My E-Bike / Pedelec project is retired for the moment. Unfortunately the motor is broken during my last Controller test, so I had to stop that project to get the Dieselbike done before the next Semester at university starts.
As you can see a small plastic sprocket is broken. I am now searching for a spare part! All I become to hear at the moment is something like: “Build in a new Motor!”
It was a little bit more to do than expected.
- rearranged the connectors, the fet and the Voltage regulator
- added one small FET for Lights
- added a fast DUAL Mosfet driver IC
- added an LM317 for the Mosfet driver IC
- build the thing into a box
- added a shunt for output current measurement
- added a voltage divisor to get the current battery voltage
Today as expected I blew up the controller!
Nature can not be fooled!
- A MosFET with 30 Vds on a 40 V power source will die immediately!
- A MosFET with 100 Vds on a 40V power source after 10 meters
I am still not exactly sure why the last law is taking action, remember, these are still my first elektronics experiments. I think it is either that I connected the FET without a driver, or it is because the starting current of the Motor is too high!
Maybe it is both, so I will go to the Metalab, making a few Modifications this Night.
The controller is complete now!
– 16 MHz Quarz
– RS232 with Max232
– padelec sensor on PC1
– throttle on PC0
– status LED on PD7
– MosFET on PD6/OC0A
Source code is available here
I will try to blow that up tomorrow! 😉
The start in the morning was great! I discovered that the RS232 on my padelec controller was not working because I set the wrong registers to enable it.
When I continued developing I discovered that the hall sensor for the pedals is getting a resistance of nearly 0 ohm on its power supply when I bring it into a magnetic field.
Because it is packaged I can’t take a look inside which sensor they used so I have to guess. I guess that using only the red and the blue wire for signalling, leaving the black wire out, is wrong, but seems to work, so I decide not to think to much about it.
Here are the first pictures of the developing of the pedelec controller:
Lessons learned: I will never do an in circuit RS232 level conversation in one of my projects again! Building one external adapter is better.
Here is a first hopefully working version of the controller:
- RS232 Debug Output
- PWR throttle
- Connection for Gas switch
- Connection for Pedal hal sensor
The batterypack that coming with my pedelec consist in 26 NiCD Cells with 7200 Ah. The first thing I did was putting all the cells apart, and charging them seperately.
Then I build 6.5 packs contaning 4 Batterys with japan plugs for each cell:
I also started building a Charger unit with two LM317 as constant current, and an Atmega Tiny for delta U switch off. I still don’t know how good that old cells will work.
As you can see, I make instant developing, the circuit develops itself instantly on the board. Also, I did not gave much love to it because I still don’t know if the batterys are working.
A friend donated me a pedelec which was lying in his garden for two years now. Actually it where two half padelecs, a front half and a back half, broken right behind the steering head, without a controller, and with an old battery pack.
The first thing I did was welding the two pieces together an weld in a top tube to make the frame stronger: