Sounds like missing the capacitor screwed up the dc conditions.
From your diagram the op-amp is already running off one 9 volt battery so that is a single supply.
You would have to bias the op-amp none inverting input to half the supply with two 100k resistors to lose the voltage splitting circuit.
OK, thanks. Yeah I guess I worded it wrong. I meant off a single rail. I will give what you suggested a try and see how it goes.
You're getting your terms mixed up... it is running "single-supply", "single-rail" - whatever you want to call it. The way you've named your voltage divider is what's throwing you.
The "top" of the supply that you have named "+5" is in fact "+9" (look at the top line - it can't be both). The center-tap of the voltage divider is "VCC/2" or "+4.5V" or 'mid-bias" not "0V". I use "VCC/2" when running off batteries because this will node will change voltage as the battery drains (which means it won't always be at "+4.5"..
An aside: to be very specific VCC/2, depending on the device can be technically incorrect as well (it could be VDD/2). Have a look here for the full scoop: IC power supply pin - Wikipedia, the free encyclopedia
"-5" is actually "ground" (or common). There is no "negative voltage" in this circuit. What you have done is created a "virtual ground. Prowl around the net searching on "virtual ground opamp". A good starting point as well is this: http://focus.ti.com/lit/an/slyt155/slyt155.pdf. Don't get thrown with using the opamp follower as they show; in a circuit such as yours you don't need anything more than the voltage divider. Millions of guitar pedals sold each year don't lie...
It looks like you've overlooked gain with the LM386; you can increase the volume with a series cap/resistor between pins 1 and 8 (look at the data sheets for how to calculate this gain).
I don't think you need R4 either (short it or take it out).
Last edited:
I was under the impression that the +4.5 and -4.5 were in relation to the center tap. I guess I labeled it wrong, but is that an incorrect thought? I added the 9v and ground symbols so that when I did the board layout the traces would be connected to the 9v source and the actual ground. 🙁
I will look that over closely this afternoon. I have a very busy day and while I would rather be at home playing, I have to bring home some bacon!😀
I started with a cap across pin 1 and pin 8, but it was too much gain for what I was after. The datasheet showed a gain of 20 with no cap in there, so I opted for that. Was that incorrect thinking?
I did reduce r4 to 1K. Should I remove it completely?
Here is the schematic as it stands now.
Ahhh... the question of "reference". You are correct; using the "0V" point from the voltage divider as your reference it does swing +/- around that. There is nothing wrong with that at all. If I were drawing the schematic I would have named the points as "+V", "Vcc/2" and "Ground" and made a note next to the battery "9V battery". If you were to distribute the schematic it might throw people having the name "+9" and "+5" on the same node.
You can use a resistor in series with the cap between pins 1 and 8 on the LM386 to tailor the gain to anything you want. Have a look again at the data sheet.
Yes, remove R4 completely; it's doing nothing in the circuit (non-inverting amp with gain). You need that resistor (and different topology) in an inverting amp. Now that I look at the schematic again, you don't need resistors R1 and R2 either (I guess you could make the argument that the 1uF and 10K make a low-pass filter with a corner at 16 Hz 🙄 ).
You can use a resistor in series with the cap between pins 1 and 8 on the LM386 to tailor the gain to anything you want. Have a look again at the data sheet.
Yes, remove R4 completely; it's doing nothing in the circuit (non-inverting amp with gain). You need that resistor (and different topology) in an inverting amp. Now that I look at the schematic again, you don't need resistors R1 and R2 either (I guess you could make the argument that the 1uF and 10K make a low-pass filter with a corner at 16 Hz 🙄 ).
Here is a new schematic. I have completely reworked the buffer section again. I got rid of the split supply and went with a voltage divider setup. I am still plagued with a bit of a hiss and some weird outer space sounds when I adjust the volume controls but overall it works pretty good.
I could use some advice on the noise issue if anyone sees a problem. I have tried several different cap values for c14, but none quieted it completely.
Thanks for all of the wonderful advice so far!
J
I could use some advice on the noise issue if anyone sees a problem. I have tried several different cap values for c14, but none quieted it completely.
Thanks for all of the wonderful advice so far!
J
Bypass R10 to gound with something like a 10uF cap.
Bypass the TLO82 supply pin with a 10uF (or bigger - electrolytic or tantalum) and a 0.1uF ceramic right at the supply pin.
What's the other half of the opamp doing? Ground the inputs if they are floating.
C14/R9 calculates to to a high pass of 398KHz 😱 Make C14 something like 800pF (or so). This works out to about 20KHz. Cf=1/2*PI*Freq*R. The resistor/cap in the feedback loop sets the high-pass filter; the resistor/cap on the input sets low-pass filter.
You still don't need R1 and R2.
Broadband noise (hiss) generally sucks. Are you point-to-point soldering this? Make sure all your grounds return to the same point. If you are doing this on a "plug in" proto-strip you're fighting yourself...
Bypass the TLO82 supply pin with a 10uF (or bigger - electrolytic or tantalum) and a 0.1uF ceramic right at the supply pin.
What's the other half of the opamp doing? Ground the inputs if they are floating.
C14/R9 calculates to to a high pass of 398KHz 😱 Make C14 something like 800pF (or so). This works out to about 20KHz. Cf=1/2*PI*Freq*R. The resistor/cap in the feedback loop sets the high-pass filter; the resistor/cap on the input sets low-pass filter.
You still don't need R1 and R2.
Broadband noise (hiss) generally sucks. Are you point-to-point soldering this? Make sure all your grounds return to the same point. If you are doing this on a "plug in" proto-strip you're fighting yourself...
Last edited:
Man I sure appreciate you sticking in with me on this project. You have been a tremendous help and source of info!!!!!!
I made the changes that you recommended. Hopefully I got all of them this time. I am not ignoring your advice so don't think for a minute that I am.
Here is the revised schematic. I will bread-board it tomorrow and see how it sounds.
No. So far I am bread-boarding it so I know it won't be perfect. I will be making a surface mount pcb, hopefully with star ground for this once I get it close.
I made the changes that you recommended. Hopefully I got all of them this time. I am not ignoring your advice so don't think for a minute that I am.
Here is the revised schematic. I will bread-board it tomorrow and see how it sounds.
No. So far I am bread-boarding it so I know it won't be perfect. I will be making a surface mount pcb, hopefully with star ground for this once I get it close.
Your latest circuit diagram is very wrong.
The 10k resistors mixing in the two top sources have been removed.
You need an input resistor on the opamp inverting input.
R12 is not required.
I would have made the opamp have the guitar input on the none inverting side with the two 100k resistors as a guitar is high impedance out. You will be killing the guitar input with the current way.
I would have done the gain through a 100k feedback resistor and small cap then through a 10k to ground through a 10uf capacitor.
The 10k resistors mixing in the two top sources have been removed.
You need an input resistor on the opamp inverting input.
R12 is not required.
I would have made the opamp have the guitar input on the none inverting side with the two 100k resistors as a guitar is high impedance out. You will be killing the guitar input with the current way.
I would have done the gain through a 100k feedback resistor and small cap then through a 10k to ground through a 10uf capacitor.
Last edited:
Hey, thanks for the drawing! I made it on the inverting input because it was the only way I could find a working example. I really am just studying this as I go, so if I do something that doesn't make sense it is because I am very new to the design part of this stuff. I am very capable when it comes to making boards, soldering, and reproducing other people's schematics, but it is another thing altogether when it comes to designing circuits. I don't ignore the advice I am given, but sometimes I don't understand it. The problem with trying to learn this way and not going to school for it, is that you get a bunch of different opinions and you have to cipher through it all and get something that works for you.
I appreciate all of the advice that I am given and I try all of it that i understand. So if something you say doesn't stick right away please don't give up on me. I am like a big sponge and I am trying to suck it all in!😀
That looks better. I might have turned around C3 as it has 4v5 on its -ve end and ground on the +ve end which is not good.
R1 and r2 are needed becasue if the MP3 player is on at the same time as the guitar and the mp3 player input is at full volume it will short out the guitar opamp output signal after the 10k.
Last edited:
OK, I didn't mean to seem like I was disagreeing with you. I just wanted to make sure I would set it up like you would!🙂
Is this Mo BETTER?
Last edited:
It looks fine now, just needs testing out to make sure it all works and sounds good.
Very good. I will test it in the morning and post back. Thanks for the help!!!!!😎
I think I have settled on THIS DESIGN unless someone sees some glaring issues with it. I have renamed the components to organize it a little. I can get it a little quieter if I raise the value of C4, but it's a trade off. I get a little clipping in the buffer if I have the volume up. not sure exactly why except maybe my battery may be getting low.
I appreciate all of your input! And if you have more I am more than happy to listen.
J
Ooops...
I missed when things flipped from non-invert, to invert and back again so the input resistor on the inverting version was needed.
C3 should be flipped around (or better yet, use bi-polar caps on all the inputs).
I'm not following; with R7 and R8 in the circuit, what's the need for R1 and R2?
I missed when things flipped from non-invert, to invert and back again so the input resistor on the inverting version was needed.
C3 should be flipped around (or better yet, use bi-polar caps on all the inputs).
I'm not following; with R7 and R8 in the circuit, what's the need for R1 and R2?
- Status
- Not open for further replies.