Yes it's for the next generation of deaf smartphones users !
600W, how much degrees on the chip ? Maybe worth it if using it at the same output power than the planed one ?
600W, how much degrees on the chip ? Maybe worth it if using it at the same output power than the planed one ?
Needs a proper heatsink. 😀 Would guess my big board would do it - peak power, not continous. (With 2oz. copper)
Nice work with laser cutter dr Mord. I think 3D prints can do similar job and have noticed natural flex in many of then shapes used for throwaway supports.
For 600w chip assuming 90% efficiency that is 60watts that needs to be pulled off. The only realistic way to do that over a 5mm square is a gold or diamond tipped heat pipe.
For 600w chip assuming 90% efficiency that is 60watts that needs to be pulled off. The only realistic way to do that over a 5mm square is a gold or diamond tipped heat pipe.
Very happy with this new chip
I like this new chip very much because it can use a high supply voltage. I do think 600W is mostly marketing, based in 10% THD and the highest supply voltage they can safely use.
In reality, when using an unregulated power supply, the maximum voltage is when idle with the mains voltage 10% high. For the TPA3251d2, this needs to be below 40V. For normal mains, the supply voltage at full power will be below 34V, then it struggles to get even 50W unclipped per channel in 8 ohms.
Of course we can use a regulated supply, but that will require a regulated switching power supply with good dynamic performance. Those are expensive...
With the new chip, I can give it 50V worst case from a standard transformer, unloaded, then the power will still reach almost 100W per channel in 8 ohms at normal mains voltage. With real music, assuming 10dB crest factor, the heat dissipation will be a reasonable 5W, stereo, 8 ohm, 90% efficiency, assuming idle loss is 3W or 7W in 4 Ohm.
Not bad at all...
Rob
I like this new chip very much because it can use a high supply voltage. I do think 600W is mostly marketing, based in 10% THD and the highest supply voltage they can safely use.
In reality, when using an unregulated power supply, the maximum voltage is when idle with the mains voltage 10% high. For the TPA3251d2, this needs to be below 40V. For normal mains, the supply voltage at full power will be below 34V, then it struggles to get even 50W unclipped per channel in 8 ohms.
Of course we can use a regulated supply, but that will require a regulated switching power supply with good dynamic performance. Those are expensive...
With the new chip, I can give it 50V worst case from a standard transformer, unloaded, then the power will still reach almost 100W per channel in 8 ohms at normal mains voltage. With real music, assuming 10dB crest factor, the heat dissipation will be a reasonable 5W, stereo, 8 ohm, 90% efficiency, assuming idle loss is 3W or 7W in 4 Ohm.
Not bad at all...
Rob
Hi,
Also received yesterday a mail from TI about the introduction of TPA3251D2 as if it was a new product.
However, all the links point to the chip we already know.
A link to the new version if there is a new one ?
TI has a preview link to this chip which I found interesting, http://www.ti.com/lit/ds/symlink/tpa3255.pdf
Think there will be more products announced soon.
Seems lots of people in many companies are working to make class-AB obsolete. 🙂
Seems lots of people in many companies are working to make class-AB obsolete. 🙂
I searched for the data sheet and found:
http://www.ti.com/lit/ds/symlink/tpa3251d2.pdf
Where its says:
ANALOG INPUTS
RIN
Input resistance
24
kΩ
//
http://www.ti.com/lit/ds/symlink/tpa3251d2.pdf
Where its says:
ANALOG INPUTS
RIN
Input resistance
24
kΩ
//
I am surprised your estimates of the max power is so low. I was able to get 60w into 5ohms out of a 3118 in PBTL with a 24v supply. I would imagine an advanced chip with 50v supply would do much better for 8ohm.
This is for audio, where we only need high power to avoid clipping on peaks, so who cares about continuous power?
I sure don't; with a very conservative crest factor of 6 dB, average power is only 1/4 of peak.
If the voltage was really 50V at full load, then the output power would be higher of course, but with an analogue supply that gives 50V unloaded at the highest mains voltage (+10%), the loaded output at normal mains voltage will be much lower, more like 43V.
Seems there are some real advantages using regulated SMPS with these chips... Time to design one...
Rob