PNP transistor current supply

Thread Starter

abc14

Joined Oct 15, 2017
123
Hi guys,


I have non-inverting digital amplifier/switch using an npn-pnp pair of transistors. Now this circuits works perfectly fine. But as soon as the vcc coming out of pnp is fed to boost converter IC, the VCC at pnp drops to .5v volts ? As if pnp is turn off ?


I am guessing this is due to pnp fails to provide enough current for boost converter IC ? your thoughts are welcome.


upload_2017-12-23_15-31-4.png
 

OBW0549

Joined Mar 2, 2015
3,566
I don't know how much current your boost converter wants to draw from Q3's collector, but very likely it's more than Q3 can provide with the component values you're using-- i.e., Q3 probably doesn't have enough base drive.

My suggestion would be to reduce the values of R2, R3 and R4 all by a factor of 20, and see what happens.
 

Audioguru

Joined Dec 20, 2007
11,248
The base current of Q3 is about (3V - 0.2V-0.7V)/20k= 0.1mA then its saturated collector current will be only 0.1mA x 10= 1mA. Almost nothing.
For Q2 to saturate with a collector current of the 0.1mA plus the 28uA in R3, its base current must be 12.8uA. Then the input to R2 must be 12.8uA x 1M= 12.8V.

1) Why do you ask a question about your problem of transistor current without telling us how much current the boost converter needs?
2) Why do you ask about transistor current when the datasheets show the transistors saturating when the base current is 1/10th the collector current?
 

Thread Starter

abc14

Joined Oct 15, 2017
123
My boost converter is driving three leds which could draw upto 90mA max current, so according to my calculations pnp is expected to provide upto 270mA.

upload_2017-12-23_15-59-46.png
 

Alec_t

Joined Sep 17, 2013
14,280
Unless your LEDs have built-in current limiters or are very well matched they should each have a series resistor. Are you relying on R5 and the gain of Q4 to provide current-limiting?
 

Thread Starter

abc14

Joined Oct 15, 2017
123
Yes, you can connect the boost converter direct to your VCC supply and turn it on or off using the enable pin.

True, but issue is VCC supply is coming form pnp. So pnp has to supply enough current for boost to work, regardless how we turn it on or off.
 

ScottWang

Joined Aug 23, 2012
7,397
For 270mA then the R2=820Ω and R4=76Ω
If the 270mA is the rating current for 3 LEDs then you can calculates 80% of 270mA *80% = 216mA then the R2=1045Ω(use 1K+47Ω) and R4=91Ω or 100Ω
R3 = 10~22K.
 

Thread Starter

abc14

Joined Oct 15, 2017
123
Then you can use your electrodes to work the enable pin.
true but two issues here, they meant to detect water. second issue I already have pcb printed along side stencil, so only thing I can change is resistor values or part numbers of transistor.

Your right, If I have the option of modyfying the circuit, I could simple feed vcc_1 to boost.

For 270mA then the R2=820Ω and R4=76Ω
If the 270mA is the rating current for 3 LEDs then you can calculates 80% of 270mA *80% = 216mA then the R2=1045Ω(use 1K+47Ω) and R4=91Ω or 100Ω
R3 = 10~22K.

I will try that, one stupid question, if pnp fails to provide required level of current, what does it do , goes to saturation or cut-of state ?

For 270mA then the R2=820Ω and R4=76Ω
If the 270mA is the rating current for 3 LEDs then you can calculates 80% of 270mA *80% = 216mA then the R2=1045Ω(use 1K+47Ω) and R4=91Ω or 100Ω
R3 = 10~22K.

Another thing according to datasheet max collector current is 200mA,

upload_2017-12-23_17-44-8.png
 
Last edited by a moderator:

ScottWang

Joined Aug 23, 2012
7,397
I will try that, one stupid question, if pnp fails to provide required level of current, what does it do , goes to saturation or cut-of state ?
I was calculated the current as the transistor get into the saturation region Ib:Ic=1:10 like 2.7mA → 27 mA → 270mA, So the Q2 should be use rating current as I>=1A like 2SA684 or others, the rating current of Ic is more big more better.
 

Audioguru

Joined Dec 20, 2007
11,248
Do you understand that your LEDs have nothing to limit their current so the boost converter and the LEDs will instantly burn out (if you fix the input circuit)?
 

Thread Starter

abc14

Joined Oct 15, 2017
123
Do you understand that your LEDs have nothing to limit their current so the boost converter and the LEDs will instantly burn out (if you fix the input circuit)?
I understand but somehow they working all fine, As i test I supplied VCC_1 from bench power supply to the input of boost, they seems to work just fine. VF forward voltage of leds is 3.3V.
 

Audioguru

Joined Dec 20, 2007
11,248
Since your boost converter and LEDs are not burning out then something must be limiting their current. Oh, the datasheet for the boost converter says it has a fixed 3.3V output. Then maybe your LEDs draw 10mA or 20mA each when powered from the 3.3V. Then the LEDs draw 30mA each when they get warm then each LED draws 100mA and burns out when it gets hot. It is a bad design with nothing to limit the current.
 

Thread Starter

abc14

Joined Oct 15, 2017
123
Since your boost converter and LEDs are not burning out then something must be limiting their current. Oh, the datasheet for the boost converter says it has a fixed 3.3V output. Then maybe your LEDs draw 10mA or 20mA each when powered from the 3.3V. Then the LEDs draw 30mA each when they get warm then each LED draws 100mA and burns out when it gets hot. It is a bad design with nothing to limit the current.

Not really, the leds are not meant to be always on, they will have flashing around 10Hz with 25% duty cycle, so I cannot see them getting warm.
 
Top