# Why my darlington pair consisted of 2 ss8050 NPN BJTs doesn't work?

#### Bowen102

Joined Aug 23, 2017
7
As you can see in the first picture, I use 1 SS8050 BJT to drive a led(which requires 500ma at max brightness).
And the power supply showed that it only consumed 130ma.

So I decided to use 2 SS8050 to make a darlington, thus hoping to amplify the output current to 500ma. (As shown in picture 2)
But strangly the output current droped to 40ma.

I checked again and again, just can't figure out how it end up like this.
Aren't it suppose to be output a larger current?

Any help will be very appreciated.

(1)

(2)

#### AlbertHall

Joined Jun 4, 2014
11,108
What exactly is the base connected to?

#### Alec_t

Joined Sep 17, 2013
11,981
What is there to limit the LED current if your micro goes ape?? I see no series resistor.

#### Bordodynov

Joined May 20, 2015
2,802
See

#### ankush_chavhan

Joined Aug 22, 2017
16
(1)Problem not because of current but it is bcz of voltage.
actually this happening because your led require 3.3v forward voltage and because of finite vce led not getting 3.3v so if you can increase vcc your problem will be definitely solved.

#### ebeowulf17

Joined Aug 12, 2014
3,275
As you can see in the first picture, I use 1 SS8050 BJT to drive a led(which requires 500ma at max brightness).
And the power supply showed that it only consumed 130ma.

So I decided to use 2 SS8050 to make a darlington, thus hoping to amplify the output current to 500ma. (As shown in picture 2)
But strangly the output current droped to 40ma.

I checked again and again, just can't figure out how it end up like this.
Aren't it suppose to be output a larger current?

Any help will be very appreciated.

(1)
View attachment 133540

(2)
View attachment 133541
Please share the LED datasheet, or at least a part number.

And to follow up on Alec's resistor question, you're also not showing resistors on the base of either bjt.

#### Bowen102

Joined Aug 23, 2017
7
What exactly is the base connected to?
Thank you, Bordodynov. According to your circuit, the one at bottom left(862.25ma), I do get a larger output current at 600ma now.
But I am wondering how much current is flowing directly through Q5, then B->E of Q6, bypassing the LED ?
Because when I try disconnecting the LED, it still ouput about 300ma . Would it be wasting too much energy?

#### Bowen102

Joined Aug 23, 2017
7
What exactly is the base connected to?
The base is connected to a GPIO of a microcontroller, which can provide 3.3v at high level.

What is there to limit the LED current if your micro goes ape?? I see no

series resistor.
The problem now is the LED can not get enough of current, when it gets larger current, I will try to add resistors.

(1)Problem not because of current but it is bcz of voltage.
actually this happening because your led require 3.3v forward voltage and because of finite vce led not getting 3.3v so

if you can increase vcc your problem will be definitely solved.
I will do some test and come back to you later.

Please share the LED datasheet, or at least a part number.

And to follow up on Alec's resistor question, you're also not showing resistors on the base of either bjt.
Sorry I don't have the LED datasheet. But I can share what it says on the LED package:
Product Name: 5730 white. LUM:50-55LM. PartNum: CB-5730QWC-PT.
WaveLength: 6000-6500K. Current: 150MA. Voltage:3.3-3.6V

Actually I parallelly connected 4 LEDs, so it needs 600ma at max brightness, but 500ma is fine for me.

And about the resistors, the problem now is that I can't get larger current instead of limiting the current.

#### ebeowulf17

Joined Aug 12, 2014
3,275
You shouldn't power LEDs without resistors unless you're using a regulated current source of some kind. Typically if you do this, the LED will draw as much current as the power source can provide, then promptly overheat and fail. I'm guessing the reason this hasn't happened is because you're barely providing enough voltage to it.

The forward voltage of your LED is anywhere from 3.0-3.2V. If you're starting with 3.3V, that leaves as little as 0.1V to drop across the ce junction of your transistor. As you can see from your transistor datasheet below, limiting your available Vce limits the current it will pass:

You need a higher supply voltage in order to operate this LED with any useful level of control over current/brightness.

Back to resistors, you especially shouldn't run multiple LEDs in parallel without current limiting resistors. Variations in forward voltage between the individual LEDs mean that generally one LED will start conducting first, then hog all the current. It will warm up, which lowers its Vf, causing it to hog an even greater share of the current... until it overheats and fails. You're lucky that your supply voltage was too low to get through a transistor unimpeded, and that saved your LEDs. If you put a separate resistor for each LED in series with the LED, that will distribute the current between the LEDs fairly evenly, avoiding the hogging/runaway heating condition.

#### Bowen102

Joined Aug 23, 2017
7
(1)Problem not because of current but it is bcz of voltage.
actually this happening because your led require 3.3v forward voltage and because of finite vce led not getting 3.3v so if you can increase vcc your problem will be definitely solved.
I think you are the winner.

I changed the vcc source of LED directely to the power supply, and I ajusted the output voltage of power supply to 5V. I don't have to worry the input voltage of MCU, because there is a LDO.

And my LED array gets 600ma@3.3v now.

Now I am going to adjust the DC-DC convertor to ouput 5V for LED, and change the power supply voltage back to 3.7V, and see if it can work as expected.

Thanks, ankush_chavhan!

#### Bowen102

Joined Aug 23, 2017
7
You shouldn't power LEDs without resistors unless you're using a regulated current source of some kind. Typically if you do this, the LED will draw as much current as the power source can provide, then promptly overheat and fail. I'm guessing the reason this hasn't happened is because you're barely providing enough voltage to it.

The forward voltage of your LED is anywhere from 3.0-3.2V. If you're starting with 3.3V, that leaves as little as 0.1V to drop across the ce junction of your transistor. As you can see from your transistor datasheet below, limiting your available Vce limits the current it will pass:

View attachment 133586
You need a higher supply voltage in order to operate this LED with any useful level of control over current/brightness.

Back to resistors, you especially shouldn't run multiple LEDs in parallel without current limiting resistors. Variations in forward voltage between the individual LEDs mean that generally one LED will start conducting first, then hog all the current. It will warm up, which lowers its Vf, causing it to hog an even greater share of the current... until it overheats and fails. You're lucky that your supply voltage was too low to get through a transistor unimpeded, and that saved your LEDs. If you put a separate resistor for each LED in series with the LED, that will distribute the current between the LEDs fairly evenly, avoiding the hogging/runaway heating condition.

Yes, yes, you are right!
By increasing the power output of DC-DC converter to 5V, now I am able to consume 600ma total.

And about the resistors, I will add resistors to the led array panel. Thanks, that's something I can't learn directly from text books.

But now I am facing another problem, as schemetaic show below, as the LED taking more current (when increased to about 300ma~600ma), the LDO supplying power for the MCU failed to output 3.3v for the MCU. I am not sure how to fix this problem right now.

#### ebeowulf17

Joined Aug 12, 2014
3,275
...the LDO supplying power for the MCU failed to output 3.3v for the MCU.
What does it output? What are the voltages at both the input and at the output of the LDO when it's failing?

#### Bowen102

Joined Aug 23, 2017
7
What does it output? What are the voltages at both the input and at the output of the LDO when it's failing?
When the current goes larger than about 300ma, the power supply shows 3.7v, but the multimeter shows that the input voltage dropped to less than 3V (about 2.3v to 2.8v, I can't remember the exact value, I am not at may work place right now. ), so the LDO fail to out put 3.3V, and the output voltage of LDO is close to the input, below 3v.

#### ebeowulf17

Joined Aug 12, 2014
3,275
When the current goes larger than about 300ma, the power supply shows 3.7v, but the multimeter shows that the input voltage dropped to less than 3V (about 2.3v to 2.8v, I can't remember the exact value, I am not at may work place right now. ), so the LDO fail to out put 3.3V, and the output voltage of LDO is close to the input, below 3v.
So it's not the LDO failing. Something else is causing you to lose voltage before it reaches the LDO. I'd trace your way back from the LDO input to the power supply output and see where the discrepancy is.

Does the power supply have a current limiting function? Could it be limiting current, causing the voltage drop?

If you don't have resistors on the LEDs yet, they might be acting as a short circuit, drawing all the current and robbing other circuit elements of power.

#### Bowen102

Joined Aug 23, 2017
7
So it's not the LDO failing. Something else is causing you to lose voltage before it reaches the LDO. I'd trace your way back from the LDO input to the power supply output and see where the discrepancy is.

Does the power supply have a current limiting function? Could it be limiting current, causing the voltage drop?

If you don't have resistors on the LEDs yet, they might be acting as a short circuit, drawing all the current and robbing other circuit elements of power.

I soldered another PCB, and everything seems to be OK now. Don't know what's wrong with that PCB, don't have time to analyze the problem right now.

So that's it, problem solved.
Thanks!

#### ebeowulf17

Joined Aug 12, 2014
3,275
I soldered another PCB, and everything seems to be OK now. Don't know what's wrong with that PCB, don't have time to analyze the problem right now.

So that's it, problem solved.
Thanks!
Glad you got it working! If you ever do go back and analyze the failed board, let us know what you find. Until then, I'm glad you have a working board now.