breadboarding a Tangent Cmoy, Power balance problems
Sep 6, 2005 at 6:48 AM Thread Starter Post #1 of 11

chuck94022

New Head-Fier
Joined
Aug 29, 2005
Posts
37
Likes
0
I'm a neophyte in electronics, but I'm trying my hand at a CMOY amp, based on the Tangent build instructions.

Here are the major variations on the plan: 1, I'm starting on a breadboard. 2, I'm using Radio Shack Capacitors (the power caps are 220uf electrolytic, the signal caps are .1uf film caps (the signal ones are 100V caps, does that matter?)).


So as far as I can tell, everything is hooked up fine. I started by building the power section, and sure enough, I get a pretty well balanced result, about +- 8.9V on the rails. I hook up the Op Amp +-V pins, and still everything is balanced.

However, as soon as I build up the input and output signal sections and connect them to the op amps, I get about +13.5V and -4.5 on the rails.

I do get a tiny bit of sound out of the headphones, but it is very quiet. I will note that if headphones are in when I power up, I get a brief period (about 1/4 second) of strong clear sound, until (I'm guessing) the power caps charge up.

Is this because it is on the breadboard, or is it because of the Radio Shack caps, or could it be something else?

I'm wondering whether I should just go about building this onto the proto board via Tangent's directions and stop screwing around with breadboarding, but I'd hate to get everything soldered up and have the same problems in a smaller footprint.

Any ideas?
 
Sep 6, 2005 at 7:46 AM Post #2 of 11
Quote:

Originally Posted by chuck94022
the signal ones are 100V caps, does that matter?)


No.

Quote:

However, as soon as I build up the input and output signal sections and connect them to the op amps, I get about +13.5V and -4.5 on the rails.


If the headphones are also hooked up when you measure this, your values may be normal.

Quote:

I get a brief period (about 1/4 second) of strong clear sound, until (I'm guessing) the power caps charge up.


More likely, you are causing the op-amp to go into overload protection mode somehow. A ground short can cause this.

Quote:

Is this because it is on the breadboard


"Breadboard" is a highly overloaded term. Please clarify.
 
Sep 6, 2005 at 3:38 PM Post #3 of 11
Quote:

Originally Posted by tangent
No.

If the headphones are also hooked up when you measure this, your values may be normal.

More likely, you are causing the op-amp to go into overload protection mode somehow. A ground short can cause this.

"Breadboard" is a highly overloaded term. Please clarify.



By breadboard I mean, based on your (very nice, btw) article, a solderless plugboard.

I get the offset with or without headphones.

I haven't detected a ground short. That's not to say one doesn't exist, but after building up the power section, I worked hard to incrementally build up the rest of the circuit. I put the chip down, then only connected +V and -V. At that point, everything is still balanced. Then I attached the source socket (unconnected to a source). Then I built up one channel, both the input side and the output side to the headphone jack. It is at this point that on checking the rails, they are out of balance. Note that neither source nor headphone is connected yet.

I separately verified that both the jacks are not shorted (they are the only components touched by a soldering iron at this point).

I will look again at the problem thinking carefully about shorts.
 
Sep 6, 2005 at 7:33 PM Post #5 of 11
Quote:

Originally Posted by chuck94022
solderless plugboard.


Ugh.

A very wise man once called those "white slabs of trouble". They have several pathologies, at least one of which (metal particles bridging two rows) could easily cause your symptom.

You can keep on fighting with it to see if you can solve the problem, or you can just build the thing on some form of perfboard. The latter is a lot harder to impeach.
 
Sep 6, 2005 at 11:35 PM Post #6 of 11
Quote:

Originally Posted by chaostic
Wait, so you only have one channel circuit built? What happens when both channels are built? And can you post pictures of the completed board?


No, the problem happened when it was fully populated, but to debug it I tore it down and started over, in order to find out when the rails get unbalanced. Everything seems fine after applying +V, -V to the chip, but only gets unbalanced when I hook up one or both of the opamps.


I'm going to rebuild it on Tangent's recommended Radio Shack proto board, exactly per instructions. Sounds like the plugboard just introduces too many possible variables.

Tangent, are you saying that the Radio Shack caps, if they are in line with the specs in the schematic, will not be a problem. One concern I have is the 20% tolerance printed on the power caps. I don't have a meter I can use to match caps, so I'm just going with what I bought, unless you suggest that I need to get the precise caps you call for in the parts list (which I will eventually do anyway, but I'm too impatient to wait for the postman! :) )

Thanks!
 
Sep 7, 2005 at 3:14 AM Post #7 of 11
Quote:

Originally Posted by chuck94022
are you saying that the Radio Shack caps, if they are in line with the specs in the schematic, will not be a problem


They're fine.

Quote:

One concern I have is the 20% tolerance printed on the power caps


All electrolytics have tolerances in that range. Electrolytics suck, hard. We use them only because there's nothing else available when you need hundreds of mics of capacitance.

Quote:

I don't have a meter I can use to match caps


Matching rail caps in a CMoy has absolutely zero benefit.
 
Sep 11, 2005 at 5:14 AM Post #8 of 11
Well, I reconstructed the amp precisely (well, as precise as a n00b like me can get) as Tangent's directions state. I soldered in a socket for the opamp.

Before plugging in the opamp, I attached batteries and measured. Everything looked good. +V and -V at 8.9 volts. Then I plugged in the op amp and powered up again. +V at 15 volts, -V -3 volts. Hmmm. On a whim, I replaced the opamp with my spare OPA2134.

Voila! +V and -V at 8.9 volts! It looks like it was the op amp. Either I bought a bad one, or I somehow managed to fry it while breadboarding on the plugboard.

In any case, that problem is now behind me! Thanks for the help!
 
Sep 11, 2005 at 7:08 AM Post #9 of 11
So the power is delivered to the Op Amp just fine. I connected up a source (headphone out of a Rio Carbon, with a very nice Ricky Lee Jones tune), and phones to the out. Total silence.

So for a newbie with just a multimeter, what are the suggested troubleshooting steps? I have gone through my solder connections and have not found any obvious shorts to adjacent pads, and it looks like my solder joints are sound.

Thoughts, suggestions? If both sides have absolutely no sound, where would you start? I will say that when I checked resistance across R4, I could hear slight amounts of static in the headphones, but other than that, the amp is totally silent in the headphones.
 
Sep 11, 2005 at 7:56 AM Post #10 of 11
Quote:

So for a newbie with just a multimeter, what are the suggested troubleshooting steps?


Pictures always help.
wink.gif


Barring that let's start with the IO jacks.
Are you sure they are wired right?
If you are....you can try to signal trace the inputs with your dmm using your soundcard as the source.
Snag TrueRTA and use the signal generator to generate say a 1Khz 1 volt sine wave.
Set your DMM on AC volts and see if that signal is getting to the opamp's + inputs.
Follow so far?
wink.gif
 
Sep 12, 2005 at 1:20 AM Post #11 of 11
Quote:

Originally Posted by MisterX
Pictures always help.
wink.gif


Barring that let's start with the IO jacks.
Are you sure they are wired right?
If you are....you can try to signal trace the inputs with your dmm using your soundcard as the source.
Snag TrueRTA and use the signal generator to generate say a 1Khz 1 volt sine wave.
Set your DMM on AC volts and see if that signal is getting to the opamp's + inputs.
Follow so far?
wink.gif



Yes, and thanks! TrueRTA is a very cool free app. I tested as you said, and that signal was getting through.

While checking this out, and then trying to find out why I had no output signal, I realized what a dummy I was! When Tangent said R5 was optional, I didn't put it in - nor did I put in a jumper! What an idiot! :)

So I quickly soldered in two resistors, and voila! Perfect sound, no hiss, no distortion, nothing but net, as it were...

Thanks all for your help!
 

Users who are viewing this thread

Back
Top