Any gotchas on using the LM3886TF?

[ventle]

Just got 6pcs of the LM3886TF bought on ebay. Wanted to hook one up just to see if it worked, using as few components as possible - see attached schematics. Omitted the decoupling capacitors, the delay capacitor on the mute resistor, the mute switch and the input potmeter. Using +/-27 Volts supply rails, 10k mute resistor, verifying that the mute resistor drops 24 volts (2.4 amps mute current), and no DC offset at the output (I count the 30-40 mVs as no offset). Source is a Creative Muvo mp3 player. Chip is attached to an aluminum heatsink.

Shouldn't this theoretically work like a charm? I can't get any sound from the speaker, which seems kind of riddiculous, isn't it just a plain op-amp circuit plus the mute pin that needs to have at least 0.5 mA drawn from it to turn the op-amp on? I know my circuit isn't optimal, but it should make some sound, shouldn't it? Are there any gotchas with this chip that I've overseen?

Tried two different chips, same results. Haven't got neither a signal generator nor a scope available right now, only test equipment available is my multimeter. I connected the mp3 player to a headset to verify it was actually playing.

Any help appreciated - feeling kind of stupid not being able to make a simple opamp circuit work like it should

[infinia]

Read the manual first and then again before touching a soldering iron. PS Youre missing about a dozen more parts. Show them on your schematic there's devils in those little details. Power supply is important too.
LM3886 - High-Performance 68W Audio Power Amplifier with Mute

[ventle]

Of course I did read the datasheet (or manual, if you prefer). But please, point out for me, which of the omitted components are making my circuit silent? As far as I see it:
- the Cs capacitors are only used for power supply filtering - the amp should amplify it's input signal even if they aren't there
- the Ci capacitor is marked as optional, and it's purpose seems to be to provide AC coupling for the feedback loop (plus some high frequency single pole compensation)
- the purpose of the Cm capacitor is to delay the turn-on of the amplifier by slowly increasing the current drawn from the mute pin
- the Rin potmeter provides volume adjustment
please, tell me, which of these components is it so inherently stupid to omit that I shouldn't be using my soldering iron?

[infinia]

all this stuff is in the data sheets consider "optional" really to be mandatory unless you know of a good reason to omit ie your an expert. RTFMS

For a starters
an example from the data sheet text not pictures
Cs is really 2 caps >470 uF + 0.1 uF
2 DC blocks cap missing and output zobels

[Sebastiaan]

Originally Posted by infinia

all this stuff is in the data sheets consider "optional" really to be mandatory unless you know of a good reason to omit ie your an expert.

For a starters
an example from the data sheet text not pictures
Cs is really 2 caps >470 uF + 0.1 uF
2 DC blocks cap missing and output zobels

That doesn't explain why his amp doesn't work it should at least produce some sound even without those optional components.

The only think I can think of is that the mute resistor should be around 48K instead

With kind regards,
Bas

[infinia]

Its probably oscillating at 10MHz

[jaycee]

It should work. Decoupling caps would help prevent oscillation. Try a mute resistor of 22K. It should be on a heatsink, even if it's only a small one.

The caps in the input and feedback paths are there to kill offset - it should work without.

[ventle]

Originally Posted by infinia

all this stuff is in the data sheets consider "optional" really to be mandatory unless you know of a good reason to omit ie your an expert. RTFMS

For a starters
an example from the data sheet text not pictures
Cs is really 2 caps >470 uF + 0.1 uF
2 DC blocks cap missing and output zobels

Thank you, I do know how to decouple supply rails, I do know how to block DC currents and I do know how to implement a zobel network to protect the output stage from the return currents of a reactive load, but none of these should make the difference between an opamp amplifying it's input signal to make some kind of sound in the attached speaker, and one that doesn't - unless I'm missing something really important regarding the differences between the LM3886 and a generic small-signal opamp like e.g. a TL072? Is it the Cm capacitor that makes the big difference?

Sebastian - the mute current should be at least 0,5 mA - anything above should be sufficient to turn it on. Initially had a 33k in place, changed it to a 10k to see if it would help.

Jaycee - the 10k resistor is rated 1W, dissipating approx. 57mW - shouldn't need a heatsink then?

[infinia]

try really good bypassing 1st & foremost, This is HF power not small signal!
Use all the parts 1st, get it working. then U can simplify

[pjp]

ventle, I think you can safely ignore infinia, he's just pretending to know what he's talking about.