• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 30 subscribers
  • Views 1910 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TPA3251D2EVM: The Sound of TPA 3251

Engne Tang34
Prodigy 40 points

Part Number: TPA3251D2EVM

After completing a 2x80W power amp design, I'm having difficulty with the way TPA 3251 sounds. This is after numerous listening tests and  comparisons with a 2x100W power amplifier (with good review by magazine), a 2x50W analog power amplifier and a 2x12W low power analog amplifier (powered by a Laptop 19V power supply).  It is the first time I use this chip for a customer and I'm looking for ways to improve the way it sounds. The deficiency is in the mid and low end frequencies.

I'm using a pair of $1500 KEF 50th anniversary Model loudspeakers.  Playing classical and soft jazz music with no EQing. My listening room is small like 20'x20'.  The bass and mid-range sounded thin, lacks deep bass punch, warmness and fullness, with OK transparency.  It is so obvious even a non-audiophile person can detect the big difference.   The frequency response of the amp built with TPA3251 is good,  -0.5dB/20HZ and -0.8dB/20KHZ with 8 ohm load. 

A question came up after studying the schematic of TPA3251D2EVM,  I am wondering if applying post filter negative feedback would solve my problem ?  Has TI ever tested the EV board with R47,49,50,51;C73,74,75,76, and R4,12,44, and R46 all inserted ?  What values are they and how do they affect the overall gain of the amplifier ?  Thank you !

Engne Tang

VAC-com, Inc.

  • 0
    TI__Expert 6195 points
    Hi Engne,

    Here's an Application Report for the TPA3251 Post-Filter Feedback.
    www.ti.com/.../slaa702.pdf

    Regards,
    Dan
  • 0
    Prodigy 40 points
    Thanks Dan.
    After reviewing SAA702 info about PPFB design, something does not add up. That it says the measured gain for circuit using TPA3255 is 12.2dB that is only 4 times gain. The datasheet says the max analog input voltage of this chip is 7Vpp which is 2.5Vrms. That means the circuit produces 10Vrms per output or 20Vrms in BTL mode. 20Vrms on 8 ohm load is a lot short of 150W rating of the chip. Am I missing anything here ?
  • 0 in reply to Engne Tang34
    TI__Expert 6195 points

    Hi Engne,

     

    In order to use the full power of the TPA3255 and achieve the 150W the device is rated for, you’re going to have some gain compensation (see section 5 of the app note). By changing some components in the setup we are able to get the system gain back to 21.5dB.

    Take a look at Figure 5 in the app note. The TPA3255 is clearly able to achieve 150W.

    As to the point about the maximum input voltage, adding gain before this isn’t necessarily going to push the input voltage of this device above its maximum. Don’t forget about R_in.

     

    Regards,

    Dan

  • 0
    Prodigy 40 points
    I see the feedback resistors were changed to 28K instead of 18K in the gain calculation in prior section of the document.  That make sense,  I'll experiment to add the external PF feedback to see if it also improve the way it sounds.  Low THD looks good on paper, if the amp does not sound musical it is still something missing.

    Engne Tang



    At 11:05 AM 3/31/2017, you wrote:

     
    A Message from the TI E2E� Community

     

    Dan Kisling replied to TPA3251D2EVM: The Sound of TPA 3251.

    Hi Engne,

     

    In order to use the full power of the TPA3255 and achieve the 150W the device is rated for, you�re going to have some gain compensation (see section 5 of the app note). By changing some components in the setup we are able to get the system gain back to 21.5dB.

    Take a look at Figure 5 in the app note. The TPA3255 is clearly able to achieve 150W.

    As to the point about the maximum input voltage, adding gain before this isn�t necessarily going to push the input voltage of this device above its maximum. Don�t forget about R_in.

     

    Regards,






    Dan

     
     

    You received this notification because you subscribed to the forum.  To unsubscribe from only this thread, go here.

    Flag this post as spam/abuse.
  • 0 in reply to Dan Kisling
    Prodigy 40 points

    Hi there,

    Experimented adding post filter feedback resulted in problem, the chip entered protection due to, I believe an oscillation. Very difficult to tame the beast.  I did not use the suggested component values in slaa702 document for the Zobel network C_Z and R_Z,  but tried with smaller capacitors for C_Z.  The reason being,  a 1.33uF cap is 6.7 ohm in impedance at 20KHZ,  with 3.3 ohm/R_Z and two sets of them at BTL outputs it's equivalent to adding a 5 ohm load in parallel with the test load.  When increase the output to 20Vrms, the 3.3 ohm resistor could not take the 56W power imposed on it and instantly the 1 watt resistor smoked while testing with 20K signal. 

    Engne Tang

  • 0 in reply to Engne Tang34
    TI__Expert 6195 points

    Hi Engne,

     

    I would not find these results unexpected if you reduced C_z. As mentioned in section 1.1 of the datasheet, the Zobel network is required for stability.

    You may find some issues with burning up the Zobel resistor burning up when exposing your system to strenuous tests even when using the suggested components. The App note was written for typical audio, so high voltage perfect sine waves may cause you issues.

    Hope this helps!

     

    Dan