Scout F - Engineering Cut

[@BPSspace]

this video is for nerds only 😎

[@chamian1167]

Do I qualify as a nerd?

[@thomas-nk7kx]

@@chamian1167 no

[@travis.napier]

Lead the way Mr. Nerdy rocket man.

[@eldoon2576]

Guess I'm a nerd :D

[@alejandrodeograciasmartin6637]

Then i know im in the right place

[@MahBor]

LOVE THESE ENGINEERING CUTS!! Pls continue this series with future rockets, they're so educational

[@gillesboulay1640]

The 🎉🎉🎉9 je suis à l'hôpital et et

[@wes9627]

I got a BS degree in EE in 1967. They were still designing some electrical circuits using vacuum tubes. We've come a long way in 55 years. Keep up the great work!

[@TheNadOby]

There are some extremely tiny "vacuum tubes" on a chip for teraherz applications already. We will stuck with them forever.

[@kurtfrancis4621]

Next BPS project - Landing a Scout using vacuum tubes!!! :)

[@index7787]

@@TheNadOby Tube amps forever.

[@c1ph3rpunk]

“Look me in the eye here, this is not a crewed vehicle … IT’S A MODEL ROCKET”. Excellent effort, you know you’ll still get folks banging on some aspect of it. Kudos to you for sharing your journey, that’s how we learn. Lifelong engineer myself, almost 30 years professionally, you are the real deal, “hacks” is what holds all of our stuff together in the early stages. Oh, all the years, all the hacks… Drop the landing footage in 60-90 minute videos in a separate series, call it “B Roll Landing Hacks”.

[@sledgebear9398]

Bro, I absolutely love these videos. I’ve actually started to make model rockets from scratch because of you. Keep it up!

[@NameThievery]

I have access to Simulink as part of Matlab while in school for Computer Science (I have an engineering background, but reactor plants). I just learned that you can get Simulink to output code and now I need to simulate something. Amazing content!

[@newmonengineering]

I would say your week was worth the effort. You learned some valuable newbie lessons about cnc machines and manufacturing. I remember my first aluminum cnc project well also. It may seem silly now but the first week is a make it or break it week for everyone in cnc land. You either learn and start enjoying it, or you learn to hate it. Either way you learn. You can read all the books you want but you learn when you try it yourself. Best lessons are failures, expensive? Sometimes, but the experience is always worth it.

[@andrewharpin6749]

First, great video, I think we need more of this for young budding engineers. 2nd, Simulation is very impressive. 3rd, look up tables are massively more computationally efficient, few instructions vs XXX depending on the complexity of your polynomial, this allows you to run the control loop much faster, also look up tables allow you to tweak the polynomial to match reality and are less restricted on the number of terms. e.g. a 7 order polynomial is typically going to be impractical for a control system due to the computation time, but with a lookup table the computational time doesn't change, whether it's 3 order or 22 order. You can pre-calculate them then depending on your interpolation algorithm and number of points you can calculate the peak quantisation error and determine if this is adequate for your system. The other advantage is you can make them part of a tune map, so you can set this as a separate the behaviour from the code, this makes the code easier to test and allows you to update with just downloading a tune, rather than modifying your code, testing your code, rebuilding, testing again. If you set the system up correctly you can also change it while it is running! ;) These add up to allow you to speed up the control system. 4th, as you are talking to the controller and moving the actuators, it's a HIL (Hardware In the Loop) test, SIL is purely on the PC. 5th, don't put code in header files, bad practice! ;) I know C/C++ build systems can be tricky to understand, but don't do it! 6th, if the ally machining is costing too much time, maybe look at forged carbon fibre, you can 3d print the moulds and may give you the strength required without the weight penalty and possibly easier production.

[@-coolerlegothings-9784]

I love the part where the rocket says "It's landing time" and proceeds to crash 4/5 of the time

[@redpug5042]

it doesn't specify a soft landing :D

[@braydonburkhardt]

💀

[@DrRusty5]

It still landed, just not pointy end up!

[@gustavkrklec2077]

@@redpug5042 q

[@gustavkrklec2077]

-

[@johngrimble3050]

Use a tapping post or arm for tapping threads! Basically mount the tap to a rod that can only spin and translate, but not side load. Free hand side load will snap a tap.

[@MinequestLeon]

As a fellow rocket hobbyist, I love these videos. Please do continue sharing information 📖✍️

[@X_Elyjo]

This is very inspiring , and we learn a lot for free . Thanks you Joe

[@Galerak1]

I love how you made those parts out of aluminum instead of aluminium. It makes perfect sense, less letters means less mass and less mass means more better 😁😂😋

[@Craftlngo]

if you keep having problems with tapping fluid. You should try brake cleaner. It's specifically designed to remove anything oily or greasy (you know, everything that could hamper with braking). It vaporizes completely, leaving a bone dry surface.

[@michaelberry950]

Joe, thanks for sharing the wonderful working of your mind with us, your thinking and process are so wonderfully precise. Truely inspirational!

[@JohnOCFII]

These engineering cut videos are excellent! So thorough and so well presented. Thanks for taking the time to create and share them

[@trailblazingfive]

Rainbow LEDs in a CNC look amazing, love the reflections

[@NASHThoughts]

I don't know what to say...but literally I'm stunned...You are a real inspiration man.♥️

[@Aldo.flores]

I love this long and over engendered videos about the design of each rocket and all the things you have to consider for the design

[@symtrial]

as an engineer, I salute you sir! This is by far the coolest project I've ever seen

[@flower-space]

I watched this video when making my flight computer. I started making pcb for it at the beginning of this video. When the video ended, i also ended making pcb :p Perfect timing :D Thanks for the support Joe!

[@adamjely]

The World needs more thinkers like you! Love it!

[@porfiunratitomas5429]

Best educational resource out there yet, pls do more.

[@AlexTaradov]

That is the most creative Tormach lighting I've ever seen.

[@darrylknowles2224]

Love the deep dive! It landed and I've been cheering for you the whole time.

[@AS_Nov.1917]

These engineering cut videos are my favorite on your channel

[@passionvinetea]

Taps break. Bits break. After tapping 96 M5 passthrough holes, 1 drop of oil on the tap threads at the start... and 1 drop of oil halfway through are all that was needed. Also, if it is a hole that is NOT passthrough, you have to clean out the chips often during tapping. For smaller taps, never let go of the tap wrench. Also, secure the part being tapped. Our hands move around alot, each movement risking lateral force on the tap... breaking it. Clean the tap often. But, the patience and slow movement as SOOOOO worth the result. My next step is to learn how to let the CNC tap holes. In 3D printing, parts are created by adding layers of material, whereas CNC routing/milling involves removing material, which requires a completely different mindset. I am three weeks into learning milling, and I've experienced many broken bits, prompting me to understand why each bit broke and solve the issues. Titan videos have been helpful, but applying First Principles and systematic thinking to the machining process has been particularly beneficial.

[@juzores1]

Fascinated by his work and equally fascinated by his flawless baby skin.

[@matterwiz1689]

Bold of you to assume that I am not willing to watch this stuff for seven more hours

[@lesumsi]

Really impressive! Love the detailed explanations! But, I want the 7 hours landing legs supercut!

[@ptonpc]

Love these longer videos, that's my day sorted.

[@behradkarimi5730]

Hi! You're absolutely wonderful! I adore the fact that you work so hard without letting your failures get in the way of advancing in your life. I'm simply in love with rockets, and I've been following your channel for several years now. I started making solid fuel rockets when I was 15 and you have always been my role model. I'm sure one day I'll be like you. Unfortunately, it's forbidden to make rockets in my country unless it's under the supervision of the military army. Now I'm 18. Unfortunately, I have to focus particularly on my school due to the university entrance exam I have to take this year. But I will start again when I got into a good university and then I may be able to immigrate and continue my research and work widely cause that's what you motivate me to do, to never give up! I wish I'll run a KZbin channel just like you and show all the practical steps of my rocket-making. I wish for the day all our wishes turn into reality. 🚀

[@patrick1532]

1:19:30 My theory on what happened was: 1. Bit creates massive thermal gradient at cutting interface: Aluminum is very conductive but the spindly shape of that part didn't leave much of it for that heat to escape through. 2. Due to that massive thermal gradient the aluminum disproportionately expands *inwards*, clamping down on the sides of the bit and applying enough friction across that surface area to pull it from the chuck. Note that 6010 has close to twice the thermal expansion coefficient of steel. 3. Since the deepest the bit managed to drill was achieved at elevated temp, the bottom-most portion of the bit forms an interference fit with the drilled shaft as the work cools and contracts, fixing it in place.

[@CDRaff]

1:35:15 FPV pilot here: They are the "RunCam Hybrid 2". Also just an FYI you could replace this with a "RunCam Split 4" which is 8g lighter(only has one lens) and it shoots in 4k30. It also uses less power, and for sure can be run off a 5v rail.

[@michaelcullen4214]

The Split 4 will only do 60fps at 1080p though - he's running the Hybrid 2 at 120fps. I might be misremembering, but I think he's used the Split 4 before, but complained about it only being 60fps

[@tehllama42]

Two hour video? MAXIMUM ENGAGEMENT

[@fbwa32nxfan29]

Can't wait to see what comes next!

[@realvideosrv1879]

As an engineer doing a master’s in control engineering this video was awesome to watch! Although some things where also very painfull like the simulink model ouchh 😅. It must not have been easy to try to model and or estimate all the dynamics of the vehicle as good as possible. Ive been following your channel for a while and it seems to me that you are a self taught control engineer ;p impressive too. Keep it up mann 🎉

[@alexanderkuhn2298]

Bless you for releasing this information for people to learn and be inspired, you and your team have clearly put a lot of work into this

[@happycoachflyer]

This was a fascinating deep dive from a historical perspective. Thanks for posting it.

[@franklinegbuche7097]

This is a brilliant video to make. Lots of people have things to learn from in it.

[@rswa]

Software nerd here. A tip for your code; you should always have arbitrary values defined somewhere as constants, even if it may seem obvious what the value represents in your code. An example in your project would be under GNC.h line 373 (at time 49:00), where you have a threshold of "6" in the if statement, but it would be better to write this line as `if (retroCurrentAccel > RETRO_ACCELL_THRESHOLD) {`... and at the top of the file you would have `#define RETRO_ACCELL_THRESHOLD 6`. There are a few reasons: - If you use a specific value in more than one place, then both will always reference the same constant, and when changed they stay synced. Even if you only use it in one place, you could expand your code in the future and reuse that constant. - All your hard-coded constants and values are very easy to locate and change (usually near the top of the file). Plus they are grouped together. - If it is not obvious what the value represents, the names of the constants themselves might help you understand. - You can give a bit more insight into the value itself by adding a comment where it is defined. This way comments in your code can stay focussed on functionality etc.

[@abubakersiddique761]

Super Building Process Bro, that's why everyone likes your KZbin channel. Thanks for Teaching Rocketry for us ❤

[@tilhoff3947]

Thank you for the huge amount of information you share here! You might want to have a look at a crimp tool for improving the servo connection cable length. It is quite cheap, easy to use and allows you to fit the cables to the exact length you want.

[@gregkrekelberg4632]

This video was just _epic_! I have two possibly helpful suggestions: 1) Avoid the pow() function and refactor polynomials into Horner's form. Matlab can do that. The pow() function can be really slow, and with the number of polynomials you use, it'll pay off in the main control loop. Horner's form reduces a polynomial into a set of nested monomials; you have nothing but addition and multiplication. Importantly, it minimizes the total number of operations. (Even if pow() is smart with integer powers, Horner's form still minimizes total operations.) 2) For integration, do a little research into 'design for manufacturability'. In your case, you'd be looking to design the ability for removable integration fixtures into the main assembly. Think of a very sparse frame that you could use to insert and align a full assembly into the flight body which you can remove to complete the rocket. These could be pushers, pullers, spacers, or whatever makes the most sense. Everything you design for a rocket doesn't need to fly on it. Nothing needs to be elaborate, but if you plan ahead, it'll make integetration go a lot easier. Thank you for sharing this. You are an absolute animal of an engineer.

[@dfgaJK]

I love your CNC enclosure RGB. I will have to copy that idea! I bet you could configure it to change depending on the machine state (tool change wait, crash, finished etc)

[@index7787]

I have used this idea, it has saved me from so many boneheaded CNC errors. Among other things, I have the speed modes for manual jogging light up red/yellow/green. It's saved me so many probes XD.

[@dfgaJK]

@@index7787 what cnc software? What rgb controller?

[@metalworker3]

Been looking forward to this nuance vid for a long time. Thanks for the time to make it!!

[@jusufsimon1993]

Finally Home alone 7 is coming : lost in space. From actor to engineer, best of you Macaulay!

[@Bizones16]

Love the work you are doing. PLEASE keep growing with this and keeping us informed. I'm downloading all of your posting and have created a filed for them. going to use them to get other young peopl interested in some type of flight ( I'm a commercial pilot).

[@AstroHopefulBenjaminGottsch]

Joe, I'm gonna be honest. I'm already watching a nearly 2 hour rocket video. If you ever post a 7 hour engineering cut I would absolutely watch it.

[@stingraymaster877]

The forbidden lightswitch in the background😂

[@arthurmiller-vl6sw]

@18:00 Surveyors equipment like levels and theodolites all use 3 footscrews to level for that reason. In the 1800’s they used to have 4 and was a nightmare to get all level without play or tension in one leg. And then you immediately say the same for your design, lol👍🙂

[@mocco4550]

You made it looks like too easy to do it😄. I prefer to have a video series where you talk about every section of the project in details, means 3D, CNC, Programming…..etc. you can use the whole 7 hours for that. Keep going and good luck 🤞

[@Galactis1]

I love this, I love the behind the curtain. This is really interesting for us, and allows you to document who this previous rocket was built. May give insight for the future. Very well done. I am just learning how to machine and use BobCad software now. And, I am a business major, not engineering at all.

[@jack_brooks]

A couple of suggestions; I do a lot of M3 hole threading in sheet aluminum. I prefer the spiral taps, example, McMaster 2673A71. They clear the chips better because they get pushed out by the spiral. I find this makes for less build up and therefore less likely to break. Also, a cheap ultrasonic jewelry cleaner would probably help with cleaning the tap oil problem.

[@gilbertanderson3456]

Don't you think a less polar solvent like heptane or even kerosene would degrease better than acetone?

[@BuffMyRadius]

@@gilbertanderson3456 I like using gasoline because it's cheap, contains all sorts of fun detergents, and I like to live dangerously

[@howlingwolven]

@@gilbertanderson3456Perchloroethylene. In the old cans of brake cleaner.

[@pacukluka]

I just watched the whole thing. This is crazy awesome!

[@dipanjanmandal1771]

Nice to see the video many times later Thanks for the video and idea sir

[@aerospaceengineering3298]

NICE VIDEO WELL DONE YOU'Ve worked so hard for this and at the end you got what you always wanted best wishes.......!

[@Trump_y_Gore_Won]

Thanks Joe! I knew a professor at UTEP (he was very mean, old, a philandering old fart, and he hated me) but he knew Wernher von Braun. I believe Dr. von Braun was in El Paso, Texas at some point, doing work there. The professor said he was very nice, and (I quote), "he was not a stuffed shirt." I don't know how someone could gather such expertise and become so distinguished, even among other scientists, but I suppose anything is possible, if the desire and curiosity is strong enough, and if a sound scientific basis is the foundation of all activities.

[@FlmngHrs]

My short attention span lasted for only about 15 minutes, but I loved it. Super interesting.

[@JaidenFloyd]

Nice job! Can’t wait for more landings

[@kilo_kilo]

popcorn ready for this one.

[@46performance92]

I tried to scroll in your premier time line several times and then I felt so with you - waaaaaaay too much footage :) but I tell you - better make it longer like you did than to cut out every thing, great video!!! and awesome project!!!

[@46performance92]

and what most ppl do not mention, if you cut a video that is 2h long, that means only final viewing will take 4+ hours! and looking over and over again your own work can be challenging!!! i wonder how long it took you to cut the video?

[@46performance92]

an idea for your arming switch would be a reed switch with a magnet attached as long as it is disabled, and on the start pad you just remove the magnet, its light and you can add a cool "remove befor flight" tag :)

[@OrangeDurito]

Excellent video! I can't put it in words how much I appreciate you making this engineering cut. Any technical person would crave for such an in-depth explanation for any system they are interested in. Also, the breadth and depth of your knowledge is astounding and it is evident from all your videos. With that being said and given that I spent the lovely evening after a rainy day inside my room hooked to this video throughout, I have a few comments/observations I would like to point out (some out of curiosity, but mostly for amusement since the explanation was very comprehensive) - 1. System Identification - You did talk about using system ID for the TVC mount (at 1:21:50) but how did you get the dynamic model for the entire rocket? In the simulation section, you had a 3 DOF Body Axes block and you measure the Mass Moment of Inertia using Bifilar Pendulum method and you did say that for a larger size, you can get away with a slightly slow control system. Do you think you would benefit from black-box system ID for the entire vehicle going forward for more mechanically complicated system? 2. 1:05:45 - It took me some time to understand when you said, since we are actively controlling the descent, we can compensate for the throttle error. Had to refer to the Simulink diagram to see you have the altitude feedback connected to PID controller to generate the throttle command. Why wasn't the PID block connected...oh right, you did the SITL simulation for that one. While the PID and the full-state feedback controller has worked great (they do a great job in most of the scenario), do you see yourself using any other non-linear control algorithm or adaptive control in the future? Just curious. 3. 1:36:30 - Thank you for talking about the power system for flight avionics. I just had realized earlier that I never saw the battery in the CAD model nor in the previous videos (I'm sure I missed it). 4. 1:23:30 - Can I edit your 6 hours landing leg development footage? Pass us your learning good sir. 5. 1:09:30 - The axes labels are inverted in the graph :) 6. 48:40 - Variable naming is an art. 7. 1:04:40 - Joey B with mustache is umm...and the camera angle did a good job on concealing your haircut disaster on the right side :P 8. 1:10:12 - That Digi-Key ruler though! 9. 1:22:31 - Good job on keeping the t-shirt consistent on subsequent recordings (the next two days) and from footage name (which shows the time there were shot), I can safely deduce that you are having your lunch at the right times, so thank you for keeping yourself healthy :) THANK YOU FOR EVERYTHING JOE! Your impact on younger (or older) generations cannot be overstated. Keep going.

[@kylejacobs1247]

The grippy pad you added to CAD is the same thing that is done with SPD bicycle cleats. They have nubs that when screwed tightly dig into the bottom of the shoe.

[@PrinceKumar-hh6yn]

This is supercool. Now u r more Engineer than a KZbinr❤️

[@kleinesfilmroellchen]

10 hours of building landing legs sounds like the kind of meme video that works well on KZbin

[@sanjivvkhanna1498]

One suggestion when you drill into some metal put continuously flow of water or inflammable thick oil for good friction and less breaking and load of bits.

[@0xElias]

love it, as a technical product designer

[@Said.Qadeer]

One man army keep it up sir! That amazing

[@ke9tv]

Rather than all those pow() functions, write your polynomials as a+x*(b+x*(c+x*d)) - for a+bx+cx²+dx³ - both faster and more stable numerically. This will be especially important on your flight software - pow() is slow and power-hungry as well as having numerical issues.

[@retronexusnet]

Please more of this series!!!!

[@thibaultschweitzer6717]

Amazing video mate, truly impressive!

[@LeiDuGodHamsterOfDune]

I know you fixed your TVC mount, but you may want to try to dither your output control signal to help remove some of the issues with backlash and general imperfections in the mechanism.

[@SkylersRants]

Fascinating and very impressive!

[@peterknoll8906]

I just found your channel! It's fantastic! Thank you for sharing your knowledge and experience!

[@picoplanetdev]

Loved the in depth look that this video provides on the work that you put into this over so many years! I'm not sure if it's just me but I think there's a slight echo in your audio, maybe consider some cheap sound absorbing panels or DIY them.

[@handsanitizermk.268]

R.I.P. fly great Video

[@adrxian]

use your drill jig for tapping, it would work great and save a ton of broken taps

[@dfgaJK]

Have you considered the accuracy of the gimbal response transfer function due to the source of the measurements being from an unloaded gimbal with less friction?

[@bjf10]

Great stuff, thanks for all the detail!

[@iwayini]

Great video summary. Thank you.

[@isakhammer6558]

oh! So that is how it sounds with a good mic and a room with echo!

[@oneaerospace6273]

Nice! Just what i wanted!!!

[@persistenceovision]

Nerds would tell you: throw the breaker & bypass the switch & install a blank on that wall plate. 😂. Just kidding. Love following your progress over the years. Well done!

[@danielberkovich2964]

You really should bring this all to maker fair in New York and demonstrate it there!

[@awesomeavionics1342]

I like how you half to trick Ava using Ava

[@Zl2mst]

Nice video and background info on decision process and evolution ...percusive persuasion excellent .......

[@nicksavage4763]

Well thought out

[@johnymodem1326]

Another great video and excellent detail and explanations. What next? Very much looking forward to your future exploits. Keep up the great work. Ad Astra! :)

[@lievenpetersen]

47:26 I'm sure these are very scientific numbers :D nice

[@heavi-armed-infadel]

I love it! Awesome job!

[@clay_lowery1]

Good luck going forward🙂

[@tuongquangphung69]

Thanks so much for your sharing!!

[@ke9tv]

You could probably get the higher supply voltage to the camera with a cheap boost converter for less mass budget than another LiPo. Just a thought.

[@alclaton9879]

Space x should be looking for you, nice work

[@porfiunratitomas5429]

you should do one of these for sprite if you ever come back to it

[@bradleywhitefield]

love this!

[@ZealChannel]

LOT OF MAGIC NUMBERS!😅 Refactor is calling you

[@eisomouwen4767]

love this

[@augustvoegeli2563]

isopropyl alcohol is a great lubricant when tapping holes in aluminum.