TPA2000D1: Pop and thump sound coming in through speaker when turning the device ON.

Hi Tapas,

the TPA2005 can only prevent the pop noise during power up, if both inputs (IN+ and IN-) see identical impedances to GND. This is the case, when the volume is turned-off. Then the slider of pot is at GND potential. But if you turn-on the volume, the impedance symmetry at the inputs is destroyed and a pop noise occurs.

Kai

Hi Kai,

Is there any way in which can reduce the pop sound even when the symmetry is not maintained at the input, because user will not care whether the potentiometer impedance is set to low or high.

Method 1:
Can we connect 100K resistor between C15 and ground ; 100K between C16 and ground, so that there is no buildup of charge in the capacitor as to reduce the pop sound in the first place.

Method 2:

We can delay the start of the TPA2005D1 using RC circuit in the shutdown pin, so that pop sound will be eliminated.

You can guide me, should I start with method1 or method2 ?

-Tapas

Hi Tapas,

The pop noise is caused by an unbalanced input.
I have some questions for you, Could you provide me the components value? what is the function of HP out signal?

Best Regards
José Luis Figueroa
Audio Applications Engineer

Hi Tapas,

I don't think that method 1 will work. And let's wait what José mentions about methode 2.

Another idea would be to insert a driver between the pot slider and C15. The driver needs to have a low ohmic output. Or R29/R30 can be used for correcting the imbalance: R29 + R(driver output resistance) = R 30.

Kai

Hi José,

The components values are
C? --------------------> 2.2uF X7R
R99 --------------------> 100K ohm 1%
O3 --------------------> 10K 1%
R28---------------------> 300 Ohm resistor 1% (to make sure volume never goes to zero)
C15, C16---------------> 1uF, X7R, 5%
R29, R30---------------> 100 K, 1%Ohm resistor (for gain of 3)
R27---------------------> 10K ohm, 1%

HP out signal goes to 3.5 mm Jack (which is goes to pre-amplifier)

Regards,

Hi Kai,

Approach 1:

"Another idea would be to insert a driver between the pot slider and C15. The driver needs to have a low ohmic output. Or R29/R30 can be used for correcting the imbalance: R29 + R(driver output resistance) = R 30."

Can you clarify, the above line specifically what driver you are talking about.
The pot slide can in any position and still while turning ON the TPA2005D1, there should be no sound. How the above system will make sure that in all the position of the pot slider inputs (IN+ and IN-) see identical impedances to GND.

Approach 2:

We can delay the start of the TPA2005D1 using RC circuit in the shutdown pin, so that pop sound will be eliminated.

Is there any specific IC from TI which can be connected to the shutdown pin so that TPA2005D1 will start with little bit delay. Or this can be implemented with the normal RC circuit with delay of 1 sec with diode across resistor (normally reverse biased) in parallel to discharge the C when the power goes off. I am wondering do we need Schmitt trigger to ensure stability of the above system.

Regards,

Tapas

Hi Tapas,

Approach 1:
Is O3 a jumper? if so, why does O3 have a value of 10K?
My recommendation would be to remove R28 because this component is adding an additional resistance on IN+ and relocate the HP Out signal before C15 as this is also producing an unbalance between the inputs.
First, you could try to do my recommendation and see if the pop noise is reduced.

Approach 2:
The shutdown pin is a digital input. When the input voltage achieves a certain level (2V according to datasheet in this device) TPA2005D1 will be activated and will start-up at 9ms according to the datasheet. TPA2005D1 has internal circuity to turn the device on at 9ms, so the rising ramp will always be the same even if you add a delay circuit.

Best Regards
José Luis Figueroa
Audio Applications Engineer

Hi José,

Approach 1:
O3 is not a jumper, it is a 10K pot (part no: RV100F-30-4K1B-B10K-B301).
The reason I have included R28 so that I volume should not become zero, it is required in the application.
I will relocated HP Out signal, will tell you about the result.

Approach 2:
Correct me if I am wrong. Even if we include delay circuit in the shutdown pin, there will always be a pop sound. Is this what you are saying in the previous thread.
If we add a delay of 1 sec, will it not solve the problem.

Approach 3:
I am thinking can we use a non-inverting op-amp, and connect the feedback resistor to 10K pot to get required level, in this approach the line will be balanced. What do you think on this.

Regards,

Tapas

Hi Tapas,

Approach 1: Please let me know if this helps you.

Approach 2: No, it means that the device starts up at 9ms as you can't add a delay to the shutdown pin as it is a digital input.
The pop noise is caused by an unbalanced input. Sometimes the single-ended input results in an impedance mismatch between two inputs and this can produce an audible pop. This pop can be reduced if there is sufficient time for both input capacitors to discharge completely, but as you can't increase the time, then you can reduce the input capacitors to minimize the discharge time of the capacitors.

Approach 3: In fact, this is something that I was going to advise you. It is always recommended using an Audio Op-Amp in order to implement a volume control. You can find an Audio Op-Amp in Product Folder.

You should relocate HP out signal, reduce the input capacitors values and use an Audio Op-Amp to implement a volume control if you follow all these points the pop noise will be reduced. Another option would be using a differential input if the system specifications permit it.

Best Regards
José Luis Figueroa
Audio Applications Engineer

Hi José,

I have made a schematic for approach 3, that we have discussed earlier (using single ended input). I have some doubt in this schematic

1) Even if the volume is full, gain = -10K/10K = -1, will it not case pop sound when the TPA2005D1 is turned ON. According to the reason given earlier, "the TPA2005 can only prevent the pop noise during power up, if both inputs (IN+ and IN-) see identical impedances to GND". How come now just by using op-amp, both the IN+ and IN- are seeing the identical impedance, because IN+ is always connected to the ground. 

2) Adding a inverting phase (inverting op-amp has been added for volume control) in the audio path will change the phase, will it affect how the sound will be perceived by the human ear.  

Regards,

Tapas

Hi Tapas,

yes, this should eliminate the pop noise. Power-on all associated circuits while the TPA2006 is in shutdown. Allow the 2µ2 cap (C) to charge up. Then, when everything is charged up, activate the TPA2006 by disabling the shutdown. At this moment the ouput of inverting OPAmp will show 0V and will provide a very low output impedance. As consequence both inputs of TPA2006 will see nearly identical impedances to GND and the pop should disappear.

Inverting the phase has no impact on the human ear.

Kai

Hi Kai,

The circuit has been designed such that all the electronics including TPA2005D1 starts at same time. It is difficult for me to change at this point in time to delay the start timing of TPA2005D1 IC. I think TPA2005D1 starts after 9ms, Well is this the sufficient time to charge the capacitor.  

Can I connect 100K ohm pull down resistor from Op-amp output, so that both IN+ and IN- will see same impedance which will be ground. (I am thinking any small voltage across this 100K ohm resistor will create impedance mismatch). What do you have to say on this.

Well, if you have any alternative approach, do let me know.

 
Regards,

Tapas

Hi Tapas,

I haven't heard back from you, I'm assuming that you were able to solve your problem. If not, just post a reply or create a new thread if the thread has locked due to time-out.

Best Regards
José Luis Figueroa
Audio Applications Engineer