January 24, 2016

inside the Sky Hawk Eye HM1315 S

Filed under: Hobby Electronics, Mechanics — Tags: , , , , , — admin @ 9:52 am

..Burned a brushed motor of the HM1315S SKY HAWKEYE.

The inside board uses an ARM mpu and a MPU6050 ( 3 axis accel + 3 axis gyro  )
The link : here

The motors gears use ball bearings.

EDIT : i received the spare motors from www.gearbest.com .

Since i finally completely removed the board from it’s housing, i can add one more picture ( 5mm squared paper at the backplane )

a 6.5 x 6.5 cm board

a 6.5 x 6.5 cm board


December 15, 2009

4 legs walker : part III

Filed under: Mechanics — admin @ 7:48 pm

Today i added 4 foots to the engine and the stability is really good now.
I added grip ability to the foot by adding some rubber bands on the inner side and 2 hot-glue points to the outer side.that looks ugly but works perfectly.

bottom view

bottom view


a closer view of one foot

a closer view of one foot

I made a footage of how the thing is really smoothly moving : click here to watch the video ( 6Mo WMV )
In addition, i made some measurements :
The complete engine weight 425grams.
At 9V, the walking sequence is pulling only 200mA from the power source - wich is a really good thing !
On the videos the engine is powered by an old 9V battery wich cannot supply the needed 200mA of current, so the engine is walking slowly than when it’s powered with the right power source (LiPo , NiMH or other thing )
Of course it cannot turn - but that point will be developped soon. stay tuned !


December 14, 2009

4 legs mechanic , part II

Filed under: Mechanics — admin @ 7:29 pm

I made another video of the walking motion after ajusting the length of one of the two arms that connects the rotation gear to the 2 elements of one leg.
That time the 2 connecting arm are not the same length, and the walking motion is less jerky and jumpy , thus we gain in stability of the engine.
At this point, the engine walks in the reverse way it should, according to the 2d simultation : i’ve checked that the length of the torque-transmission arms are very important, and modifying the length is reversing the walk motion,thats why i had a reverse-motion instead of the awaited one.

The video (poor quality…i’ll made a better one next time ) is available here : 4leg_test_2 (WMV , 4.6Mo video )

After that tests, i’m going to build some solid and strong plexiglass links between the main shaft and the arms (the uncentered wheels just pushed on the shaft are not a good idea), and check for a phi-ratio based configuration :


So now i got to test this water turned to wine. stay tuned !


December 13, 2009

A four legs, Theo Jansen based robot mechanic

Filed under: Mechanics — admin @ 2:21 pm

As you’ll notice, i decided to write my posts in english as i noticed that a not-so-small part of my visitors come from outside France.
So that one will be the first , and honestly i hope my english will not be too bad to be right understood by most of the people falling here on my website.

-now… back to the main subject !-

After the tests of the ‘spider6‘ , wich was a 6 legged robot, with 2 servos at each arm, i was not happy about the total power consumption of the system.
In fact, the current drawn from the supply was very high, with peaks at each start of the servos,when moving the robot (about 1 to 2 amps @ around 6 .. 8 volts ).
That thing was a major problem at my eyes, because i think that power consumption of robots must meet the Nature standards rules, who says that the more must be done with the less power possible.

So i decided to test other mechanical configurations for a ‘leg’ based robot mechanics, and after some times of search , i found out a webpage and videos dealing with Theo jansen , who is a kinetic sculptor (do a search on your favorite search engine).

The Strand Beasts ’robots’ he created are amazing, and i’ve been seduced by the smooth moves of that creatures on the sand of the beaches.
After a few tests on a 2d simultation software,i decided to test in real-life my configuration (ratios of the moving parts) of the theo-jansen based legs.

I decided to built a quickly-cut prototype made of plexiglass and decided to use a 30:1 motor gear to drive the thing.
The engines looks rough and dirty but my main goal was to check for stability, reliability, and efficience of the engine before starting a clean project.

As you will see on the pictures and on the video, i choosed to mount the motor on the top of the engine, to provide a strong weight at it’s top , then allowing to check if the thing is stable , or falling on the sides when just a tilt of the floor is simulated.

At this point, let me talk about the engine :
The motor with the ‘reducer’ 30:1 (modelcraft) weights about 180 g.
The total engine (motor + frame and legs) weights 380 g.
Viewed from top, the engine is about 15×15cm , and about 20cm tall.

I added later 4 ‘foots’ to the engine, to help it stabilize (and it needed it !) , because the motor position was a really ‘bad’ choice , pushing the frame of the engine at it’s physical limits, falling on one side or the other when just a very little tilt was there.


The results are impressive !

  • First of all, the power consumption is really low. as far as i’ve measured on my power supply, the engine sucks only a few 100 or 200mA under 7V, to move this 380g body at a speed thats not too low (check the video).
    The Jansen mechanism as many rotating joints, but i used simple double nuts on screws, and even with that assembling, power consumption is excellent.
  • The legs are strong and can bear heavy load (thats a point Mr Jansen has talked about).
    My prototype is far away of the plexiglass assembly - breaking limits, and far away the Jansen system’s break-up.
    Notice :An important thing is that the shaft-to-uncentred-arm link has to be strong ( i just plugged the things by force for my test, but the link needs to have a screw squizzin the shaft to avoid wiping of it when the engine walks on rough floors like sand , then needing greater torque to move the legs)
  • The frame of the engine is stable , despite the very bad position of the motor mounted at the top, and the fact that it has 4 legs and not 6.

Here are the videos :

  1. walking, with a 9V battery in my hand 4legs jansen based prototype walk (AVI 4Mo movie)
  2. another one to check the moves of the legs : Movement of the legs ( AVI 5Mo movie)

Now i’m able to start a serious project using that mechanics, thanks to Mr Jansen (stay tuned ! )

Older Posts »

cat{ 3 } { post_43 } { } 2009-2015 EIhIS Powered by WordPress