LM3886t Turn 0n Buzz

[willy]

I've built 2 LM3886t amps according to datasheet schematic. Both amps have about a 2 second strong buzz in speaker at turn on. It's not a regular pop, but it sounds like a DC buzz? So I put a 4.7 cap on the input. They both run quite as a mouse once this 2 second buzz stops. My meter reads .080mv on output.

I replaced the output 10ohm resistor & coil with a 1000uf cap. The buzz is gone. So - is this a DC problem? Is .080 too high?
Where would I be picking up this DC?

Only difference in my circuits and the datasheet, is I didn't include the Ci cap on the feedback and I have a 15K resister to neg. rail from the mute pin. Could one of these be the problem?

Common sense says it's my fault because both circuits built identical and they both do the same thing.

(note: I'm an old solid state guitar amp. repairman from the 60s & 70s) Not exactly a newbie - but new to chipamps.

Thanks for your patience & advice. Much appreciated.

[pacificblue]

One thing has not changed since the 60s and 70s, you cannot hear DC. The capacitors on the positive and negative rail probably do not charge equally fast, so you get an imbalance in the rails, which makes the ripple voltage heard in the speakers. Or ther is a point, where the undervoltage protection circuit in the chip is at its threshold and switches on and off until the supply voltage is high enough. Did you check, if the smoothing capacitors are okay?

Assuming they are, the first step would be to increase the mute resistor values as much as possible (|Vee-2,6|/0,0005). That way the amplifier starts to work only, when the rail voltages are already very high and nearly stable. Second step is to use the muting capacitor and delay the amplifier's activation for those two problematic seconds.

80 µV? You have a good meter.

By the way, there are several schematics in the datasheet.

[AndrewT]

80uVac or 80uVdc?

Delay the output using the cap + resistor that Pacific mentions.
It is detailed in National's datasheet.
my turn on delay is <<500ms and I don't hear any buzz.

[ionioni2000]

Include de Ci DC blocking cap in NF network, is a must.
II

[digi01]

I think it is a dc problem caused by power supply.have you check the +/- rails voltage?is it balance during the first 2 second?I think one rail is weak(lower or higher than usual).


Zang

[vincent s]

Originally posted by ionioni2000
Include de Ci DC blocking cap in NF network, is a must.
II

yes,it is a must unless you have a dc servo.
in my opinion,no ci and dc servo is dangerous.If the unit works for a long time,then chip's temperature rise,dc swings,will make your unit broken.

Ci is a coupling capacitor also,remove it will make it listening better if you add a dc servo.

I think your power on buzz is not strange(most units have this probelm,but most units have a long relay time,so they can not hear the buzz),increase the value of Cm and Rm to increse the signal output ralay time.

[willy]

Thanks so much. Really appreciate the help. Maybe this will help someone else.

The mute circuit was my problem. I went up to 20K resistor and added the cap, the buzz is gone.

My power supply might have added to the problem. I used a xformer from my storage bin. Voltage to the chip is +/- 19v That's probably a little low for this chip, but it sounds great.

Also thanks for the tip on the Ci cap. Listed as an option on the page 1 schematic, I didn't think is was necessary.

These amps are being used as low volume guitar practice amps.
They seem to be doing their job now. Thanks again.