except maybe Nylon and TPU.

I've had good luck with PLA landing gear in the past. I figure that if it worked with my 3500 gram 80mm F-104, it should work for the smaller 750 gram 50mm F-104. While it is more brittle than ABS, it is still a plastic and so not that brittle and it has much better layer adhesion which was the point of failure for both the nose gear and the main gear. PLA is also somewhat stronger in terms of both yield and ultimate tensile strength when 3D printed. The reason I wanted to go with ABS is its better wear resistance since the landing gear on my big F-104 has had problems with the wheels wearing into the struts. The big issue with this gear was the hinge on the main gear which is pretty thin, so if I can't make it work with the new print orientation, then I'll just print rigid gear without the suspension (the 80mm F-104 had those and it's been flying off them since April without any issues apart from wear caused by the wheels). One possibility is just printing the lowest section of the gear in ABS since I haven't had issues with that part to improve wear characteristics.

In terms of lining up the stresses, the problem is that the load is in multiple axes so you can't really get it so the layer lines are not loaded. The gear failed in bending backwards at the hinge which is not a highly loaded direction, and it probably would have been fine if it had not gone off runway. The new orientation is slightly weaker with the plane coming down vertically on the gear but should be much better with that bending axis backwards where the hinge failed.

 

[telnar1236]

Ok, it's finally time. The F-104 is ready for its maiden flight. There are a few minor things to finish up after the test (gear doors and some aesthetic things like painting) but everything is working, and it passed its new taxi test with flying colors. The new PLA gear struts printed in the new orientation are much stronger and also more rigid resulting in a slightly better stance while on the ground and tighter steering (I attribute this to the tighter tolerances I was able to achieve this time around). Weather and time permitting, I hope to attempt to fly it sometime tomorrow or the following day so hopefully I'll update with results once it has flown.

 

[telnar1236]

It flies! I had a mostly successful maiden flight this morning. Other than being a bit underpowered, it flies very nicely. It is able to slow down extremely well, but unfortunately it won't be breaking any speed records. It's a lot more agile than I expected and can actually turn pretty tightly, and it doesn't even bleed speed very badly in the turn. The landing gear servo was not strong enough to lift the gear against the force of the air, so I will be replacing it with a stronger one. With the flaps all the way deflected, the EDF is barely strong enough to keep it in the air on a 3s 1500 mAh pack, and even without the flaps I was full throttle the entire flight. On landing, I got a bit too slow, and the plane pitched up (it's an F-104 after all) and it came down from maybe 1.5 ft on its tail, but the only damage was it breaking the nose gear out of the retract and breaking the elevator bell crank, both of which I can fix pretty easily.

To fix the issue with it being underpowered, I am going to try flying it on a 4s 1300 mAh pack next time I get it out to the field. I did some bench testing, and the EDF should be fine with the 4s pack (it is rated to draw 38 A on 3s, but only draw about 25 A, even from a very large capacity pack with a high discharge rating, and it only draws 40 A max on 4s; I think they might have messed up and sent me the wrong EDF since it pretty closely matches the rated specs of the 4s EDF). Otherwise, now that I know it flies, I am going to go ahead and finish up the few remaining items and then paint it. One additional change I am going to make is to cut off the bottom of the fuselage where the skid plate will go and screw the skid plate in place so that I have an access hatch in case I need to change out servos or mess with the control linkages.

 

[telnar1236]

I flew the F-104 again and the power from the 4s pack definitely helps a lot, both with thrust to weight ratio and speed. It's still not a rocket ship, but it actually feels like an F-104 now. I was able to actually get it on the runway this time too, which was nice. The flight times are a bit short on the 1300 mAh pack, though so next time it's up to a 1500 mAh pack.

It was also pointed out to me that I never posted the file for the rudder, so here it is, along with the update control horn, fairing, a little piece to replace the rudder hinge point built into the fuselage/vertical stabilizer in case you break it off like I did, and finally the 1:14.7 scale sidewinders for the wingtips.

 

[telnar1236]

I finally got around to starting to paint the F-104. The splash of contrast definitely helps offset the aluminum.

Obviously there is some painting left to do, but I think it's coming together.

A more substantive change was a redesign of the elevator bell crank 2. The original method of torque transmission was a glue joint to the carbon fiber rod in the elevator. However, that broke after the first hard landing and then again in normal use. In retrospect, it was a bad design. The new bell crank also connects to the fairing at the top of the elevator making it much stronger. The repeated breakage and repairs meant that I lost some material from around the pivot, so I also installed two small plywood reinforcements.

 

[telnar1236]

It's all painted up now. The paint on the horizontal stabilizer didn't really match the color of the aluminum on the fuselage, so I probably need to repaint it, but otherwise I'm happy with how it is turning out.

 

[gk4rpv]

Congratulations on an amazing accomplishment! I'm sure everyone would love to see flight video!

 

[telnar1236]

Congratulations on an amazing accomplishment! I'm sure everyone would love to see flight video!

Unfortunately it's probably not in the cards. It crashed a while back due to signal loss (probably because of the aluminum foil). With how time intensive it was to build, I'm probably not going to do so and instead will stick with some of my newer projects.