Need input on design for LED circuit

[Doctor Hufflepuff]

Hello
For a project I am creating a plasma speaker complete with lights (specifically RBG LEDs) which are supposed to flash with the music. Rather than creating the speaker circuit and the three channel color organ circuit, which are well beyond my capabilities to make from scratch, I am using ones found online (http://www.instructables.com/id/LED...om/id/LED-Color-Organ-Triple-Deluxe/?ALLSTEPS and http://www.instructables.com/id/Make-a-Plasma-Globe-out-of-a-Light-Bulb/)

To make this project unique, I plan to utilize the multi-color ability of the LEDs to have two main settings: one for customizing preferred color for each of the three music frequencies (for example, setting the LEDs for low frequency to red, medium to purple, and high to blue, or any other combination), and the other for using the proportions of low, medium, and high frequencies to select the proportion of red, blue, and green in each LED, effectively playing a different color for each sound.

To do this I have developed the following circuit, which is essentially the color organ circuit connected to an array of voltage dividers. My question: can anyone who has experience/knowledge of how this circuit would behave tell me whether or not it will work, and does anyone have any suggestions on how to improve the design, preferably without increasing the price of components?

Notes: The switch I plan to use to alternate between modes is in the top right corner of the picture. I do not know if I need to add resistors between the switch and the TIP31 transistors I am using as electronic switches to direct the flow of power. If anyone knows, please teach me HOW to find the resistor values, as I am using this project for a learning experience as much as a fun way to pass the time.

If I left anything out please just ask. Thank you to anyone who can offer any help/input. Thanks

 

[Alec_t]

Welcome to EP!
Your circuit seems to have all red anodes connected to ground? It's unclear whether all the cathodes are connected to ground or to the tops of the rheostats? Where do the blue and green anodes connect?
LED brightness would normally be controlled by varying the current through the LED, not the voltage across it.
To get a decent visual effect you probably will need to add sample-and-hold circuits for the three channels; otherwise each LED will respond only briefly to signal peaks and will appear very dim.

 

[AnalogKid]

It looks like the switch on the far right and the six TIP31's form a multiplexer. If that is the intent, this circuit has several problems.

Once the TIP31 base is connected to Vcc, its emitter will be fixed at about 0.8 V below that and will not move regardless of the varying analog signal on the collector.

The three 2907's can modulate the current through the TIP31's, but they have only about 0.6-0.7V of headroom to work with.

The amount of current available will be limited by the 2907's. The fact that the TIP31's are power transistors will not help, because a common base amplifier acting as a diode switch has no current gain.

T5, 6, and 7 drive LEDs with no current limiting resistors, not good.

T8, 9, and 10 drive LEDs through 4.2K resistors, probably too large to give you any real brightness. Also, pots R11-R28 shunt current away from the LEDs, further decreasing brightness.

Since all LEDs have one end connected to GND, all outputs are driving the anodes, not the cathodes.

Since the emitter of T5 connects to all red anodes, they all are connected together even when T5 is off and T8 is driving things. This means there is no isolation among the LEDs in the T8-T10 output mode.

I'm not trying to rain all over your idea. There is a good approach underneath this circuit, but the implementation needs work. If you draw in the LED and trace the current paths, you will see some of the problems.

ak

 

[Doctor Hufflepuff]

It looks like the switch on the far right and the six TIP31's form a multiplexer. If that is the intent, this circuit has several problems.

Once the TIP31 base is connected to Vcc, its emitter will be fixed at about 0.8 V below that and will not move regardless of the varying analog signal on the collector.

The three 2907's can modulate the current through the TIP31's, but they have only about 0.6-0.7V of headroom to work with.

The amount of current available will be limited by the 2907's. The fact that the TIP31's are power transistors will not help, because a common base amplifier acting as a diode switch has no current gain.

T5, 6, and 7 drive LEDs with no current limiting resistors, not good.

T8, 9, and 10 drive LEDs through 4.2K resistors, probably too large to give you any real brightness. Also, pots R11-R28 shunt current away from the LEDs, further decreasing brightness.

Since all LEDs have one end connected to GND, all outputs are driving the anodes, not the cathodes.

Since the emitter of T5 connects to all red anodes, they all are connected together even when T5 is off and T8 is driving things. This means there is no isolation among the LEDs in the T8-T10 output mode.

I'm not trying to rain all over your idea. There is a good approach underneath this circuit, but the implementation needs work. If you draw in the LED and trace the current paths, you will see some of the problems.

ak

I have tried to take your comments into consideration, but I need some clarification on a few points. FIrst, I have changed the annode/cathode problem so that all inputs connect to annodes and the common cathode of each RBG LED connects to ground. I have also reduced the 4.2k resistors to 900 ohm resistors, and the 47k potentiometers to 10k potentiometers. I realize that this will result in a heavier current draw when set to zero, but I am not concerned with the effieiency of the design. Also, I did add resistors to T5, 6, and 7, as suggested.

The clarification I need is on why the 2907's will limit the current: I am trying to use them to sumply regulate the amount of voltage from 0 to 12(minus voltage drop across transistor) in order to power the rest of the circuit. How can I fix this, or what alternative is best?

Also, I do not know why the TIP31 emmitter will be fixed at .8v below the base, rather than the collector. Could this be changed by using a different transistor (I changed them to 2n222's in the diagram below), or is this a characteristic of all transistors?