Видео

It is way out of balance, the shaft adapter, the motor with the wires hanging off of it, they all need to be balanced and any force applied is going to induce procession. My favorite gyroscopes are aircraft attitude indicators and vcr tape motors. Aircraft attitude indicators come in different voltages and phases, some require 120v 400hz 3 phase while others need just 12v dc. A good one will run nearly silent all day and all night. They can be found on E-bay for $25.00 to $1500.00 depending on the age and quality. They also come vacuum powered, it takes a little more time to spin them up. The VCR motors can be found cheap at thrift stores or even for free on Craig's list or Offer Up. I've got about 6 different attitude indicators, a slip and turn gyro, and a directional gyro, they are a bunch of fun. On the inside they use an interesting set of slip rings and very thin contacts, everything is precision balanced and built with every other component in mind on how it will affect the balance of the unit.

thanks for the vid friend appreciate it

There is also the issue that if it is concistently falling in one direction, the gyro will fall to 90 degrees as you are essentially changing whats gaining kinetic energy.

A good looking watch. Dial, hands and case shape inspired by Seiko snxs79.

great project!

Big fan

Kp

Here is complete Arduino code with huge comment section to explain setup and details. Hope you make one! // jim demello jun 2024 - short Arduino code for SBR (self balancing robot) using 48:1, 150 rpm yellow motors with no encoders // majority of code taken from midhun_s www.instructables.com/Arduino-Self-Balancing-Robot-1/ // // hardware: 1.a. L298N motor driver - be sure jumpers are on enable pins or will not work. These drivers are cheap, I had a bad one // so buy more than one. // 1.b. Arduino Uno - also had a bad one where one wheel would only turn in one direction. Replaced Uno and it was good. // 2. I used two potentiometers to tune targetAngle and Proportional term but you don't have to, but it makes life easier to do so // 3. MPU6050 - be sure you calibrate it, don't use someone elses values, like mine. // MPU6050 - be sure you mount it with the X axis parallel to the axis of the wheels. // 4. power - I used 2s lipo to power L298N. So a fully charged 2s lipo, 8.4 volts will deliver about 6.4 to 7 volts to motors. // And I used rechargeable 9 volt battery to power the Arduino Uno when not connected to USB. Lasts about 15 minutes. // 5. motors - 48:1 geared yellow motors, 150 rpm at 6 volts. No encoders required. Encoders will be necessary to // control tracking and fast robot response. Will do that someday. // 6. wheels - larger and heavier wheels make for better control. 65mm are ok but 80mm are great. // 7 counterweight - mount lipo on top of SBR for better control - SBR falls over slower with weight on top - // larger wheels and counterweight can compensate for slower motors. // // Notes: 1. be sure wheels are turning towards direction of lean. // 2. do not use low baud rate for Serial which will result in a timing loop too slow to work. // 3. according to this article: polyengineer.wordpress.com/2014/08/08/self-balancing-robot-pid-control/ // you should use 5 times Proportional term for Integral term value. It works. I use up to 7 times. // This prevents drifting (except on uneven surface - need encoders to stop that). Not using Derivative so this is a PI controller. // 4. targetAngle is another term for setPoint. // 5. don't use delay() - it will mess up the timing. // // My opinion: 1. an inverted pendulum like an SBR has a natural frequency that needs to be matched by the timing of the loop // and the Proportional term. If P is too high then it becomes a bang-bang machine, on and off - effective // but ugly. The timing loop must be consistent for good balancing or it will jump all over the place. // So print out how long the loop takes to determin its Hz. A Hz of 50 to 60 is ideal - about 20 milliseconds. // Be sure all of your code is inside the timing loop or strange things can happen. // Timing loop can run from 30 down to 5 milliseconds. // 2. The high Integral value (5 times P) prevents drifting. // 3. This is the shortest, simplest and most effective code to self balance a two wheeled robot - I think. // 4. This robot will balance and stay in place pretty much on a smooth surface like a stone floor. Most robots // with these motors will not balance on a stone floor and/or they will drift off and fall over. // They will balance only on carpet or rough surfaces. I think it is due to bad timing loop. // 5. MPU6050 Complementary filter code here is effective, maybe more effective than MPU6050 DMP values which seem to jump // around more than these. Haven't tried Kalman filter - looks too complicated for me. // 6. The motors can be optomized by matching the point at which they both begin to move at the same time by // adding or subtracting from one of the motors motorSpeed values. Haven't done that here as I wanted the code // stripped down to it's essentials. // 5. A great thanks to midhun_s. // 6. I started work on this robot 6 years ago and finally got it working to my satisfaction this year. // I accidently had it working 6 years ago on a smooth surface but never knew why. Now I think I know. Never give up. // 7. Opinions are overrated by their holders. // #include "MPU6050.h" // vars for mpu6050 // int16_t AcX,AcY,AcZ,Tmp,GyX,GyY,GyZ; #define leftMotorPWMPin 6 #define leftMotorDirPin 7 #define rightMotorPWMPin 5 #define rightMotorDirPin 4 float sampleTime ; MPU6050 mpu; int16_t accY, accZ, gyroX; volatile int motorPower, gyroRate; volatile float accAngle, gyroAngle, currentAngle, prevAngle=0, error, prevError=0, errorSum=0; ////////////////////////////////////////////////////////// volatile float Kp=25,Ki=0,Kd=0,targetAngle=0; ////////////////////////////////////////////////////////// float adjustAngle = 0; unsigned long currentMillis; long previousMillis = 0; long loopTimer; #define PRINTLN(...) Serial.println(__VA_ARGS__) void setMotors(int leftMotorSpeed, int rightMotorSpeed) { if(leftMotorSpeed >= 0) { analogWrite(leftMotorPWMPin, leftMotorSpeed); digitalWrite(leftMotorDirPin, LOW); } else { // if leftMotorSpeed is < 0 then set dir to reverse analogWrite(leftMotorPWMPin, 255 + leftMotorSpeed); digitalWrite(leftMotorDirPin, HIGH); } if(rightMotorSpeed >= 0) { analogWrite(rightMotorPWMPin, rightMotorSpeed); digitalWrite(rightMotorDirPin, LOW); } else { analogWrite(rightMotorPWMPin, 255 + rightMotorSpeed); digitalWrite(rightMotorDirPin, HIGH); } } void setup() { Serial.begin(115200); // do not go low as in 9600 or timing loop will be very slow // set the motor control and PWM pins to output mode pinMode(leftMotorPWMPin, OUTPUT); pinMode(leftMotorDirPin, OUTPUT); pinMode(rightMotorPWMPin, OUTPUT); pinMode(rightMotorDirPin, OUTPUT); // initialize the MPU6050 and set offset values Serial.print("Initialize MPU at: ");Serial.println(millis()); mpu.initialize(); int con = mpu.testConnection(); // test MPU6050 - I had some problems so this was necessary for me PRINTLN(con ? F("connection successful") : F("connection failed")); mpu.setYAccelOffset(2067); // from calibration routine - use your own calibration values mpu.setZAccelOffset(1207); mpu.setXGyroOffset(4); Serial.print("End Initialize MPU at: ");Serial.println(millis()); delay(1000); loopTimer = 15; sampleTime =0.015; } // end setup() void loop() { int noMillis; currentMillis = millis(); noMillis = currentMillis - previousMillis; // main timing loop if(noMillis > loopTimer) { targetAngle= float(map(analogRead(A0), 0, 1023, -600,600))/100; //Kp = map(analogRead(A1), 0, 1023, 0,800); Kp = 60; Kd = .00; Ki = 500; // read acceleration and gyroscope values accY = mpu.getAccelerationY(); accZ = mpu.getAccelerationZ(); gyroX = mpu.getRotationX(); previousMillis = currentMillis; doPID(); motorPower = constrain(motorPower, -255, 255); if (abs(error) > 15) motorPower = 0; // if fall over then shut off motors Serial.print("Kp= ");Serial.print(Kp); Serial.print(" currentAngle= ");Serial.print(currentAngle); Serial.print(" error= ");Serial.print(error); Serial.print(" errorSum= ");Serial.print(errorSum); Serial.print(" adjustAngle= ");Serial.print(adjustAngle); Serial.print(" targetAngle= ");Serial.print(targetAngle); Serial.print(" noMillis= ");Serial.print(noMillis); Serial.print(" motorPower= ");Serial.println(motorPower); setMotors(motorPower , motorPower ); } // end timer } // end main loop void doPID() { // calculate the angle of inclination accAngle = atan2(accY, accZ)*RAD_TO_DEG; gyroRate = map(gyroX, -32768, 32767, -250, 250); gyroAngle = (float)gyroRate*sampleTime; // complementary filter /////////////////////////////////////////// currentAngle = 0.9934*(prevAngle + gyroAngle) + 0.0066*(accAngle); ////////////////////////////////////////////////////////////////// error = currentAngle - targetAngle + adjustAngle + 0; errorSum = errorSum + error; errorSum = constrain(errorSum, -250, 250); // clamp Integral //calculate output from P, I and D values /////////////////////////////////////////////////// motorPower = Kp*(error) + Ki*(errorSum)*sampleTime - Kd*(currentAngle-prevAngle)/sampleTime; ///////////////////////////////////////////////////////////////////////////////////////////// prevAngle = currentAngle; }

Ik denk dat hij begint te wippen omdat de magneten niet correct180 graden van elkaar verwijderd zijn, mooi ontwerp

Stand it vertical and replace the hand with a head and it is also a great neck mechanism

Cool watch, congrats!

Did u still have the code for his project ?

Crap

Nice watch and thank you for the review. I noticed that you are a very patient man, but I have to say that your woman is disrespectful toward you.

Nice work!! I am really fascinated by gyros too. I am going to attempt this with a brushless torque motor and encoder. Also I will invest in a dynamic balancer.

This video doesn’t show any repairs.

Thank you Jimmy for the video! I was having issues finding out why my motor wasn’t running and it’s all because of the VDD pin. Nowhere in the data sheet did it say I had to connect the VDD to the 5V pin on my Uno. I was stuck for several days on this issue. You have my greatest thanks!

Робат палесос

Very cool. I'd at least use an USB hub to power those servos. You notebook might not burn out right away, but those servos can pull a surprising about of current. Thanks for sharing this video. I really enjoyed it.

Sir have you tested on an inclined surface

Awesome, thankyou for your kind generosity with all you share, many hours research, thankyou, all the best

What motor are u using

有難う御座居ました。

Very cool, thank's for this Video!

I like this in white but not sure i want another chinese watch.

🥺🤲 i love it, thank you for this video! ❤

Did you machine your own flywheel or did you purchase one? Also, what motor are you using?

There is also a kit from these people: ruclips.net/video/pbCuqW50T-M/видео.html

Very nice. It’s what you see with your eyes

It it possible to keep the HDD in its case? This way it's safer, and one can take advantage of the vacuum inside for more efficient operation. As long as one can control the RPMs through the HDD controls. I am thinking to try this with a self-balancing monorails.

How do you make the motion smooth?

ruclips.net/video/kUYzuAJeg3M/видео.html&ab_channel=PrimalSpace

It's so cute!

Have had for couple weeks now and found that the mat finish on the band scratches and polishes into a shiny patina. Had to adjust the number of links and reduce the ones to the left of the clasp so that the clasp is centered on the bottom of wrist or else it tends to bite into my arm if under any pressure. Still, very comfortable and my favorite watch now of my collection.

This is the most stupid video in the world ! Man you lost my time to watch your stupid watches! Don't do video again, please!

This will never be self balancing.

Merry Christmas to you and Angie

Just an update on having this watch for a year now: one time it got tangled up in my coat and the band became unlatched and the watch fell and hit our stone floor. Put a notch in the case but crystals and rest of watch intact. I still like wearing it but got a nice nylon band which helps it wear better and more secure. Wish they had beveled the top of the case then the damage would have been less. My most colorful watch for sure and it is a pleasure to have it.

Love the interaction with your assistant .

I got here by searching on your unicycle curb mounting. I learned to ride a unicycle the same time as you. Only you were 55 and I was 65. If you wear enough protection it's pretty safe. I am 78 and the only problems I have are static mounting. I use to be good at mounting but after my wife died I stopped riding for a year and then was unable to static mount anymore. I am working on a training device for that. I just wondered if you were ever able to static mount. We do a lot of the same things. You should look into getting a 3d printer. It's good for making projects. You draw better than me. Play instruments better than me and juggle better than me ha ha. But I have tried all of those things plus a lot more over the years. I am working on a bunch of projects and have a habit of adding too many of them. I was working on a radio controlled lawn mower using a hacked hover board the prototype works but since I don't know my left from my right an have to reverse some of the controls. I am working on a project to measure how much weight i put on the pedals when I mount my unicycle. I usually ride my 24 inch one but I think I will try the 29 inch one again 'cause I want to ride it on a 50 mile ride. I got a quad copter but kept crashing it so I got a simulator and keep crashing it there too ha ha.

Nice work! 4 balls is a real step up from 3.

This is such fun to watch, Jim. Thanks for sharing. (Denny Noyes)

I used solar cells on mine, so it runs only during the day.

Beautiful ❤

So cool!

Here is the code for PID control but does not have the potentiometer code in it, have lost that one. wordpress.com/post/demej00.wordpress.com/2901

I bought three of these trying to own one that had glow in the dark hands. On two, only the minute hand was luminous, on the third, neither the hour hand, nor the minute hand would glow. To me, a watch that doesn’t function as designed is worthless. Contacted Timex; their customer service is as bad, or worse, than their quality assurance.

is the watch still working after two weeks of usage?

Neat! And thanks for showing the failed version too. It should have worked! I'm also a bit surprised that the center of the pen stays between the fingers.

I really like this model. Simple and nice. The blue face with red second hand is a nice touch.