Virtual ground in power amp applications

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
the National schematics are quite confusing.
Most designers and builders will recommend fitting an RF filter to the Input of a Power Amplifier. I go a bit further. Fit an RF attenuating filter to the Input of every bit of Audio equipment.
You use R1 as part of the filter. Add a small cap across R2 to put the other half of the RF attenuating filter in place.

The amplifier is DC coupled at the input. It will pass any DC presented to the input right through to the output, your speaker.
In addition the +IN and -IN inputs are loaded with different biasing resistances. This will induce an input offset. That input offset translates into an output offset. Worse still, the unbalanced input offset increases the sensitivity of the chipamp to temperature changes. The offsets are worse with changes in temperature than when the biases to the inputs are set up equally.
The terrible schematic that you have posted presents 22k to the -IN input.
The resistance presented to the +IN can vary from 1k to 5k45. This difference can be eliminated by adding a DC blocking capacitor to the input, either before or after R1.
This cap does two jobs:
a. Blocks DC that could come in from a Source
b. Allows the biasing to the input stage to be fixed at the correct values.

These two added components can be omitted by very experienced builders who know what to look for and how to eliminate the potentially damaging consequences.
Newcomers and Beginners should never omit any of the "National optional components".

Further I recommend a small air cored inductor should be placed in the speaker feed.
This is the R//L often seen in schematics and converts the Output Zobel to the Full Thiele Output Network.
 
Please bear with me as I try to understand your comments. I appreciate the input and want to fully understand it so that I can come up with the best possible results.
the National schematics are quite confusing.
Most designers and builders will recommend fitting an RF filter to the Input of a Power Amplifier. I go a bit further. Fit an RF attenuating filter to the Input of every bit of Audio equipment.
You use R1 as part of the filter. Add a small cap across R2 to put the other half of the RF attenuating filter in place..
By small cap, I assume you mean something on the order of 0.1uF film cap?? Or a "small" ~47uF electrolytic?

The amplifier is DC coupled at the input. It will pass any DC presented to the input right through to the output, your speaker.
In addition the +IN and -IN inputs are loaded with different biasing resistances. This will induce an input offset. That input offset translates into an output offset. Worse still, the unbalanced input offset increases the sensitivity of the chipamp to temperature changes. The offsets are worse with changes in temperature than when the biases to the inputs are set up equally.
The terrible schematic that you have posted presents 22k to the -IN input.
The resistance presented to the +IN can vary from 1k to 5k45. This difference can be eliminated by adding a DC blocking capacitor to the input, either before or after R1.
This cap does two jobs:
a. Blocks DC that could come in from a Source
b. Allows the biasing to the input stage to be fixed at the correct values.
These two added components can be omitted by very experienced builders who know what to look for and how to eliminate the potentially damaging consequences.
Newcomers and Beginners should never omit any of the "National optional components"..
For a DC blocking cap I assume you mean something like a 10uF high quality film cap?

Further I recommend a small air cored inductor should be placed in the speaker feed.
This is the R//L often seen in schematics and converts the Output Zobel to the Full Thiele Output Network.
I have seen this done, usually a small bit of wire wrapped around a 3 or 5 W low ohm resistor that is in series with the Speaker. Is this what you mean?
Finally, what do you mean by "The terrible schematic that you have posted"? Am I to assume you take issue with the values chosen? or is there an issue you have with the design? I am working from a PC board with the shown schematic, but could easily change values if I can understand why other values would be better.
Again, I am totally new to this chipamp thing and appreciate any help. Just please don't yell at me for being "dumb."
 
The input RF filter is to attenuate the RF interference without attenuating any of the audio signal.
Builders will adopt a turn over frequency (F-3dB) from 20kHz to 1Mhz.
Most aim for around 100kHz to 300kHz.
The frequency is predicted using the formula F-3dB = 1 / { 2 * Pi * R * C).
If your R=1k and you use 1nF (0.001uF or 1000pF) then F-3dB ~ 1/2/3.14/1000/0.000000001 ~ 160kHz
i.e. small = < 0.01uF Use a plastic film.

The DC blocking cap can be anywhere from 10nF to 1000uF, depending on what frequencies must pass through the capacitor since it forms a high pass filter.

Using the same formula as above you can determine a suitable value for your equipment and your ears.

The National schematic and the one shown omits all the optional components. This is inviting disaster. Experienced builders can properly test a stripped down chipamp implementation and know what to look for to ensure correct behaviour. Beginners should never omit the optional components.
 
The National schematic and the one shown omits all the optional components. This is inviting disaster. Experienced builders can properly test a stripped down chipamp implementation and know what to look for to ensure correct behaviour. Beginners should never omit the optional components.
OK. I guess I'm being dumb, but what "optional components" are you referring to? (other than the RF filter and DC blocking cap)? The "Typical Application" from the data sheet doesn't even have R2 (+ input to gnd) or Rz/Cz (Zobel).
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.