On to my next project. I'd like to build a simple electric guitar practice headphone "amp". Something to use on my wife's guitar later at night. I've seen this recommended several places:
http://www.generalguitargadgets.com/richardo/practiceamp/lm386.gif
Seems easy and small. The crutch is I'd like to be able to also mix in a cd/mp3 input for playing to. Can I use a simple passive mix at the lm386 parts output, using resistor (ive seen 680ohm to 4k7recommended for this) andd a 10k pot to mix the cd/mp3 source? If not, what is the simplest method. I don't need the cd/mp3 amped or with hifi sound.
Part 2 of all this is I have a couple opa2132/4 and extra CMOY parts laying around. I tried the guitar on a 2228 based CMOY tonight and it worked so-so. I know the impedance and such was wrong and the gain was insufficient, but it sounded good actually with little noise. Could i modify a typical CMOY to work for this? Perhaps changing the input resistor of the CMOY to 1M ohm and boosting the gain to say 20x or even higher? I know 21x works "fine" in a CMOY as I tried I for fun.
Same question applies redarding mixing in an aux source if theodified CMOY will work.
Thanks in advance!
http://www.generalguitargadgets.com/richardo/practiceamp/lm386.gif
Seems easy and small. The crutch is I'd like to be able to also mix in a cd/mp3 input for playing to. Can I use a simple passive mix at the lm386 parts output, using resistor (ive seen 680ohm to 4k7recommended for this) andd a 10k pot to mix the cd/mp3 source? If not, what is the simplest method. I don't need the cd/mp3 amped or with hifi sound.
Part 2 of all this is I have a couple opa2132/4 and extra CMOY parts laying around. I tried the guitar on a 2228 based CMOY tonight and it worked so-so. I know the impedance and such was wrong and the gain was insufficient, but it sounded good actually with little noise. Could i modify a typical CMOY to work for this? Perhaps changing the input resistor of the CMOY to 1M ohm and boosting the gain to say 20x or even higher? I know 21x works "fine" in a CMOY as I tried I for fun.
Same question applies redarding mixing in an aux source if theodified CMOY will work.
Thanks in advance!
I've used one of those LM386 chips to make an amp for a friend... realy easy, but I suggest you look at the datasheet and add the optional gain boost components and a switch....Description"The LM386 is a power amplifier designed for use in low voltage consumer applications. The gain is internally set to 20 to keep external part count low, but the addition of an external resistor and capacitor between pins 1 and 8 will increase the gain to any value from 20 to 200."LM386 - Low Voltage Audio Power Amplifier
Cmoy can put out like 40mA combined into a short circuit... not alot of power to drive a speaker...
Cmoy can put out like 40mA combined into a short circuit... not alot of power to drive a speaker...
Last edited:
Thanks. So, what makes a good buffer. Ive found this on on this forumhttp://www.italentshare.com/storage/Jam%20Along%20Headphone%20Amp/Jam%20Along%20Amp.jpg.
I've also seem a simpled buffer for the cd/line in that did not use an opamp. Most designs I've seen show using 10k resisters in series with the aux lines and paralleled with the guitar into the lm386. I gather 10k is sufficient to prevent signal mixing problems while adjusting for differences in input strength between the line in and pre-amplified guitar signal.
What do you mean by buffer?
Thanks again!
I've also seem a simpled buffer for the cd/line in that did not use an opamp. Most designs I've seen show using 10k resisters in series with the aux lines and paralleled with the guitar into the lm386. I gather 10k is sufficient to prevent signal mixing problems while adjusting for differences in input strength between the line in and pre-amplified guitar signal.
What do you mean by buffer?
Thanks again!
Would a good solution be to combine the top right, simple virtual ground active mixer with buffer at this link:
http://www.all-electric.com/schematic/simp_mix.gif
and feed the output of that in parallel with the guitar to the lm386 amp? I would connect tip and ring of aux in together for mono.
If I went that route, can I use a 2132/4 for the buffer opamp, using one side of the 2134 to buffer the other?
The problem is, won't that cause lots of distortion to the aux input If I use high gain on the lm386? I'd like to avoid that if possible. Using the 2134 buffered to mix after the lm386 would also work right, and solve the cd distortion issue?
In reality, since the low end, undistorted gain of 20 on the lm386 is easily achievable gain on the 2132/4, wouldn't the simplest sulution be to use a normal CMOY circuit but change the input impedance of it, R2 on tangent style layouts? Maybe 1M resistance tongroumdnat the guitar input to the + pin on the opamp. I lile that solution since I have all the parts and boards I need to do it
. If that is a solution, should I only use one channel of the 2134, buffer with the other side, or feed the guitar into both halves of the 2134?
Sorry for all the questions
Would the output of the lm386 and and aux source like a CD player have similar levels/impedance, allowing for passive mixing after the lm386?
Thanks!
http://www.all-electric.com/schematic/simp_mix.gif
and feed the output of that in parallel with the guitar to the lm386 amp? I would connect tip and ring of aux in together for mono.
If I went that route, can I use a 2132/4 for the buffer opamp, using one side of the 2134 to buffer the other?
The problem is, won't that cause lots of distortion to the aux input If I use high gain on the lm386? I'd like to avoid that if possible. Using the 2134 buffered to mix after the lm386 would also work right, and solve the cd distortion issue?
In reality, since the low end, undistorted gain of 20 on the lm386 is easily achievable gain on the 2132/4, wouldn't the simplest sulution be to use a normal CMOY circuit but change the input impedance of it, R2 on tangent style layouts? Maybe 1M resistance tongroumdnat the guitar input to the + pin on the opamp. I lile that solution since I have all the parts and boards I need to do it
. If that is a solution, should I only use one channel of the 2134, buffer with the other side, or feed the guitar into both halves of the 2134?Sorry for all the questions
Would the output of the lm386 and and aux source like a CD player have similar levels/impedance, allowing for passive mixing after the lm386?
Thanks!
Last edited:
A guitar is a low level high impedance source, while a CD/MP3 player would be more like a line level low impedance source. Buffers would be ideal, but you may be able to work without them.
The LM386 has a 50K input impedance, so using summing resistor of any more than 50k is not ideal as the input voltage would be cut in half when using 50k resistor (neglecting guitar's impedance). Using the 386 in high gain mode, try around 20k summing resistors. The line input for the player would need an additional voltage divider or pot to reduce its level prior to summing.
The LM386 has a 50K input impedance, so using summing resistor of any more than 50k is not ideal as the input voltage would be cut in half when using 50k resistor (neglecting guitar's impedance). Using the 386 in high gain mode, try around 20k summing resistors. The line input for the player would need an additional voltage divider or pot to reduce its level prior to summing.
Thanks John for the reply.
Why can't I use summing resistors after the lm386, to connect the 386 output and the cd player output? I don't really need to amp the cd/mp3 player and if I did, I could use a CMOY. Since no one seems to donor this way, there must be a reason, I'm just not experienced to know why.
Why can't I use summing resistors after the lm386, to connect the 386 output and the cd player output? I don't really need to amp the cd/mp3 player and if I did, I could use a CMOY. Since no one seems to donor this way, there must be a reason, I'm just not experienced to know why.
You can do this, but the isolation between the driving amps will be poor, each will drive the other, particularly as you will have to keep the resistances in the passive mixer small in order to avoid wasted power, excess heat generated and poor damping factor in the speaker drive. Best is to make an active mixer (summing opamp) as in the diagrams you have shown, before the 386, be aware that they require a split-rail power supply, although you can use a single rail with virtual ground, but again this is not ideal.
w
w
Wakibaki (great name by the way), thanks for your reply.
A couple of questions, regardless of the mixer type ( I can accept that active buffered opamp is best), why place it before the lm386. That could add lots of distortion to the cd signal.
Second, will a 2134 work for the mixer using it buffered and obviously in mono?
Thanks again.
A couple of questions, regardless of the mixer type ( I can accept that active buffered opamp is best), why place it before the lm386. That could add lots of distortion to the cd signal.
Second, will a 2134 work for the mixer using it buffered and obviously in mono?
Thanks again.
Because although it's low power (125mW), it's the power amplifier in this context, opamps are in the 10s of mW range (although there are exceptions to this).
Yes, it's more than adequate.
w
I'd say to try mixing on the output side. Use 10 Ohm resistors on the L & R out of the player and 10 on the 386. What happens is sort of a bridge amp effect with half the bridge acting as a sink for the active side. The resistors act as a voltage divider and the signal is reduced. Not a problem for the LM386, but the music player will need to be turned up louder to make up for the loss. If the song is not recorded loud enough, you may run out of volume. It is very easy to breadboard the circuit and try.
Last edited:
I must say I'm a bit confused, likely because I don't understand what a buffer is/does. I've read not to put the guitar and cd parallel, even with series resistors, as they are different signals. But, the active buffered opamp mixers ive seen and linked have both in parallel with series resistance before the chip. Why don't they try to feed each other this way?
So to clarify, my order of "sections" should be:
Guitar parallel with cd, into active buffered mixer, into lm386
Or
Guitar into lm386 paralleled with cd into buffered opamp mixer
Or
Cd into buffered opamp, parallel with guitar (after lm386)
Thanks.
So to clarify, my order of "sections" should be:
Guitar parallel with cd, into active buffered mixer, into lm386
Or
Guitar into lm386 paralleled with cd into buffered opamp mixer
Or
Cd into buffered opamp, parallel with guitar (after lm386)
Thanks.
Wikipedia is my friend. If understand correctly, the buffer converts a high impedance source to a low impedance signal while conserving the voltage. That is what the buffered opamp mixer part does correct? I still don't get how that would allow two different impedance sources to be in parallel before the buffer.
The lm386 does not buffet either, correct since there is no feedback loop between the inverting pin and the output.
This all leads me to believe the correct order should be:
Guitar into lm386, into buffer, mix with cd. But, I don't see it ever in that order.
The lm386 does not buffet either, correct since there is no feedback loop between the inverting pin and the output.
This all leads me to believe the correct order should be:
Guitar into lm386, into buffer, mix with cd. But, I don't see it ever in that order.
True. But the mixer is generally considered in terms of it's summing action, not as a buffer, although it does have (depending on design) a high input impedance and a low output impedance. Buffers generally have no voltage gain (gain of 1).
It allows the summing of 2 voltages while preserving the isolation between them. The guitar really wants to see 500k input impedance, otherwise the output voltage will collapse, the CDP doesn't care. The guitar has a high intrinsic or Thevenin resistance. (See Thevenin's Theorem) If it is loaded with a low impedance most of the output voltage appears across the internal impedance, not the load. Other than that the impedances are irrelevant, it's the relative voltage amplitudes that matter.
The 386 has internal feedback. It has a high input impedance and a low output impedance, but is not thought of as a buffer because it offers voltage gain.
The 386 is designed to drive a speaker as low as 8 ohms, most opamps will struggle with a load as low as this, and will only output 10s of milliwatts into any load. I mistakenly said 125mW for the 386 but some flavours will deliver up to 1 watt.
No, you don't, it's the wrong way to do it. Low power operations precede high power operations as a rule, because of efficiency and heat considerations. The 386 can handle power levels which would destroy most opamps, if circumstances permitted them to pass such levels, which is called 'letting out the magic smoke'.
Electronics is a complex and subtle business. It's easy to get hold of the wrong end of the stick. Many of the things that happen in it are not what one would expect based on simple commonsense, because there are hidden features which are not immediately obvious, because unlike in mechanics, they are not visible. The easiest way to come to terms with them is to allow yourself to be taught the subject by someone who already understands it.
w
Thanks for the detailed reply wakibaki. I greatly appreciate it!
Everything you said makes sense to me. Since my last post I've done a bit more reading and coupled with your reply I think I'm starting to understand a little better. I think I will attempt to combine the various things I've learned here, and elsewhere, to build this amp. I've tried an mp3 player through the lm386 with a gain of 20 and 200 and it sounds fine for the intention of this device. I think I will essentially follow the final schematic in another thread on this forum, http://www.italentshare.com/storage/Jam Along Headphone Amp/Jam Along Amp.jpg
However, I will use only one lm386 and simply connect the cd/mp3 tip and ring to make it mono, doing the same at the headphone jack. Additionally, I'll make the lm386 gain variable and I'll and a volume out on the lm386.
Does that seem reasonable? Thanks again everyone for your time!
Everything you said makes sense to me. Since my last post I've done a bit more reading and coupled with your reply I think I'm starting to understand a little better. I think I will attempt to combine the various things I've learned here, and elsewhere, to build this amp. I've tried an mp3 player through the lm386 with a gain of 20 and 200 and it sounds fine for the intention of this device. I think I will essentially follow the final schematic in another thread on this forum, http://www.italentshare.com/storage/Jam Along Headphone Amp/Jam Along Amp.jpg
However, I will use only one lm386 and simply connect the cd/mp3 tip and ring to make it mono, doing the same at the headphone jack. Additionally, I'll make the lm386 gain variable and I'll and a volume out on the lm386.
Does that seem reasonable? Thanks again everyone for your time!
Here is the schematic I have come up with.
Thoughts? If I ditch the MP3/CD volume pot, does that change the value of R1?
C1 should read 1uF. C1, C4, C6, C7 will be polyester, rest electrolytic.
Z1 should go to negative, not ground. It is a 12V/1W zener.
[/url]
Guitar practice amp with aux input by Dennis Dietz, on Flickr[/IMG]
Thoughts? If I ditch the MP3/CD volume pot, does that change the value of R1?
C1 should read 1uF. C1, C4, C6, C7 will be polyester, rest electrolytic.
Z1 should go to negative, not ground. It is a 12V/1W zener.
Guitar practice amp with aux input by Dennis Dietz, on Flickr[/IMG]
I would sum the other source in after the lm386, when you open up the high gain setting it will crunch the sound of whatever else is comming through with the guitar too...
I saw a nice driverstage chip in TI's newsletter this morning.
[FONT=Verdana, Arial, Helvetica, sans-serif]The DRV632 is a 2-VRMS, stereo line driver designed to replace line driver circuits created with an op amp and passives. The device is ideal for single-supply electronics where size and cost are critical design parameters.- More[/FONT]
Line Driver / Receiver - 2V/3Vrms Line Driver - DRV632 - TI.com
I saw a nice driverstage chip in TI's newsletter this morning.
[FONT=Verdana, Arial, Helvetica, sans-serif]The DRV632 is a 2-VRMS, stereo line driver designed to replace line driver circuits created with an op amp and passives. The device is ideal for single-supply electronics where size and cost are critical design parameters.- More[/FONT]
Line Driver / Receiver - 2V/3Vrms Line Driver - DRV632 - TI.com
Thanks for the reply digits! I tried an experiment last night however, I built a quick lm386 circuit and ran an MP3 player through it on both 20 and 200 gain. The source had to be almost all the way down for this, but it sounded fine and did not add to much distortion, in my opinion.
That being said, would you recommend:
guitar through the TL082 buffer then through the lm386, summing the cd in afterwards with resistors, but I don't know. Reading around the internet will give almost every solution. And I still don't get how summing two source if very different impedance and power together with resistors before running both into a buffered opamp prevents the issue of one trying to drive the other.
Thanks!
or,
Put the guitar through the lm386 and sum the cd in parallel to the TL082 for mixing?
Sorry to be so dense. I feel like the second answer is correct
That being said, would you recommend:
guitar through the TL082 buffer then through the lm386, summing the cd in afterwards with resistors, but I don't know. Reading around the internet will give almost every solution. And I still don't get how summing two source if very different impedance and power together with resistors before running both into a buffered opamp prevents the issue of one trying to drive the other.
Thanks!
or,
Put the guitar through the lm386 and sum the cd in parallel to the TL082 for mixing?
Sorry to be so dense.
I feel like the second answer is correct- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.