Author Topic: NPN Transistor Base Resistor Calculations  (Read 3546 times)

0 Members and 1 Guest are viewing this topic.

Offline TerraByterTopic starter

  • Contributor
  • Posts: 11
  • Country: gr
NPN Transistor Base Resistor Calculations
« on: January 03, 2021, 12:32:18 pm »
Hi all and Happy New Year,

i want to learn how to find the Base Resistor for a NPN Transistor, i search and i found a lot of information but also found different methods to find the Base Resistor with different results.

i make 3 excel with a 3 different methods how to calculate the Base Resistor.

Can you help me to find the right one?





« Last Edit: January 03, 2021, 12:38:41 pm by TerraByter »
 

Offline Heisen

  • Contributor
  • Posts: 42
  • Country: in
Re: NPN Transistor Base Resistor Calculations
« Reply #1 on: January 03, 2021, 12:49:40 pm »
You need to know the minimum beta value of the transistor you are using (available in the datasheet), as beta changes with temperature that's why we look at the minimum beta value to make the circuit conservative.

Assume the minimum beta value to be 30. Current at base times the beta value is the current that is allowed through collector to emitter. With the VCC 12v a 10k resistor at the base is enough to flow roughly 1mA into base, which will allow 1x30 = 30mA to flow between Collector and emitter in the worst case, which is higher than your 20mA requirement for the led.



« Last Edit: January 03, 2021, 12:56:45 pm by Heisen »
 
The following users thanked this post: TerraByter

Offline Peabody

  • Super Contributor
  • ***
  • Posts: 2095
  • Country: us
Re: NPN Transistor Base Resistor Calculations
« Reply #2 on: January 03, 2021, 04:03:35 pm »
Method C gives a much different result than the other two, apparently because of the "overdrive factor".  I'm not sure what that is.  The other two methods are ok provided the gain is actually 100 at the current you will be drawing.  Gain can vary with temperature, and tends to be lower at high current levels.  Generally, you take the worst case based on the datasheet, then give yourself a little leeway.  That means you may be wasting a little base current, but for a product you need to be sure you've covered the individual variation among parts.  For a single project, you can just test it, and find the highest base resistor that still works.

I think Method B is technically correct.  The calculation of collector current in Method A appears to be wrong.  The collector current is (Vcc - Vf - VceSat) / 480.  That gives 20mA, and at a hFE of 100 that's a 57K base resistor.  But that's before you give yourself any room, so you'd probably end up with 47K.

The other thing you have to consider is speed.  The faster you have to switch the LED on and off, the higher the base current you may need.
 
The following users thanked this post: TerraByter

Offline TerraByterTopic starter

  • Contributor
  • Posts: 11
  • Country: gr
Re: NPN Transistor Base Resistor Calculations
« Reply #3 on: January 03, 2021, 04:43:58 pm »
Thanks Peabody for the detail reply,

Method C gives a much different result than the other two, apparently because of the "overdrive factor".  I'm not sure what that is.

I read that article https://www.nutsvolts.com/magazine/article/may2015_Secura and make the calculations for Method C. If you read it in the middle say :

Quote
Therefore, in order to eliminate this possibility, we use what is known as an “Overdrive Factor” (ODF). This is an arbitrary number between 2 and 10 that is used to insure that the transistor is driven hard into saturation (fully turned on) — and where temperature changes fail to drop the transistor out of saturation

I think Method B is technically correct.  The calculation of collector current in Method A appears to be wrong.  The collector current is (Vcc - Vf - VceSat) / 480.  That gives 20mA, and at a hFE of 100 that's a 57K base resistor.  But that's before you give yourself any room, so you'd probably end up with 47K.

I make the correction and now Method A give the same Result with method B.



The other thing you have to consider is speed.  The faster you have to switch the LED on and off, the higher the base current you may need.

Can you please explain little more that?
« Last Edit: January 03, 2021, 04:46:14 pm by TerraByter »
 

Offline Peabody

  • Super Contributor
  • ***
  • Posts: 2095
  • Country: us
Re: NPN Transistor Base Resistor Calculations
« Reply #4 on: January 03, 2021, 05:46:56 pm »
There are capacitances involved with both bipolar transistors and mosfets.  To change the state of something, you have to charge or discharge that capacitance, and to do that more rapidly, you have to source or sink more current.  We think of things turning on or off instantaneously, but that's really not the case.  Every change takes time.  For a simple power switch, that's not something you have to worry about, but if the transistor is part of an oscillator, then switching time comes into play.  So if you want something to switch fast, you have to drive it hard.
 
The following users thanked this post: TerraByter

Online exe

  • Supporter
  • ****
  • Posts: 2599
  • Country: nl
  • self-educated hobbyist
Re: NPN Transistor Base Resistor Calculations
« Reply #5 on: January 03, 2021, 07:21:04 pm »
If it's a switching application, then the value is not that important, just add a bit of "overdrive" current. Like, two-three times more base current than, say, minimum beta requires. If it's an amplifier, well, then things become more complicated, but in any case a good bjt circuit should work in with a wide range of betas.

It's also easy to breadboard and test it.
 
The following users thanked this post: TerraByter

Offline srb1954

  • Super Contributor
  • ***
  • Posts: 1110
  • Country: nz
  • Retired Electronics Design Engineer
Re: NPN Transistor Base Resistor Calculations
« Reply #6 on: January 04, 2021, 03:41:45 am »
If it's a switching application, then the value is not that important, just add a bit of "overdrive" current. Like, two-three times more base current than, say, minimum beta requires. If it's an amplifier, well, then things become more complicated, but in any case a good bjt circuit should work in with a wide range of betas.

It's also easy to breadboard and test it.
Typically, using bipolar transistors in a simple LED driver or relay driver circuit, the transistor is over-driven with excess base current to ensure that the transistor is fully turned on (saturated) to achieve the minimum possible voltage drop across the transistor (VCEsat). The overdrive of the base current is usually determined as a ratio; Ic=Ib*20 for small signal transistors or Ic=Ib*10 for larger transistors. If you check your transistor data sheet it will typically include various specification figures or graphs for the saturated switching parameters e.g. Vce(sat), Vbe(sat) at specified collector currents and a particular Ic/Ib ratio.

Knowing your required Ic you can quickly determine the Ib from the specified Ic/Ib ratio. The base resistor can then be easily calculated Rb <= (Vin - Vbesat)/Ib.

 
The following users thanked this post: TerraByter

Offline gcewing

  • Regular Contributor
  • *
  • Posts: 197
  • Country: nz
Re: NPN Transistor Base Resistor Calculations
« Reply #7 on: January 04, 2021, 06:01:50 am »
The symbol Ω on it own already means "ohm". If you're spelling out in full, it's "ohm" (with a small o), not "Ωhm".
 
The following users thanked this post: TerraByter

Offline emece67

  • Frequent Contributor
  • **
  • !
  • Posts: 614
  • Country: 00
Re: NPN Transistor Base Resistor Calculations
« Reply #8 on: January 04, 2021, 11:07:41 am »
.
« Last Edit: August 19, 2022, 04:08:35 pm by emece67 »
 
The following users thanked this post: TerraByter

Online Zero999

  • Super Contributor
  • ***
  • Posts: 19690
  • Country: gb
  • 0999
Re: NPN Transistor Base Resistor Calculations
« Reply #9 on: January 04, 2021, 01:37:51 pm »
If it's a switching application, then the value is not that important, just add a bit of "overdrive" current. Like, two-three times more base current than, say, minimum beta requires. If it's an amplifier, well, then things become more complicated, but in any case a good bjt circuit should work in with a wide range of betas.

It's also easy to breadboard and test it.
Typically, using bipolar transistors in a simple LED driver or relay driver circuit, the transistor is over-driven with excess base current to ensure that the transistor is fully turned on (saturated) to achieve the minimum possible voltage drop across the transistor (VCEsat). The overdrive of the base current is usually determined as a ratio; Ic=Ib*20 for small signal transistors or Ic=Ib*10 for larger transistors. If you check your transistor data sheet it will typically include various specification figures or graphs for the saturated switching parameters e.g. Vce(sat), Vbe(sat) at specified collector currents and a particular Ic/Ib ratio.

Knowing your required Ic you can quickly determine the Ib from the specified Ic/Ib ratio. The base resistor can then be easily calculated Rb <= (Vin - Vbesat)/Ib.
Yes, the base current is often selected to drive the transistor hard on, into saturation, but that's normally unnecessary. Quite often, a considerable power saving can be made by not driving it so hard. This of course has to be ballanced with the the slightly higher voltage drop across the transistor. For example, the BC337 is specified with a minimum hFE of 60, when VCE = 1V & IC =300mA and 100, when IC = 100mA, so I normally use those figures when choosing a base resistor, for that transistor. Another good thing is hFE has a positive temperature coefficient, so the the voltage drop will reduce, as it warms up. Relays will normally activate at slightly below the voltage rating, so a voltage drop of 1V is no problem and drivng it hard, will only save a few hundred mV, so it generally isn't worth it.
https://www.onsemi.com/pub/Collateral/BC337-D.PDF

In this case it's an LED, which has a votlage drop of around 2V and the supply voltage is 12V, which drives both the base and LED, I'd set RB to roughlly 100 times the current limiting resistor value. There's no need to be accurate. Use the nearest standard E24 value.

What is the LED being used for and what type is it? If it's used as an indicator, under normal office lighting conditions and it's a modern high efficiency red LED, then 20mA is probably overkill and 5mA will probably be bright enough. If it's a crappy old, orange/yellow/green LED, then you probably will need to drive it with 20mA.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf