TDA2003 Amplifier

[VictorPS]

Hi,
Speaker connect to TDA2003 Amplifier shall be connect to output (pin 4) and refer to ground.
What happend for abnormal design, where the speaker is refer to +12V?

**broken link removed**
I have test it, it work well.
But I would like to know if I will get into problem after a long run for this abnormal circuit?
( I connect to +12V is for compatible on existing speaker setup which prefer not to change.)

Thank you,

 

[akg]

i don't think it will cause problems ..
since it is dc decoupled/ ac coupled . only difference shld be that when viewed thru oscilloscope , the waveforms are 180 degree out of phase.

 

[Nigel Goodwin]

Hi,
Speaker connect to TDA2003 Amplifier shall be connect to output (pin 4) and refer to ground.
What happend for abnormal design, where the speaker is refer to +12V?

**broken link removed**
I have test it, it work well.
But I would like to know if I will get into problem after a long run for this abnormal circuit?
( I connect to +12V is for compatible on existing speaker setup which prefer not to change.)

It makes NO difference if you connect the speaker to positive or negative (signal wise they are both identical) - BUT you MUST reverse the direction of the speaker coupling capacitor C14, as it's then connected the wrong way round!.

BTW, there's nothing 'abnormal' connecting a speaker this way, it's fairly common practice!.

 

[akg]

BTW, there's nothing 'abnormal' connecting a speaker this way, it's fairly common practice!.

is there any preferance/ advantage when connecting like this?..or is it done when the layout matters ?

 

[Nigel Goodwin]

BTW, there's nothing 'abnormal' connecting a speaker this way, it's fairly common practice!.

is there any preferance/ advantage when connecting like this?..or is it done when the layout matters ?

The usual advantage (although it doesn't apply in the case of this IC) is that it can save you a boot-strap capacitor, the speaker coupling capacitor can be used to boot-strap the top output transistor, saving a seperate capacitor. You used to see it done a lot in low priced record players and radios.

 

[audioguru]

The TDA2003 is designed for high output current. An 8 ohm speaker doesn't need much current.
With the 12V supply, an 8 ohm speaker gets only 2W at an awful-sounding 10% distorted overdrive. At clipping it will be only about 1.5W.
Into a 1.6 ohm load (five paralleled 8 ohm speakers), the total power with 10% distortion is 8W.

 

[Roff]

Is this done this way to minimize the pop when the power is switched on?

 

[Nigel Goodwin]

Is this done this way to minimize the pop when the power is switched on?

No, just to save a boot-strap capacitor (on amps that use them), there's no advantage on the TDA2003 circuit.

 

[VictorPS]

It makes NO difference if you connect the speaker to positive or negative (signal wise they are both identical) - BUT you MUST reverse the direction of the speaker coupling capacitor C14, as it's then connected the wrong way round!.

Thank you, will reverse polarity of C14.
How about C15 & R12, just leave it as it or refer to +12V?

 

[audioguru]

You can minimize the turn-on pop of a single supply amp like this one if you allow the input capacitor to charge slower than the feedback capacitor but only if the load is cap-coupled to ground.
I think this amp would have a terrible BANG when it is turned-on.

 

[akg]

You can minimize the turn-on pop of a single supply amp like this one if you allow the input capacitor to charge slower than the feedback capacitor but only if the load is cap-coupled to ground.
I think this amp would have a terrible BANG when it is turned-on.

Is that pop caused due to initial charging of the i/p cap?, or does other capacitor in the ckt take its part ?

 

[audioguru]

Since the input cap is discharged before the amp is turned on, when you turn it on then the input cap keeps the input voltage low until it charges, therefore the amp's output voltage is also low charging the big output cap suddenly through the speaker.
The pop will be much less if the speaker was connected to ground.

 

[Nigel Goodwin]

It makes NO difference if you connect the speaker to positive or negative (signal wise they are both identical) - BUT you MUST reverse the direction of the speaker coupling capacitor C14, as it's then connected the wrong way round!.

Thank you, will reverse polarity of C14.
How about C15 & R12, just leave it as it or refer to +12V?

It doesn't really matter!.

 

[Roff]

I believe that we are assuming here that +12V is well-regulated and ripple-free. Any noise on +12V will show up across the speaker.

 

[audioguru]

I believe that we are assuming here that +12V is well-regulated and ripple-free. Any noise on +12V will show up across the speaker.

Certainly!
And if the supply has some internal resistance then the amp will make motorboating sounds like crazy due to positive feedback.

 

[akg]

I believe that we are assuming here that +12V is well-regulated and ripple-free. Any noise on +12V will show up across the speaker.

Certainly!
And if the supply has some internal resistance then the amp will make motorboating sounds like crazy due to positive feedback.

hmm..that's why the spkrs are grounded

 

[Nigel Goodwin]

I believe that we are assuming here that +12V is well-regulated and ripple-free. Any noise on +12V will show up across the speaker.

Certainly!
And if the supply has some internal resistance then the amp will make motorboating sounds like crazy due to positive feedback.

hmm..that's why the spkrs are grounded

I would disagree! - in BOTH cases you have exactly the same current through both supply rails, so connecting the speaker to either makes no difference. If your supply is poorly decoupled and has a high source impedance (or the battery feeding it is flat!) then it will motorboat regardless of where the speaker is connected!.

 

[audioguru]

With the speaker connected to the supply, if the supply has enough internal resistance then the resulting positive feedback to the non-inverting amp's input causes motorboating.

With the speaker grounded, then resistance in the positive supply causes negative feedback for reduced gain, and no motorboating.
Actually, I think the reduced gain would be half-cycle, causing severe 2nd harmonic distortion (valve-sound!). :lol:

 

[Nigel Goodwin]

With the speaker connected to the supply, if the supply has enough internal resistance then the resulting positive feedback to the non-inverting amp's input causes motorboating.

With the speaker grounded, then resistance in the positive supply causes negative feedback for reduced gain, and no motorboating.
Actually, I think the reduced gain would be half-cycle, causing severe 2nd harmonic distortion (valve-sound!). :lol:

You are imagining the IC as a 'perfect' opamp, which of course doesn't exist, motorboating will occur in either case - NOT via feeds to the inverting or non-inverting inputs, but via the supply pins!.

 

[audioguru]

Hi Nigel,
With an opamp usually the input voltage divider has a bypass capacitor to reduce hum and noise from the supply.
The TDA2003 car amp has an internal voltage divider for its input which has fairly poor supply noise rejection (-36dB), and unlike an opamp it doesn't have a balanced differential input that would provide excellent rejection.

Exactly like an opamp, the datasheet shows that it has the same voltage gain with different supply voltages.

With the speaker grounded, current is drawn from the supply and the ouput capacitor is charged into the speaker when the input and output go positive. This causes the supply voltage to drop if it has resistance, resulting in negative feedback to the input voltage divider.
When the input and output go negative, the output capacitor discharges into the speaker without affecting the supply current very much. This causes severe 2nd harmonic distortion.

With the speaker connected to the supply, current is drawn from the supply and the output capacitor is charged into the speaker when the input and output go negative. This causes the supply voltage to drop if it has resistance, resulting in positive feedback to the input voltage divider.
This results in motorboating. :lol: