DIY Electric Car Forums banner
1 - 7 of 7 Posts

·
Registered
Joined
·
230 Posts
Discussion Starter · #1 ·
Here is my friends explanation of it-"cycle by cycle" current limit that does not require manual reset. It is reset each PWM cycle. The PWM clocks a flip flop. The output of the flip flop is ANDed with the PWM. The flip flop is cleared by the over current signal. So the PWM is started on the leading edge of the PWM. It is ended by the end of the PWM or the over current signal - whichever comes first. You need to make sure the flip flop clocks on the leading edge of the PWM signal and a few other such details. If you try to command more than the current limit this limits to the current limit and will run at the current limit rather than shutting off.

He showed me a schematic of it and it looked like an AND gate and a latch. The pwm signal and the latch both went to the AND gate and the circuit would operate as normal but whenever your lem current sensor goes to a certain voltage it triggers the reset of the latch to turn the circuit off for sometime.

I would just like to know has anyone tried this? What are the pros and cons?
 

·
Registered
Joined
·
118 Posts
Surely if your controller goes over current you want it to switch off until a system reset can be done. If a mosfet fails closed and causes an over current failur then you want the system shutdown ASAP before you get a cascade failure of the mosfets. Check out the open revolt cougar controller overcurrent system. Pretty straight forward with a power down to reset system.
 

·
Registered
Joined
·
126 Posts
Here is my friends explanation of it-"cycle by cycle" current limit that does not require manual reset. It is reset each PWM cycle. The PWM clocks a flip flop. The output of the flip flop is ANDed with the PWM. The flip flop is cleared by the over current signal. So the PWM is started on the leading edge of the PWM. It is ended by the end of the PWM or the over current signal - whichever comes first. You need to make sure the flip flop clocks on the leading edge of the PWM signal and a few other such details. If you try to command more than the current limit this limits to the current limit and will run at the current limit rather than shutting off.

He showed me a schematic of it and it looked like an AND gate and a latch. The pwm signal and the latch both went to the AND gate and the circuit would operate as normal but whenever your lem current sensor goes to a certain voltage it triggers the reset of the latch to turn the circuit off for sometime.

I would just like to know has anyone tried this? What are the pros and cons?
What you have described is exactly what is already in place on the open source cougar motor controller. It is very effective that limiting current to what the trim pot is set for. When it is limiting current, it makes the motor make a growling noise from it switching the pwm on and off so quickly.


-Adam
 

·
Registered
Joined
·
230 Posts
Discussion Starter · #4 ·
Maybe I am wrong but from what I understand the p and s motor controller board relies a lot on software whereas this would be strictly a hardware thing that uses just a pwm code with a micro or analog board. If I could build this with 3 or 4 parts why buy that board with 100 or so parts on it?
 

·
Registered
Joined
·
126 Posts
Maybe I am wrong but from what I understand the p and s motor controller board relies a lot on software whereas this would be strictly a hardware thing that uses just a pwm code with a micro or analog board. If I could build this with 3 or 4 parts why buy that board with 100 or so parts on it?
The open source controller takes the reading from the LEM current sensor and pass' it to the LM393 comparator. The other side of the comparator is a voltage reference generated from a trim pot. If the sensor is above the setting of the trip pot (495 amps to 776 amps) it causes the lm393 to disable the PWM through the NAND gates. The controller then re-enables this on every 4th PWM cycle I believe. But in the end, It is hardware over-current as well as software as a second.

-Adam
 

·
Registered
Joined
·
118 Posts
If I could build this with 3 or 4 parts why buy that board with 100 or so parts on it?
Because the other 96 parts do other things, from opto isolation, stabilised power supplies correct signal formatting (with pull ups, pull downs, diodes, etc) and many other things. Good luck with the 4 parts. you'll need it! :eek:
 

·
Registered
Joined
·
230 Posts
Discussion Starter · #7 ·
I have just had a lot of bad luck with kits whether they be a lie detector, snooper ear, or a robot power motor controller and my soldering is above average. But usually when I solder something up myself on a perf board then I know its right and I only have one person to blame if its wrong and I know damn well what's wrong with if it doesn't work. My micro that I soldered already has opto isolation and an isolated power supply. What I mean by four parts is just for the over current protection but your right there may be a few more parts.
 
1 - 7 of 7 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top