I redesigned the circuit and have all the valus for resistors inputed.

I have assumed that Vbe is 1.2V
All the LEDs labeled G are at 3.2V, LEDs labeled Y 2.2V
strings are made of 2 G LEDs w/ 180ohm resistors and 3 Y LEDs w/ 170 ohm resistors and also a 2 Y LED string w/ 280ohm resistor.

I'm not exactly sure if everything works out, but I think it does.

I'm just having some trouble taking it from here to a PCB designer program w/ netlists, but just take it one step at a time and see if this circuit is alright?

 

Nothing in your circuit will turn on the LEDs because nothing applies a positive voltage to them.

When the transistor conducts then it grounds the anodes of the LEDs. The other end of the LEDs is also grounded by the resistors.

Since the output of the 555 does not go high enough to turn off a PNP transistor then it would need a voltage divider on its base which might reduce the base current to be too low.

Redesign the circuit so that the anodes of the LEDs connect to +10V and the collector of the NPN transistor grounds the resistors to turn on the LEDs.

 

Sorry wrong connection I messed up the resistor connections they should have been connected to the power supply. Here's the new one. I just realized that all my LEDs are flipped, they should be the other way around, so that current flows through them towards the transistor.

 

Why is R3 280 Ohms?

 

Because its connected to two Y LEDs which is a drop of 4.4 over both LEDs. Leaving a drop of 5.6V across the resistor and then you got 20mA running through the resistor and the LEDs.

So 5.6V/20mA = 280ohms

 

Okay... then "G" is for green and "Y" is for yellow?

 

Yes...which is why the LED voltages are different. The voltages are based off of the LEDs I will be purchasing very soon.

 

To avoid having to calculate the resistor values depending on LED fV and supply voltage you could drive the LEDs from constant current sources.

The attached circuit will drive each LED string at around 20mA regardless of supply voltage, LED type or number of LEDs (as long as total string fV is less than the supply by at least 1V). It looks complicated but would be very easy to build on stripboard.

Just repeat the driver stage for more LEDs, but for a very large number of LEDS a smaller value of resistor on the output of the 555 will be required to supply the extra base current.

Any small general purpose NPN transistors will be suitable but if you want to be really picky, the transistors should be sorted so that they have closely matching base-emitter drops but with unmatched transistors the differences in current will be small.

The 555 connections are not detailed.

 

Would what I had drawn not work?

 

Yes, probably exactly as you want it too. Mr. Spirale design has one advantage though, he has wired the transistors as constant current devices, which will equalize the brightness on the LEDs quite well.

Basically it is your call. I love how the numbers jump out at you on your schematic, but it makes it hard to proof read, as the number of diodes on each leg is pretty important. The other side is I suspect you will be able to tweak it easily if something isn't quite right. I love LED projects, I think you have a working schematic for what it's worth.

 

Would what I had drawn not work?

This is just another way of doing things which addresses most of the problems associated with a plain old resistor current-source. The downside is that more transistors are required.

 

I was just wondering if I was anywhere close, out of curiosity.

So Mr. Spirale's circuit just adds constant current sources (transistors) to each LED string.
What's with the resistor and diodes after the 555 timer?

 

They set the reference voltage for the current limiter.

Like I said, your existing schematic would work. All the current limiters add is a certain consistency. When using long chains of LEDs they compensate for variations in the voltage drops on LEDs.

 

I was just wondering if I was anywhere close, out of curiosity.

So Mr. Spirale's circuit just adds constant current sources (transistors) to each LED string.
What's with the resistor and diodes after the 555 timer?

The diodes are used to create a reference voltage and fix all the transistor bases at about 1.2V (2 x 0.6V diode fV).

Due to the way transistors work, the emitters are fixed at 1 diode drop (0.6V) below the bases.

Therefore, we have around 0.6V across each emitter resistor and this never changes. The emitter resistors never change either which means that the emitter current also doesn't change - it is constant.

Since collector current = emitter current, by defining the emitter current we have also defined collector current.

 

But to limit the current the resistors attached to each transistor needs to be
0.6V/20mA = 30ohms. Correct?

So what is Ib or what state is the transistor in? Does it even matter to know what state it is in?

 

Yes you are correct, it would be 30 Ohms with those values. A few ohms here and there aren't important and neither is calling a diode voltage drop 0.6 or 0.65 Modern LEDs are pretty robust so a bit more current is not a problem.

The transistor will be partially on. Ib will be Ic/hFE per transistor.

 

Why would Ib = Ic/hFe?

 

Why would Ib = Ic/hFe?

That's basically the way transistors work.

Note that hFE is not flat across the entire Ic range of a transistor. It will help you a good deal to look at charts in the datasheets of various transistors.

With small signal transistors, it's typical for hFE to be moderate for low Ic, then to peak somewhere between 1mA-20mA, and then fall off as Ic increases. You can get inexpensive DMM's that have built-in testing for NPN/PNP transistor's hFE. You get an hFE readout at a fixed collector current, generally around 10mA. Very handy tool to have.

 

I'm guessing by saying the transistor is on, you are saying it is in forward active? where BE acts as a battery, and CE acts as a current source, thats the way I learned it in class, the only thing they told me to watch out for in the real world, is interchangeable use of Beta and hFE.

If the transistor is in forward active, then it would mean that Ib = BIc (Ib=hFEIc).

One thing I did not learn was that hFE changes.

About the entire project, I'm putting it on hold for a little bit because I got a new job, and school is very hectic, but I really appreciate all the help I recieved. Whenever, I get the chance I'll try to redesign the circuit, I'm thinking of reducing the number of LEDs, and I'll try to get all the equipment I need to build this.

And lastly, anyone know where I can get the schematic for the 555 Timer on Eagle, I have'nt fully learned how to make individual component schematics in Eagle.

Thanks,
PNB