Author Topic: Simple current limiter  (Read 13821 times)

0 Members and 1 Guest are viewing this topic.

Offline cyberfishTopic starter

  • Regular Contributor
  • *
  • Posts: 240
  • Country: gb
Simple current limiter
« on: November 01, 2010, 08:17:07 pm »
I need a simple current limiter for a project (a brushless motor driver in particular, as I'm pretty sure I will accidentally short the supply sooner or later).

To my surprise, I couldn't find one in 15 minutes of Googling, so I decided to design my own.

It should just sit in series with the load, and if current goes above some limit, it will drop more voltage to limit the current to the set limit.

I don't really care about voltage regulation much. The load is designed to work with large variations in supply voltage (from a battery).

Attached is what I came up with. Haven't built it yet, but appears to work in simulation.

Can I get some feedback please? Is there a way to make it simpler/better/both?

Better voltage regulation is nice, but only if the circuit is not significantly more complex.

In my case, the parameters are - Vcc = 12V, Ilimit = 10mA.

At high current, this probably won't work. The resistor will blow up before dropping 0.7V.

Thanks
 

Offline Zad

  • Super Contributor
  • ***
  • Posts: 1013
  • Country: gb
    • Digital Wizardry, Analogue Alchemy, Software Sorcery
Re: Simple current limiter
« Reply #1 on: November 01, 2010, 11:02:39 pm »
You could always use an LM317 regulator or similar in current configuration.


Offline cyberfishTopic starter

  • Regular Contributor
  • *
  • Posts: 240
  • Country: gb
Re: Simple current limiter
« Reply #2 on: November 01, 2010, 11:24:01 pm »
Thanks! That does look simpler.

The voltage drop is greater (2V dropout + R * I), but that doesn't matter for me.
 

Offline TechGuy

  • Regular Contributor
  • *
  • Posts: 79
Re: Simple current limiter
« Reply #3 on: November 02, 2010, 12:22:10 am »
PTC fuses work pretty well. A PTC or Positive Temperature Coefficient Fuse works by changing from a state of low resistors to a state of very high resistance when accessive current flow occurs. The Advantage of using PTC is that they are cheap. The disadvantage is there slow response, They transistion in a matter of hundreds to thousands of milliseconds.

Some Motor Controllers such as the L6201 have built in current sensing to limit current. L6201 is a 2A full bridge driver and the L6203 is a 4A full bridge driver. With a Full bridge driver, you can operate the motor forward and in reverse.
http://www.st.com/stonline/books/pdf/docs/1373.pdf

In your circuit. I am not sure why you need the Rsense resistor. If I understand your circuit, you're forcing the base current to remain fixed so that it will limit the amount of current flowing between the emitter and collector. The pair of diodes is to limit the voltage. Should a higher Vcc be used the pair of diodes would clamp the voltage and limit the base current. The Rsense resistor is not providing any feedback. If you want a feedback current limiter you need to use a something like a TL431 adjustable zenor (replacing the pair of diodes). The TL431 adj. pin is connected to the trace between the NPN emitter and the Rsense transistor so that as the voltage drop across the Resense resistor increases, TL431 zener voltage falls reducing the NPN base current. That said, you may run into problems running a motor that presumably has moderate current levels. The NPN transistor will act as a resistor dispating a lot of energy as heat, and has the protential of failure, if it gets too hot. BJT have a negitive thermal coefficient, meaning that the hotter they get, the more current they will pass. For instance an NPN transistor at 25C with a static base current might permit 1A to pass. At 150C it might permit 2A to pass. For example see this datasheet for the 2N3055 on page 3 showing the hfe plot at -55C 25C and 150C.  http://www.onsemi.com/pub_link/Collateral/2N3055A-D.PDF

Another issue is compensating for motor start up current. Electric motors need considerable more current to startup, You can address this by using a capacitor connected from the NPN collector to ground, so that when you turn on your motor the capacitor will charge up providing a current boost to get the motor started. Once the cap. is fully charged it won't pass any significant current.

 

Offline cyberfishTopic starter

  • Regular Contributor
  • *
  • Posts: 240
  • Country: gb
Re: Simple current limiter
« Reply #4 on: November 02, 2010, 01:00:08 am »
Yeah PTC is another option. Or a fuse. But I won't be able to easily change the current limit, for example. This is a circuit I would like to set up on a breadboard quickly when I need them, so I'll preferably not have to rely on specific component values (unless I stock a big range of PTCs).

I had to build my own motor driver because 1) it's brushless (not brushed DC, where you just apply a voltage across them - for brushless you need to supply 3 phase power), and 2) it draws about 10A.

Quote
In your circuit. I am not sure why you need the Rsense resistor. If I understand your circuit, you're forcing the base current to remain fixed so that it will limit the amount of current flowing between the emitter and collector. The pair of diodes is to limit the voltage. Should a higher Vcc be used the pair of diodes would clamp the voltage and limit the base current. The Rsense resistor is not providing any feedback. If you want a feedback current limiter you need to use a something like a TL431 adjustable zenor (replacing the pair of diodes). The TL431 adj. pin is connected to the trace between the NPN emitter and the Rsense transistor so that as the voltage drop across the Resense resistor increases, TL431 zener voltage falls reducing the NPN base current. That said, you may run into problems running a motor that presumably has moderate current levels. The NPN transistor will act as a resistor dispating a lot of energy as heat, and has the protential of failure, if it gets too hot. BJT have a negitive thermal coefficient, meaning that the hotter they get, the more current they will pass. For instance an NPN transistor at 25C with a static base current might permit 1A to pass. At 150C it might permit 2A to pass. For example see this datasheet for the 2N3055 on page 3 showing the hfe plot at -55C 25C and 150C.  http://www.onsemi.com/pub_link/Collateral/2N3055A-D.PDF
Not quite. When the voltage across Rsense increases to 0.65V, the 2 diodes (one of them to cancel out the 0.65V drop in base-emitter) will start to conduct, and turn off the transistor by taking away the base current, hence limiting current (or in other words, limiting the drop in Rsense to 0.65V).

It's not usually a good idea to design with the precise gain of the transistor, because the gain changes with a gazillion things. It's usually a better idea to design for the worst case gain (which can be max gain or min gain, depending on application).

This current limiter won't be there when I connect the motor. This is just for development/testing, in case a coding mistake causes a short in the MOSFETs.

The "real thing" will be run off a lithium polymer battery that can supply about 70A.
 

Offline scrat

  • Frequent Contributor
  • **
  • Posts: 608
  • Country: it
Re: Simple current limiter
« Reply #5 on: November 02, 2010, 05:56:30 pm »
I'd use the sense voltage with a comparator towards a precise threshold. It then could switch on a MOSFET or BJT, depending on which makes the lower power dissipation for your rated current/voltage.
It could be designed such that the "switch" is normally open unless you give supply to the comparator.
Besides being more precise, this will allow you to use a smaller sense resistor.

Will you let me know about the inverter and controller development? I'm involved mostly in synchronous motor drive research, and so I'd like to hear from the application field.
Thanks.
One machine can do the work of fifty ordinary men. No machine can do the work of one extraordinary man. - Elbert Hubbard
 

Offline cyberfishTopic starter

  • Regular Contributor
  • *
  • Posts: 240
  • Country: gb
Re: Simple current limiter
« Reply #6 on: November 02, 2010, 06:11:34 pm »
Quote
I'd use the sense voltage with a comparator towards a precise threshold. It then could switch on a MOSFET or BJT, depending on which makes the lower power dissipation for your rated current/voltage.

Oh yeah, that would be a lot better regulation, lower voltage drop, with just little higher complexity. Scales far better into high current land, too. Will try it out next time.

The controller is just good old BLDC with back-emf sensing. Mostly just following AVR444 (http://www.atmel.com/dyn/resources/prod_documents/doc8012.pdf). 3 half-bridges, one high, one low, one high-Z. Just hobbyist-grade stuff.
 

Offline scrat

  • Frequent Contributor
  • **
  • Posts: 608
  • Country: it
Re: Simple current limiter
« Reply #7 on: November 02, 2010, 06:51:17 pm »
Thanks. I'm always interested in how people implement control of motors very simply, while to achieve slightly better performances you have to struggle so much.
For example, many motors controlled as BLDC are sinusoidal ones (while BLDCs have trapezoidal B-EMF).
They are simply used below their possibilities, but since they work... Keep It Simple, Stupid, someone said.
One machine can do the work of fifty ordinary men. No machine can do the work of one extraordinary man. - Elbert Hubbard
 

Online Zero999

  • Super Contributor
  • ***
  • Posts: 19344
  • Country: gb
  • 0999
Re: Simple current limiter
« Reply #8 on: November 02, 2010, 10:21:36 pm »
I need a simple current limiter for a project (a brushless motor driver in particular, as I'm pretty sure I will accidentally short the supply sooner or later).

*snip*
Quote
In my case, the parameters are - Vcc = 12V, Ilimit = 10mA.

10mA? Are you sure? How small is your motor?

The LM317 idea will drop between 2.8V to 4V depending on the current but has good current regulation.

The transistor and diodes idea has poor current regulation but a lower voltage drop, typically 0.8V when limiting, less if it's operated way below the limit. The voltage drop can be reduced by replacing one of the diodes with a Schottky diode.

I think you should revise your current limit upwards and use a PTC resistor or bi-metal circuit breaker.
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 37661
  • Country: au
    • EEVblog
Re: Simple current limiter
« Reply #9 on: November 03, 2010, 12:00:55 am »
Can I get some feedback please? Is there a way to make it simpler/better/both?

That is quite a common circuit.
The diodes can be replaced with a LED. One less part and you get a bonus power LED.

Dave.
 

Offline cyberfishTopic starter

  • Regular Contributor
  • *
  • Posts: 240
  • Country: gb
Re: Simple current limiter
« Reply #10 on: November 03, 2010, 03:55:55 am »
Quote
10mA? Are you sure? How small is your motor?
This is just for testing without motors. I don't need a current limiter when the motors are attached.

Schottky is a good idea.

Quote
That is quite a common circuit.
The diodes can be replaced with a LED. One less part and you get a bonus power LED.
That's great! I just reinvented a common circuit, usually meaning I'm doing something right (doesn't happen too often). LED is nice (over-current indicator), at the cost of higher voltage drop.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf