Is my LM308N hosed?

[mgwhit]

A few weeks back I wanted to experiment with diode clipping in a simple distortion circuit, so I breadboarded Dead-Easy Dirt (http://www.forrestwhitesides.com/node/93), set up an eight-position dipswitch for different diode combinations, added a few resistors and capacitors for tone shaping and learned quite a bit.

Since I've always wanted a RAT, I ordered an LM308N.  I still had the Dead-Easy Dirt breadboard laying around, so I replaced the LM386 with my new LM308N, rewired for the different pin-outs,removed most of the unnecessary components, plugged in and...nothing.  Here's the basic gist of what I've got on the board:



Voltages are like so:

1: 9V || 8: 9V
2: 0V || 7: 9.5V
3: 0V || 6: 8.9V
4: 0V || 5: 0V

I've tested with a signal tester and  I get clean signal on pin 3, no signal on pin 6.  I should probably say no guitar signal -- it picks up and amplifies some crazy church radio station just fine (as do all of my breadboarded and unboxed effects).  That makes me think that the LM308N is still functional and that I'm just missing something.

Any help much appreciated.  Thanks!

[jkokura]

Missing something I'd say. The voltages look as they should to me!

Jacob

[gtr2]

Remove pin 2 from ground.  Pin 6 needs to connect back to the pin 2 in the neg feedback loop.  You'll probably need to add in a little resistance there as well.  Also throw a 50pF-150pF cap between pins 1 and 8.

Josh

[mgwhit]

Thanks for the input, guys!  I followed gtr2's suggestions -- 47pF across the 1-8 compensation loop and 22K on the 2-6 negative feedback loop -- but still no noise.   I even reseated all of my components just to be sure they were making good contact.  Sigh....  Do you think I need to put some voltage on pin 3?

(Interestingly, if I sub a 470R in the NF loop, the radio interference is very low level on pins 2 and 6, somewhat amplified on pins 1 and 8, and fully amplified on pin 3.  It goes away completely with a large value resistor in the NF loop.)

[gtr2]

Sigh....  Do you think I need to put some voltage on pin 3?

haha, yeah that might help    You'll need to apply a bias voltage of 4.5V to pin 3.  It's starting to defeat the quick and dirty dirt though 

Check out madbeans slow loris schematic, that should point you in the right direction.

Josh

[mgwhit]

Yeah, it's way past quick and dirty now.  I went back to the RAT schematic and implemented enough of the power section to get 4.5V on pin three, and I added all the components on the input, too.  I'm getting sound now, but it's really only about unity gain and, consequently, doesn't clip on the diodes at all.  I'm going to go to bed with the datasheet and an electronics textbook and try to figure out what I'm doing wrong.  I have a strange feeling I'm going to be doing more between pins 2 and 6 tomorrow....

Thanks for all of your help!

[mgwhit]

I got frustrated yesterday, pulled everything and started from scratch on the breadboard, implementing the first half of the Slow Loris schematic with a slightly simplified feedback loop.



I subbed the 47K in place of the A100K Gain pot assuming it would be a nice middle of the road value, and I put 47R on R5 as it was close enough for the value of 47R and 560R in parallel (43R).

Voltages are:

1: 8.82V || 8: 9.37V
2: 0.01V || 7: 9.54V
3: 0.58V || 6: 8.78V
4: 0.00V || 5: 0.00V

Voltage is 4.5V at VB until I connect it up to the input and pin 3 via the 1M resistor -- then it drops like a rock.

Signal is present at pin 3, lower on pins 1 and 8, and totally undetectable at 2 and 6.

If someone could compare my voltages to a real Slow Loris, I'd appreciate it.  In the meantime I'm going to run to Radio Shack on my lunch hour and pick up an LM741CN, which I think I should be able to sub in here.

[oldhousescott]

You don't still have pins 2 and 4 jumpered, do you? If so, separate them. Pin 2 should be sitting at about 4.5vdc. Pin 3 will measure below 4.5vdc because your meter will load the circuit through R3. Pin 6 should measure about 4.5vdc as well.

[mgwhit]

Nope.  Pin 4 is going straight to ground.  Pin 2 is connected to pin 6 via the feedback network.  Unless I've miswired something, what is on the schematic is what I've got on the breadboard.

So it sounds like pins 2 and 6 are dramatically off.  I'll triple check the components and make sure nothing is accidentally shorting.

[oldhousescott]

Your second drawing looks righteous. The grounded sides of C2 and C5 could instead return to VB -- in that case, C11 wants to be at least 47uF -- but they should be fine grounded as shown. Do you have another LM308 to test with?