Help with circuit to drive LED lights

[Firethelaser]

Hello,

(Please see attached schematic)

I have a set of 6 simple LED's that are each hooked to the single pole side of their own SPDT switch. The other poles are connected to a 5v source and 0-5v square wave signal source fed from a ZTX458 transistor. The switch selects between these two sources with the LED's activating when the square wave signal goes low (they blink).

My problem is that when the switch selects the square wave source, the fourth LED (from the left) fails to blink. The rest of them do. I checked the LED and it was not defective. When I measure the voltage drop on this LED I get 1.7v as compared the 0.4v I am getting with the other LED's.

When I remove this fourth LED, then the fifth one fails the blink. I used a 330 ohm resistor for this one because I ran out of 380 ohm resistors. I am wondering if this is the problem because the currents are not evenly distributed to the LED's? Any insight is greatly appreciated.

[IanB]

I don't think you need to have your 117 ohm resistor on the collector of the transistor. What happens if you remove it?

[bilko]

I don't think you need to have your 117 ohm resistor on the collector of the transistor. What happens if you remove it?

You will probably have to buy another transistor 

[T4P]

Eh ... where do i start?
Oh , did you only have the ZTX458 ? That's a pretty high voltage and high cost transistor 
Err ... i'm not entirely sure the transistor is happy driving 5 equal LED's then suddenly 1 unequal LED . LED's require constant current and i have never actually heard of somebody who placed a unequal LED before ....
But what staggers me is why would the 5th LED not blink? That i don't know .

[TerminalJack505]

The 117 ohm resistor is way too small.  It isn't going to allow the transistor's collector to get down near 0V.  Use something like a 10k resistor.

[MikeK]

I don't think you need to have your 117 ohm resistor on the collector of the transistor. What happens if you remove it?

It depends on the transistor and the collector supply.  If the supply can provide more current than the limit of the transistor, the collector current will only be limited by the gain.  If it's around 200, the collector current will be around 0.5A, which is too much if it's a TO-92 package.

EDIT: Also, what Jack said. 

[Firethelaser]

I had the ZTX lying around so i used it. I have it biased so that I am getting 40mA in the collector. The load for the sources require only microamps of current so it is more than enough.

I did not think the unequal resistance would be much of a problem since they do not differ by that much (50 ohms). Now I have to search for a 380 ohm to replace it with.

[TerminalJack505]

If you tie the transistor's base (through the 2.2k resistor) to 5V and measure the voltage at the collector what do you get?

I'm guessing based on the behavior you are reporting that it is going to be in the neighborhood of 3V to 3.5V.  You want it to be near 0V.

The reason that particular LED isn't coming on is that there isn't enough voltage to forward bias it.  This is due to a large chuck of voltage being dropped across that 117 ohm resistor.  Use a much larger resistor.

[Firethelaser]

If you tie the transistor's base (through the 2.2k resistor) to 5V and measure the voltage at the collector what do you get?

I'm guessing based on the behavior you are reporting that it is going to be in the neighborhood of 3V to 3.5V.  You want it to be near 0V.

The reason that particular LED isn't coming on is that there isn't enough voltage to forward bias it.  This is due to a large chuck of voltage being dropped across that 117 ohm resistor.  Use a much larger resistor.

I am getting 0.95v at the collector when the base is at 5v. This is enough to turn on the LED's but for some reason only 5/6 turn on.

[Kremmen]

If the circuit is really put together exactly as in the picture, then the 117 ohm resistor on the collector is meaningless. The 4th led fails because somewhere it is connected wrong.
The exact value of the series resistors is not critical and cannot be the reason for failure.

[TerminalJack505]

If the circuit is really put together exactly as in the picture, then the 117 ohm resistor on the collector is meaningless. The 4th led fails because somewhere it is connected wrong.
The exact value of the series resistors is not critical and cannot be the reason for failure.

Connecting each of the switches to 5V is unnecessary as well.

[IanB]

I don't think you need to have your 117 ohm resistor on the collector of the transistor. What happens if you remove it?

You will probably have to buy another transistor 

I'm not sure why? It seems to me if you remove the resistor and leave the transistor collector open, the circuit will still work fine. In addition there will be no parasitic current flow through the collector resistor.

[Kremmen]

Connecting each of the switches to 5V is unnecessary as well.

True, forgot to add that point. But it does not actually hurt and can't explain the failure.

[bilko]

I'm not sure why? It seems to me if you remove the resistor and leave the transistor collector open, the circuit will still work fine. In addition there will be no parasitic current flow through the collector resistor.

Oops, sorry misunderstood your comment. I thought you meant to connect the collector direct to the 5v rail. Removing the resistor is a good idea.

[Firethelaser]

You were right Kremmen, there was an open in the circuit. I had the entire circuit already on a PCB an apparently there was a solder joint not making proper contact with a trace. That was careless of me.

About the 117 ohm resistor. I want to get around 40mA of current in the transistor. VCE for saturation was min 0.2 volts. Using KVL:

0.2+Rc*(40mA)=5

Rc is around 117 ohm

Im still a bit of a novice in electronics but I always thought there had to be a resistor in either the collector or emitter. Would'nt the transistor have burnt out if there was no resistor?

[bilko]

Would'nt the transistor have burnt out if there was no resistor?

See IanB's posts, I've edited your schematic

[Kremmen]

Ok, no surprise there then. Maybe a lesson to learn: double checking is not enough, at least triple check
How many times we all have heard the words "i've tried everything" when in fact 2 things were tried - and the second only because the first didn't immediately work.

That reasoning around the collector resistor is curious and confusing. The collector doesn't "need" anything in particular. You obviously don't want to short circuit the transistor but the extra resistor doesn't help there. When switched on, the leds all have the series resistor already so that takes care of that. The only thing you need for the transistor itself is to drive it sufficiently to bring it to reasonable saturation when all leds are connected, and secondly that the Icmax is enough to take the total current. (plus cooling of course if that is needed).

[Firethelaser]

@Kremmen: i see what you mean. It makes sense.

@yachtronics: I was confused with the edit at first but after staring at it for some time I caught the redundancy.

The circuit is already soldered up on the PCB. It works as it should now but I learned a couple of things 

Thanks guys!