Author Topic: Audio Amplifier Power Output Cunundrum  (Read 1876 times)

0 Members and 1 Guest are viewing this topic.

Offline DW1961Topic starter

  • Frequent Contributor
  • **
  • Posts: 753
  • Country: us
Audio Amplifier Power Output Cunundrum
« on: October 06, 2021, 07:42:25 am »
If you have a class D amplifier, such as the Texas instruments TPA3116D2, can you get any benefit from using two amp chips, unabridged, with the same power supply (given that the power supply isn't even strong enough to drive a single amp chip to full power)?

Specifics
Power Supply: 19V @ 4.7A

I'm asking if you have one amp chip and the power supply isn't even powerful enough to power the single amp chip to 100% of its specifications, could using two of the same chip create anymore power output? And, the reason I'm asking is that Chinese vendors are putting two of these chips on their "new" amps, but using the same power supply. So, they are advertising twice the power output.

I'm not understanding how they can get more peak or RMS using two chips with the same power supply, given that the power supply isn't even driving a single amp chip to 100% of its spec.

The 3116D will put out about:
 47.5 watts at 8 ohms and 24V
86 watts at 4 ohms and 24V
(page 9 from the spec sheet below)
But the amp manufacturer is only using a 19V 4.7A power supply.

Texas Instruments TPA3116D2 amp chip.
https://www.ti.com/lit/ds/symlink/tpa3116d2.pdf?ts=1596921462063&ref_url=https%253A%252F%252Fwww.ti.com%252Fproduct%252FTPA3116D2
« Last Edit: October 07, 2021, 01:55:01 am by DW1961 »
 

Offline bob91343

  • Super Contributor
  • ***
  • Posts: 2675
  • Country: us
Re: Audio Amplifier Power Output Cunundrum
« Reply #1 on: October 06, 2021, 05:51:35 pm »
Your poor grammar makes the question ambiguous.
 

Offline DW1961Topic starter

  • Frequent Contributor
  • **
  • Posts: 753
  • Country: us
Re: Audio Amplifier Power Output Cunundrum
« Reply #2 on: October 07, 2021, 01:39:26 am »
Your poor grammar makes the question ambiguous.

LOL, yeah that was very unclear. Sorry about that. I've edited the question, and it was a really late night.

See above.
« Last Edit: October 07, 2021, 01:50:31 am by DW1961 »
 

Offline bob91343

  • Super Contributor
  • ***
  • Posts: 2675
  • Country: us
Re: Audio Amplifier Power Output Cunundrum
« Reply #3 on: October 07, 2021, 05:18:49 am »
You reaction is a proper one.  I mean, how can you get 86 Watts from an amplifier that is run from a power supply only capable of 80 Watts?  Even at 100% efficiency it's not possible.

Perhaps in the fine print they specify a different supply.  How much output do they claim?

You are also correct that using two chips won't magically double the power unless they are running it way below ratings.  If they claim 10W with one chip maybe they can get 20 W with two chips.
« Last Edit: October 07, 2021, 05:22:03 am by bob91343 »
 

Online Zero999

  • Super Contributor
  • ***
  • Posts: 20357
  • Country: gb
  • 0999
Re: Audio Amplifier Power Output Cunundrum
« Reply #4 on: October 07, 2021, 05:57:04 pm »
It's BS. It won't deliver that much RMS power into a 4 Ohm load, with a 19V power supply, even at 100% efficiency.

P = V2/R

For a sine wave:
PRMS = PREAK/2

Therefore:
POUT = VSUPPLY2/(2R)
POUT = 192/(2*4) = 361/8 = 45.125W
 

Offline DW1961Topic starter

  • Frequent Contributor
  • **
  • Posts: 753
  • Country: us
Re: Audio Amplifier Power Output Cunundrum
« Reply #5 on: October 07, 2021, 06:12:11 pm »
You reaction is a proper one.  I mean, how can you get 86 Watts from an amplifier that is run from a power supply only capable of 80 Watts?  Even at 100% efficiency it's not possible.

Perhaps in the fine print they specify a different supply.  How much output do they claim?

You are also correct that using two chips won't magically double the power unless they are running it way below ratings.  If they claim 10W with one chip maybe they can get 20 W with two chips.

They are claiming the amp chips continuous power, which is 100W per chip at 4 ohms. (Check the amp PDF file I linked for amp chip specs.)
https://www.amazon.com/gp/product/B096KMY6WW

" 2PCS TPA3116 chip, max output power can reach 200W, comes with 19V power supply, can drive 200W (8Ω) or 300W(4Ω) speakers. " Uh, what?

The Chinese electrical engineers that do these amps are no dunces, so I'm trying to understand how they get their power ratings.

Using ohms law (https://www.rapidtables.com/calc/electric/ohms-law-calculator.html):

8ohms: 200 watts = 40V / 5A
4ohms: 200 watts = 28V / 7A

The Texas Instruments spec sheet says:
– 2 × 50 W Into a 4-ΩBTL Load at 21 V
(TPA3116D2)

Which would be (without efficiency loss added in):
5A
20V
= 100 watts
So yes, the spec sheet is right on.

At first I thought maybe they could under power each chip and get the same volume, but with less distortion, since the chips are not running up against increasing THD, which starts at are round 50% power. But, does that reasoning even work? I mean, 25% power, regardless of whether it is going through two chips or one chip is still the same power and volume, right?
 

Offline DW1961Topic starter

  • Frequent Contributor
  • **
  • Posts: 753
  • Country: us
Re: Audio Amplifier Power Output Cunundrum
« Reply #6 on: October 07, 2021, 06:22:02 pm »
It's BS. It won't deliver that much RMS power into a 4 Ohm load, with a 19V power supply, even at 100% efficiency.

P = V2/R

For a sine wave:
PRMS = PREAK/2

Therefore:
POUT = VSUPPLY2/(2R)
POUT = 192/(2*4) = 361/8 = 45.125W

Yep, when I got their first amp like the one above, I asked them about it and they said it does not provide 100% power, but in their tests, the 19V 4.75 power supply was indistinguishable sound wise from the 21V power supply, and the reason they did that was to save money. So they were forthcoming about that. And, on the other product page, they actually changed the description to add that explanation.

I'm not worried about their 19V PSU. I'm just wondering how they can add 2 of the same chip, use the same power supply (even at 21V) and get more power?
 

Online TimFox

  • Super Contributor
  • ***
  • Posts: 9000
  • Country: us
  • Retired, now restoring antique test equipment
Re: Audio Amplifier Power Output Cunundrum
« Reply #7 on: October 07, 2021, 07:14:30 pm »
Again, the equations above are for mean power, not rms power, despite incorrect nomenclature found in some regulations.
Anyway, legitimate amplifier sales engineers have often used the theoretical maximum mean power obtainable from the no-load DC power supply voltage to estimate the power available on musical peaks, since the peak to average ratio on normal music (as opposed to key-down sine waves for non-musical purposes such as test and measurement) is high, and the peak levels found on sforzandos generally do not last long enough to discharge the power supply capacitors.  In the past, these values were confirmed by replacing the amplifier's built-in power supply with an external regulated lab supply for the tests, especially since traditional THD testing required more than one second of measurement time.
My point about power:  if you measure the rms voltage and know the load resistance, then the mean power equals (rms voltage)2/Rload.  If you calculate the actual rms power (which is a useless parameter), it is (3/2)1/2 = 1.225 times higher than the mean power. 
 

Offline DW1961Topic starter

  • Frequent Contributor
  • **
  • Posts: 753
  • Country: us
Re: Audio Amplifier Power Output Cunundrum
« Reply #8 on: October 07, 2021, 09:33:43 pm »
Again, the equations above are for mean power, not rms power, despite incorrect nomenclature found in some regulations.
Anyway, legitimate amplifier sales engineers have often used the theoretical maximum mean power obtainable from the no-load DC power supply voltage to estimate the power available on musical peaks, since the peak to average ratio on normal music (as opposed to key-down sine waves for non-musical purposes such as test and measurement) is high, and the peak levels found on sforzandos generally do not last long enough to discharge the power supply capacitors.  In the past, these values were confirmed by replacing the amplifier's built-in power supply with an external regulated lab supply for the tests, especially since traditional THD testing required more than one second of measurement time.
My point about power:  if you measure the rms voltage and know the load resistance, then the mean power equals (rms voltage)2/Rload.  If you calculate the actual rms power (which is a useless parameter), it is (3/2)1/2 = 1.225 times higher than the mean power.


So can they get twice the power using 2 chips and the same power supply?

Note: The power specifications from Texas Instruments says it is Bridge Tied Load.
 

Offline DW1961Topic starter

  • Frequent Contributor
  • **
  • Posts: 753
  • Country: us
Re: Audio Amplifier Power Output Cunundrum
« Reply #9 on: October 13, 2021, 12:06:44 am »
It's BS. It won't deliver that much RMS power into a 4 Ohm load, with a 19V power supply, even at 100% efficiency.

P = V2/R

For a sine wave:
PRMS = PREAK/2

Therefore:
POUT = VSUPPLY2/(2R)
POUT = 192/(2*4) = 361/8 = 45.125W

What about two of the same amp chips running in BTL mode?
 

Online Zero999

  • Super Contributor
  • ***
  • Posts: 20357
  • Country: gb
  • 0999
Re: Audio Amplifier Power Output Cunundrum
« Reply #10 on: October 13, 2021, 07:41:35 am »
It's BS. It won't deliver that much RMS power into a 4 Ohm load, with a 19V power supply, even at 100% efficiency.

P = V2/R

For a sine wave:
PRMS = PREAK/2

Therefore:
POUT = VSUPPLY2/(2R)
POUT = 192/(2*4) = 361/8 = 45.125W

What about two of the same amp chips running in BTL mode?
My calculations are for the amplifier configured in BTL mode, with a single supply.
 
The following users thanked this post: DW1961


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf