Not sure about the EVM board restrictions. Is it for both channels driven simultaneously?
The attached is directly from TI's spec sheet on the chip. It's my understanding that you can't drive both channels simultaneously to 70w each due to the chip's heat dissipation restrictions. But for a single chip in PBTL mode your only driving one output to 70w. I'm at 26v with fresh batteries so I get about 78w @ 1% before clipping the output with the resulting high distortion levels.
Try to drive one output with pbtl 😀
Early there were all kind of power measurements for evm, heat with 4 ohm the limiting factor, max output into 4 ohm 112 watt, continuous around 56 watt, so if you want you can say you have a 112 watt amp, no music lasts 1 second, not counting some kinds of noise 🙂 Somewhere around 24V the 3116evm reaches maximum, increasing PSUvoltage doesn't bring higher output.
Early there were all kind of power measurements for evm, heat with 4 ohm the limiting factor, max output into 4 ohm 112 watt, continuous around 56 watt, so if you want you can say you have a 112 watt amp, no music lasts 1 second, not counting some kinds of noise 🙂 Somewhere around 24V the 3116evm reaches maximum, increasing PSUvoltage doesn't bring higher output.
Not sure if I understand what your saying ...
Were the measurements at 4 ohm, continuous 56w, for each channel with both channels driven?
Increasing the supply voltage ups the "the rail", the upper limit of the electrical wave form before clipping the output, thus increasing the usable power. Again, not sure about the EVM module, but the chip is spec.'ed to 26v.
I do agree on music not being as demanding as a sine wave (e.g., a sine wave is the only true "rms" power rating).
PBTL 4 ohm subwoofer unstressfull subwoofer 78Watt 25V is what I replied to, PBTL is both channels driven, all "4 amps" combined drive one channel.
Around 24V TI EVM reaches maximum output, ~56Watt. Datasheet shows parts EVM can't reach, especially big part of 2 ohm power graph, TIsupport usually replies "you need an awesome cooling strategy" to get there.
Increasing PSU voltage increases heat in tpachip and heat is the limiting power factor for 4 ohm load.
Around 24V TI EVM reaches maximum output, ~56Watt. Datasheet shows parts EVM can't reach, especially big part of 2 ohm power graph, TIsupport usually replies "you need an awesome cooling strategy" to get there.
Increasing PSU voltage increases heat in tpachip and heat is the limiting power factor for 4 ohm load.
24V vs. 26V..
Hmm,
Rds_on = 0.12R
Rdc_L = 0.1R
"Load": 4R + 2*Rds_on + 2*Rdc_L = 4R + 0.24R + 0.2R = 4.44R
24V*24V/4.44R = 129W or ~ 65Wrms
26V*26V/4.44R = 152W or ~ 76Wrms
So +3dB is double the power. To have +3dB signal compared to 24V you'd need PVCC = 33.8V. For the numbers above, 26V is 1.08 * 24V. Or in dB, you'll gain 0.67dB.
Would you hear that? 😀
Hmm,
Rds_on = 0.12R
Rdc_L = 0.1R
"Load": 4R + 2*Rds_on + 2*Rdc_L = 4R + 0.24R + 0.2R = 4.44R
24V*24V/4.44R = 129W or ~ 65Wrms
26V*26V/4.44R = 152W or ~ 76Wrms
So +3dB is double the power. To have +3dB signal compared to 24V you'd need PVCC = 33.8V. For the numbers above, 26V is 1.08 * 24V. Or in dB, you'll gain 0.67dB.
Would you hear that? 😀
(efficient amp)
gmarsh3118 at 11:00 volumecontrol with turned down Meanwell 24V(22V), set at 32dB gain, 4 ohm speaker is much louder than
Sanwu identical system volumecontrol at 16:00. (source is 2+Vrms cdplayer, something is off with the little chinese board here, way off, voltagedrop diode is ~0.25V??)
gmarsh3118 at 11:00 volumecontrol with turned down Meanwell 24V(22V), set at 32dB gain, 4 ohm speaker is much louder than
Sanwu identical system volumecontrol at 16:00. (source is 2+Vrms cdplayer, something is off with the little chinese board here, way off, voltagedrop diode is ~0.25V??)
Likely not discernible. But hey, I lose several dB here and there (e.g., baffle step) so adding back 2/3 dB can't hurt. And the difference between 78w and 56w is another 1.5dB.
"A billion here, a billion there, pretty soon, your talking about real money." Evertt Dirksen, US Senator.
I'm running portable outside at the beach so I'm of the "every bit I can extract" philosophy. In the end it's all dictated by running for long periods on two SLA-AGM batteries - starts at 26v and ends at 24v.
I don't understand this. What I meant is the reduction in SPL for higher mids/ highs off-axis. I'm running 300W out of a 12V SLA (9Ah) stepped up to 36V for roughly 7hours at "one louder up to eleven" on my bike. Bass compensates for that. 😀
Huh, I'm in urbia on my bike (cycle) and the local drivers have no conscience. So I couldn't enjoy the tunes even if they sounded great.
I'm on the beach with the box pointing at my crowd so designed as such. People adjacent and behind hear what frequencies they get as a byproduct. Lots of complements on the tunes from all around with no complaints (except one recurring a$%h#*l, year after year).
With the current YJ BB output filter (10uh, 0.68uf)...
If i were to use a 4ohm speaker (currently 8ohm), does the existing filter have to be tweaked ?
If i were to use a 4ohm speaker (currently 8ohm), does the existing filter have to be tweaked ?
I simulated a number of filters (using Online Circuit Simulator with SPICE ) and couldn't work out why people are using 680n at all.
The best results I got were with 220n, which worked for 4-16 ohms with only 1dB of variation at 20kHz. I tested this with real hardware and it sounded good too with both 8ohm and 4 ohm speakers.
See my coil winding thread for the schematic I used, which was basically from Trevor Marshalls page: Trevor Marshall - Class D Audio Amplifier Design - TDA7498 Output filters
Use decent polyproylene caps, panasonic do a nice range I'm rather liking.
I've had a look at :
(Sample)RLC Low-pass Filter Design Tool - Result -
*i'm assuming* this is the idea behind the output filter and speaker. With the 220nF (i used 10uH value with this), are the high frequencies (beyond 20k) sufficiently attenuated ?
If i'm reading things correctly... the frequency response doesnt seem as flat with a 1uF.
(Sample)RLC Low-pass Filter Design Tool - Result -
*i'm assuming* this is the idea behind the output filter and speaker. With the 220nF (i used 10uH value with this), are the high frequencies (beyond 20k) sufficiently attenuated ?
If i'm reading things correctly... the frequency response doesnt seem as flat with a 1uF.
Hello. Need help guys.
I want to set the TPA3116D2 Official Version Amp in the car. There was a question on the power unit, I tried different but a bad sound with distortion or poor transmission of sound at medium and high frequencies. Help please with a good car for the power supply. Tell me where you can buy it. It is advisable to aliexpress.
More TK2050 I have also wanted to try it in the car, but the noise from a very strong car. Help.
Maybe there are well-regulated power supplies would be interested.I want to get a decent sound in the car, at home all happy over.
I want to set the TPA3116D2 Official Version Amp in the car. There was a question on the power unit, I tried different but a bad sound with distortion or poor transmission of sound at medium and high frequencies. Help please with a good car for the power supply. Tell me where you can buy it. It is advisable to aliexpress.
More TK2050 I have also wanted to try it in the car, but the noise from a very strong car. Help.
Maybe there are well-regulated power supplies would be interested.I want to get a decent sound in the car, at home all happy over.
Check your car doesn't connect the -ve of the speakers together somewhere, use a multimeter to confirm that each speaker lead is independent. I had a car (old E30 BMW) where there was a common ground for all the speakers....
Filterfrequency with 220nF becomes 107kHz, so not enough attenuation.
For 4 ohm resistor, use 2 ohm/1uF/10uH to simulate single ended