Author Topic: Some guidance on an adjustable Bench PSU  (Read 3321 times)

0 Members and 1 Guest are viewing this topic.

Offline antolanca

  • Contributor
  • Posts: 12
  • Country: gb
Some guidance on an adjustable Bench PSU
« on: January 26, 2014, 10:24:31 am »
Hi EEV bloggers!

I need your guidance on building the adjustable power supply that David did on video 221 to 223.
First off, have a look at the PDF below. You will have the schematic (did my best copying it from the video), a drawing and a 3D Rendering of the board.

I went for a dual layer design (I'd prefer a single layer but coudn't figure out how to do it) with 500mW 5% resistors X7R (apart from the 1R which is a 7W for more amps according to the video) and 3 surface mount ceramic caps (soldered to the bottom layer).

Could someone have a look and tell me if there are any modifications that I should do to the circuitry and how could I go from 7V input to 12V in the future?

Thanks
Yesterday is yesterday. If we try to recapture it we will only lose tomorrow.
 

Online Rerouter

  • Super Contributor
  • ***
  • Posts: 3816
  • Country: au
  • Question Everything... Except This Statement
Re: Some guidance on an adjustable Bench PSU
« Reply #1 on: January 26, 2014, 10:34:28 am »
first modification is cathode to cathode 7.4V zener diodes across the set and output pin, as later on it was found out these chips get fried if the difference between the 2 is greater than 10V,

You have bugger all input capacitance, and output capacitance for that matter? if your pulling more than 10mA you will have problems there, it needs to be closer to 220nF minimum on the output, and it never hurts to have a large input capacitance

your op amp can already handle a higher supply voltage so i am left wondering what your concern is with a higher voltage?
 

Offline antolanca

  • Contributor
  • Posts: 12
  • Country: gb
Re: Some guidance on an adjustable Bench PSU
« Reply #2 on: January 27, 2014, 03:59:56 am »
My worry with the higher voltage is  that going from 7v to 12v input is a big leap and this system was designed and tested to have a 0-7V output with a 7v input by David himself. But my main worry is can the circuitry handle 12v input?

Ok. Let's see if I understood you. You would put 2 1N5343BG zener diodes in series cathode to cathode? Is that by the LM334, LT3080...?

What if I would put 22uF ceramic capacitors all round, would that suffice?

Thank you for helping a newbie like me.

Ant

PS: Made slight changes to the drawing. Forgot to insert a resistor that goes from pin 7 of the bottom Op Amp to the 2N3904 transistor. New picture attached.
Yesterday is yesterday. If we try to recapture it we will only lose tomorrow.
 

Online Rerouter

  • Super Contributor
  • ***
  • Posts: 3816
  • Country: au
  • Question Everything... Except This Statement
Re: Some guidance on an adjustable Bench PSU
« Reply #3 on: January 27, 2014, 07:12:57 am »
My worry with the higher voltage is  that going from 7v to 12v input is a big leap and this system was designed and tested to have a 0-7V output with a 7v input by David himself. But my main worry is can the circuitry handle 12v input?
To me it would only be your capacitors that would need to be swapped out to suit, as ceramic capacitors loose capacitance as the voltage across them increases, to counteract this the general rule of thumb is get them in twice to 2.5 times the voltage rating you are applying to the circuit (in this case 25-40V)

Quote
Ok. Let's see if I understood you. You would put 2 1N5343BG zener diodes in series cathode to cathode? Is that by the LM334, LT3080...?

That would be across the LT3080's set pin and output pin, as a difference in either direction of 10V will kill the chip

Quote
What if I would put 22uF ceramic capacitors all round, would that suffice?

For input that is OK, but for output the ESR of the capacitor would be too low, (its expecting an electrolytic or tantalum)



Also at a glance your LED dropping resistors value changes between schematic and pcb, so you may wish to double check and validate just what values your using (470K is way high for an LED)
 

Offline antolanca

  • Contributor
  • Posts: 12
  • Country: gb
Re: Some guidance on an adjustable Bench PSU
« Reply #4 on: January 28, 2014, 01:13:10 am »
Quote
To me it would only be your capacitors that would need to be swapped out to suit, as ceramic capacitors loose capacitance as the voltage across them increases, to counteract this the general rule of thumb is get them in twice to 2.5 times the voltage rating you are applying to the circuit (in this case 25-40V)
Quote
For input that is OK, but for output the ESR of the capacitor would be too low, (its expecting an electrolytic or tantalum)

I've got a 3300uF 50v electrolitic from an old printer transformer. Would that be good?

Quote
That would be across the LT3080's set pin and output pin, as a difference in either direction of 10V will kill the chip.

I've drawn 2 layouts below. Which do you think is best? And if you have the time, why that way?
By the way, the LT3080 I'll be using will be the 5 pin version.

Quote
Also at a glance your LED dropping resistors value changes between schematic and pcb, so you may wish to double check and validate just what values your using (470K is way high for an LED)

Whoops. I've put the wrong amount in the schematic. All corrected. It's supposed to be a 1K.
Yesterday is yesterday. If we try to recapture it we will only lose tomorrow.
 

Offline phil

  • Regular Contributor
  • *
  • Posts: 67
  • Country: de
Re: Some guidance on an adjustable Bench PSU
« Reply #5 on: January 28, 2014, 03:03:45 am »
Both layouts are wrong. You wired the zeners parallel, not in series. It's supposed to be like this:

Code: [Select]
*-----|<|-----|>|------*
1     ZD1     ZD2      2

That limits the voltage difference between the two pins to zener voltage + diode drop, i.e. 7,4V+0.7V = 8.1V.
If the voltage difference rises above that (1=10V, 2=0V for example), the reverse biased zener (ZD1) will break down and the other one (ZD2) will conduct anyway. So they just short out any voltage greater than 8.1V.

Thats a common technique used to protect parts of a circuit that must not go above a certain voltage, i.e. MOSFET gates.


A 3300µF output cap is very risky and defeats the purpose of your current limit. The current limit can limit (duh) the current that flows through your regulator, but it cannot limit the cap discharge current. That means, if you test a prototype and limit the current in case of a short, your circuit will blow up nontheless, because the output cap will discharge (read: tens or even hundreds of amps!) through your faulty circuit. AFTER that, the usual current limit kicks in. That short discharge is enough to kill most semiconductors and can even damage small tracks on your PCB.

In short: 3300µF is too much, better use ~100µF or less if possible.

Phil
 

Offline antolanca

  • Contributor
  • Posts: 12
  • Country: gb
Re: Some guidance on an adjustable Bench PSU
« Reply #6 on: January 28, 2014, 04:07:50 pm »
Ok. I think I've got it.

Zeners in series in the picture below. Between set pin and output pin.

Caps of choice: 22uF 35V Electrolytic all round.

http://www.farnell.com/datasheets/1558295.pdf

I know they are not the best but I can always upgrade the caps to ceramics or film later.

What do you think?
Yesterday is yesterday. If we try to recapture it we will only lose tomorrow.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf