Author Topic: LM336 to replace LM4040C25  (Read 2403 times)

0 Members and 1 Guest are viewing this topic.

Offline belzrebuthTopic starter

  • Frequent Contributor
  • **
  • Posts: 487
  • Country: gr
LM336 to replace LM4040C25
« on: November 26, 2019, 10:04:37 am »
I'd like to try a board while waiting for the correct part to arrive and thought since I've got a LM336 I could use that instead of the LM4040.
To my understanding both parts are zeners but the 336 also has an adj pin.
Problem is that 336 needs more current to operate and is less precise but of the circuit can handle the current am I going to be okay temporarily?
Schematics to follow since I'm on the road..
« Last Edit: November 26, 2019, 11:13:58 am by belzrebuth »
 

Offline newbrain

  • Super Contributor
  • ***
  • Posts: 1801
  • Country: se
Re: LM336 to replace LM4040C25
« Reply #1 on: November 26, 2019, 05:12:53 pm »
I see no problems as long as the 3300 \$\Omega\$ resistor is brought down to ~1000 \$\Omega\$.

This will provide (3.3-2.5)/1000 - 150E-6 = 650 µA to the LM336, enough to have it working correctly.
The ADJ pin can be left unconnected.

Nandemo wa shiranai wa yo, shitteru koto dake.
 

Offline not1xor1

  • Frequent Contributor
  • **
  • Posts: 716
  • Country: it
Re: LM336 to replace LM4040C25
« Reply #2 on: November 26, 2019, 05:21:25 pm »
Neither LM336 nor LM4040 are zener diodes. They are shunt regulator ICs with a bandgap reference.

A proper design of a shunt regulator circuit must take into account:
- maximum load current
- minimum working current of the shunt regulator IC
- power dissipation in the shunt regulator IC and resistor

so the value of R8 in your circuit must be:

(3.3V - 2.5V) / (max_load_current + minimum_regulation_current)

for instance:
if your 2.5V load never needs more than 10mA and you use an LM336 you have to use a:
(.8 / (10e-3+400e-6)) resistor i.e a 76.9 ohm one

taking into account resistor and voltage reference tolerance (and current leakage through C17) you have to use a standard and lower value resistor i.e. 68 or even better 56 ohm

BTW if your load just takes few µA you have to take into account the capacitor leakage that would probably be the same order of magnitude

and of course power dissipation in the resistor and shunt regulator would be:
.8^2 / R8 and 2.5 * (.8 / R8)
 

Offline newbrain

  • Super Contributor
  • ***
  • Posts: 1801
  • Country: se
Re: LM336 to replace LM4040C25
« Reply #3 on: November 26, 2019, 05:37:27 pm »
Neither LM336 nor LM4040 are zener diodes. They are shunt regulator ICs with a bandgap reference.

A proper design of a shunt regulator circuit must take into account:
- maximum load current
- minimum working current of the shunt regulator IC
- power dissipation in the shunt regulator IC and resistor

so the value of R8 in your circuit must be:

(3.3V - 2.5V) / (max_load_current + minimum_regulation_current)

for instance:
if your 2.5V load never needs more than 10mA and you use an LM336 you have to use a:
(.8 / (10e-3+400e-6)) resistor i.e a 76.9 ohm one

taking into account resistor and voltage reference tolerance (and current leakage through C17) you have to use a standard and lower value resistor i.e. 68 or even better 56 ohm

BTW if your load just takes few µA you have to take into account the capacitor leakage that would probably be the same order of magnitude

and of course power dissipation in the resistor and shunt regulator would be:
.8^2 / R8 and 2.5 * (.8 / R8)
All true, but the schematics indicates a 150µA load, dutifully taken into consideration.

Plus I do not expect anything close to a µA o.o.m. leakage from a ceramic 100nF cap...Ma dove li compri condensatori così scamuffi?  ;)

A 56  \$\Omega\$ resistor with no load (or a light one) would force 0.8/56 ~ 14 mA through the LM336 - exceeding recommended operating conditions.
Nandemo wa shiranai wa yo, shitteru koto dake.
 

Offline not1xor1

  • Frequent Contributor
  • **
  • Posts: 716
  • Country: it
Re: LM336 to replace LM4040C25
« Reply #4 on: November 27, 2019, 05:54:39 am »
All true, but the schematics indicates a 150µA load, dutifully taken into consideration.

Plus I do not expect anything close to a µA o.o.m. leakage from a ceramic 100nF cap...Ma dove li compri condensatori così scamuffi?  ;)

A 56  \$\Omega\$ resistor with no load (or a light one) would force 0.8/56 ~ 14 mA through the LM336 - exceeding recommended operating conditions.

I was misled by the polarized capacitor symbol (with reverse polarity) and that led me to think it was a 100 µF capacitor...  :palm:

I would not trust much such schematic...

You're right about the LM336 limit... I forgot to check the datasheet  |O

I think it is better to provide some more information to beginners so one may understand how circuits work and when such information is wrong (like the oversight on LM336 current of my example) somebody would correct that  :)
« Last Edit: November 27, 2019, 05:59:51 am by not1xor1 »
 

Offline newbrain

  • Super Contributor
  • ***
  • Posts: 1801
  • Country: se
Re: LM336 to replace LM4040C25
« Reply #5 on: November 27, 2019, 07:52:28 am »
I was misled by the polarized capacitor symbol (with reverse polarity) and that led me to think it was a 100 µF capacitor...  :palm:
That is Style 2 from the IEEE/IEC/ANSI standard. It might have been more common in days past, and it's not the preferred one, but still perfectly acceptable.
See Chapter 2.2 at page 24 (44 in pdf).
I don't understand what you mean with reverse polarity, even considering it polarized.

I would not trust much such schematic...
Probably due to the above misreading; I see nothing wrong with it.

I think it is better to provide some more information to beginners so one may understand how circuits work and when such information is wrong (like the oversight on LM336 current of my example) somebody would correct that  :)
This is true, I should have made the calculation explicit.
Nandemo wa shiranai wa yo, shitteru koto dake.
 

Offline not1xor1

  • Frequent Contributor
  • **
  • Posts: 716
  • Country: it
Re: LM336 to replace LM4040C25
« Reply #6 on: November 28, 2019, 06:04:45 am »
I was misled by the polarized capacitor symbol (with reverse polarity) and that led me to think it was a 100 µF capacitor...  :palm:
That is Style 2 from the IEEE/IEC/ANSI standard. It might have been more common in days past, and it's not the preferred one, but still perfectly acceptable.
See Chapter 2.2 at page 24 (44 in pdf).
I don't understand what you mean with reverse polarity, even considering it polarized.

I had not noticed the "+" on the top as it was partially hidden by "C17".
That's probably due the fullHD resolution of my laptop and the poor state of my eyes.  ;D
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf