First time build - Little Gem - not working - now with pics

[rockhorst]

Holidays finally arrived, so I decided to start building! Hurray! Decided to postpone the mechanical difficulties of a volume pedal for a while and choose the well known Little Gem and Ruby practice amps as a good simple start. Interesting too to hear the difference between the circuits. But the Little Gem I've assembled this afternoon isn't working so far, so maybe some help? Basically, there's nothing coming out of the speaker, not even static.

Following R.G.'s 'what to do'-list:
I'm new to building and soldering. Using a 25 W iron and resin cored solder. Working on perfboard cut down to size. Little Gem schematic:

http://www.runoffgroove.com/littlegem.html

I didn't really plan the soldering before hand, but decided to keep the perf rather large and make it up on the fly. The biggest problem design wise for me was what to do with all the grounds. I grounded everything to the input jack, since there is no output jack.
I started by mounting/soldering the OpAmp and noticed that it was quite hot after soldering (initial thought was that I might have fried it, but I am getting Volt readings, see below). Then I installed everyting around the IC. I substituted a 10K logpot in stead of the 25 Ohm reostat. Turning the pot in either direction didn't make any difference. No other substitutions made.

Voltage measurements of OpAmp IC:

Pin #

Voltage (V)

1	1.30
2,3,4	0
5	3.6
6	7.6
7	3.8
8	1.3

The voltage on pin 7 strikes me as odd, since it had no use.

I wired the output to the positive terminal of an old 8 Ohm stereo speaker and it's negative terminal was connected again to the input jack. I first had the polarity of the speaker wrong, but switched it around when I found that out. The output pin of the opamp (pin 5) runs to the 3rd pin of the output pot, speaker connected to 2nd potpin and 1st potpin grounded.

Polarity of the polarized caps has been checked.

I also checked all connections for continuity with the multimeter. As a closing remark I should add that the board+components have been dropped from about 1m height a few times  No on sight damages though.

Pictures, though I doubt them being of much use:







Hope someone can help otherwise I'll just start over (might not be such a bad idea).

ps. any tips on cutting/breaking perfboard? I tried the exacto knife, but that took a long while before I could break it off, and ruined the knife bit.

[Peter Snowberg]

First off I want to say thank you for what could easily be a model "how to get something debugged here" post.

Let's see, no sound at all...

First, your voltages all look good. Pins 1, 7, and 8, will all have voltages on them that depend on what's happening inside the 386.

Pins 1 & 8 are for adding a cap to increase gain. If you put a resistor in series you'll reduce the amount of boost and a pot will give you a gain control. The ROG circuit leaves this cap out and I'm a big believer in having it in for the sound qualities in both clean and overdriven modes, but omitting it won't make the sound go away.

Pin 4 is ground so it will always be 0. Pin 6 is power so we can see that your battery is putting out 7.6V which should get you plenty of sound.

Pins 2 & 3 are the inputs and are biased to ground in the 386 which is one thing that makes it different from a typical opamp. (the other big difference is the internal feedback path which sets the gain)

Pin 7 is labeled "bypass" on the datasheet and placing a cap from that pin to ground, you can make the power a bit cleaner for the input section of the chip. That is an optional part, and another place you can customize your amp. Depending on what you want out of the circuit, you may or may not want to add that cap.

Pin 5 is the output and with no input signal it should be sitting at roughly 1/2 of the supply voltage so that looks good too.

Check. Great! 

On the output side your pot is correctly wired. To make sure the pot isn't broken you can turn it all the way up and then short the wires you labeled "in" and "speaker". At this point it could still be a bad output cap, but let's look toward the input.

I can't quite tell from the last photo, but I suspect that the wiring error is there. Those jacks have 6 terminals in two rows of three. One of those rows is a set of switched contacts that connect to the contact across from them in the other row when nothing is plugged in. When a plug is inserted these connections become open. The other side has the contacts you need which are the ones that the spring contacts connect directly to. The circuit ground should connect to the contact closest to the jack, the input goes to the one on the other end, and the battery minus lead goes to the one in the center. It looks like the red wire in the last photo is connected to the wrong side.


On cutting perf, I use either a little triangle shaped file to score a trench into one side and snap to separate, or I use tin-snips.

[GibsonGM]

I frequently use a Dremel tool with a heavy-duty cutoff wheel to cut PCB (wear a mask!)...another way is to use a coping saw with the board clamped in a vise...the fine blade makes quick work of it.

I agree with peter, most likely place for an error to exist in that circuit is at those jacks...you can use a continuity tester at the other end of a plugged-in patch cord to figure out the pinout if you need to.

[TheBigMan]

Peter's nailed it.  The input signal wire needs to be on the other side of the jack.

[rockhorst]

Dang, you're correct Peter!  I hadn't noticed the input jack was asymmetrical. in any way. And indeed hadn't thought of continuity checking that part. Thanks a lot It's working now, indeed making the predicted amount of noise of 'just above conversation level'. I like the clean a lot, the gain is a bit fuzzy to my taste, but that might also be due to the small nonguitar-speaker.

Now to modify it a little bit (putting in a LED, DC power jack and pretty on/off switch) and giving it a nice comfy home. Thinking of multiplexing a little combo cabinet, tolex and all, just for fun.

Could you please elaborate a bit more on pin #7?

[GibsonGM]

I bet Peter knows more than I do about 386's, Rock, but pin #7 allows you to bypass the power supply.  You'd place a .1uf or so cap from pin 7 to ground.  This lets AC spikes go to ground, while blocking DC. Cleans up the power supply (a form of regulation).  Additionally, the cap acts like a 'little battery'; when the IC loads the power supply, it will discharge, keeping power more or less continuous.   

Think of when an oscillator discharges...it may pull on the PS, causing it to drop momentarily and cause a ticking sound.  The cap is there to keep the power continuous & adds stability to the IC's operation.   Go ahead & put it in, and see if it does anything for the sound! You can always take it out again.  I use them by default most of the time.

[rockhorst]

Thanks, one more question (didn't want to start a new topic for it):

I'm trying to switch the battery off as soon as a 9V adapter is plugged in, a little trick I saw at GGG:
http://generalguitargadgets.com/diagrams/power_jack_switching.gif
Didn't fully understand how to connect it up. I this the right way?

[Peter Snowberg]

I can't tell from the photo... are the red wires going to different lugs? If not, and if the circuit works when you unplug the battery and plug the AC adapter in, all you need to do is have the battery red connected to the third lug.

[rockhorst]

In that case I'll have to switch that battery wire . Have some difficulty understanding the function of the three lugs. One is the inside pin, you have a second lug that connects when the supply is plugged in, cuz the space on the other side (with no lug) is pushed up like in a normal jack...What does the third one do? Couldn't really tell from the datasheet.

[Connoisseur of Distortion]

the third one is a normally closed switch connecting to your outside pin. basically, your negative battery lead attaches to that third pin and the positive lead attaches to the inside pin.

the idea is that the battery circuit will close only in the event of an open socket.

i'll admit to having difficulty with these for about a year, before realizing it was switching the negative, and not the positive...!

[Peter Snowberg]

i'll admit to having difficulty with these for about a year, before realizing it was switching the negative, and not the positive...!

Whoops... yes...  it depends on the polarity of your power supply. 

I always assume center negative for my stuff. When you have a center negative jack situation, one lug connects to the center alone, one connects to the barrel of the plug, and the third connects to the barrel connection when no plug is present. The battery positive connects to that third lug. The positive supply to the circuit connects to the second lug, and of course the first luug is for ground. The battery ground then connects to the ring contact on the signal input jack.

As Connoisseur of Distortion just said, the third contact on the power jack makes sure that the battery is not connected if the AC adapter is. The input jack must also be occupied for the battery to power the circuit so the battery wires both go directly to switched contacts.

Here's one picture reference: http://www.tonepad.com/getFile.asp?id=76  (look at diagrams 2 through 5)

I know J.D. has a good picture at GeneralGuitarGadgets, but I couldn't find it.

Plug your adapter into the jack without any wires attached (assuming a center negative plug!) and measure the lugs with a meter. You will find two that have power between them and one that isn't connected. The third lug will connect to the lug that shows positive when the plug is removed.

[rockhorst]

I'm sticking to center negative, like most commercial pedals, to keep things nice and uncomplicated.