I can hear Christian music through my power supply!

[Illum]

I recently completed a LM386 Stereo "pocket amplifier" for my MP3 which does not last long with the volume up.
I followed the schematic here: http://www.quasarelectronics.com/3115v2-1w-stereo-amplifier-module-lm386.htm
and came up with my own orientations on PCB.

I'm bothered by two issues, one being slightly more immediate than the other. The immediate one is that, while it functions correctly via 9V battery, any time I put my hand on the pot axis I hear really bad noise that overwhelms the music that's being outputted. Its leading me to believe that, judging from the noise [or better yet hum], is caused by some unknown oscillation in itself rather than picking up something externally. Its not static as there is no crackles or pops.

It draws 10ma+/-5% on a 9V battery, but I wondered whether that current value would vary if I biased 5V or 12V to it. So I had it hooked up to a power supply in series with my digital ammeter, upon powering on I hear christian music worshiping god in the background of my harp music. Trying to turn my music low to hear it better I touched the dial [didn't turn it] and there goes my harp music completely, its now "Jesus Oh! Jesus" without a hint of static or distortion. I have nothing against Christian music but it wasn't the intention of this project. If I take my hand away then I can only hear it faintly in the background. What is odd about this is that radios don't always work correctly in my workshop, everything from 90.7 to 107.1 FM has issues with static on a continuous basis, so its been awhile since I heard radio [I presume radio] interference that sounded this great on my workbench.

My issue is, how do I go about mediating the potentiometer issue? use a rubber stopper as the knob?

the circuit I based it off of looks like this, two LM386s were used along with a 100K dual gang pot. rather than buy new parts I used what I have and listed below

Changes:
Pot1 = 100K Audio taper, dual gang
C1, 5 = 10uF 25V Ecap
C3, 7 = 0.1uF 100V Ceramic
C2, 9 omitted
C4 = Paired a 0.1uF Ceramic and a 220uF 35V Ecap
C6, 8 = 220uF 35V Ecap
C10, 11 = 47nF Mylar
R1, 2 = 10 ohm 1/4W

I used LM386N-3 DIP-8
Input is a Stereo 3.5mm line
Output is a Mouser Part 35RAPC2BH3, 3.5mm Stereo PCB Jack

I was up until 1AM this morning cramming everything in, drilling holes and cutting out material for things to fit without pinching anything. I'll take awhile for me to figure out how to disassemble it

 

[TooManyGizmos]

You need to put a different "Breed" of cat hair in it !

To get a different station .
~

 

[cimino]

I wonder if there's a market for Jesus-powered radios...

 

[louie]

I'll make some educated guesses.

I conjecture that your touching the pot shaft is coupling your body as an antenna and picking up electric fields (fluorescent lights?) or RF transmitters. You might try grounding the body of the pot (assuming it's metal). Also, small ferrite beads on the input leads near the IC can help cut non-audio interference. Sometimes small disc capacitors on the inputs to ground can do the thing. You could also have a layout and/or shielding problem. Input too close to output? How high is the gain? Maybe too high. How near are you to transmitters?

Note in the National spec sheet that they recommend twisting the power supply leads very tightly - apparently they expect this to be a source of interference. In addition, the ferrite beads or small bypass caps on the PS leads might cut the radio interference.

 

[Illum]

I wonder if there's a market for Jesus-powered radios...

Pops isn't doing this, because testing it in my room I'm picking up what appears to be Spanish channel

You might try grounding the body of the pot (assuming it's metal). Also, small ferrite beads on the input leads near the IC can help cut non-audio interference. Sometimes small disc capacitors on the inputs to ground can do the thing. You could also have a layout and/or shielding problem. Input too close to output? How high is the gain? Maybe too high. How near are you to transmitters?

Note in the National spec sheet that they recommend twisting the power supply leads very tightly - apparently they expect this to be a source of interference. In addition, the ferrite beads or small bypass caps on the PS leads might cut the radio interference.

the Pot is a radioshack 271-1732 Stereo Volume Control, the shaft does not extend past the second PCB, only way I can ground it is probably solder a wire on the mounting hex nut.

The one thing that I do not have readily on hand is ferrite beads, probably I need some at this point. :candle:
Input and output are thankfully at far ends of the enclosure, gain preset to 20, the jumpers shown on that schematic can enable a gain of 200, which I omitted. Disc capacitors to ground from the input? hmm, I might have some 33nF ceramic disc caps, I wonder how high does the value have to be or if the values are even critical. :thinking:

I went ahead and took pictures, the layout is terrible as I built it from draft A rather than working out the components on the PCB first.

Draft A_Rev3, the border tells me what space I can use in the Radioshack 3" x 2" x 1" enclosure. the block at left was what the Pot would look like footprint wise, but it didn't turn out like that

Externals look okay, offset knob on top, output port on right, power from top, input from bottom. Yes, that "knob" is a rubber stopper

Internals to me looks awful, it was because I thought I had 470uF caps available for output, but I was left with one instead of a pair, so using 220uF caps as substitutes left a clearing on the right and a traffic jam on the left. No you are not seeing things, I sawed a DIP-16 machine socket in half

Top view

Closeup

At this point you can see that the green caps are C10, 11 on that schematic I posted earlier, inputs from the bottom goes through 10uF caps and goes to the pot as red, yellow lines are the siphon that goes into the LM386. Black lines are ground or signal low and I have bridged two pins together as one for ground. the wires are 24AWG, I have 33AWG but I lost my small wire stripper somewhere and had to use this.

Dual LM386N-3 =P, Green and purple = Vss, Blue = Gnd, Brown = left out, white = right out...none of my wiring is by convention, just by what is available

 

[Lynx_Arc]

did you try grounding the case of the pot itself?

 

[TorchBoy]

The juxtaposition of harp music and Christian worship music is quite funny. Does it sound like a commercial radio station or a local MP3 transmitter?

 

[Illum]

did you try grounding the case of the pot itself?

I'm going to get that part done sometime tomorrow and hopefully add a ferrite bead on the yellow wires, even if it doesn't help it'll be a step in the right direction :thanks:

The juxtaposition of harp music and Christian worship music is quite funny. Does it sound like a commercial radio station or a local MP3 transmitter?

it reminds me of a commercial radio station, unless present day MP3 transmitters have narrators between songs as well

 

[fyrstormer]

The juxtaposition of harp music and Christian worship music is quite funny. Does it sound like a commercial radio station or a local MP3 transmitter?

Harp music AND Christian worship music? Sounds to me like he's found a way to spy on Heaven.

Hey Illum, if you get any messages from God on that thing, be sure and post 'em, okay? :huh:

 

[Lynx_Arc]

Harp music AND Christian worship music? Sounds to me like he's found a way to spy on Heaven.

Hey Illum, if you get any messages from God on that thing, be sure and post 'em, okay? :huh:

most likely it is the closest station to him with the most powerful signal strength if some other station was more powerful he could be listening to something else like rap music

 

[Per-Sev]

I had a tube guitar amp that I would pick up a local radio station on and was told that it was operating at the same hertz as the station and I could try some noise filters but since it was under warranty I just returned it for a different brand and no more problem.

 

[Illum]

grounding the pot did it!

this little amp works well for small speakers as a preamp of sorts but horrible for earphones, there's this "windy mic" shhhh noise over everything, no amount of add-on's could help with the oscillations having several wires on top of each other

As much as I would like to travel with it I highly doubt I can get through FAA unnoticed

Harp music AND Christian worship music? Sounds to me like he's found a way to spy on Heaven.

Hey Illum, if you get any messages from God on that thing, be sure and post 'em, okay? :huh:

, if my amp starts prophesying while off I will be sure to notify you.

I had a tube guitar amp that I would pick up a local radio station on and was told that it was operating at the same hertz as the station and I could try some noise filters but since it was under warranty I just returned it for a different brand and no more problem.

I was thinking of something like that, but such that a ferrite cylinder pulled through a coil of wire could pickup too, but my issue began with the absence of ferrite material used so I was baffled

 

[Lynx_Arc]

one other idea would be to take some aluminum foil and shield components. Old 9v transistor radios used to do that at times to block stray signals.

 

[JohnR66]

The problem is the high impedance input. Using the 100K pot allows the amp inputs to pickup electrical noise and feedback on itself to oscillate. As you have seen, grounding the pot will help as will shielded leads on the input circuit. It is much better to use a 10K pot or if this is only to be used with an MP3 player, use a 1K pot.

One error in the design is the input AC coupling capacitor should be between the pot and input pin. The way it is here makes it sensitive the the impedance changes as to pot is adjusted.

With a music player, I think gain would be enough keeping pins 1 and 8 unconnected. It will help reduce the noise and hiss you hear.

HTH, John

 

[paulr]

Illum, If you're doing that as an educational DIY project, that's great, you might like Horowitz and Hill's "Art of Electronics" book. If you just want an amplifier for your MP3 player, there are tons of them you can already buy, and in fact I have a few that I got on woot specials and can send you one if you want.

 

[Illum]

The problem is the high impedance input. Using the 100K pot allows the amp inputs to pickup electrical noise and feedback on itself to oscillate. As you have seen, grounding the pot will help as will shielded leads on the input circuit. It is much better to use a 10K pot or if this is only to be used with an MP3 player, use a 1K pot.

One error in the design is the input AC coupling capacitor should be between the pot and input pin. The way it is here makes it sensitive the the impedance changes as to pot is adjusted.

With a music player, I think gain would be enough keeping pins 1 and 8 unconnected. It will help reduce the noise and hiss you hear.

HTH, John

hmm, if dual gang pots are available for a lesser value I'd certainly opt for it. So the caps are better placed after the pot huh...hmm...thats one thing I'm willing to try next time

Illum, If you're doing that as an educational DIY project, that's great, you might like Horowitz and Hill's "Art of Electronics" book. If you just want an amplifier for your MP3 player, there are tons of them you can already buy, and in fact I have a few that I got on woot specials and can send you one if you want.

I find soldering very therapeutic and I'm running out of thru-hole projects to solder
Buying something is certainly nice, but debugging something built first hand has a remarkable feeling to it. I've started building audio amplifiers with discrete transistors...I gave up that route because they have had issues, this was the first audio oriented circuit I've built in awhile that works so I'm excited

 

[JohnR66]

I'm with you there. I never put my soldering iron away.

Building an audio amp from discrete components is a challenge. The amplifier is made of several modules that have to be engineered to work together for a complete unit. It is something I haven't tried yet either. There are plenty of well designed ICs to use.

The LM386 IC is an old but functional low power audio amp. There are many newer ICs out there than can perform its function requiring less parts yet deliver more power to the speaker at the same supply voltage.

 

[Illum]

I'm with you there. I never put my soldering iron away.

When I feel ambitious enough I try to solder deadbug style on surface mount parts, but I still prefer thru-hole. To demonstrate the ease of use I once soldered up PWM generator using a 555 using thru-hole components and a piece of heavy weight paper as the PCB and had the circuit taped to the wall with the wires dangling:laughing:

Building an audio amp from discrete components is a challenge. The amplifier is made of several modules that have to be engineered to work together for a complete unit. It is something I haven't tried yet either. There are plenty of well designed ICs to use.

I started with this: http://www.eleccircuit.com/portable-headphone-amplifier-by-bc337bc327/

I've must have resoldered this 4-5 times until I was left with a solder plane, I use 0.022" solder so you can probably imagine the amount of solder that left the spool. I then unsoldered all the parts and stick everything back on the breadboard and played around with the resistor values using low value pots thinking the transistors aren't being biased correctly, but it just will not work what so ever, not even a static.

I then moved the parts around on my own schematic to facilitate easier soldering, I used resistors probably dug out of several stereo/CD player/Clock radios back when I was dumpster diving on a monthly basis...which aren't consistent in type and probably was the reason for the failure. come to think of it, the wiring overlap is the other concern. My goal was to have this thing be slightly bigger than a CR123A and slight smaller than my MP3...so I designed my placement around a 1.97" x 1.38" x 0.8" enclosure so...

The component designators are consistent with the schematic posted,
Top left to right R7, C3, R4. R5, R1
Middle left to right Q3, Q2, Q1
Bottom left to right R8, R3, C2 [laid on top] R6, R2

Power comes in from the top left, Neutral on bottom left, input [capacitors mounted out of the enclosure because I ran out of space] on top right, output from the bottom, rerouted using a yellow wire to bottom left.

The LM386 IC is an old but functional low power audio amp. There are many newer ICs out there than can perform its function requiring less parts yet deliver more power to the speaker at the same supply voltage.

It doesn't necessarily have to be new, the old painless TDA2822M comes in DIP-8 and its stereo with two inputs and outputs...its available at mouser, but not stores around here. Uses 11 other passive components [as compared to 12:nana:] and good for headphone or other 16 ohm devices.