can anyone confirm is this schematic "should" work?

[methodofcontrol]

http://farm3.static.flickr.com/2774/...fbc6d5a5_o.gif

i just built this and it's not working. any ideas? i double checked my connections and layout. wondering
if it's just a wrong schematic. i'm inclined to think the chip is the culprit based on the fact pin 7 has voltage.
my chip voltages look like this w/ a 12v power supply:
1) 1.2v
2) 390mv
3) 0v
4) 0v
5) 1.5v
6) 11.94v
7) 6v
8) 1.4v

[methodofcontrol]

also... my 2n3904 has 8v on the base. and all of these measurements were taken w/ the pots in fully ccw position, (wiper shorted to ground on the 10k, and wiper shorted to R1 on the 100k)

[Enzo]

Look up the data sheet on the IC, and look at its internal circuit. We would expect some voltage on pin 7, roughly half way between what is on pins 6 and 8. And that is what you have.

I don't see how you could have 8v on the base of the transistor if it is wired to ground like your drawing says.

Looking at the flat face of the 2N3904, left to right, the leads are emitter, base, collector

That is different from the common asian arrangement as on 2SC945, E-C-B.


Your transistor is being used as a diode, by leaving the collector unused.

[JHow]

It definitely works as drawn - I made one. If volume is all the way down it is silent. To Enzo's point something must be hooked up wrong. Base should be at 0 dc.

[methodofcontrol]

i think i misspoke, the EMITTER has 8v on it. could i replace that w/ a standard 1n4148 diode? i assume the cathode would go to ground in place of the base and the anode would go to the C1/R1 connection?

[JHow]

i think i misspoke, the EMITTER has 8v on it. could i replace that w/ a standard 1n4148 diode? i assume the cathode would go to ground in place of the base and the anode would go to the C1/R1 connection?

You could try it. It should be working now with 3904, you can get it to work with 3906 too if you flip it around. When I made one I tried some diodes 1n914, leds and such, but transistors were noisier (in a good way).

[methodofcontrol]

well, i tripled checked my work, tested the individual components, unsoldered/resoldered/swapped out the chip and still ABSOLUTELY NO SOUND. it's almost as if my signal is being grounded...i don't even have a faint sound or anything. thoughts? this is such a simple circuit, i'm really scratching my head.

[km6xz]

Are you sure pin 2 is grounded to pin 4? If it is floating, as 390mv suggests, it won't work as an amplifier. The output should be a multiple of the difference between the signals on pin 2 and 3. The schematic might be confusing in that the connection points and where the wires merely cross over each other without touching is not clear. Pin 3, from the noise source pot only connects to the wiper of the pot and pin 2 does have a small dot on the junction of lines from 2 and pin 4, so 2 and 4 need to be shorted together, but not 3.

[methodofcontrol]

somehow i didn't see that pin 2 and pin 4 are supposed to be connected. oops.
but, the result is not white noise. it just sounds like a motorboat w/ no volume control?

[methodofcontrol]

which pot is supposed to work the volume? the 10k seems to be affecting the tone somewhat, the 100k appears to have no affect. i'm looking for the type of white noise that a static-y tv would produce. this one sounds like someone left an electric motor in a desk drawer. can anyone suggest quick mods to the existing circuit?

[JHow]

The 10K pot labelled "VR2" is the volume control. You are right the 100k var res doesn't do much. I used a trimmer there. Mine "turns on" with about 7.78 V on the emitter and the 100k pot at about 9.5K ohms. You could probably just ditch it for a fixed resistor. The sound is not like an electric motor but rather a shushing sound. Perhaps you have a picture of your build?

[Martin Pedder]

This won't stop it working, but should be corrected. C3 is drawn the wrong way round. The + end should go to pin 5.

[J M Fahey]

Both symptoms are reasonable.
Motorboating means your amp is working , but unstable.
Probably layout or grounding problems, but we're guessing here, of course.
No hiss means either whatever there is gets swamped by motorboating and also maybe your transistor is not acting as a Zener.
Replace it with a 3 to 6V Zener, Cathode to Emitter, Anode to Base (ground).
I do not trust much the drawing either, compare it to the official LM386 datasheet pinout.
Fo testing lower the LM386 gain, pull the capacitor from pins 1-8 .
At this point I'd rather prefer to hear not too loud hiss (confirming it basically works) than loud motorboating.
Good luck.

[methodofcontrol]

finally got back to this project. found that the machine only made white noise in a very limited dial settings. fix was to filter the power supply through a 100uf cap. now i have near perfect functionality on volume and "tone". add that cap if you're using this schematic for your noise machine.