The Other XB-70 Build | Valkyrie V2
CarolineTyler
Legendary member
Excellent maiden. I think the unusual fuselage of the plane makes the AOA look more steep than it was.
Congratulations!
Pieliker96
Elite member
It's about time I fly this thing again. It's currently ready to go, with retracts and canard flaps functioning, but there are some things I want to do before I take it for a spin again.
What I've done:
-Added actual wheel collars
-Removed the unnecessary gear door servos, retract sequencer, and associated wiring
-Repaired the nose retract mount
-Replaced the faulty mains retract unit
-Made the landing gear holes larger and removed some ducting for better airflow to the EDFs
In its current state, the plane's all-up-weight is 5.225kg (11.5lbs). I retested static thrust, resulting in 3.5kg ( 34.3N) of thrust compared to 3kg (29.4N) prior. Keep in mind, this is with a ~3000ft density altitude increase (~8% air density decrease) from January to June and July. I'm hoping this extra thrust and the canard flaps can compensate for the density altitude and shorten my takeoff run to where I'd be comfortable flying it at FFE 2019.
What's left to do:
-Make it look better! Tail number, USAF logo and text, the black parts of the nose, etc.
-Un-warp the wings from storage
-Taxi Tests
-Send it!
What I've done:
-Added actual wheel collars
-Removed the unnecessary gear door servos, retract sequencer, and associated wiring
-Repaired the nose retract mount
-Replaced the faulty mains retract unit
-Made the landing gear holes larger and removed some ducting for better airflow to the EDFs
In its current state, the plane's all-up-weight is 5.225kg (11.5lbs). I retested static thrust, resulting in 3.5kg ( 34.3N) of thrust compared to 3kg (29.4N) prior. Keep in mind, this is with a ~3000ft density altitude increase (~8% air density decrease) from January to June and July. I'm hoping this extra thrust and the canard flaps can compensate for the density altitude and shorten my takeoff run to where I'd be comfortable flying it at FFE 2019.
What's left to do:
-Make it look better! Tail number, USAF logo and text, the black parts of the nose, etc.
-Un-warp the wings from storage
-Taxi Tests
-Send it!
bracesport
Legendary member
@Pieliker96 - This has been a pleasure to follow - a marvellous achievement!
Pieliker96
Elite member
Pieliker96
Elite member
Yeah, that'll be fantastic!
I looked on Google Maps and compared it with the video of the maiden flight. My takeoff roll was around 135 feet. I think I can get closer to 100 feet with the higher TWR (0.66 vs 0.53) and canard flaps, but there's only one way to find out.
Does anyone know how long the hard runway was at FFE last year? Not the black blanket stuff, but the small stretch at show center? I need to figure out my margins for takeoff and landing distance.
Pieliker96
Elite member
Here's some pics of the XB-70 in its final livery. All "decals" are done in sharpie with the help of a straightedge and laser-cut stencils.
More pics and a taxi test tomorrow, assuming the weather cooperates.
More pics and a taxi test tomorrow, assuming the weather cooperates.
The hard surface at show center was only maybe 50ft. You definitely don’t want to bank on that. I hope it can still roll well enough on the Geotex, it all depends on your weight and wheel loading.
Looking great man, can’t wait to see it!
Pieliker96
Elite member
Smoked a motor bearing during another static thrust test (I made some minor modifications to the inlets). At least it didn't happen in the air, but now I have to wait for a new motor to arrive. And of course, that doesn't inspire confidence for the reliability of the other motors.
The plan is to do another flight ASAP to make sure the canard flaps and retracts are working properly - probably the 6th or 7th, assuming the motor comes on time. Then it's off to FFE.
The plan is to do another flight ASAP to make sure the canard flaps and retracts are working properly - probably the 6th or 7th, assuming the motor comes on time. Then it's off to FFE.
Pieliker96
Elite member
Photo dump!
Flight soon™.
Flight soon™.
Pieliker96
Elite member
Not exactly the update I want to be making, but it's not exactly the worst-case scenario either. Maybe doing the vehicle #2 paint scheme was a bad omen.
Yesterday, I did some taxi tests, broke the nose gear mount, then rebuilt it to be stronger.
I went out and flew it today, then crashed. From the video it looks like I got myself into a stall(?) at a high bank angle and low altitude, without enough altitude to properly recover.
It hit the ground with the wings level, nose down. The nose gear sheared clean off, ripping out the steering servo and associated wiring in the process. One of the mains buckled, and the other one somehow managed to hang on inside the fuse. The batteries slid all the way forwards into the nose, and disconnected in the process. The plane bounced, glided another 50 feet or so, and came to rest on its remaining landing gear and the wingtip.
The wings and fuselage look exactly like before the flight, save some serious grass and dirt streaks on the bottom of the latter. The front of the nacelle took most of the impact, crumpling a bit. The neck also separated a bit around halfway down the length of the plane from the torque of the nose and batteries. The landing gear is pretty mangled.
The good news is that all of the electronics power up fine. The EDFs are all in working condition, despite inhaling a bit of dead grass. All of the major structural bits (Wing spars, neck attachment) are fine as well. I'm planning on stripping out the gear, repairing the front nacelle, and then bringing it to FFE. If I can find a way to get it in the air there (ex. ATV/buggy launch), then I will. And if I do, it'll be nice not having to worry about landing on a runway!
I'm also considering giving away the airframe (or at least parts of it) at FFE, then building a B-70 V2. There's a lot that I can improve on with this design - in terms of weight reduction, thrust-to-weight, power, and overall aesthetics. Maybe the folding tips and chutes, If I'm feeling up for it. If not an XB-70 V2, then a Lancer, Concorde, or 7-4.
Remember - "Just a Plane".
Yesterday, I did some taxi tests, broke the nose gear mount, then rebuilt it to be stronger.
I went out and flew it today, then crashed. From the video it looks like I got myself into a stall(?) at a high bank angle and low altitude, without enough altitude to properly recover.
It hit the ground with the wings level, nose down. The nose gear sheared clean off, ripping out the steering servo and associated wiring in the process. One of the mains buckled, and the other one somehow managed to hang on inside the fuse. The batteries slid all the way forwards into the nose, and disconnected in the process. The plane bounced, glided another 50 feet or so, and came to rest on its remaining landing gear and the wingtip.
The wings and fuselage look exactly like before the flight, save some serious grass and dirt streaks on the bottom of the latter. The front of the nacelle took most of the impact, crumpling a bit. The neck also separated a bit around halfway down the length of the plane from the torque of the nose and batteries. The landing gear is pretty mangled.
The good news is that all of the electronics power up fine. The EDFs are all in working condition, despite inhaling a bit of dead grass. All of the major structural bits (Wing spars, neck attachment) are fine as well. I'm planning on stripping out the gear, repairing the front nacelle, and then bringing it to FFE. If I can find a way to get it in the air there (ex. ATV/buggy launch), then I will. And if I do, it'll be nice not having to worry about landing on a runway!
I'm also considering giving away the airframe (or at least parts of it) at FFE, then building a B-70 V2. There's a lot that I can improve on with this design - in terms of weight reduction, thrust-to-weight, power, and overall aesthetics. Maybe the folding tips and chutes, If I'm feeling up for it. If not an XB-70 V2, then a Lancer, Concorde, or 7-4.
Remember - "Just a Plane".
Pieliker96
Elite member
Took the crash frame-by frame and holy cow look at that flex (it's a gif)
Fuselage attachment area will need some more inspection.
Fuselage attachment area will need some more inspection.
Last edited:
Pieliker96
Elite member
Was able to recover some of the underbelly camera footage, from takeoff until a couple seconds before the crash.
I also measured the takeoff distance, which was almost exactly on my estimate of 100ft (Just 5 feet over).
I also measured the takeoff distance, which was almost exactly on my estimate of 100ft (Just 5 feet over).
Last edited:
OliverW
Legendary member
That sucks! Keep in mind, it can be fixed and flown again! My dad built a 120 inch wingspan Cessna skymaster from a kit, and spent 3 years building it! That included fiberglassing and painting it. On the 15th or so flight, the wing folded and it hit the ground only leaving behind toothpicks.
Pieliker96
Elite member
Don't know why I was so afraid of crashing. An hour or two with the glue gun and some foam and it's all good! I was even able to bend the retract pins back, and swapped some parts from the old set to get a somewhat working set of retracts (which intermittently get stuck up/down due to bad limit switch tolerances and the motor bottoming out the jack screw). I also made some makeshift bogies as the solder joints all cracked on the previous one - they'll be fine for a takeoff or two.
Pieliker96
Elite member
Footage from Flite Fest. It started feeling weird in the air because of the wind so I opted to put it down early.
Pieliker96
Elite member
TL;DR: This design was cool but not the best, might do XB-70 V2: Electric Boogaloo.
Time to bring this project back from the dead.
During my first year at college I've been studying to be an Aerospace Engineer. I know a bit more about aircraft design than I did when I started this project. Combined with recognising the various shortcomings of the original design, I've got to thinking about what I could do better.
The original suffered from being heavy, overbuilt, and underpowered. These issues stemmed from gross underestimation of the final weight, structural inefficiency / overbuilding, excessive ducting losses, excessive wiring losses, battery voltage drop, and more. The result was a plane which had to be flown hot and fast while struggling to do so, causing lots of stress on the components and the pilot. Here's what I would change in a version 2 to mediate that:
Wings
The wings were a simple flat-bottom delta with an eyeballed trapezoidal cross-section and < 3% thickness. There was no intentional washout added, although the wings conveniently warped a couple of degrees in storage and retained some washout after de-warping. Moving to a thicker aerofoil would generally increase the lift at a given angle of attack, leading to slower flight speeds. Thicker wings are also easier to deal with structurally and may resist twisting and deformation in storage. It may be prudent to move to an actual aerofoil instead of an eyeballed trapezoid for an improvement in lift and drag. Intentional washout is also a must for stability reasons.
Wiring
The main power wiring to and from the ESCs is undersized on the battery end and oversized on the motor end, resulting in electrical and mass inefficiency. They're also solid-core, which is a pain to deal with. Moving to properly sized stranded core could decrease wiring losses and mass while being easier to work with. The same can be said of the servo extensions, although to a lesser degree. Having a dedicated servo extension crimping tool and a roll of standard servo wire is probably better over soldering whatever wire I could find in the scrap bin to the two ends of an existing servo extension.
Batteries
In an effort to economize legal tender, I opted to get another set of the batteries I had for my RC truck instead of get an entirely new set of batteries. This meant that one of the pairs was a bit old and had been abused previously. Combine that with the high-discharge torture I put them through and you've got a nice voltage drop under load. Higher quality batteries of the same capacity weigh less (due to not being hardcase) and have higher discharge ratings. Another idea is to add an extra cell to combat the wiring inefficiency, add more effective capacity, and give a bit more "oomph" for takeoff. Whether the motors will be happy with that is a different story.
Ducting
A combination of voltage sag under load and ducting inefficiencies led to a near halving of thrust from on the test stand to installed in the plane. Some thrust was recovered on later flights by massively enlarging the landing gear / cheater holes and adding some directing plenums inside the duct. Adding larger bellmouths or properly converging the duct into the fan's mouth instead of the current airbox setup could lead to greater static thrust. I have also wanted to investigate the effect of the geometry of the thrust tube on static thrust, not just the outlet size in terms of fan swept area.
Construction
The attachment of the front fuselage to the main fuselage was so strong that it did not need to be repaired even after the crash on flight 2, during which it visibly flexed to a ridiculous extent. This tells me it is quite overbuilt, which is a bunch of excess mass that isn't needed. This is true of other parts of the craft, such as all of the sheeting of the fuselage and wings. There is great potential for weight reduction in the use of alternate materials - I'm thinking about thinner sheeting like balsa and depron as well as some techniques like hot-wire. The spars attaching the four main parts and the methods of locking those parts in place are also not ideal from a structural or aesthetic perspective.
Aesthetics
It looks good from afar, but far from good. There are numerous gaps in the foam sheeting, rough edges, and wrinkles. It's beneficial in that it gives me less qualms about the potential of crashing it and ruining what would be a perfect airplane but detrimental in that it's a bit of drag and aerodynamic inconsistency.
These are just the broad topics in the bits that I'd like to change. Will I actually get to changing them? Well, I've got a long summer with not much to do at the moment, so we'll see. This time around, I make no promises.
Time to bring this project back from the dead.
During my first year at college I've been studying to be an Aerospace Engineer. I know a bit more about aircraft design than I did when I started this project. Combined with recognising the various shortcomings of the original design, I've got to thinking about what I could do better.
The original suffered from being heavy, overbuilt, and underpowered. These issues stemmed from gross underestimation of the final weight, structural inefficiency / overbuilding, excessive ducting losses, excessive wiring losses, battery voltage drop, and more. The result was a plane which had to be flown hot and fast while struggling to do so, causing lots of stress on the components and the pilot. Here's what I would change in a version 2 to mediate that:
Wings
The wings were a simple flat-bottom delta with an eyeballed trapezoidal cross-section and < 3% thickness. There was no intentional washout added, although the wings conveniently warped a couple of degrees in storage and retained some washout after de-warping. Moving to a thicker aerofoil would generally increase the lift at a given angle of attack, leading to slower flight speeds. Thicker wings are also easier to deal with structurally and may resist twisting and deformation in storage. It may be prudent to move to an actual aerofoil instead of an eyeballed trapezoid for an improvement in lift and drag. Intentional washout is also a must for stability reasons.
Wiring
The main power wiring to and from the ESCs is undersized on the battery end and oversized on the motor end, resulting in electrical and mass inefficiency. They're also solid-core, which is a pain to deal with. Moving to properly sized stranded core could decrease wiring losses and mass while being easier to work with. The same can be said of the servo extensions, although to a lesser degree. Having a dedicated servo extension crimping tool and a roll of standard servo wire is probably better over soldering whatever wire I could find in the scrap bin to the two ends of an existing servo extension.
Batteries
In an effort to economize legal tender, I opted to get another set of the batteries I had for my RC truck instead of get an entirely new set of batteries. This meant that one of the pairs was a bit old and had been abused previously. Combine that with the high-discharge torture I put them through and you've got a nice voltage drop under load. Higher quality batteries of the same capacity weigh less (due to not being hardcase) and have higher discharge ratings. Another idea is to add an extra cell to combat the wiring inefficiency, add more effective capacity, and give a bit more "oomph" for takeoff. Whether the motors will be happy with that is a different story.
Ducting
A combination of voltage sag under load and ducting inefficiencies led to a near halving of thrust from on the test stand to installed in the plane. Some thrust was recovered on later flights by massively enlarging the landing gear / cheater holes and adding some directing plenums inside the duct. Adding larger bellmouths or properly converging the duct into the fan's mouth instead of the current airbox setup could lead to greater static thrust. I have also wanted to investigate the effect of the geometry of the thrust tube on static thrust, not just the outlet size in terms of fan swept area.
Construction
The attachment of the front fuselage to the main fuselage was so strong that it did not need to be repaired even after the crash on flight 2, during which it visibly flexed to a ridiculous extent. This tells me it is quite overbuilt, which is a bunch of excess mass that isn't needed. This is true of other parts of the craft, such as all of the sheeting of the fuselage and wings. There is great potential for weight reduction in the use of alternate materials - I'm thinking about thinner sheeting like balsa and depron as well as some techniques like hot-wire. The spars attaching the four main parts and the methods of locking those parts in place are also not ideal from a structural or aesthetic perspective.
Aesthetics
It looks good from afar, but far from good. There are numerous gaps in the foam sheeting, rough edges, and wrinkles. It's beneficial in that it gives me less qualms about the potential of crashing it and ruining what would be a perfect airplane but detrimental in that it's a bit of drag and aerodynamic inconsistency.
These are just the broad topics in the bits that I'd like to change. Will I actually get to changing them? Well, I've got a long summer with not much to do at the moment, so we'll see. This time around, I make no promises.
-------------------- XB-70 V2: More Power, More Lift, More Engineering. --------------------
F106DeltaDart
Elite member
Will be looking forward to the new and improved version! Always love seeing XB-70s!
Pieliker96
Elite member
V1 Is no more
I have made a worthy sacrifice to the foamy gods. The wiring monster, with its many insulated tendrils, has been slain and broken back down to its seperate components. In time, a new airframe will rise from the ashes, incorporating new-to-me building techniques and composite materials.
Thrust Test Results, Cause of propulsion inefficiency
TL;DR: The primary constituent of thrust loss was ducting losses followed by battery voltage droop under load.
This post is mostly about data and quantifying the severe amount of thrust loss experienced in V1 in order to find the necessary areas of improvement. If numbers aren't that interesting to you (which is quite understandable), don't read the rest of the post. It's pretty bland and reads like a lab report. This is for the engineer in me (and maybe you), we like numbers or something.
Operational definitions
I performed three thrust tests today, the last three entries in the table. The first three are derived from prior data. "V1" refers to the configuration used on flight 1, whereas "V1 modded" refers to the configuration used on flights 2 and 3. The modifications consisted of some small bits of ductwork and massively enlarged landing gear holes. All values are per fan unless noted otherwise. No thrust tubes were installed for the last four tests. "Double wire" refers to two runs of twelve-guage (used in the airframe), whereas "single wire" refers to a single run. The wire is 3.2 feet in length. "Single EDF" and "Double EDF" denote the number of EDFs running off the pair of batteries, which power half the overall propulsion system.
Procedure
Thrust was measured by a digital scale and a custom-built thrust test stand. The batteries were topped off to 16.81v when in series between runs. A test run consists of a ramp up to full throttle over a period of around 5 seconds, sustaining full throttle for another 5 seconds, and then cutting the throttle. Voltage was measured at the batteries and at the ESCs with two multimeters. Average Vbatt and Vesc readings were taken over the course of the full-throttle interval as recorded by video. The voltage difference across the wire was used to calculate the amp draw based on the wire's resistance which was determined from an online calculator - the resistance is far to small for the meters to measure accurately or precisely. The wires from the ESC to the fan were not considered as they have an extremely low resistance compared to the motor windings and do not carry a high amount of current; the energy lost in the phase wires is negligible.
Data
Data Analysis:
The battery's effective internal resistance, calculated from an average of the last three trials, was 18.9 milliohms.
The total thrust loss in the modded version as compared to the ideal single EDF with minimal wiring was 3056g, yielding a combined ducting and electrical efficiency of 53%.
Of that thrust loss,
- 42.9% was due to electrical (9.3% wiring, 33.6% battery droop)
- 57.1% was due to ducting
Conclusions
The majority of the propulsion inefficiency experienced in V1 was due to ducting and a good bit was due to battery droop. The wiring losses can be considered acceptable. As a result,
V2 will have a larger intake area and smoother ducting. Modifications to the ducting will be tested experimentally to minimize losses.
V2 will use batteries with a lower internal resistance and higher current delivery capability.
V2 will use the same effective guage over a shorter distance on the battery side and a smaller guage over a shorter distance on the EDF side to reduce mass and maintain performance.
I've set a goal to get 4.25kg of total static thrust out of V2, a ~25% increase over V1 in its final form. The nacelle is first on the list of priorities and will allow for tweaking of ducting before finalizing the design.
I have made a worthy sacrifice to the foamy gods. The wiring monster, with its many insulated tendrils, has been slain and broken back down to its seperate components. In time, a new airframe will rise from the ashes, incorporating new-to-me building techniques and composite materials.
Thrust Test Results, Cause of propulsion inefficiency
TL;DR: The primary constituent of thrust loss was ducting losses followed by battery voltage droop under load.
This post is mostly about data and quantifying the severe amount of thrust loss experienced in V1 in order to find the necessary areas of improvement. If numbers aren't that interesting to you (which is quite understandable), don't read the rest of the post. It's pretty bland and reads like a lab report. This is for the engineer in me (and maybe you), we like numbers or something.
Operational definitions
I performed three thrust tests today, the last three entries in the table. The first three are derived from prior data. "V1" refers to the configuration used on flight 1, whereas "V1 modded" refers to the configuration used on flights 2 and 3. The modifications consisted of some small bits of ductwork and massively enlarged landing gear holes. All values are per fan unless noted otherwise. No thrust tubes were installed for the last four tests. "Double wire" refers to two runs of twelve-guage (used in the airframe), whereas "single wire" refers to a single run. The wire is 3.2 feet in length. "Single EDF" and "Double EDF" denote the number of EDFs running off the pair of batteries, which power half the overall propulsion system.
Procedure
Thrust was measured by a digital scale and a custom-built thrust test stand. The batteries were topped off to 16.81v when in series between runs. A test run consists of a ramp up to full throttle over a period of around 5 seconds, sustaining full throttle for another 5 seconds, and then cutting the throttle. Voltage was measured at the batteries and at the ESCs with two multimeters. Average Vbatt and Vesc readings were taken over the course of the full-throttle interval as recorded by video. The voltage difference across the wire was used to calculate the amp draw based on the wire's resistance which was determined from an online calculator - the resistance is far to small for the meters to measure accurately or precisely. The wires from the ESC to the fan were not considered as they have an extremely low resistance compared to the motor windings and do not carry a high amount of current; the energy lost in the phase wires is negligible.
Data
Data Analysis:
The battery's effective internal resistance, calculated from an average of the last three trials, was 18.9 milliohms.
The total thrust loss in the modded version as compared to the ideal single EDF with minimal wiring was 3056g, yielding a combined ducting and electrical efficiency of 53%.
Of that thrust loss,
- 42.9% was due to electrical (9.3% wiring, 33.6% battery droop)
- 57.1% was due to ducting
Conclusions
The majority of the propulsion inefficiency experienced in V1 was due to ducting and a good bit was due to battery droop. The wiring losses can be considered acceptable. As a result,
V2 will have a larger intake area and smoother ducting. Modifications to the ducting will be tested experimentally to minimize losses.
V2 will use batteries with a lower internal resistance and higher current delivery capability.
V2 will use the same effective guage over a shorter distance on the battery side and a smaller guage over a shorter distance on the EDF side to reduce mass and maintain performance.
I've set a goal to get 4.25kg of total static thrust out of V2, a ~25% increase over V1 in its final form. The nacelle is first on the list of priorities and will allow for tweaking of ducting before finalizing the design.
-------------------- XB-70 V2: More Power, More Lift, More Engineering. --------------------
