Voltage Regulator Getting HOT

[Box_Stuffer]

I put together this tremolo that uses an LED run by a 555 timer using a 5 volt regulator. The first board that I built works fine, but on the second one I made the regulator gets really hot whenever it is hooked to the 9v source. I thought the regulator was bad, but I have tried 2 others and it keeps happening.

Could I just use a resistor to cut down from the 9v to 5v?






This is the board I built that works. The second one is EXACTLY the same - as far as the power hookup. Nothing is shorted. The joints are good. The power is the yellow wire and the regulator is to the lower left of the middle pot. 9v runs through the red wires to the LM386 and to the transistor between the 555 and the LED. Then, through the regulator and to the left pot.

[ElectricDruid]

According to the schematic, that's a BC547 NPN transistor, not a 5V voltage regulator. If you put a voltage regulator in there instead, then it might well get hot.

Try it with the right part and see what happens.

Using a resistor to drop voltage is not generally a good idea since the amount of voltage dropped depends on the current drawn (V=IR, remember!). Since the current is very rarely completely mixed and stable, the voltage will also vary, and this can cause all sorts of problems.

That schematic is a bit odd, tbh. Using a transistor to drive an LED like that is common enough, but you'd usually expect the LED and it's series resistor to be connected to the emitter. Like this there's a 1K+470R+LED chain connected from 9V to Ground when the transistor is off, and then the 470R+LED are shorted out to ground when the transistor is on.
Edit: I went to watch the video, and he says he did it like that because he wanted the output inversion it gives. Fair enough, I suppose.

[Box_Stuffer]

According to the schematic, that's a BC547 NPN transistor, not a 5V voltage regulator. If you put a voltage regulator in there instead, then it might well get hot.

Try it with the right part and see what happens.

Using a resistor to drop voltage is not generally a good idea since the amount of voltage dropped depends on the current drawn (V=IR, remember!). Since the current is very rarely completely mixed and stable, the voltage will also vary, and this can cause all sorts of problems.

That schematic is a bit odd, tbh. Using a transistor to drive an LED like that is common enough, but you'd usually expect the LED and it's series resistor to be connected to the emitter. Like this there's a 1K+470R+LED chain connected from 9V to Ground when the transistor is off, and then the 470R+LED are shorted out to ground when the transistor is on.
Edit: I went to watch the video, and he says he did it like that because he wanted the output inversion it gives. Fair enough, I suppose.

I changed it up a little from the original example. I used the 78L05 regulator into the 555 because I had done some LED projects that way and I knew what resistors I would need to get the desired brightness. I guess I can re-do it more like in the video, but I will have to do some trial and error with the resistor values to find the sweet spot because the intensity and color of the LED seem to have an effect on the LDR. At 5v the yellow LED with a 200ohm resistor works best. Like I said - the first one I built works fine somehow.

[antonis]

Is your regulator TO-92 (78L05) or TO-220 (7805)..??

[Elektrojänis]

If it gets hot on one build and not on the other there is a difference. However it might not even be an "error" on the one that makes the regulator heat up. It could also be that the one where it doesn't heat up has an "error" that somehow makes it behave.

If you want to lover tho voltage with just a resistor you need to experiment with the value. Correct value depends on the current and the current might not even be stable and the added resistance might make the whole circuit unstable and act weird in some way.

I'd say it would be far easier to build it how it was designed and experiment with the led series resistor to get the desired effect. The series resistance on other circuits might not match with this one anyway because here the real series resistance is 1kohm+470ohm=1470 ohm. If something doesn't use that "ground it to turn it off" -scheme, it probably doesn't have that extra 1kohm there anyway.

[DIY Bass]

A regulator getting too hot is being asked to deliver more current than it can handle.  Number one cause in my experience is an unintended short somewhere.

[ElectricDruid]

I changed it up a little from the original example. I used the 78L05 regulator into the 555 because I had done some LED projects that way and I knew what resistors I would need to get the desired brightness.

Aah! Ok, I now I get it. You didn't mention that before.

+1 what DIY Bass said, commonest cause of a regulator getting hot is a short in my experience too. Second-commonest is putting the regulator in back-to-front.

[antonis]

A regulator getting too hot is being asked to deliver more current than it can handle.

Actually, more power to dissipate..
(which is the product of current times In-Out voltage drop..)

@OP: Could you plz post a schematic of what exactly regulator feeds..??

[Box_Stuffer]

A regulator getting too hot is being asked to deliver more current than it can handle.

Actually, more power to dissipate..
(which is the product of current times In-Out voltage drop..)

@OP: Could you plz post a schematic of what exactly regulator feeds..??

Its just the most basic variable 555 setup I could find. The only thing added is the NPN transistor (as a "phase inverter"). It makes it flash differently than if you have it connected directly to the output pin. This setup is slightly different from what they use in the original design, but I had done it this way before (minus the transistor). All it has to do is flash the LED, there are probably several other ways to set it up.

They are probably right - I may have a short. I havent been able to find it yet.

[Rob Strand]

Wiring the 10k pot like that is a bad practice.  You should add a 470R or 1k fixed resistor in series with it.

As shown if you set the 10k pot to minimum, zero ohms, pin 7 shorts out the power rail.

Also the regulator should have caps on the input and output.

[antonis]

I think you need a 10k or so(*) resistor between pin 3 and BJT Base..
(when pin 3 is at "high" state, its voltage level is close enough to that of pin 8..)

(*) of value lower than (4.3V X h_FEmin) / 900μA, here..


P.S.1
For a fully saturated BJT (V_CE almost zero volts), Base resistor value should be much lower than calculated above but here you simply need a V_C-E equal to or lower than LED forward voltage drop..

P.S.2
Without that resistor, pin 3 draws about 100mA current (it looks like a 2.9V(**) voltage source  with a 30R resistor across +/-) forcing 78L09 to excessive current draw when LED is OFF..
(400mW power dissipation without heat-sinking can bring a TO-92 device close to its limits..)

(**) 4.3V is  exaggereted a bit 'cause output voltage depends on totem-pole transistors configuration current demand..
(see below for 10mA current through a grounded load..)

[duck_arse]

is the 10k on the npn and 200R still connecting to +9V, or the new 5V? if the 9V, there really is no point to running the 555 on 5V, because the led is still drawing its current, and limiting its current, from the 9V you started with. the npn is shunting 200R and the led, no matter what the led supply is.

[antonis]

the npn is shunting 200R and the led, no matter what the led supply is.

But 555 power supply DOES matter..
(at least, as it is - without BJT Base resistor..)

In case of LED & power supply same voltage level, say both at 9V, BJT Collector can't go lower than Base (9V) hence can't shunt the LED.. 

[Rob Strand]

But 555 power supply DOES matter..
(at least, as it is - without BJT Base resistor..)

The schematic at the top of the thread and the video use 1k but the OP's hand drawing is missing the base resistor.

The whole output stage is a bit weird.

The NPN transistor needs a low collector resistor in order to drive the LED.  In the original schematic the effective resistor is 1k + 470ohm = 1.47k.   The OP's ckt changes the 1k resistor to 10k, dropping the current.

The NE555 has plenty of drive to drive an LED directly and the whole NPN thing isn't adding anything.

You can drive LEDs directly like this:


On the OP's schematic the LED is on when the NE555 output is low (and the transistor is off).  So for the directly driven case you want to wire the LED to the positive rail via a resistor like the green LED in the linked schematic.

[antonis]

On the OP's schematic the LED is on when the NE555 output is low (and the transistor is off). 

Agree..

What I said to Stephen is that for both LED and 555 +9V supply and without Base resistor, LED should be permanently ON..

[Box_Stuffer]

But 555 power supply DOES matter..
(at least, as it is - without BJT Base resistor..)

the whole NPN thing isn't adding anything.

I don't know what I am doing. I have no engineering education or experience. I just thought it would be cool to buy a bunch of parts and start with really simple projects. I have started to mix elements of different circuits that I have seen and it obviously isn't always right.

I did get this to work on the breadboard somehow and the first board I wired up works. Somehow. Even with the 9v power to the NPN it doesn't burn out the LED. Also, running the signal through the transistor DOES make a difference in the way it flashes and it makes an audible difference in the effect. I don't know the physics behind it, but I have tried it both ways and it sounds different.

[Rob Strand]

I did get this to work on the breadboard somehow and the first board I wired up works. Somehow. Even with the 9v power to the NPN it doesn't burn out the LED.

The NPN transistor (and NE555) is copping a beating but the LED doesn't see much different.

Also, running the signal through the transistor DOES make a difference in the way it flashes and it makes an audible difference in the effect. I don't know the physics behind it, but I have tried it both ways and it sounds different.

It shouldn't make much difference if things are made to be the same (ie. LED current and on/off timing).

Just wiring the LED and resistor to NE555 doesn't make an apples to apple comparison.   When you change the configuration you need to adjust the LED resistor so the LED current in both configurations are the same.   Also, you need to wire the LED+resistor from NE555 pin3 to the positive rail.   If you wire the LED+resistor from NE555 pin 3 to ground it will sound completely different because the LED on and off times are then completely different.

You need to work out the LED current in you original circuit.  When the LED is on the current passes through both collector resistor and LED resistor, as if the two resistors are in series.

LED current I_LED = (9 - 1.7) / (Rcollector + RLED1)   ;   1.7V is the voltage drop for a RED LED.

For the NE555 with the LED + resistor wired from pin 3 to the positive rail the LED current is:

               I_LED = (9 - 1.7) / RLED2

For the same LED current  you need to choose RLED2 = Rcollector+RLED1

Rcollector    RLED1   RLED2 = Rcollector+RLED1
10k             200         10.2k
1k               1k           2k

I'm not sure what part values you have in your working circuit because the resistor values are different throughout the thread.

[antonis]

I don't know what I am doing. I have no engineering education or experience. I just thought it would be cool to buy a bunch of parts and start with really simple projects.

Good..!! 

I have started to mix elements of different circuits that I have seen and it obviously isn't always right.

No problem, as long as you understand individual building block function..