3D Printed 64mm MiG 21

telnar1236

Elite member
I've had a lot of time when I couldn't get to my workbench again recently, so I decided my F-104 needs an adversary. This MiG 21 is a much simpler design with only 3ch controls plus a steerable nose gear. It is also designed to be printed in normal PLA, mostly without supports. The thick airfoils and large cheater inlet mean that it probably won't be that fast, but it should be easy to fly, and the 64mm EDF and 4s 2200 mAh battery mean that it should have plenty of power. This should be a pretty quick project (it's 90% designed already and should be easy to assemble) so hopefully I can print and build this in the next few weeks while I finish the scheme for my F-104.

Side View.png Top View.png Bottom View.png Front View.png


Iso View.png
 

telnar1236

Elite member
Will you release files? I would love to build one. also, pla or lwpla?
I plan on releasing the files in this thread. The material is normal PLA. I've done some testing, and a well optimized PLA design is only barely heavier than a well optimized LWPLA one. The final AUW should be about 800 g. In terms of a timeline, I'll release finalized versions once I've gone through and printed/tested everything. Maybe in a couple of weeks. If anyone is interested in testing them in the meantime, let me know and I can post them now, but they come with the caveat that they are preliminary versions and may not work.
 

FlyerInStyle

Elite member
I plan on releasing the files in this thread. The material is normal PLA. I've done some testing, and a well optimized PLA design is only barely heavier than a well optimized LWPLA one. The final AUW should be about 800 g. In terms of a timeline, I'll release finalized versions once I've gone through and printed/tested everything. Maybe in a couple of weeks. If anyone is interested in testing them in the meantime, let me know and I can post them now, but they come with the caveat that they are preliminary versions and may not work.
Is there any way it can be modified for a prop or scaled down for a 50mm edf. With just trailerons? If so I would love to test it
 

Inq

Elite member
won't be that fast, but it should be easy to fly

:ROFLMAO::ROFLMAO: I think there must be a relatively thing going on, on my monitor. On your side (of my monitor) planes fly with postage stamps for wings. On my side I need big honk'n wing area and something slow for my skill level (thinking about claiming I'm a novice but that might jinx my whole world). I hope to go flying today... where is that wood table to knock on?

So when I brave an EDF plane, I'm thinking of an F-104 sibling... maybe a U2. :cool: Goes well with my black site background.

So... questions
  1. Do you always put the EDF in the same relative spot (front/back) or do you move it around based on plane?
  2. Have you ever done any tests to see if there are any advantages one way or another... EDF at the rear and sucking through the fuselage, or EDF at the front and blowing through the fuselage?
  3. I was thinking of putting the EDF under the tail surfaces (U2) and although the primary inlets would be the scale ones, that it could also have spring loaded inlet doors just in front of the EDF (still under the tail surfaces) that would open up when the vacuum pressure was high like at take-off? Any thoughts / insights?
  4. What kind of thrust loss do you get from your EDF on a stalk test versus being in a scale body like this?
  5. For this '21, where are you going to put all the hardware? It seems to me you have one long tube that needs to be un-obstructed and you don't really have any place to put it. Or are you talking a fairly large plane where the battery can sit up in the turtle back and be out of the flow?
  6. What kind of thrust to weight ratio are you typically dealing with?
  7. I have a pair of cheap 70 EDF's (4S)... Do your designs scale (say 70/64 worth) or is the internal structure optimized so tightly for the 0.4 mm walls that scaling of 109% would fail?
 

Inq

Elite member
Is there any way it can be modified for a prop or scaled down for a 50mm edf. With just trailerons? If so I would love to test it

That might make things a lot easier actually. With the axes-symmetric nose, it would seem easy to put the prop in the nose and 3D Print a spinner in the shape of the inlet with shock cone and all. It might not be as fast as the F-104, but I bet it would be easy to give it a > 2:1 thrust to weight ratio and climb like a rocket (leaving the F-104 down in the dust). In a shoot out, I think the vertical axis would trump the speed axis. ;)
 

FlyerInStyle

Elite member
That might make things a lot easier actually. With the axes-symmetric nose, it would seem easy to put the prop in the nose and 3D Print a spinner in the shape of the inlet with shock cone and all. It might not be as fast as the F-104, but I bet it would be easy to give it a > 2:1 thrust to weight ratio and climb like a rocket (leaving the F-104 down in the dust). In a shoot out, I think the vertical axis would trump the speed axis. ;)
I was actually thinking more of the 50mm edf route. not sure if that is easy or hard to do. I want somethign that flies not too fast, as it will be my first plane faster than an intermediate trainer
 

Inq

Elite member
I was actually thinking more of the 50mm edf route. not sure if that is easy or hard to do. I want somethign that flies not too fast, as it will be my first plane faster than an intermediate trainer

You must be way ahead of me... going from an intermediate trainer to a high wing loaded jet. If it warms up a little today (it's suppose to) I hope to get some more time on my Turbo Storch which I'd say is only an intro trainer. If I haven't crashed the Storch and I'm feeling cocky, I might cut loose with the Hybrid F-22. But it's very lightly loaded and has 1.8:1 thrust to weight. It'll be more than enough for me for a while.
 

FlyerInStyle

Elite member
You must be way ahead of me... going from an intermediate trainer to a high wing loaded jet. If it warms up a little today (it's suppose to) I hope to get some more time on my Turbo Storch which I'd say is only an intro trainer. If I haven't crashed the Storch and I'm feeling cocky, I might cut loose with the Hybrid F-22. But it's very lightly loaded and has 1.8:1 thrust to weight. It'll be more than enough for me for a while.
I really want to try making an edf, and I am a fan of the fishbed, with its relatively simplistic design, so it might be agood first jet, as well as me wanting to 3d print a plane, this should be perfect. I jsut want somethign different from all of hte regular high wing slower foam prop planes that I have
 

telnar1236

Elite member
Is there any way it can be modified for a prop or scaled down for a 50mm edf. With just trailerons? If so I would love to test it
The plane currently uses tailerons, and I don't plan on having a version with ailerons at all, so you're all good there. 3D printed planes can be hard to hand launch because they're a bit heavy, so I personally prefer gear, but the gear is optional and with this plane hand launching shouldn't be too bad. I might program a launch mix that gives a bit of up elevator, but tailed deltas are very stable when slow and you might not need it.

In terms of a 50mm EDF, it depends on your EDF unit. I am using one of the worst 64mm EDFs I am aware of, so if you have a decently strong 50mm one like the FMS or Powerfun 4s unit, then it should pretty much be a drop in replacement. The printed weight is about 430 grams, so any EDF with enough thrust to propel that weight plus itself, 2-3 servos, and a battery should work so long as you can get the plane to balance. If you tell me what EDF you want to use, if it will work, I can CAD up an adapter pretty easily and I would recommend making a thrust tube out of poster board or similar to get an exhaust about 90% FSA. This plane is to the same scale (1:14.7) as my 50mm EDF F-104 and with a 64mm EDF and 2200 mAh 4s pack will only be about 6% heavier, so it's a small 64mm or large 50mm scale. It would actually be lighter than the F-104 with the same power system, but it is draggier, so do with that what you will.

A prop might be possible, but it would run into more problems. A pusher prop would have to be back past the swept tail fins and could cause CG problems as well as being dangerous to launch. A tractor prop would ruin the looks (for me at least) and could shift the CG too far forward, but would avoid the clearance issues with the tail and be safer to launch. The short wings would also cause all sorts of problems with torque from a larger prop, so something like a racing drone motor and prop would be the way to go. That said, adapters for either the tail or nose should be easy to draw up in CAD, so if you want to try it, I'll make them with the warnings I just gave.
 

telnar1236

Elite member
I really want to try making an edf, and I am a fan of the fishbed, with its relatively simplistic design, so it might be agood first jet, as well as me wanting to 3d print a plane, this should be perfect. I jsut want somethign different from all of hte regular high wing slower foam prop planes that I have
Tailed deltas are very stable, but you do have to stay on the right side of the power curve. They produce a ton of drag when slow, so you can get into a bit of trouble. Once I've built and flown it, I'll have a better idea of the level of difficulty. I expect it to be fairly easy to fly, but as a 3D printed plane it will have a higher wing loading so a light foam board EDF (like the Viggen) or a straight wing EDF might be a better first choice before moving into something like this.
 

FlyerInStyle

Elite member
Tailed deltas are very stable, but you do have to stay on the right side of the power curve. They produce a ton of drag when slow, so you can get into a bit of trouble. Once I've built and flown it, I'll have a better idea of the level of difficulty. I expect it to be fairly easy to fly, but as a 3D printed plane it will have a higher wing loading so a light foam board EDF (like the Viggen) or a straight wing EDF might be a better first choice before moving into something like this.
I am already building an edf trainer (have built, but trying to figure out a way to make strong tailerons) and fly that a couple times, but it doesnt particularly look good. I might want t o beuild this scaled down fro a 50mm edf as those are cheaper, and so I can fly it at a smaller field to beta test it. I would be happy to beta test it. You should make a phantom to rival this next
 

telnar1236

Elite member
:ROFLMAO::ROFLMAO: I think there must be a relatively thing going on, on my monitor. On your side (of my monitor) planes fly with postage stamps for wings. On my side I need big honk'n wing area and something slow for my skill level (thinking about claiming I'm a novice but that might jinx my whole world). I hope to go flying today... where is that wood table to knock on?

So when I brave an EDF plane, I'm thinking of an F-104 sibling... maybe a U2. :cool: Goes well with my black site background.

So... questions
  1. Do you always put the EDF in the same relative spot (front/back) or do you move it around based on plane?
  2. Have you ever done any tests to see if there are any advantages one way or another... EDF at the rear and sucking through the fuselage, or EDF at the front and blowing through the fuselage?
  3. I was thinking of putting the EDF under the tail surfaces (U2) and although the primary inlets would be the scale ones, that it could also have spring loaded inlet doors just in front of the EDF (still under the tail surfaces) that would open up when the vacuum pressure was high like at take-off? Any thoughts / insights?
  4. What kind of thrust loss do you get from your EDF on a stalk test versus being in a scale body like this?
  5. For this '21, where are you going to put all the hardware? It seems to me you have one long tube that needs to be un-obstructed and you don't really have any place to put it. Or are you talking a fairly large plane where the battery can sit up in the turtle back and be out of the flow?
  6. What kind of thrust to weight ratio are you typically dealing with?
  7. I have a pair of cheap 70 EDF's (4S)... Do your designs scale (say 70/64 worth) or is the internal structure optimized so tightly for the 0.4 mm walls that scaling of 109% would fail?
Well-designed wings will generate a ton more lift than most people expect, and deltas basically never stall so you can high alpha them in to land. My F-104 has a stall speed not that much faster than a scale warbird with the flaps down, and this one will have a stall speed of about 25mph When I'm talking easy to fly, I'm talking easy to fly for model of a 1950s/60s supersonic fighter. This plane is by no means a trainer, but it might be a second EDF (I'll let you know once I fly it). I'm actually planning on taking my 50mm F-104 and changing some things around to make a printable U-2 as well since the early U-2s were essentially heavily modified F-104s anyway.

Now for the questions:
1. EDF placement varies. Mostly I try to position it to attain a good CG with a reasonable battery position
2. EDFs are most efficient with a short thrust tube, so it's good to place them as near the back as reasonable. That said, efficient ducting is much more important, so avoiding sharp corners and sudden changes in area will go a long way. The outlet should typically be about 90% FSA so there needs to be some length for the area to decrease smoothly. The ducting in this MiG 21 is terrible and should not be used as an example. The ducting in my F-104 is better, but still has room for improvement, largely because of trying to package 6ch worth of electronics into a 50mm EDF. The ducting in the FT Viggen is the golden standard for foam board, in my opinion.
3. In the U-2 I would generally expect the EDF to go about halfway between the wing and tail or maybe a bit more forward, largely for CG reasons. So long as the plane will balance there is no reason it should not go more aft, but under the tail planes is probably too aft (see 2). Spring loaded cheater inlets work and look really scale compared to other types of cheater inlets. I personally find them too mechanically complex for too little benefit, especially when NACA inlets work efficiently through the whole flight, but especially on larger planes it's a matter of personal preference.
4. Losses can be anywhere from near zero to as high as 50% with very poor ducting. I expect the MiG 21 to lose about 10-15% thrust when static and more when at high speed because of the duct design. In contrast, my F-104 lost a similar amount of thrust when static, but would actually gain some thrust back past about 80mph (if it could get that fast in the first place) because of much better designed ducting. When I've really messed something up, I've seen losses as high as 40-50%, but that is rare, even with horrible ducting design.
5. Most of the hardware sits on top of the duct on a platform under the canopy. You can kind of see where the canopy will be accessed in the pictures I posted. There is barely room for a 2200 mAh 4s pack, a receiver, one servo, and the mechanism to steer the nose gear. The nose gear strut and wing spar extend through the duct and will just result in some losses. The elevon servos aren't modeled yet but will go right below the horizontal stabilizer in the fuselage. I expect the EDF on this plane to mostly breath through the cheater inlet anyway since the nose inlet is far too small for the size of EDF and is mostly there for aesthetics, so the losses from the strut and spar protruding through the inlet should be minimal.
6. Thrust to weight ratio varies hugely. I have an old Phase 3 U-2 that can't have a TWR more than about 0.3, while my 80mm EDF speed-optimized F-104 is a bit over 1. This plane will have a TWR of about 0.8 and my 50mm F-104 flew with a TWR of about 0.6 on 3s, although that should exceed 1 on 4s.
7. This model should probably scale up just fine. There isn't that much to go wrong with the print. The only concern would be the wing structure printing separately from the wing skin since there are some small clearances to make sure the skin turns out smooth, but that shouldn't be too bad if you turn off retraction and print it one part at a time which is the recommended method anyway. I tried slicing the nose in Cura scaled by 109% and it didn't seem to cause anything to go wrong. I cannot speak to how the structures would stand up on a larger model, but I don't think anything would go too badly wrong there either. You would definitely need some shims for the servos and some of the non 3D printed parts (CF rods mainly) would be weird sizes like 11mm as opposed to 10 for the wing spar. Other models could have substantially larger problems especially when scaled up by more than you're talking about.
 
  • Like
Reactions: Inq

telnar1236

Elite member
I am already building an edf trainer (have built, but trying to figure out a way to make strong tailerons) and fly that a couple times, but it doesnt particularly look good. I might want t o beuild this scaled down fro a 50mm edf as those are cheaper, and so I can fly it at a smaller field to beta test it. I would be happy to beta test it. You should make a phantom to rival this next
Cool, I'll post the STLs once I've printed them in that case, just so I know they print at all. For flying in small spaces, it probably isn't the best idea to scale this thing down. As a highly loaded jet, it will require some turning room, and reducing the size will increase the wing loading further. It's already the size of a large 50mm design and a strong 50mm EDF will be plenty to power it, so I would strongly recommend just printing it as is and then putting the 50mm EDF in it. Larger 64mm EDFs actually will not fit in this one. It is about 39" long, but it only has a 21" wingspan, so it isn't a large RC jet by any means. It seems to work in Cura when scaled down, but I can't promise it won't cause structural issues either.

I do eventually want to get around to a Phantom, but I think after this one, my next build will probably be an F-106 since I have a spare 80mm EDF that is exactly the right size for one at this same scale (the F-106 was HUGE).
 

FlyerInStyle

Elite member
Cool, I'll post the STLs once I've printed them in that case, just so I know they print at all. For flying in small spaces, it probably isn't the best idea to scale this thing down. As a highly loaded jet, it will require some turning room, and reducing the size will increase the wing loading further. It's already the size of a large 50mm design and a strong 50mm EDF will be plenty to power it, so I would strongly recommend just printing it as is and then putting the 50mm EDF in it. Larger 64mm EDFs actually will not fit in this one. It is about 39" long, but it only has a 21" wingspan, so it isn't a large RC jet by any means. It seems to work in Cura when scaled down, but I can't promise it won't cause structural issues either.

I do eventually want to get around to a Phantom, but I think after this one, my next build will probably be an F-106 since I have a spare 80mm EDF that is exactly the right size for one at this same scale (the F-106 was HUGE).
OK. Thanks. I thought this was a beast, and the size it is now is a good size fro the future. what batt. will this be for? I will make sure to build it though, once I geet to finishing most of my current projects.
 

Inq

Elite member
This model should probably scale up just fine. There isn't that much to go wrong with the print. The only concern would be the wing structure printing separately from the wing skin since there are some small clearances to make sure the skin turns out smooth, but that shouldn't be too bad if you turn off retraction and print it one part at a time which is the recommended method anyway. I tried slicing the nose in Cura scaled by 109% and it didn't seem to cause anything to go wrong. I cannot speak to how the structures would stand up on a larger model, but I don't think anything would go too badly wrong there either. You would definitely need some shims for the servos and some of the non 3D printed parts (CF rods mainly) would be weird sizes like 11mm as opposed to 10 for the wing spar. Other models could have substantially larger problems especially when scaled up by more than you're talking about.

3DLabPrint uses a technique where they predominantly use single walls and they'll use a reversed strip patch offset to the inside by 0.8 mm so that it will give a small region of a double wall. They use it often in their wing design to form "ribs". Scaling it up or down (even 109%) is enough to break that concept and either not form the second wall (down scale) or make it so the second wall doesn't attach well (up scale).

Good to know that you don't use those techniques.
1673420426691.png
 

telnar1236

Elite member
3DLabPrint uses a technique where they predominantly use single walls and they'll use a reversed strip patch offset to the inside by 0.8 mm so that it will give a small region of a double wall. They use it often in their wing design to form "ribs". Scaling it up or down (even 109%) is enough to break that concept and either not form the second wall (down scale) or make it so the second wall doesn't attach well (up scale).

Good to know that you don't use those techniques.
View attachment 233728
Yeah, I model everything as solids, not surfaces. It doesn't print as nicely, but I like the flexibility it gives me. There is a risk with upscaling the MiG that the ribs in the wing will not attach well on my design as well (I think I mentioned it in my previous post). Like the 3D lab print design, there is a small offset on the ribs so upscaling could result in too large of a gap. That said, I'm realizing that I designed this plane way too heavy as I start to print it. It should fly in this first version, but there is a ton of structural strength it doesn't need. I'm planning on messing with different wing internals on the redesign, and the way I'm looking at changing them should have no problem being scaled up.
 

Inq

Elite member
@telnar1236,

While we're on the subject. I bought 2 of these years ago because they were cheap. They've never been out of the box. I know they're not the best, but with your knowledge of these things in general, could you estimate a thrust I'd get if they were in a long fuselage F-104 / U2 type design. I just want a ball-park. I don't know if I need to design for 500 mg trust or 5000 mg thrust. I just have no clue.

70mm, 6 blade EDF, 4S

https://www.amazon.com/dp/B099S781Q6/?tag=lstir-20
 

telnar1236

Elite member
@telnar1236,

While we're on the subject. I bought 2 of these years ago because they were cheap. They've never been out of the box. I know they're not the best, but with your knowledge of these things in general, could you estimate a thrust I'd get if they were in a long fuselage F-104 / U2 type design. I just want a ball-park. I don't know if I need to design for 500 mg trust or 5000 mg thrust. I just have no clue.

70mm, 6 blade EDF, 4S

https://www.amazon.com/dp/B099S781Q6/?tag=lstir-20
If I had to guess, maybe somewhere between 700 grams and a kilogram, but I don't have anything to back that up. It claims 640 W as the max power, so if that is correct, you should be fine powering up to about a 1600 g AUW for something like a U-2 (ideally you want at least 180 W/lb for an EDF, but for a U-2 you should be able to go a bit lower without many issues) but I am always very skeptical of claims I see online and typically multiply by between 0.5 and 0.8 when making an estimate for my designs (the EDF in my F-104, for example claims a power consumption of 479 W on 3s online, but on measuring it only uses about 350 W on 3s, so a factor of 0.73). A simpler approach is to try and keep the aircraft AUW below the rated thrust since that should give a flyable TWR for most designs. The best option would be to do a thrust test. A mechanism that takes the force from the EDF and uses it push down on a scale is pretty simple to build and extremely useful.
 
  • Like
Reactions: Inq