I’ve reached the limit of my knowledge, and could use some help tweaking an amp.
I’m building a headphone amp that can mix line-level signals with the signal from the headphone output of a stereo. I need a way to practice guitar and bass along with a CD without waking up everyone in my house.
My plan was to take the LM386 headphone amp designed by Alan Gary Campbell (http://headwize.com/projects/showfile.php?file=guitar_prj.htm) and add a mixing stage at the front end. I have a design that works, but there are problems that need to be fixed before the amp is useable.
The first thing I did was build Campbell’s amp configured for a line-level signal (since I run my guitars through a compressor/limiter before going into the amp). The amp worked just fine. (I later read where someone said that you should not bypass the inverting input (#2) with a capacitor, so I removed the 0.0047uF capacitor (C3 in the schematic) and sent the inverting input straight to ground. Nothing changed as far as I could tell.)
The next thing I did was to try to figure out how to reduce the signal from my CD player's headphone jack to around the same level as the line-level signal from the guitar. I used 100k resistors to sum the left and right channels into a single signal, and then attenuated the mono signal with a 100k/4.7k voltage divider. According to the ear test, the resulting signal through the amp seemed to be at around the same level as the line-level signal. So far, so good.
The problem began when I tried to put it all together and mix the guitar and music signals. I tried summing the guitar signal with the music signal (already summed and attenuated as described) using 100k resistors. The resulting signal through the amp was not what I expected. The music level was a little loud and the quality seems to have been affected. The level of the guitar was lower than the music level, and began to distort as the volume was increased. I used my bass guitar and there was more distortion with the low E string than with the other three strings. (I only used the 100k ohm volume control for the guitar because I only had one audio pot; in the final, tweaked version of the amp, I would like to have a volume control for the music, too).
My schematic is attached.
I don’t understand impedance as much as I’d like to, and I suspect this is part of the problem. The LM386 datasheet (http://www.national.com/ds/LM/LM386.pdf) says that the input resistance of the opamp is 50k ohm. I’ve seen posts that say that the pickup impedance for electric guitars “is over 20K and really wants over 100k loading … 100k – 470k are typical values.” I don’t know what this means, or if it also applies to bass guitars. Or if it even makes a difference in what I’m doing. I suspect that it does.
This is where I could really use some advice. Thanks.
I’m building a headphone amp that can mix line-level signals with the signal from the headphone output of a stereo. I need a way to practice guitar and bass along with a CD without waking up everyone in my house.
My plan was to take the LM386 headphone amp designed by Alan Gary Campbell (http://headwize.com/projects/showfile.php?file=guitar_prj.htm) and add a mixing stage at the front end. I have a design that works, but there are problems that need to be fixed before the amp is useable.
The first thing I did was build Campbell’s amp configured for a line-level signal (since I run my guitars through a compressor/limiter before going into the amp). The amp worked just fine. (I later read where someone said that you should not bypass the inverting input (#2) with a capacitor, so I removed the 0.0047uF capacitor (C3 in the schematic) and sent the inverting input straight to ground. Nothing changed as far as I could tell.)
The next thing I did was to try to figure out how to reduce the signal from my CD player's headphone jack to around the same level as the line-level signal from the guitar. I used 100k resistors to sum the left and right channels into a single signal, and then attenuated the mono signal with a 100k/4.7k voltage divider. According to the ear test, the resulting signal through the amp seemed to be at around the same level as the line-level signal. So far, so good.
The problem began when I tried to put it all together and mix the guitar and music signals. I tried summing the guitar signal with the music signal (already summed and attenuated as described) using 100k resistors. The resulting signal through the amp was not what I expected. The music level was a little loud and the quality seems to have been affected. The level of the guitar was lower than the music level, and began to distort as the volume was increased. I used my bass guitar and there was more distortion with the low E string than with the other three strings. (I only used the 100k ohm volume control for the guitar because I only had one audio pot; in the final, tweaked version of the amp, I would like to have a volume control for the music, too).
My schematic is attached.
I don’t understand impedance as much as I’d like to, and I suspect this is part of the problem. The LM386 datasheet (http://www.national.com/ds/LM/LM386.pdf) says that the input resistance of the opamp is 50k ohm. I’ve seen posts that say that the pickup impedance for electric guitars “is over 20K and really wants over 100k loading … 100k – 470k are typical values.” I don’t know what this means, or if it also applies to bass guitars. Or if it even makes a difference in what I’m doing. I suspect that it does.
This is where I could really use some advice. Thanks.
Attachments
Hi,
if you use shunt feedback (as opposed to series feedback)
then the virtual earth at the input becomes a mixer stage.
Consequently you can change the gain for each input by
setting its input resistor value, (determines current through
to the virtual earth point), using potentiometers you can
make each input variable.
Note the input resistor value for shunt feedback is input resistance.
This should not be an issue if you use the compressor/limiter for
both guitar and bass.
/sreten.
if you use shunt feedback (as opposed to series feedback)
then the virtual earth at the input becomes a mixer stage.
Consequently you can change the gain for each input by
setting its input resistor value, (determines current through
to the virtual earth point), using potentiometers you can
make each input variable.
Note the input resistor value for shunt feedback is input resistance.
This should not be an issue if you use the compressor/limiter for
both guitar and bass.
/sreten.
Thanks, sreten.
Confession: I don't know what "shunt feedback" is. Your suggestion sounds much cleaner, though (i.e., fewer parts in the signal path), so I would like to understand.
Last night I tried using 10k resistors to sum the stereo music signals to mono, then attenuated the music signal with a 10k/480 ohm voltage divider. The resulting music signal was then summed with the guitar signal using 10k resistors (after the guitar signal ran through the 100k audio pot). This seemed to work, although I have no idea whether something else, such as your idea, might work better.
Even if I use shunt feedback, will I still need to sum the stereo music signals to mono and then sum the mono music signal and guitar signal? My question stems from an article I read recently entitled, “Why Not Wye” (http://headwize.com/tech/wye_tech.htm), in which the authors state:
“A wye-connector used to split a signal into two lines is being used properly; a wye-connector used to mix two signals into one is being abused and may even damage the equipment involved…. Outputs are low impedance and must only be connected to high impedance outputs -- never, never tie two outputs directly together -- never. If you do, then each output tries to drive the very low impedance of the other, forcing both outputs into current limit and possible damage. At a minimum, severe signal loss results.”
The authors then go on to describe how to use resistors to sum two signals.
I would like to hear your thoughts on all of this. And get tutored on shunt feedback. Thanks again.
Ravelo
Confession: I don't know what "shunt feedback" is. Your suggestion sounds much cleaner, though (i.e., fewer parts in the signal path), so I would like to understand.
Last night I tried using 10k resistors to sum the stereo music signals to mono, then attenuated the music signal with a 10k/480 ohm voltage divider. The resulting music signal was then summed with the guitar signal using 10k resistors (after the guitar signal ran through the 100k audio pot). This seemed to work, although I have no idea whether something else, such as your idea, might work better.
Even if I use shunt feedback, will I still need to sum the stereo music signals to mono and then sum the mono music signal and guitar signal? My question stems from an article I read recently entitled, “Why Not Wye” (http://headwize.com/tech/wye_tech.htm), in which the authors state:
“A wye-connector used to split a signal into two lines is being used properly; a wye-connector used to mix two signals into one is being abused and may even damage the equipment involved…. Outputs are low impedance and must only be connected to high impedance outputs -- never, never tie two outputs directly together -- never. If you do, then each output tries to drive the very low impedance of the other, forcing both outputs into current limit and possible damage. At a minimum, severe signal loss results.”
The authors then go on to describe how to use resistors to sum two signals.
I would like to hear your thoughts on all of this. And get tutored on shunt feedback. Thanks again.
Ravelo
Hi,
I've just started experimenting with electronics. I'm a bass player and built the CMOY headphone amp over the weekend. It works well enough. I think I might increase the caps to boost the bass a bit. But my question: If I elimenate the music-in (from your stereo) functionality would I have a wicked little headphone amp for my bass?
Hope your project is going well. My amp works, by the way, but I didn't build it effeciently enough to stuff it into an Altoid tin. But this success has me all fired up.
Thanks!
-Mark
I've just started experimenting with electronics. I'm a bass player and built the CMOY headphone amp over the weekend. It works well enough. I think I might increase the caps to boost the bass a bit. But my question: If I elimenate the music-in (from your stereo) functionality would I have a wicked little headphone amp for my bass?
Hope your project is going well. My amp works, by the way, but I didn't build it effeciently enough to stuff it into an Altoid tin. But this success has me all fired up.
Thanks!
-Mark
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.