I made Christmas LEDs but I'm not sure how to amplify the output to support more leds!!!

Thread Starter

babaliaris

Joined Nov 19, 2019
160

I just made an oscillator using not gates and a capacitor. Then I could not support more than 1,2 LEDs so I connected the output of the oscillator
to the base of a transistor, short-circuited the controller with the 5vDC source and connected 5 LEDs in parallel with current input the current of the emitter.

It seems that I did an improvement but did I? Also can I do better?
 

dl324

Joined Mar 30, 2015
16,845
You've been told that you should drive LEDS low side if you're going to use an NPN transistor. If you're going to ignore what we try to teach you, we'll start ignoring you.

This hasn't come up yet, but you shouldn't wire multiple LEDS in parallel without using ballast resistors to account for mismatched forward voltages.

You need a base resistor and base current needs to be 5% of the emitter current.
EDIT: We know he's using BC547 from another thread.

LS04 can't source much current. You can use a darlington to increase current gain.
 
Last edited:

ci139

Joined Jul 11, 2016
1,898
drive multiple such LED modules (?with PNP), or use higher Wattage darlington or mosfet
/!\ but you do need to limit the LED-s' current by separate resistors or with a "constant current swith" (and still balancing resistors for leds ~ depends from avg. forward current of your LED-s) it's near the conduction slope of the led ~e.g. R.dynamic=∆U/∆I like that
 

Thread Starter

babaliaris

Joined Nov 19, 2019
160
Ok, I understand why is best to use a resistor for each led and one on the base but, the reason I didn't use any was that the output voltage was already low, and I used 5 LEDs. This is the reason I did not use any resistors. If I used any the LEDs would barely light up.

In my current set up the max I of the emitter is about 30mA (I probably need an oscilloscope to see it correctly, I'm struggling with a multimeter now, which also does not have AC ampere meter only DC). Also, the voltage in each led is about 2.5v.
 

Tonyr1084

Joined Sep 24, 2015
7,853
LED's are not voltage based components, they're current based components. Meaning you can supply a single LED with 5 VDC and 250Ω and drive the LED with 20 mA. (Ignoring forward voltage drops for the moment here) - OR you can supply that same single LED with 12 VDC and 600Ω and still drive 20 mA. OR you can supply 100 VDC and 5KΩ. You STILL get 20 mA.

I've ignored the forward voltage drop for the sake of demonstrating how voltage does not affect how the LED works, but it's the amount of current through the LED. IF I started with a 5 volt source and a 3 volt forward voltage drop I'd be left with 2 VDC for the resistor. In that case you'd use a 100Ω resistor. With 12 volts, 450Ω and lastly with 100 volts, 4.85KΩ. The higher the voltage the less significant the forward voltage is. But the bottom line is that LED's are driven by the amount of current flowing through them.
 

ci139

Joined Jul 11, 2016
1,898
what the graph below shows is LED forward voltage drop against LED forward current
e.g. a slight drift in voltage forces a large change in current -- is the reason for untying separate parallel leds (coz their currents at the same terminal voltage may differ signifficantly - even when they are all the same type . . .) /// using them in series string puts the same current through all the LED-s and the voltage drop differences (the differences of the power dissipated on each LED) would be insignifficant

OSPW5111P - pinge - vool.gif
 

TeeKay6

Joined Apr 20, 2019
573
Ok, I understand why is best to use a resistor for each led and one on the base but, the reason I didn't use any was that the output voltage was already low, and I used 5 LEDs. This is the reason I did not use any resistors. If I used any the LEDs would barely light up.

In my current set up the max I of the emitter is about 30mA (I probably need an oscilloscope to see it correctly, I'm struggling with a multimeter now, which also does not have AC ampere meter only DC). Also, the voltage in each led is about 2.5v.
Look again at the circuits I showed. You have failed to understand the advantage of using an NPN transistor with its emitter connected to ground. Using an emitter follower, as you did, has disadvantages. One disadvantage is that your circuit cannot provide enough voltage to drive the LEDs with current limiting resistors. The "better" and "best" circuits provide greater voltage to drive the LEDs and their resistors. Using the last "alternative best" circuit (with an N-channel MOSFET) you can drive hundreds of LEDs with a single 74LS04.
 

Thread Starter

babaliaris

Joined Nov 19, 2019
160
Look again at the circuits I showed. You have failed to understand the advantage of using an NPN transistor with its emitter connected to ground. Using an emitter follower, as you did, has disadvantages. One disadvantage is that your circuit cannot provide enough voltage to drive the LEDs with current limiting resistors. The "better" and "best" circuits provide greater voltage to drive the LEDs and their resistors. Using the last "alternative best" circuit (with an N-channel MOSFET) you can drive hundreds of LEDs with a single 74LS04.
You were right! I did the "best" approach, without using a transistor at all, (I just took the output from the oscillator and feed it to each resistor) and now I have 6 LEDs working!

So what was the mistake before? (Let's forget about the transistor). When I was connecting the LEDs in parallel without resistors why they were not working? I mean if the current was high enough they would be dead, so that means that the current was not enough? But now with the resistors which they limit the current more, they light up?

It does not make too much sense to me.
 

TeeKay6

Joined Apr 20, 2019
573
You were right! I did the "best" approach, without using a transistor at all, (I just took the output from the oscillator and feed it to each resistor) and now I have 6 LEDs working!

So what was the mistake before? (Let's forget about the transistor). When I was connecting the LEDs in parallel without resistors why they were not working? I mean if the current was high enough they would be dead, so that means that the current was not enough? But now with the resistors which they limit the current more, they light up?

It does not make too much sense to me.
You say "I did the "best" approach, without using a transistor at all." NO, you did not do the best approach because the best approach does use a transistor! You say "if the current was high enough they would be dead, so that means that the current was not enough." If you forced 10amps through one diode, it would surely die. However, an 74LS04 is not capable of providing even 0.1amp, for all six diodes combined. You say "It does not make too much sense to me." Well, it doesn't make much sense to me either, but since you are misusing the electronic devices, I do not care if it makes sense. I do not know exactly what is the schematic of what you are now doing; I do not know the value of the current limiting resistors that you added; I do not know what is the value of your "5V" power supply when 6 LEDs are lighted; I do not know what is the voltage on the anode and cathode of each lighted LED; I do not know what is the voltage at the output of the 74LS04 when the LEDs are lighted. Therefore, I am not surprised that it makes no sense to me. Electronic design is NOT about making random connections among components; electronic design means that you first understand a basic principle and then attempt to connect components in accord with that principle. You can continue to randomly connect any components you wish, but do not think you are doing design, you are merely playing with components.
 
Top