Author Topic: Constant current load Help needed again  (Read 1660 times)

0 Members and 1 Guest are viewing this topic.

Offline LukeBTopic starter

  • Contributor
  • Posts: 46
  • Country: au
Constant current load Help needed again
« on: September 28, 2018, 10:12:34 am »
Hello again,

I have another really newbie question, sorry, I have this constant current load that I am building as a learning experience and as a tool. I think most of the circuit is all good but I want to know what you think about C2. Is that the best way to smooth out the load?? As you can tell I am still learning but I would like some criticism on my project here so that I can learn. Hopefully this doesn't waste anyone's time but I am here to learn.

Thanks for the help in advance.

BTW, VCC is probably going to be 12v or 9v and the load will probably be for a lead acid car battery. The max current really only needs to be 1A but up to 4 would be nice. Probably depends on mosfet temp.
« Last Edit: September 28, 2018, 01:48:58 pm by LukeB »
 

Offline Prithul0218

  • Contributor
  • Posts: 48
  • Country: us
Re: Constant current load Help needed again
« Reply #1 on: September 28, 2018, 02:58:00 pm »
I am new here. I am not an expert, just a young hobbyist. So judge my words before applying in your project.

It seems to me like what you're doing is smoothing the feedback signal. But that's not what you want to do. You want a fast changing feedback but your capacitor is actually slowing it down. Maybe a tiny capacitor would be ok, like, a few pFs but not necessary.

However, in order to make the load current smooth, what you should do is make the output of the op amp stable. Check the attached photo. It is from the part 1 of Scullcom Hobby Electronics DC load project which I highly recommend you take a look at, here's the full YouTube playlist with 10 videos: https://www.youtube.com/playlist?list=PLUMG8JNssPPzbr4LydbTcBrhoPlemu5Dt

You would probably want to use that type of RC filter configuration.

Correct my mistakes and help me grow.
Thanks.
 

Offline Dave

  • Super Contributor
  • ***
  • Posts: 1356
  • Country: si
  • I like to measure things.
Re: Constant current load Help needed again
« Reply #2 on: September 28, 2018, 05:03:29 pm »
Correct my mistakes and help me grow.
No mistakes, your advice is solid.
<fellbuendel> it's arduino, you're not supposed to know anything about what you're doing
<fellbuendel> if you knew, you wouldn't be using it
 

Offline Prithul0218

  • Contributor
  • Posts: 48
  • Country: us
Re: Constant current load Help needed again
« Reply #3 on: September 28, 2018, 06:00:30 pm »
Thank you Dave.  :)
 

Offline LukeBTopic starter

  • Contributor
  • Posts: 46
  • Country: au
Re: Constant current load Help needed again
« Reply #4 on: September 29, 2018, 12:40:32 am »
So why do you need to tie it back to the inverting input?

With my method wouldn't smoothing the feedback out a bit cause the output to be smooth anyway?
 

Offline Prithul0218

  • Contributor
  • Posts: 48
  • Country: us
Re: Constant current load Help needed again
« Reply #5 on: September 29, 2018, 03:05:45 am »
So why do you need to tie it back to the inverting input?

With my method wouldn't smoothing the feedback out a bit cause the output to be smooth anyway?

The op amp configuration with resistor and capacitor you see in the picture is probably called "Inverting Integrator". I suggest you to look it up in order to understand why the resistor is connected to the Inverting input.

Making the feedback smooth, the op amp will not be able to detect the noises. It will "fool" the op amp into thinking that it's already noiseless.  But your op amp needs to detect the noise and control the output accordingly in order to get rid of the noises. That's why making the feedback smooth will not work.

Somebody correct me if I'm wrong.
Thank you.
« Last Edit: September 29, 2018, 03:12:24 am by Prithul0218 »
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 15149
  • Country: de
Re: Constant current load Help needed again
« Reply #6 on: September 30, 2018, 10:34:01 am »
The capacitors in the feedback loop are not primarily about noise, but to avoid oscillations. The capacitor around the OP provides local feedback that is not effected by the possibly slow MOSFET. The actual feedback from the shunt is only for the lower frequencies, were the power part is well behaved.
The capacitor in parallel to the shunt does the wrong thing, it slows down the feedback signal and this way causes a phase shift that promotes oscillation.

One can also see the OP with local FB via the RC as a kind of PI regulator. The capacitor sets the I term while the series R (not always found in current sinks) sets the P term. So the capacitor is there to slow down the regulator enough that it does not oscillates. If chosen right the addition P term can help with stability and the response to a changing set current.
 

Offline Prithul0218

  • Contributor
  • Posts: 48
  • Country: us
Re: Constant current load Help needed again
« Reply #7 on: September 30, 2018, 10:51:44 am »
The capacitors in the feedback loop are not primarily about noise, but to avoid oscillations. The capacitor around the OP provides local feedback that is not effected by the possibly slow MOSFET. The actual feedback from the shunt is only for the lower frequencies, were the power part is well behaved.
The capacitor in parallel to the shunt does the wrong thing, it slows down the feedback signal and this way causes a phase shift that promotes oscillation.

One can also see the OP with local FB via the RC as a kind of PI regulator. The capacitor sets the I term while the series R (not always found in current sinks) sets the P term. So the capacitor is there to slow down the regulator enough that it does not oscillates. If chosen right the addition P term can help with stability and the response to a changing set current.

I have used this method before but never thought of it that way. It really is like a PI controller.

I have a side question. How does one select the right values for the capacitor and the resistor? I just used the same values as the schematic.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf