2026 Speed Challenge - Telnar1236

telnar1236

Master member
That thunderbird theme is awesome too!
My biggest concern is how poorly silver paint hides damage - with a 3D printed airplane the occasional crack in the skin from handling it or bumping it is inevitable and while it's easy to repair them with super glue, the super glue ruins gloss silver spray paint where with white paint it's much less obvious. This is an example of a super glue repair on some silver spray paint, and it makes it this ugly gray instead
1784490582928.png

I like the bare aluminum paint schemes better - they might just not be practical
 

telnar1236

Master member
The Super Duper Sabre is officially ready for its maiden flight! I got it ready yesterday evening, but when I woke up this morning, the wind was gusting up to 17 mph and despite hoping for it to get calmer all day, it has just gotten worse, so I can't fly it this weekend sadly.
1784490708181.png

1784490735722.png

1784490804495.png

I got it painted in its first layer of primer so I'm not trying to fly a matte black airplane (from experience it's almost impossible to maintain orientation when your plane is solid black) and installed the remaining control linkages and the rudder.
1784491079930.png


Unfortunately, it doesn't quite balance on a 4000 mAh pack, so I needed to use a 5000 mAh battery. With this all the way forwards flying weight is 2,833 g though that might go down a few grams as the primer fully dries. Since the end goal was always to fly it on a 9,000 mAh Li-Ion battery at 2900 g, for which the projection hasn't changed, I'm ok with flying it at 2,800 g. The landing speed on paper is still a couple mph slower than my 90mm F-106 and only a few mph faster than the Ghost, so it should be completely manageable. And of course, I expect thrust in flight at slower speeds to be in the ballpark of 3 kg so it won't be lacking for power.

The only remaining design change was a set of small fairings for the stabilator control horns. With my original design, when I gave the plane full down elevator, the cover flapped up off the side of the plane by about an eighth of an inch, so it needed just that little bit more space.
1784491570914.png

However, with the combination of ball links, removal of the rubber grommets from the stabilator servos, and additional thickness for the stabilator control horns, the linkages are now very stiff and feel quite reliable.
 

telnar1236

Master member
The Super Duper Sabre flies! I flew it for the first time today with mixed results. The short version is that it flies beautifully and will be pretty near perfect with a bit more tuning in the air. However, it inherited all the landing gear problems from the Ghost and 2 of 3 struts collapsed on a landing that was admittedly harder than it should have been. Fortunately, the required repairs are pretty minor. Here's a picture post maiden flight:
1784678663209.png

In the air, the Super Duper Sabre is surprisingly docile for being such a heavy jet with such small wings. Roll control is good at all speeds with the combined ailerons and elevon stabilators. I need to turn up the elevator rates a bit, but overall, it responded well to what I told it to do. It also slows down shockingly well with flaps - it's still a heavy plane but with the flaps down it slows down as well as many lighter EDFs. There was a weird uncommanded left roll right after takeoff, but I think it must have just been wind or me not letting go of the rudder quickly enough since it went away very quickly in flight. This is really the Reynolds number (size and speed of airplane) I designed this airfoil for, in this plane, and it shows in being much less stall-prone than any of my previous laminar flow wing designs. It's been interesting flying these more recent designs derived from my experimental planes thread since none of them (the Ghost, Rapier and now Super Duper Sabre) really feel like they're flying fast even when they clearly are. I'm guessing I got it ballpark 120-130 mph on this maiden flight with no gear doors and no attempt to optimize passes and only maybe 10 seconds of full throttle in the full flight, but that's really just based on going, wow, it's getting small quick a couple of times. Control response felt the same with full flaps at landing speed as it did at full throttle.

I'm also really very happy with the CFD trim predictions. The plane needed next to no elevator trim and had no trim change with the flaps thanks to the (fairly aggressive) elevator mix I had preprogrammed on the ground. I do want more elevator, but that's a matter of preference - it's hitting the angles of attack I predicted, I just want more ability to get it to higher angles of attack if I want to. My original tune was meant to avoid pitch up but with the 50mm version, I proved that was pretty manageable if you gave it more throw so now I want to be able to maneuver a bit more aggressively.

However, the landing gear and surrounding structure are a problem. My landing must have been harder than I thought since both wings cracked just outboard of the gear struts and one of the main struts broke off right at the retract unit. The nose gear also pulled a new trick and punched its way through the bottom of the inlet duct although the strut itself survived.

To fix this, I plan to add carbon fiber rods running through the section of the fuselage around the nose gear and the full span of the wings. I also need to look at redesigning the main gear struts, maybe with a trailing link design. After all the pain with these gear struts in the Ghost, I intentionally designed in extra space in case the gear needed to get a bit bigger in this one. Fortunately, the parts that broke are the easiest ones to fix - the wings are only a few parts each, and the nose fuselage is grand total of 3 parts so I can reprint them pretty fast. And everything unbolts so I can just remove the wings with 3 screws each and the nose fuselage with 8 screws and not need to risk damaging anything else.
 

telnar1236

Master member
Got the new v2 wings designed. They have two 6mm carbon fiber spars, one running most of the length of the leading edge and one running the full length of the wing near the trailing edge.
1784758043524.png

The nose gear mounts are also now resting on two carbon fiber spars which are much better tied in with the fuselage structure to hopefully prevent the nose gear from punching through the wing.
1784758158431.png

And finally, I'm experimenting with using lateral honeycomb infill with some settings changed instead of gyroid infill for the wings. It seems to be equally good or better at resisting warping but much stronger in the directions the wing is loaded. It's also somewhat lighter which helps make up for the weight of the added spars.
1784758392892.png
 

Mr Man

Mr SPEED!
Got the new v2 wings designed. They have two 6mm carbon fiber spars, one running most of the length of the leading edge and one running the full length of the wing near the trailing edge.
View attachment 259464
The nose gear mounts are also now resting on two carbon fiber spars which are much better tied in with the fuselage structure to hopefully prevent the nose gear from punching through the wing.
View attachment 259465
And finally, I'm experimenting with using lateral honeycomb infill with some settings changed instead of gyroid infill for the wings. It seems to be equally good or better at resisting warping but much stronger in the directions the wing is loaded. It's also somewhat lighter which helps make up for the weight of the added spars.
View attachment 259468
How does a spar one each edge (like what you did) compare to 2 in the center?
 

telnar1236

Master member
Not 100% sure - I put my new spars where they were because there wasn't space for them anywhere else without a more substantial redesign of the wing. Guessing the spars further out probably increase the torsional stiffness of the wing for a given spar size, but mean that the 3D printed structure has to work harder to transmit the load from the landing gear than if the spars were right next to it. However, if you can put the spar somewhere, it's generally better to put in near the thickest part of the wing, since a spar with a larger diameter can be lighter to carry the same spanwise bending load, or can be much stronger for only a bit more weight. An I beam would really be a more efficient way to do this than the tube spars we typically use, but it would also be a lot more expensive and carbon fiber spars are already very strong and light.
1784764046878.png


There are also two more spars in the wing that have been there from the beginning so it's really more of a 4 spar wing at this point - the two new ones are meant to stop the wings shearing off on hard landings while the two original ones in red are what carry the majority of the in flight loads.
1784763803972.png
 

Mr Man

Mr SPEED!
Not 100% sure - I put my new spars where they were because there wasn't space for them anywhere else without a more substantial redesign of the wing. Guessing the spars further out probably increase the torsional stiffness of the wing for a given spar size, but mean that the 3D printed structure has to work harder to transmit the load from the landing gear than if the spars were right next to it. However, if you can put the spar somewhere, it's generally better to put in near the thickest part of the wing, since a spar with a larger diameter can be lighter to carry the same spanwise bending load, or can be much stronger for only a bit more weight. An I beam would really be a more efficient way to do this than the tube spars we typically use, but it would also be a lot more expensive and carbon fiber spars are already very strong and light.
View attachment 259478

There are also two more spars in the wing that have been there from the beginning so it's really more of a 4 spar wing at this point - the two new ones are meant to stop the wings shearing off on hard landings while the two original ones in red are what carry the majority of the in flight loads.
View attachment 259477
Ah ok, thanks 👍
 

telnar1236

Master member
Got the new landing gear designed. To try and make it stronger, the struts themselves are now made with a carbon fiber tube and a carbon fiber rod. I'm using two scissor links to hopefully hold the wheels straight. There wasn't quite space to fit a trailing link design so it's still a standard compression strut.
1785000865432.png
1785000833925.png

Depending on how torsionally stiff the struts are, I may or may not install the rear scissor link since it almost interferes with the wing spar when the wheels are retracted. The wheels are also moved aft by 5mm to hopefully make the airplane handle a tiny bit better on the ground - it was totally manageable but the nose was a bit bouncy with the original CG and gear placement.
1785001112965.png
1785001160160.png
 

telnar1236

Master member
Another day, another set of landing gear problems. The good news is that the carbon fiber landing gear struts work perfectly - the scissor link design locks them in the right orientation far better than my key design ever did and they are very strong. I'm going to be replacing the nose gear strut with one of these too since it's a much more compact layout and should let me remove that bulged fairing from the gear doors. The reinforcement to the nose fuselage also worked great - no more nose gear punching up through the duct.

The bad is that the change to the wing infill did not work great. Looking at the sliced design in retrospect, it's very obvious that I removed most of the infill supporting the landing gear mounts. So predictably the gear mounts ripped right out of the wing as soon as it hit the ground. I'm still figuring out how this plane will like being landed and when to flare and how much so the landing was harder than I would have liked, but it shouldn't have punched the gear up most of the way through the wing. I also realize I didn't get any pictures of the struts before the flight, so here's a picture of the strut after being ripped out of the wing. Strut is totally fine, the wing is less so.
1785092718547.png

I also got a speed measurement from this flight and the Super Duper Sabre tied with the Rapier at 133 mph with no gear doors and with only a single speed run with a mostly drained battery. Not an official speed for the contest, but it gives me a sense of what to expect and that number is still in the 140-160 range.
 

telnar1236

Master member
New nose gear design
1785540460458.png

1785540480620.png
1785540513431.png
1785540535733.png

This takes the same carbon fiber compression strut idea as the main gear and adapts it to the nose gear. It both fits better inside the fuselage and should give more suspension than the original design while being stronger too, so it's a win-win situation.
 

telnar1236

Master member
Stormy weather this weekend, so I've been working on the gear doors. Once these are designed, the whole design will be "done" and it will just be a matter of refining things and fixing issues as the appear.
1785688192358.png

1785688214302.png

The nose gear strut still needs a bit of a bulged fairing, but it can be much smaller, sleeker, and prettier so I'm much happier with this arrangement.
1785688316287.png
 

telnar1236

Master member
Used the extra time from the bad weather to do what I should have done from the very beginning and figure out how to build low profile trailing link struts for the main gear.
1786058511933.png

At their max deflection, they provide 45mm of travel vs. the 20mm of travel in the scissor link struts which should hopefully help the airframe absorb bad landings better. Even with the strengthened wings, I'm probably just transmitting these loads to the fuselage and at this point I'm starting to be worried about the weight increase (~100g of strengthening so far).
1786058890156.png

The outer gear door is also now directly integrated into the strut which should make it easier to build and avoids my worries about the wheel hitting the gear door with the scissor link design. The wheel is moved back by 30mm as well. The nose was pretty light when taxiing the previous version and it seems to fly best with the CG 10mm behind my initial target for most of the envelope though it does make it so that pitch up is noticeable if you push it too hard, so that makes the nose even lighter on the ground which makes taxiing in any wind at all quite difficult. The more aft main wheels should fix that.

The core fuselage did need to be redesigned to accommodate the further aft wheels, but with how modular this plane is, it's just 26 screws that need to undone to make the swap.
1786059403856.png
 

telnar1236

Master member
The new landing gear design is a big improvement. The plane taxis much better and the gear doesn't break off the instant the plane touches the ground on landing. With the CG 10mm behind the original location and the elevator rates turned down a bit to compensate, the plane also handles a lot nicer and responds to control inputs a lot better with the flaps down at the cost of further adding to the pitch up tendencies inherent in the design (I actually started to experience a Sabre Dance today but was able to get out of it thanks to the very high thrust to weight ratio and it remains much less severe than the real F-100 and fairly easily controllable).
1786323843787.png

And, better yet, the integrated main gear doors seal the wheel wells and let me hit 138 mph this flight which officially makes this the fastest plane I own (and possibly the fastest plane I have ever flown - I've gotten a 140 and 150 before, but I'm not sure I believe those readings since the planes they came from averaged much slower and the GPS unit was an older less reliable one and was crashed prior to both those readings).

With me needing to fight the airplane less, I can also cruise at lower throttle settings (this thing still cruises as fast as the top speed of most EDFs) which gives me much better flight times. On a 5000 mAh battery, I flew a 3 minute time on throttle flight and landed with 47% remaining and a 4:30 time on throttle flight and landed with 34% which makes a 5 minute time on throttle flight completely reasonable on that battery. With a 9000 mAh Li-Ion pack, flight times should be in the 8-10 minute range which I'm very happy with. The laminar flow airfoil and very clean airframe with very efficient ducting are working perfectly to give me a very fast plane with long flight times.

At this point, there are a few repairs I need to make after running off the runway into the grass, and a couple minor updates, and then I'm ready to really push for top speed. The plane needs the remaining gear doors and I want to rework the flap control linkages a bit to make them stiffer and more precise. And then, after a nice paint job, it's done.
 

Mr Man

Mr SPEED!
For a 0.4 mm nozzle, this is what I've found works well for different filaments

  • PLA: 220 C nozzle, 60-65 C bed, 0.98 extrusion ratio, 0.38 mm line width, 100% fan, no enclosure
  • LW-PLA: 255 C nozzle, 60 C bed, 0.6 extrusion ratio, 0.4 mm line width, 100% fan, no retractions, no enclosure
  • ABS: 250-270 C nozzle (hotter is typically better but can be issues with heat build up), 100 C bed, 1 extrusion ratio, 0.38 line width, 5-10% fan (more fan only for steep overhangs or very short layer times), print in enclosure
  • ASA: same as ABS
  • CF-ABS: 270 C nozzle, 95 C bed, 0.98 extrusion ratio, 0.38mm line width, 5-10% fan, ideally print pretty slowly, print in enclosure
  • TPU: 235 C nozzle, no bed heating required, 0.4mm line width, 100% fan, no retractions to avoid jams/clogs, no enclosure, good for tires on planes
  • LW-ASA: who knows: 260-270C nozzle, 90 C bed, 0.7 extrusion ratio (otherwise it's too weak), 0.4mm line width, fan speed is a lottery, no enclosure, no retractions, super prone to jams/clogs
  • PA6 nylon: 280 C nozzle, 85 C bed, 0.98 extrusion ratio, 0.4mm line width, 15% fan, needs enclosure, 8mm^3/second max volumetric flow rate, good for wheels but warps too much with single wall to use for a whole plane
  • PA612-CF nylon: 300 C nozzle, 50 C bed, 0.98 extrusion ratio, 0.4mm line width, 20% fan, no enclosure, 5mm^3/second max volumetric flow rate - almost PLA easy to print if you can get the nozzle hot enough and keep the filament dry, but pretty expensive - I only use this for gear struts
  • PETG-GF: 260 C nozzle, 85 C bed, 0.98 extrusion ratio, 0.4mm line width, 30% fan, no enclosure, 8mm/second max volumetric flow rate, good for wheels on planes up to ~1.5 kg
  • PA6-GF nylon: impossible to print with a 0.4mm nozzle, needs a 0.6mm nozzle minimum, otherwise same as carbon fiber filled nylon
  • PETG: not suitable for RC planes due to weight and lower strength than PLA - can used for decorative parts and canopies
  • PP-CF: not suitable for RC planes due to being too flexible, also a huge pain to print
I think that about sums up the various filaments I've messed around with for rc planes
Is the LWPLA settings you posted for prefoamed, or non foamed?