What I think I know so far: I = V/R. And also P = VI
Looking at the TDA7377 spec sheet, I see Vop = 18, Vs = 28, Vs(peak) = 50.
http://www.e-ele.net/DataSheet/TDA7377.pdf
But I will be using Vs = 10 to get 10W (2 x 10W into 4ohm speakers).
Using I = V/R, I get: 10 volts / 4 ohm = 2.5Amps
Using I = P/V, I get: 20W/10V = 2 Amps (20W = 10W x 2 speakers)
How best to calculate it using the chip's spec sheet as a reference?
Obviously I'm not going to be playing sine waves through it - is there a rule of thumb I should take into account?
Looking at the TDA7377 spec sheet, I see Vop = 18, Vs = 28, Vs(peak) = 50.
http://www.e-ele.net/DataSheet/TDA7377.pdf
But I will be using Vs = 10 to get 10W (2 x 10W into 4ohm speakers).
Using I = V/R, I get: 10 volts / 4 ohm = 2.5Amps
Using I = P/V, I get: 20W/10V = 2 Amps (20W = 10W x 2 speakers)
How best to calculate it using the chip's spec sheet as a reference?
Obviously I'm not going to be playing sine waves through it - is there a rule of thumb I should take into account?
Last edited:
28 V and 50 V are voltages it can survive permanently and for 50 ms, respectively, but without any guarantee that it will work. If it has to work, you have to stay below 18 V. Its intended application is car radio, and there can sometimes be nasty overvoltages on the 12 V supply in a car, for example when the battery gets disconnected while the alternator is running or when someone tries to jump start the car from a 24 V lorry battery. It is nice when the car radio survives such conditions, but it is no big deal if it temporarily stops working during these conditions.
Anyway, when you have a power P dissipated in a resistor R, the current follows from:
P = V I = I R I = I^2 R, hence
I = sqrt(P/R)
That means that the RMS current through a 4 ohm resistor that on average dissipates 10 W is sqrt(10 W/4 ohm) ~= 1.5811 A.
When the waveforms are sinusoidal, the peak value is sqrt(2) times the RMS value and the average absolute value is 2 sqrt(2)/pi times the RMS value.
Hence, the peak current is 2.2361 A per loudspeaker and the average absolute current 1.4235 A per loudspeaker.
In a bridge tied load configuration, the current through the loudspeaker flows through the amplifier's supply at all times, so with two loudspeakers, the theoretical peak supply current is 4.4721 A and the average supply current when playing loud sine waves is 2.8471 A. In practice the TDA7377 requires some current of its own and the loudspeaker can have impedance minima to below 4 ohm, so the currents can be somewhat higher.
Anyway, when you have a power P dissipated in a resistor R, the current follows from:
P = V I = I R I = I^2 R, hence
I = sqrt(P/R)
That means that the RMS current through a 4 ohm resistor that on average dissipates 10 W is sqrt(10 W/4 ohm) ~= 1.5811 A.
When the waveforms are sinusoidal, the peak value is sqrt(2) times the RMS value and the average absolute value is 2 sqrt(2)/pi times the RMS value.
Hence, the peak current is 2.2361 A per loudspeaker and the average absolute current 1.4235 A per loudspeaker.
In a bridge tied load configuration, the current through the loudspeaker flows through the amplifier's supply at all times, so with two loudspeakers, the theoretical peak supply current is 4.4721 A and the average supply current when playing loud sine waves is 2.8471 A. In practice the TDA7377 requires some current of its own and the loudspeaker can have impedance minima to below 4 ohm, so the currents can be somewhat higher.
Last edited:
That 10V is the peak voltage, so indeed you get 2.5A peak into 4 ohms.
But power in a speaker load is normally calculated in Wrms assuming a sine wave, where the sine wave Vrms is the Vpeak/SqRtof2. In this case, the 10V peak represents about 7V rms so your power into the speaker would be just a trifle over 12W (49/4).
The peak current into the speaker is 2.5A as you calculated, and the rms current about 2.5*0.7=~1.75A.
But music has a very low average level so you won't need 2.5A for each channel; it may peak there but the average will be much lower.
Note also that with Vs=10V (assuming you mean supply voltage) you won't get 10V on your speaker, more like 8V max. Then your rms power in the speaker comes to about (8*0.7)^2/4=~7.8W
Jan
OK that's very helpful, thank you. It seems that many chip spec sheets state a value for I peak, normally for sinusoidal wave of a certain frequency.
Does this mean that if my calculated I peak or RMS for a my of speakers & voltage exceeds this, then I should look at adjusting the speaker R value (different speakers) or voltage in order to keep things safely within the stated max I peak value?
Otherwise I'm guessing the amp will not operate correctly and overheat/perform badly?
Thanks for helping.
Cheers,
Chris.
Does this mean that if my calculated I peak or RMS for a my of speakers & voltage exceeds this, then I should look at adjusting the speaker R value (different speakers) or voltage in order to keep things safely within the stated max I peak value?
Otherwise I'm guessing the amp will not operate correctly and overheat/perform badly?
Thanks for helping.
Cheers,
Chris.
Yes, although the datasheet usually already specifies for what combinations of nominal loudspeaker impedance and supply voltage the chip is meant. Most chip amplifiers are very well protected against excessive loads or poor cooling, so usually the worst that can happen is gross distortion due to some protection circuit kicking in.