Debug first build (Ruby Amp)

harmonic · July 18, 2008, 03:26:42 PM

Hi everyone!

I've been lurking for a month or so now -- love all the discussion and the helpfulness everyone exhibits.

I decided to build a ruby amp (http://www.runoffgroove.com/ruby.html) as my first effort and, much to my surprise, it worked first time on the breadboard. I mean, literally just shoved everything in there, plugged in and bam! -- nice sound straight away! I was so impressed -- simple circuit notwithstanding -- that it all went so smoothly. So, onto the 'proper' version. Enclosure bought and drilled; pots and jacks mounted; components soldered. Hooked everything up and ... nothing. Spent 3 days going over everything, couldn't get it to work.

Fortunately, I bought two of everything so I thought of just going on and building the board again. This time I get hiss through the speaker, but nothing else.

I've verified that the transistor is the right way round (worked on the breadboard).
I've tried to follow the circuit around and check that I get readings on the meter everywhere that isn't grounded.

I may have fried the elecrolytics but don't know how to test them as my meter has no capacitance test.
I may have fried the transistor or the IC with my newbie soldering skillz (but the fact I get readings on all the pins seems to indicate no?)

I've taken some measurments on the transistor and the IC -- would some kind soul have a quick peek to see if these readings are reasonable?

IC (LM386N1) Measurements based on a 12v DC supply:
Pin 1: 7.35v
Pin 2: 7.21v
Pin 3: 0.00v
Pin 4: 7.96v
Pin 5: 8.57v
Pin 6: 12.08v
Pin 7: 9.00v
Pin 8: 7.36v

Transistor (2N5457) readings
No input signal:
D: 12.09v
S: 1.09v
G: 0.00v

Touching the other end of the guitar cable jack's tip:
D: 12.09v
S: 1.19v
G: 0.06v

Does this look reasonable?

Many MANY thanks!!!

Auke Haarsma · July 18, 2008, 03:45:03 PM

Hi, nice post! Kudo's for including all the info.

According to the schematic at ROG pin 3 and 4 of the LM386 should be connected to ground. The readings on Pin 3 (0V) indicate that this pin is connected correctly. However, Pin 4 should also read 0V.

Hope this helps,

Auke

frequencycentral · July 18, 2008, 04:05:23 PM

Is your FET pinout correct?

harmonic · July 18, 2008, 04:23:45 PM

Quote from: Auke Haarsma on July 18, 2008, 03:45:03 PM
According to the schematic at ROG pin 3 and 4 of the LM386 should be connected to ground. The readings on Pin 3 (0V) indicate that this pin is connected correctly. However, Pin 4 should also read 0V.

GAH!!! *slaps head* Thank you, Auke -- that was it!

The circuit fired up so I got my guitar out and plugged in. The sound is very different from the breadboard version which was just kike how it was described on ROG. This circuit has ZERO clean headroom. Gain fully down, volume at half, guitar volume starting at zero and slowly increasing. It jumps immediately to distorted sound. But not the nice overdrive I was getting before. This just sounds like farty nasty fizzy with zero sustain.

Any ideas what to go for next?

Thanks!

ambulancevoice · July 18, 2008, 11:47:23 PM

did you solder the ic and transistor to the board???

harmonic · July 19, 2008, 04:10:21 AM

Quote from: ambulancevoice on July 18, 2008, 11:47:23 PM
did you solder the ic and transistor to the board???

Hi! I used an IC socket, but the transistor is directly soldered (didn't have a socket for that).

ambulancevoice · July 19, 2008, 05:00:43 AM

well, the ic should be fine
get a socket (small bear or something) and try another transistor
or, use an alligator clip as a heat sink on the legs of the transistor
try and solder it in quickly

harmonic · July 19, 2008, 07:18:30 AM

Quote from: ambulancevoice on July 19, 2008, 05:00:43 AM
well, the ic should be fine
get a socket (small bear or something) and try another transistor
or, use an alligator clip as a heat sink on the legs of the transistor
try and solder it in quickly

Yeah, it's already on my next order list!!

Could a transistor be damaged by heat cause that change in the sound?

ambulancevoice · July 19, 2008, 08:57:18 AM

could be, seems to add up
since it was not soldered to the breadboard, it worked
when it was, it didnt

harmonic · July 19, 2008, 12:20:16 PM

Quote from: ambulancevoice on July 19, 2008, 08:57:18 AM
could be, seems to add up
since it was not soldered to the breadboard, it worked
when it was, it didnt

I guess! Could the sound be from a messed up electrolytic? I _may_ have fried one by accidentally reversing the polarity on the power supply ...

frequencycentral · July 19, 2008, 12:52:14 PM

Quote from: harmonic on July 19, 2008, 12:20:16 PM
I guess! Could the sound be from a messed up electrolytic? I _may_ have fried one by accidentally reversing the polarity on the power supply ...

That would do it. Did you get smoke too? Change that cap!

harmonic · July 19, 2008, 01:16:53 PM

Quote from: frequencycentral on July 19, 2008, 12:52:14 PMThat would do it. Did you get smoke too? Change that cap!

Yeah. I was avoiding facing the truth.

There was very definately a wisp of smoke.

Is it likely that it was both electrolytics in the circuit? How do I test for a dodgy capacitor if my dmm doesn't have a capacitance test?

frequencycentral · July 19, 2008, 01:35:17 PM

Quote from: harmonic on July 19, 2008, 01:16:53 PM
Quote from: frequencycentral on July 19, 2008, 12:52:14 PMThat would do it. Did you get smoke too? Change that cap!

Yeah. I was avoiding facing the truth. There was very definately a wisp of smoke.

Is it likely that it was both electrolytics in the circuit? How do I test for a dodgy capacitor if my dmm doesn't have a capacitance test?

The 220uf at the output should be ok as its not connected across the power rails, and its unlikely the 386 was outputting anything.

To test the blown cap - replace it - if the new one works the old one was blown!

I'm not too sure how 386's react to being incorrectly powered - you may have to replace that too - definately use an IC socket.

ambulancevoice · July 19, 2008, 06:47:09 PM

incorrectly powered?

? the 100uf cap is just a filter cap, if it blew, it wouldnt really effect the voltage amount... right???

harmonic · July 19, 2008, 07:17:02 PM

Hi all,

Firstly, thanks very much for all the help and ideas -- I really appreciate it!

I've had a really frustrating evening (but good for all that -- always learning!). I replaced the two electrolytics, no change. I removed the transistor and replaced it with a sliced-in-half dip socket. Tried both transistors I have in there, no change. I tried the other 386 in the socket, but it is definately dead.

So, anything else I can try?

ambulancevoice · July 19, 2008, 07:18:37 PM

checking for cold solder joints or wrong connections
check your resistor values and cap values just in case something is in the wrong place

harmonic · July 20, 2008, 04:59:46 PM

Ah rats. Still sounds nasty. I resoldered all the connections making sure non were a cold joint.

I think I've messed up either both my transistors or the single 386 that still works. I used all the remaining parts (I bought double of everything) to re-breadboard the circuit and it sounds the same as the soldered version.

I'll need to get another shipment of parts to check any further.

Thanks very much for all your help -- I'll update with progress when there is some!!

harmonic · July 22, 2008, 01:42:59 PM

Yay! Success!!! I sucumbed to impatience and bought a couple of replacement 386s from Maplin, bunged one in the breadboard version, plugged in and switched on: 1 watt sonic nirvana!! Put the other one in the boxed up version and the same thing.

So, in the end it was all down to one missed ground connection and a dead IC. I'm pleased with that -- could've been way worse!

Thanks for all your help.

harmonic.

News:

Debug first build (Ruby Amp)