• State Resolved
  • Locked Locked
  • Replies 12 replies
  • Subscribers 29 subscribers
  • Views 4523 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.

Audio Power Amplifier IC with TPA3116D2 audio quality but more output power

Tomasz Chumiecki
Genius 3060 points
Other Parts Discussed in Thread: TPA3116D2, TPA3118D2, TAS5614, TAS5630, TAS5756M, TAS5766M, CC8530, PCM5242, PCM5142, TPA6120A2

Hi TI Audio-Team,

I am looking for an audio power amplifier IC, doesn’t matter with analog or digital input, with similar audio quality to the TPA3116D2 / TPA3118D2 but with more power. The TPA3116D2 / TPA3118D2 ICs sound fantastic. I could reach output idle noise level below 40µV A-weighted RMS – the best IC audio amplifier from TI I ever implemented :)
I use the chips in small active wireless speakers and it works perfect.

Now we want to make our middle size speakers active too but even the TPA3116D2  power is no more sufficient to use it as a bass driver :(

Are you planning to bring on the market any stronger version of the TPA3116D2 / TPA3118D2 family?
Is there any other audio amplifier IC from TI having more power but idle noise level of TPA3116D2 / TPA3118D2 family? I guess TPA3116D2 / TPA3118D2 with supply voltage of 35V would be a perfect product for today’s audio market :)

Cheers,
Tomasz

  • 0
    TI__Expert 4320 points

    Tomasz

    Thank you for the nice words on the tpa3116.

    Have you looked at using lower impedance woofers? like a 2.5R - then a PBTL connected TPA3116D2 willl do 100W, and the benefit is that you can use a single 24V voltage rail for both the woofer and mid/tweeter.

    i have a set-up running with a dual 2.5R voice coil then 200W is possile with 2 devices in PBTL - still from the same 24V power supply.

     for higher power devices look at the TAS5614/24 for 35V and TAS5630/31 for 50V devices - these devices have a litle different loop gain topologies, with higher noise due to the need to work with higher supply voltages.

    for I2S input devices we do have a MCM device that use the TPA3118 output stage. they are TAS5756M and TAS5766M

    rgds,

    Kim

  • 0 in reply to Kim Nordtorp Madsen
    Genius 3060 points

    Kim,

    Thank you for your fast response and all the information. I have to talk to the customer I do the project for, if they have such a woofer available. As far as I know the standard 4R woofer they use goes to 3.5R and they should have also a 3R woofer on the shelf. You right with the 24V single rail supply – it is very comfortable and cost effective for the system to use. Paralleling two devices in PBTL mode I do need two output filter sections, am I right?

    I used TAS5614 and TAS5630/31 in the past but I was really unsatisfied with the idle noise level. The TAS3116D2/18D2 is way better – I was really surprised how good the IC is implementing it for the first time. Interesting is, that I measure much lower idle noise on my board as specified in the data sheet… do I make something wrong? I use NTi XL2 Audio Analyzer for RMS noise and THD measurement. If you like I can share with you my setup and measurement results.

    TAS5756M looks also very interesting too, but I will first stay with TAS3116DA/18D2. I changed my signal path right and currently I have CC8530 -> PCM5242 -> 2x TAS3116DA/18D2 for stereo PBTL amplifier. Is it possible to use 2 TAS5756M devices parallel to drive 2.5R woofer? Do I need just to feed the TAS5756M digital input with this same audio data? Do you know is the miniDSP section of TAS5756M similar to PCM5242?

    Thanks for your support!
    Best regards
    Tomasz

  • 0 in reply to Tomasz Chumiecki
    TI__Expert 4320 points

    Tomaz

    the TPA3116D2 can be connected in parralle before the filter, but then the filter inductors need to 15A saturation current, so often we see that it is cheaper to use same inductor as for BTL and then connect in parrallel after the inductor.

    yes correct the TAS5614 & TAS5630 is gain optimized in different way that leads to higher noise - the SNR is in the same 102-104dB range

    with the 2 chip solution you have now: PCM5242+TPA3116/18 you have the most flexibility, as you can use pad down or pad up for more or less power - also using PBTL connected TPA3116's giver higher power options than can be acheived with TAS57xx devices.

    yes TAS5756M and TAS5766M can be used in PBTL mode, but cost vise PCM5242+2xTPA3118D2 will be lower cost than 2xTAS5756M

    yes we use same good Dsp & DAC in the TAS5756M and TAS5766M as in PCM5242.

    The ROM/RAM codes are different for PCM5242 & TAS5756/66 so a different micro code is needed

    rgds,

    Kim

  • 0 in reply to Kim Nordtorp Madsen
    Genius 3060 points

    Kim,

    Thank you for your response and answering all my questions. I will stay with my 2 chip solution: PCM5242 + TPA3118D2 / TPA3116D2 for the moment because it fits perfectly into my project now and as you said gives the flexibility.

    I have to play a little bit more with the PurePath Studio because the new chips with included DSP section look more and more interesting :) We use a different DSP platform in our projects right now but for simple active loudspeakers the PCM5242 or the TAS5756M can bring significant cost saving keeping the same performance and functionality.

    Thanks for your support!
    Best regards
    Tomasz

  • 0 in reply to Tomasz Chumiecki
    Genius 3060 points

    I have one more question regarding TPA3118D2 / TPA3116D2. Do the both ICs use this same die? TPA3118D2 in a regular way with thermal pad down and TPA3116D2 reversed with thermal pad up? On a data sheet they look exactly this same? Is it possible, providing sufficient cooling, getting this same output power from TPA3118D2 as from the TPA3116D2?

    Tomasz

  • 0 in reply to Tomasz Chumiecki
    Genius 3060 points

    Kim,

    Have you got any reference design ore just schematic diagram with PCM5242 + TPA3116D2 you could share with me?

    Tomasz

  • 0 in reply to Tomasz Chumiecki
    Genius 3060 points

    Hi TI Experts,

    just refreshing my question regarding TPA3118D2 / TPA3116D2. Do the both ICs use this same die? TPA3118D2 in a regular way with thermal pad down and TPA3116D2 reversed with thermal pad up? On a data sheet they look exactly this same? Is it possible, providing sufficient cooling, getting this same output power from TPA3118D2 as from the TPA3116D2?

    Tomasz

  • 0 in reply to Tomasz Chumiecki
    TI__Expert 4320 points

    Tomasz

    Its not the same die's its mirrored

    yes with the same cooling you will get the same power..

    If you do not need full power continiously the TPA3118d2 is quite powerfull on the EVM..

    rgds, kim

  • 0 in reply to Kim Nordtorp Madsen
    Genius 3060 points

    Hi Kim,

    Thanks a lot for your replay.

    This is a very satisfying answer I was expected. I have a single side assembled PCB laying flat on aluminium heat sink with some thin GapPad in between to provide good electrical isolation. TPA3118D2 is thermally very good connected with the PCB bottom side. The PCB bottom side is nearly whole ground plane with wide area under the TPA without soldering mask to provide optimal heat transfer to the heat sink. The TPA3118D2 in this configuration performs really outstanding. This mechanical concept is also very simple for manufacturing and inexpensive. I considered to replace the TPA3118D2 with TPA3116D2 just to reach more impulse power, but if booth dies perform this same in case of impulse power I will stay with my concept. In fact, playing music, even with lot of low frequencies, the impulse power is often more important as the continuous one.

    Thanks,
    Tomasz

  • 0 in reply to Tomasz Chumiecki
    TI__Expert 4320 points

    Tomasz

    no special filters are needed between PCM5242 & TPA3118D2 - we have atest board with PCM5142+TPA3118 where we just have a single ac-coupling cap and it measures very well, both noise and THD vise.

    if you want to optimize for lowest possible noise put a small RC filter in, like 100R 1nF, with diff in, the lowest gain setting, 20dB, of the TPA3118 is actual too high so you could look at dividing the DAC output by 6dB.

    rgds, kim

  • 0 in reply to Kim Nordtorp Madsen
    Genius 3060 points

    Kim,

    thanks a lot, i will try. I already placed footprints for similar filter like you have on TPA6120A2 evaluation module - just in case I need it. You right, the PCM5242 output is powerful, I have to take care not to overdrive the TPA3118D2. As soon I have some first result I will try to share some here.

    Cheers,
    Tomasz

  • 0 in reply to Kim Nordtorp Madsen
    Genius 3060 points
    Hi Kim,

    Long time, a lot of work and finally my Amplifier works :) I have one short question. Could you explain me a little bit, why do I need the AC-coupling connecting PCM5242 output to TAS3118D2? I thought PCM5242 has the line driver without any DC offset? Do I need the AC coupling any time implementing the TAS3116/TAS3118 family?

    Best regards,
    Tomasz