Lately, I have been trying to make a graphic equalizer with a built in audio analyzer for my electric guitar. Unfortunately, the two features do not want to work together. What I have done, more or less, is make 10 bandpass filters using an LM248 op amp (I'm pretty sure that's a quad version of an LM741) in what I think is an inverting version with a potentiometer for feedback/volume control, and then send the output to the input of an LM3915 LED bargraph display driver. I used the standard circuit found in the datasheet of the LM3915, except I adjusted the values of the resistors to lower the signal voltage to about 1.25 V. Oh, and the LM3915 works PERFECTLY if I just skip the filtering and op amps and send in the signal from my electric guitar directly.
My problem, as you may guess from the title, is that I am getting too much DC noise from the op amp. I've tried putting a capacitor between the op amp and the LM3915, and that works eventually. It takes a while to fill up the capacitor (all I have are 1000 pF (the 1000 pF doesn't seem to do anything), a single 1 uF unpolarized electrolytic that is rather bulky, 10 uF, 22 uF, 47 uF, 100 uF and 220 uF). If I increase the output from the op amp by adjusting the potentiometer, I get more DC noise until the capacitor absorbs it. I have tried placing a resistor to ground after the capacitor (only later did I realize that this was actually a high pass filter), and the DC noise disappeared MUCH more quickly. . .as did the AC signal I wanted after a while. I tried a 1M resistor, which had a fairly slow rate of getting rid of the DC noise but delayed the inevitable AC signal loss, but a 100K resistor worked too fast. I could no longer get a reaction from the LM3915 after enough time had passed with this method.
I have included my basic plans as a pdf file. I originally made them with DIYLC, so it's basically the physical parts themselves. It's a bit messy and has some unnecessary things in the bottom corner I just put there for reference.
Am I just using a noisy and ancient op amp model? Would I be able to get rid of the noise by using a less noisy and more modern op amp (please tell me they have the same pinouts, because I really don't want to have to unsolder my filters I assembled)? I can't exactly afford new parts to try willy nilly now, but I do have some pieces like resistors (I've got plenty of those), some diodes, some LM386s, some transistors, and some other assorted pieces.
Oh, by the way, will I need to separate the sound filtering from the inputs for the LM3915s? I only thought about it a couple days ago, but I realized that the outputs of the filters would all be connected at the sound output and the higher signals could probably go backwards and overpower the weaker signals when they get sent to the LM3915. Would putting an op amp with a gain of 1 (a buffer?) between the input of the LM3915 from the output of the filter and the signal output solve this?
My problem, as you may guess from the title, is that I am getting too much DC noise from the op amp. I've tried putting a capacitor between the op amp and the LM3915, and that works eventually. It takes a while to fill up the capacitor (all I have are 1000 pF (the 1000 pF doesn't seem to do anything), a single 1 uF unpolarized electrolytic that is rather bulky, 10 uF, 22 uF, 47 uF, 100 uF and 220 uF). If I increase the output from the op amp by adjusting the potentiometer, I get more DC noise until the capacitor absorbs it. I have tried placing a resistor to ground after the capacitor (only later did I realize that this was actually a high pass filter), and the DC noise disappeared MUCH more quickly. . .as did the AC signal I wanted after a while. I tried a 1M resistor, which had a fairly slow rate of getting rid of the DC noise but delayed the inevitable AC signal loss, but a 100K resistor worked too fast. I could no longer get a reaction from the LM3915 after enough time had passed with this method.
I have included my basic plans as a pdf file. I originally made them with DIYLC, so it's basically the physical parts themselves. It's a bit messy and has some unnecessary things in the bottom corner I just put there for reference.
Am I just using a noisy and ancient op amp model? Would I be able to get rid of the noise by using a less noisy and more modern op amp (please tell me they have the same pinouts, because I really don't want to have to unsolder my filters I assembled)? I can't exactly afford new parts to try willy nilly now, but I do have some pieces like resistors (I've got plenty of those), some diodes, some LM386s, some transistors, and some other assorted pieces.
Oh, by the way, will I need to separate the sound filtering from the inputs for the LM3915s? I only thought about it a couple days ago, but I realized that the outputs of the filters would all be connected at the sound output and the higher signals could probably go backwards and overpower the weaker signals when they get sent to the LM3915. Would putting an op amp with a gain of 1 (a buffer?) between the input of the LM3915 from the output of the filter and the signal output solve this?
Attachments
-
311.9 KB Views: 20