Author Topic: Understanding a weird schematic [SOLVED]  (Read 2401 times)

0 Members and 1 Guest are viewing this topic.

Offline toto83Topic starter

  • Contributor
  • Posts: 19
  • Country: fr
Understanding a weird schematic [SOLVED]
« on: February 27, 2018, 01:14:00 pm »
Hello,
I need a symetrical PSU
I wanted to do it using an op amp, and a transistor.
While doing some research, I found this one, however, i dont understand the mosfet and led stuff. Can somebody tell me if it is desirable ? I would like a LED to tell me when it is working.
« Last Edit: February 27, 2018, 04:44:53 pm by toto83 »
 

Offline frozenfrogz

  • Frequent Contributor
  • **
  • Posts: 936
  • Country: de
  • Having fun with Arduino and Raspberry Pi
Re: Understanding a weird schematic
« Reply #1 on: February 27, 2018, 01:28:22 pm »
N-channel JFET can be used as voltage-controlled constant-current generators.

Read more about this over at Nuts and Volts magazine for example :)
He’s like a trained ape. Without the training.
 

Offline Wimberleytech

  • Super Contributor
  • ***
  • Posts: 1134
  • Country: us
Re: Understanding a weird schematic
« Reply #2 on: February 27, 2018, 01:29:46 pm »
It is a JFET, not a MOSFET. Configured as a current limiter.
 

Offline toto83Topic starter

  • Contributor
  • Posts: 19
  • Country: fr
Re: Understanding a weird schematic
« Reply #3 on: February 27, 2018, 02:44:50 pm »
Okay, but what kind of diode can I use, i there any voltage range, or i can use any led?
 

Offline Wimberleytech

  • Super Contributor
  • ***
  • Posts: 1134
  • Country: us
Re: Understanding a weird schematic
« Reply #4 on: February 27, 2018, 03:05:36 pm »
Okay, but what kind of diode can I use, i there any voltage range, or i can use any led?

The BF244 at room temp, VGS=0 provides between 6 and 14mA.  You can use any LED that you have lying around.
 

Offline toto83Topic starter

  • Contributor
  • Posts: 19
  • Country: fr
Re: Understanding a weird schematic
« Reply #5 on: February 27, 2018, 03:49:47 pm »
Thank you for your answer.
Okay, how did you do to calculate the current?
Is it not needed that you add a resistor, what about the diode voltage?
 

Offline Benta

  • Super Contributor
  • ***
  • Posts: 6268
  • Country: de
Re: Understanding a weird schematic
« Reply #6 on: February 27, 2018, 04:32:13 pm »
Okay, how did you do to calculate the current?
Is it not needed that you add a resistor, what about the diode voltage?

From the datashhet, parameter "Zero voltage gate drain current".
You don't need a resistor, the JFET works as a constant current source.
 
The following users thanked this post: Wimberleytech

Offline SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15444
  • Country: fr
Re: Understanding a weird schematic [SOLVED]
« Reply #7 on: February 27, 2018, 05:22:17 pm »
The JFET with its gate tied to source forms a decent approximation of a constant current source. Since an LED brightness depends on the current, that would make sense to ensure a somewhat constant brightness when the voltage it's powered from can vary.

But in this case, the voltage comes from a linear regulator and is fixed. I don't see any benefit compared to a simple resistor.

It will even be worse than a resistor, since there is a wide dispersion from one JFET to another (and it's also dependant on temperature), much wider than for the treshold voltage of most LEDs. The current flowing through the LED will thus actually be much more predictable (within the tolerance of the resistor and the error of the regulated voltage) with a resistor here, than with a JFET.

Now if the output voltage was variable, that could make sense, but you would still have significant dispersion in LED brightness from one unit to another.
« Last Edit: February 27, 2018, 05:26:19 pm by SiliconWizard »
 

Offline Benta

  • Super Contributor
  • ***
  • Posts: 6268
  • Country: de
Re: Understanding a weird schematic [SOLVED]
« Reply #8 on: February 27, 2018, 05:38:04 pm »
But in this case, the voltage comes from a linear regulator and is fixed. I don't see any benefit compared to a simple resistor.

I think you've overlooked RV1. It is indeed a variable supply.
 

Offline SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15444
  • Country: fr
Re: Understanding a weird schematic [SOLVED]
« Reply #9 on: February 27, 2018, 05:50:16 pm »
But in this case, the voltage comes from a linear regulator and is fixed. I don't see any benefit compared to a simple resistor.

I think you've overlooked RV1. It is indeed a variable supply.

Ouch, indeed!  ;)

Since the author said he needed a symmetrical PSU, he may just need a fixed voltage one, which I unconsciously assumed. So that would depend on his needs. I assumed the schematic was just one he wanted to base the design on.

If it's variable, it makes sense as I said, even though there would still be wide dispersion in brightness from one JFET to another. That's of course no problem for a single-unit prototype, but would not be that great for a commercial product.
 

Offline toto83Topic starter

  • Contributor
  • Posts: 19
  • Country: fr
Re: Understanding a weird schematic [SOLVED]
« Reply #10 on: March 03, 2018, 10:57:42 pm »
I need a fixed voltage, but that schematic is not mine.
I will use it as a base, but I will not use this jfet.
 

Offline jb79

  • Contributor
  • Posts: 27
  • Country: at
Re: Understanding a weird schematic [SOLVED]
« Reply #11 on: March 04, 2018, 01:14:06 am »
Hi,

the JFET is necessary because of the voltage range of the powersupply: 1,27 (Vref) to about 35V: the led would have the same brightness over nearly the same range.
If you are using a fixed voltage a simple resistor would do the job.

What voltages are we talking about:
input from powersupply, output voltage,...
 

Offline Neomys Sapiens

  • Super Contributor
  • ***
  • Posts: 3268
  • Country: de
Re: Understanding a weird schematic [SOLVED]
« Reply #12 on: March 04, 2018, 01:20:25 am »
But you are well aware that the LED, whether used with a resistor or the FET, will not indicate if the VOLTAGE SPLITTER (the lower part of your circuit, generating the virtual ground) works properly?
 

Offline toto83Topic starter

  • Contributor
  • Posts: 19
  • Country: fr
Re: Understanding a weird schematic [SOLVED]
« Reply #13 on: March 04, 2018, 03:54:09 pm »
I am aware of that, however, as this is a PSU, a fail of the op-amp would not go unoticed, as there will be a Voltmeter after that.
 

Offline toto83Topic starter

  • Contributor
  • Posts: 19
  • Country: fr
Re: Understanding a weird schematic [SOLVED]
« Reply #14 on: March 04, 2018, 03:55:33 pm »
I will use the IC as a fixed voltage regulator, so the JFET is not necessary in my case, I think I'm gonna go with zener + resistor to fullfil the voltage requirement for the LED
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf