diyAudio (
-   Everything Else (
-   -   Circuit Help needed for graphic equalizer using LM3915 (

sandiegoman 9th September 2013 04:20 AM

Circuit Help needed for graphic equalizer using LM3915
2 Attachment(s)
I have a great graphical EQ that displays 8 freq using 6 led's. It currently uses 5mm leds and each gets around 2 volts. Most likely they are .20MA

I would like to take away a couple bands, maybe the highest, lowest and one in the middle and go down to 5 led's. So it will be a 5x5 graphical eq (spectrum analyzer)

I would like to change all the LEDs to 8mm Straw Hat Wide Angle White LED - Ultra Bright LEDs using 100-120ma and 3.2 - 3.6v.

I can't believe it, but they actually light up dimly when I swapped 1 row of them out, I would just like for them to have the max brightness.

I have attached the schematic and parts list, any help would be greatly appreciated.


DF96 9th September 2013 01:18 PM

'Bright LED' has two meanings:
1. brighter than an average LED at a particular lowish current.
2. capable of being very bright given a big enough current (that would kill a normal LED).

I fear you may be confusing these two meanings. The circuit diagram is difficult to read, but somewhere in it there will be some means to limit the current in the LEDs. You need to modify this, while keeping within the capabilities of the driving chips.

sandiegoman 9th September 2013 02:00 PM

It appears the driver is a CD4028BF, I see resistors and such all over the place, not sure which ones I should be looking in to. Any actual help would be greatly appreciated.

sandiegoman 9th September 2013 02:14 PM

I think what might actually be driving the LEDs is the CD4028BF. That seems to have plenty of power, but I am not 100% sure that is the role of that IC. It looks like the lm3915 is switching the negative on and off and the cd4028BF is providing the + voltage. There is a 10k resistor between each of the pins and a bc337 transistor:

  • Collector-Emitter Volt (Vceo): 45V
  • Collector Current (Ic): 0.8A
  • hfe: 100-250 @ 100mA
  • Power Dissipation (Ptot): 625mW
  • Current-Gain-Bandwidth (ftotal): 210MHz
  • Type: NPN
which then runs to the LED columns.
I am not sure if that will help.


DF96 9th September 2013 02:42 PM

My guess is that the 10k base resistors are doing the LED current limiting, via transistor current gain. You could try reducing the resistor value a little, but watch for overheating in the transistors.

sandiegoman 9th September 2013 05:16 PM

I added a jumper on each side of the 10k resistors and it actually made them less bright. I shorted out a row of LEDs and that made the rest brighter... I guess if the power dissipation is 625mW and there are 6 LEDs in a column, that could explain why the brightness is low? Each LED I am trying to use consume between 100-120ma. The voltage required is 3.2 - 3.6, but it seems like they are only getting around 2v. Could that be the problem? I am surprised they even lit up honestly. Not sure how to get the voltage and Ma up.

dangus 9th September 2013 09:34 PM

The LM3915 has built-in current limiting. So, to drive 100 mA you'd need to buffer the LM3915 outputs, and probably also do something about the row select transistors. Is there a version of the schematic that is more readable?

sandiegoman 10th September 2013 06:17 PM

I have uploaded them here. Any help would be GREATLY appreciated.

So the problem is not on the power end, but the grounding end on the lm3915 because it will only go up to 30ma?

I am wondering what resistors I can use for R45 and R46 to get the max

sandiegoman 10th September 2013 07:02 PM

Schematic is as good as it gets, still working to get one directly from the company...


dangus 11th September 2013 05:07 AM

Those scans are much more legible. The use of the transistors seems odd; they can't turn on fully. If the gain is 100, you'll see at least a couple of volts across them when driving an LED at 10 mA or so. I suppose that's "good enough" for the stock application, and maybe it helps to drop some voltage across the transistors rather than heating up the LM3915.

So, I'd add an open-collector inverting buffer to the 4028 outputs, and use those to pull down the bases of a set of PNP transistors, possibly TO220 package power transistors so as to have the option of driving 1W LEDs.
The LM3915 outputs would need pull-up resistors, then an inverting buffer could drive a set of NPN transistors. If the display operates in dot mode, a resistor would be sufficient to set the LED current, but in bar mode it would be best to modify the transistor drivers to act like constant current sources or sinks.

Alternatively, maybe it would be more sensible to omit the inverters, and let the LM3915 pull down the PNP transistors coming from +12, and the 4028 drive the NPN transistors. Unless I've overlooked some reason this won't work. It's late and I haven't had my dinner yet.

I will compare your RTA display with the Goldline RTA that I've been slowly building. Maybe they're similar enough that I can create a circuit that will allow either of them to drive higher power LEDs. I'd like to have the option of plugging it into a large wall-mounted display.

All times are GMT. The time now is 09:03 PM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 18.75%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio