Strange LED driver circuit

Thread Starter

AlbertHall

Joined Jun 4, 2014
12,345
This is the LED driver circuit from a thing I am trying to repair. I am attempting to draw a schematic. Q1 is marked S8050 and Q2 is marked S8550. I have checked this drawing multiple times because it didn't make sense. Checking the data I can find for the transistors they both seem to be EBC. I put the circuit into LTspice and it doesn't seem to work but does LTspice handle that PNP transistor with reversed polarity?
1652436267654.png
 

Attachments

Audioguru again

Joined Oct 21, 2019
6,674
The PNP will be damaged since it is upside-down.
The 3300 ohms resistor has a current in it of only (12V - 1.5V)/3300 ohms= 3.2mA for one IR LED and less current with two and for for colored LEDs, then two transistors are not needed.
 

Thread Starter

AlbertHall

Joined Jun 4, 2014
12,345
The PNP will be damaged since it is upside-down.
I don't think that's right. As you say the current is limited.

"The reverse hfe value is typically less than 0.1, more like 0.02 or so. In this mode, the bjt makes a lousy amplifier, but an outstanding switch. In the reverse mode, when driven into saturation, the collector-emitter voltage drop, Vcesat, can be tens of millivolts. That is hard to achieve in the normal, or forward mode."

Reference: https://www.physicsforums.com/threads/what-is-reverse-hfe.305553/
 

eetech00

Joined Jun 8, 2013
3,859
This is the LED driver circuit from a thing I am trying to repair. I am attempting to draw a schematic. Q1 is marked S8050 and Q2 is marked S8550. I have checked this drawing multiple times because it didn't make sense. Checking the data I can find for the transistors they both seem to be EBC. I put the circuit into LTspice and it doesn't seem to work but does LTspice handle that PNP transistor with reversed polarity?
View attachment 267096
1. Remove the "1 cycle" on the Pulse source so it will become a clock generator, then change .trans duration to 300mS so you can see the switching action of the LED's in the waveform viewer.

For experimentation, use native LTspice components:
2. Change the NPN=2n3904 PNP=2n3906
3. Change the LED's to NSCW100
4. Swap the C,E of PNP
5. Adjust R1 for the desired LED V/I.
 

Audioguru again

Joined Oct 21, 2019
6,674
Ii the fault of the circuit you are trying to fix caused because the manufacturer or a repair wrongly used a PNP instead of an NPN?
Is the problem fixed when the PNP is correctly replaced with an NPN?
 

Thread Starter

AlbertHall

Joined Jun 4, 2014
12,345
Ii the fault of the circuit you are trying to fix caused because the manufacturer or a repair wrongly used a PNP instead of an NPN?
Is the problem fixed when the PNP is correctly replaced with an NPN?
The transistor type numbers are marked on the board so they are the correct transistors in the correct place as the designer intended.
 

crutschow

Joined Mar 14, 2008
34,285
The transistor type numbers are marked on the board so they are the correct transistors in the correct place as the designer intended.
Then the designer has designed a strange circuit.

The LTspice simulation shows the correct operation of the circuit, which is that it does not work.
I suspect there is an error somewhere in your reverse engineering of the circuit.

How exactly, did the circuit work when it was operating correctly?
 

Thread Starter

AlbertHall

Joined Jun 4, 2014
12,345
Believe me I have checked the reverse engineering many times 'cos I didn't believe it either - but that is how it is built.
It showed red, yellow, or green (the two leds are in one package) depending on the state of charge of the battery.
 

Audioguru again

Joined Oct 21, 2019
6,674
Then the circuit is a battery voltage indicator.
The red LED indicates very dimly when the battery voltage is about 2.6V and gets a little brighter at higher battery voltages.
The yellow indicates very dimly when the battery voltage is 2.8V.
The green uses the reverse emitter-base of the PNP transistor as a 6V to 8V zener diode and the LED indicates dimly when the battery voltage is about 7.7V.
 
Top