Відео

Hi there! Thanks for your elaborate comment and the compliments. The approach taken for calculating the area per rectangular cross-section sounds good. I don't remember exactly why I used nStream=30, probably I was just experimenting. One remark is that the computation for the reynolds number might be different for a circular and a rectangular cross-section, but this you'd have to look up somewhere. I've forgotten most of my fluid dynamics courses ;)

@@kloppertje Thanks for the reply! Im glad the 30 value was just experimenting, took me a long time to figure out why it was there. As for the reynolds number, yeah there is a separate equation for rectangular cross sections, but I cant figre out the wall thickness of the infill so I cant calculate it. Im a fourth year mech. eng. student, and this project is really helping me out so thanks again!

Hello again @kloppertje ! I'm currently working on a design project for my course at school and Im using a similar design to the nozzle you have in your videos. I've also modified the python code to better fit my project. I just wanted to thank you again for both the video and taking the time to answer my previous question regarding the python code! Its been nice virtually meeting you, and I hope you have a wonderful rest of your day :)

Thank you :)). My plan is to sell some lamps by the end of this year, either as full product or as a kit. I'll post a video at that point

Nice! Do you have a link?

Yes a mobile phone can be used. The robot connects to a wifi network, and the robot can be controlled by accessing a web page

I appreciate your work! I'm starting to build this robot, I'm absolutely blown away. Thank you very much my friend!

@@Technical_stuff86 thank you and have fun :)

Yes that's how I do it, see elexperiment.nl/2018/11/high-speed-balancing-robot-introduction/ For the new robot the outer loop isn't implemented yet as there is no position sensor in the wheels

Thanks for mentioning; I'll try to push the latest code to the git repository this weekend. The code isn't spectacular (on purpose); I simply turn on the PD controller while the robot is lying on the ground. The code is here: gitlab.com/kloppertje/balancingrobotdc/-/blob/devel/src/main.cpp?ref_type=heads#L159 but since this video I made some changes to it regarding self righting. I'll notify you if I push the code

@@kloppertje just another question, so if all you are doing is turning on the pid controller while the robot is down, why does it not automatically self right after it falls if you don't turn off the pid controller

@@abdulwaheedouabihi5903 To be exact: I simply disable the motors, the PID controller keeps running in software while the robot is lying down :). There is a state machine in the code. If the angle is below a certain value, the robot is probably lying flat. The robot is enabled either when it is near vertical (so, someone put it upright), or when a self-right command is received

@@abdulwaheedouabihi5903 The version of the code used for this video is now pushed, see gitlab.com/kloppertje/balancingrobotdc/-/blob/devel/src/main.cpp?ref_type=heads#L190

@@kloppertje thanks

Thanks for your reply! I did a quick experiment with this; a quite large capacitor would be needed. A solution could be to have a diode followed by a buffer capacitor to power just the ESP32. However, this would be a "patch" instead of solving the underlying problem (voltage drop of the source); I'd rather solve the underlying problem than patching it. Experience has learned me that stacking patches instead of solving root causes leads to unmanageable complexity.

You might want to supply the microcontroller via a separate power supply with a filter for conditioning. Inductive loads, like motors, can cause big voltage spikes, positive and negative, which can play havoc with digital electronics. Cool project. Good luck!

@@tolkienfan1972 I'll keep it in mind and test for voltage spikes when making a PCB for this project. Thank you :)

A small diode will probably explode (been there done that). I believe that, if the LCR circuit is designed right, the current will have an "overdamped" shape i.e. the current will not become negative. You could simulate (for example in LTSpice) your circuit to see if this belief is correct

@@kloppertje Mm...yeh I kinda thought same , like using some method so the capacitor would not discharge completly , for example using some kind of fuse in series that would blow up and open the circuit before reaching 0 v , that would make me feel more confident

@@rodriguezfranco3839 I don't fully grasp/remember the workings of a spark gap (as used in this video); perhaps they stop conducting electricity when the current goes through zero? Although this is not good advice, I have never had an electrolytic capacitor explode with my setup 😅 And please, please be aware of the dangers of a setup like this; it's no toy

You are right brother , I am sorry I forgot about the sparkgap , I was thinking about some other chanel I saw wich was manually shorting the gap with a chicken stick , yeh it would stop probably before 0v

Thanks for the comment :). My idea was that the resistor works as a voltage divider. I have *assumed* the protection features are still active but have not tested this (shame). This is bad design; however at the moment I'm using a protected 18650 cell so this doesn't worry me. For later on, when I'll be using an unprotected 18650, this'll be definitely relevant. A second pair of MOSFETs would definitely be a solution; would even be possible on a small footprint :).

Hello, good to hear you're still having fun from the robot :). For setting up a compliation environment, see gitlab.com/kloppertje/balancingrobot/-/tree/ps3control/Software?ref_type=heads For implementing something with an ESP and a joystick, I'd recommend first testing the code on two ESP boards without all the balancing robot code, then when it works integrate it into the balancing robot. That shouldn't be too hard (once you get the compilation environment up and running), the code has a "remoteControl" struct with which the robot can be controlled. gitlab.com/kloppertje/balancingrobot/-/blob/ps3control/Software/src/main.cpp?ref_type=heads#L897 Let me know if you make something and perhaps I can integrate it in the code

Thanks for responding so fast! I am gonna try getting it to work!

Hi, tuning the robot requires some trial and error. Also, regarding your other questions, the documentation can be found in the gitlab repository, my website, and the youtube videos. You will require a certain amount of problem solving yourself. This is a free, open source project (and quite some time goes into making things open source); it is not a complete guide on how to build a robot.

gitlab.com/kloppertje/balancingrobot/-/blob/ps3control/Software/src/main.cpp?ref_type=heads#L154

@@kloppertje I'm also not able to connect to wifi by using my credentials in html (although esp32 as AP is visible to me) Its 4am in my country btw🥲

i'm using tmc2208 please tell me whats the microstep resolution i should choose, my robot seems to work at 1/4 but u r saying its 1/16 ?@@kloppertje

Good question. The stepper motors are always on and yes they do consume quite a bit of current. A 4 wheeled robot could be made more efficient. With the battery used in this video battery life is > 5 hours so still pretty good. Also, I'm working on a version that doesn't use stepper motors but DC motors, resulting in a much lower power consumption. From some first tests it appears you can have a battery life of say 5 hours on a single (cheap) lithium cell :)

The STL is very simple: it's just a cylinder, and the mesh is the infill generated by the slicer :)

The biggest version I've used is I think the 17HS4801. Later on I switched to the 17HS4401 and then 17HS3401, because they are cheaper and lighter and apparently still have sufficient torque.

@@kloppertje thank you for the answer, do you have any resources that could help me make a web I terface for the esp32 so parameters can be modified on a servers.

@@arbymohamed6136 I explain some things that use the web interface in this video: ua-cam.com/video/_FOYB26Yeow/v-deo.html In the description of that video is a link to my git repository, which contains code for live setting of parameters using html and websocket servers running on the ESP32

@@kloppertje thank you very much. I have another question. While reading your code I saw that PID_ANGLE_MAX 12, how is this so low while the maxstepspeed is 1500. I know there something I'm missing. Maybe the output of anglepid is multiplied by some factor somewhere in the code ?

@@arbymohamed6136 if I remember correctly it's because the angle pid controller outputs acceleration

That might work. Do you have an idea how to keep the sensor parallel to the ground, while the robot is tilting?

@@kloppertje do you need to? You only need accurate measurements while upright. If tilted, ignore the pos sensor

Haha yup, I've added something in the title. Thanks!

@@kloppertje love your vids man keep it up especially the high speed balancing robot very impressive

My thoughts exactly :)

Might need a 3rd "passive" wheel, though, like in a ball-mouse.

Yes you would need a construction to keep the ball in place. Another solution would be to actuate the bolt linearly. Then you would be able to steer and balance in 2 directions.

Thank you :)

drive.google.com/file/d/1htTO05NlbKhhiemlGtUqAxqWo7ozHWwT/view?usp=drive_link

Thanks, will try it in the next iteration!

Correct! When power is disconnected, or when a fault situation occurs (for example, over voltage), and after every shot (in case the spark gap doesn't trigger the main charge), the bleeder resistor is active.

@@kloppertje Wow, thanks! That's a very beefy resistor then :P. Necessary of course if you want a quick discharge.

Lol, good question :p