I don't know your specific objection to what I said, but I don't like power outputs quoted against distortion. The figure we really want is for unclipped power and the distortion figure above that is largely a function of how much of the sine wave has been sliced off.
I thought we were trying to find out if we could get 15W using a 12V battery.
Let's do it the other way around. Power = V^2/R and to get 15W from 4 ohms the RMS voltage will be sqrt(60) = 7.746V. Times sqrt(2) for the peaks = 10.95V, or 22V peak to peak. With a bridged amp we have nominally 24V available, so I reckon that is pretty much doable. With a standard output stage we will have to rely on the battery charging to higher than nominal, but for something modern with a charge pump boosting the driver voltages I would have thought it's eminently possible. Whether the losses are quite this low at full current I haven't looked, but I'd be surprised if there weren't something close among TI's offerings.
Then, of course, there is always the possibility of boosting the supply voltage for the whole thing and you can have any power you want. Hence 500W automotive systems ...
I don't think he needs to.
http://www.nxp.com/documents/data_sheet/TDA1562Q_ST_SD.pdf
Go to page 13, third line down.
I'm under the assumption that the lift supply in the TDA1562 can only raise the rails temporarily for transients requiring more power. If a continuous waveform is applied, the lift supply is not able to sustain the higher voltage. I could be wrong.
My method to determine amplifier power is to measure the power at just before clipping using a continuous sine wave into a non inductive load resistance. If the amplifier has more than one channel, all channels must be driven and loaded. If the lift supply cannot maintain the higher voltage, it cannot maintain the higher output in a proper amplifier power measurement.
My method to determine amplifier power is to measure the power at just before clipping using a continuous sine wave into a non inductive load resistance. If the amplifier has more than one channel, all channels must be driven and loaded. If the lift supply cannot maintain the higher voltage, it cannot maintain the higher output in a proper amplifier power measurement.
Last edited:
surely will not qork with a sqare wave, and the lift capacitrs must be high quality.
most probably the limiting factor would be the RC time.
on the otherhand with audio signals i do not see why should it not work properly.
someone has put quite some effort to pack a full class H amp in a chip.
i would think this effort did not go to waste.
Supposedly the lift capacitors can charge while the output signal is achieveable with the normal supply rails and the caps only kick when the rail is not high enough for the output signal.
so supposedly, at least what i think, the caps have time to charge. question is if the powersupply rail will handle the extra load
most probably the limiting factor would be the RC time.
on the otherhand with audio signals i do not see why should it not work properly.
someone has put quite some effort to pack a full class H amp in a chip.
i would think this effort did not go to waste.
Supposedly the lift capacitors can charge while the output signal is achieveable with the normal supply rails and the caps only kick when the rail is not high enough for the output signal.
so supposedly, at least what i think, the caps have time to charge. question is if the powersupply rail will handle the extra load
I don't see any reason why it shouldn't work continuously at that power and would imagine that, like most of us (and JohnR66), they define power output as continuous at a given frequency. The size of the capacitor at 4700uF would also seem to point to this. (Besides I don't think you would want to try to sell an amp that collapsed after a few seconds of being pushed hard.)
Nor do I see any reason for it not to work with a square wave since all that is happening is that the rails move up in voltage.
It's just a question of efficiency and keeping the chip cool or the heatsink size down (in this case down to something practically achievable). The price you pay is an extra pair of caps. It can otherwise be thought of as just an amp with an on-board DC-DC converter (which will be perfectly able to deliver the current needs)
Nor do I see any reason for it not to work with a square wave since all that is happening is that the rails move up in voltage.
It's just a question of efficiency and keeping the chip cool or the heatsink size down (in this case down to something practically achievable). The price you pay is an extra pair of caps. It can otherwise be thought of as just an amp with an on-board DC-DC converter (which will be perfectly able to deliver the current needs)
at first my plan it was build a amplifier with tda7297 but with a 12-13 volt power supply i can't have more than 10 watt per channel.
so i decide to use lm 1876 that with a 13/0/-13 power supply give about 10 watt per channel and quality of sound is better.
now i need inverting my 12v power supply with about 400-500 mA in 13 and -13 line.
You can't. And you will need more than the 400-500mA that you mention also. The choices are either that you bridge two amplifiers, that you buy a bridged chip like the TA2020 or that you buy a supply boosted chip like the TDA1562 which will give you 40W.
ookay.
why is it so important to have it run on a battery?
is it going to be used in a car?
Switchmode Power Supply For Car Audio
this is what you need.
but keep in mind the foot note:
"That it does work if built as described is certain, that you will be able to achieve the same results is not. If you do not have (or at least have access to) an oscilloscope - don't even think about trying to make the supply, as it will not be possible to ensure that the duty cycle of the controller is exactly 50%, or that there is no severe overshoot or ringing at the output."
a nother option is to salvage a car amp with a blown power amp, but working powersupply.
then you can "simply" mesure its output voltage and find a suitable chipamp it can power.
building an SMPS requires .. practice. Chipamps do not like to be powered with rails that can overshoot the absolute maximum supply voltage. they simply die without any warning.
why is it so important to have it run on a battery?
is it going to be used in a car?
Switchmode Power Supply For Car Audio
this is what you need.
but keep in mind the foot note:
"That it does work if built as described is certain, that you will be able to achieve the same results is not. If you do not have (or at least have access to) an oscilloscope - don't even think about trying to make the supply, as it will not be possible to ensure that the duty cycle of the controller is exactly 50%, or that there is no severe overshoot or ringing at the output."
a nother option is to salvage a car amp with a blown power amp, but working powersupply.
then you can "simply" mesure its output voltage and find a suitable chipamp it can power.
building an SMPS requires .. practice. Chipamps do not like to be powered with rails that can overshoot the absolute maximum supply voltage. they simply die without any warning.
Well, with around 700mV signal (or very little less, sine wave, 30Hz) I had constant 13V output and around 6A power consumption, it was very hot at that power tho.
Thanks for confirming your measurements.
The 13Vac is equivalent to 36.77Vpp
12Vdc and 6Adc input is equivalent ot 72W of input power giving an efficiency at full power into 4r0 of ~57%. Certainly plausible.
What is does confirm is that the amplifier is not operating from a single polarity 12Vdc source. You can't get 36Vpp from 12V supply.
The 13Vac is equivalent to 36.77Vpp
12Vdc and 6Adc input is equivalent ot 72W of input power giving an efficiency at full power into 4r0 of ~57%. Certainly plausible.
What is does confirm is that the amplifier is not operating from a single polarity 12Vdc source. You can't get 36Vpp from 12V supply.
would someone care to explain why would that be so impossible?
the chip if i remember is bridged internaly. the "lift" caps can provide 2x supply voltage when needed. (for the "top" part of the sine wave)
Dear Andrew, they use a 12V battery but it's not just "a 12V supply " by any means.
The trick is quite clever.
The amp is Class H (or F, don't remember which letter applies here) and for low level signals it is fed straight (probably through a Schottky diode plus a MOS switch or similar) from the 12V supply.
Once output rail approaches 12V, it's switched to the + side of a capacitor, charged to 12V, whose - side gets connected to +12V , so for a brief instant the amplifier is fed from 12+12V=24V.
Since the amp is bridged, it's equivalent to a single amplifier fed from +/-24V with a theorical capability of 48V PP.
We are quite far from that, even if fed from a stiff, high current capability , ripple less car battery, because there are many junctions in the way.
But the idea is real bright.
Problem is it requires a car battery (the intended application) or at least a good, fresh, well charged lead gel battery quite close to the amplifier board or all that switching and pumping is heard as a horrible rhythmic buzzing noise, which some describe.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.