choosing transformer VA rating

[Tonge]

Hey.

I have a question about how to choose the VA rating of a transformer, for a given power amplifier, and i am hoping that some of you would be kind enough to help me

I am pretty sure i already know the answer to this, but i am a bit confused, and i just want to check if i really am right.

Let's assume i have build an amplifier that runs on +/- 35V supply rails and puts out +/- 24,7V RMS, to the connected load.

In this case, a 4R speaker is connected as load, and for simplicity, let's say this speaker is purely resistive (it is not, i know).
At max. output this would result in the speaker drawing: 24,7V/4R = ~6,2A RMS, or 24,7V * 6,2A = ~150W, from each of the power supply rails, through the amplifier of course.

Therefore, if i want to run the amp at full power continously, i would need to use a transformer with a VA-rating of at least 300VA (150VA each secondary winding), right?

Since i didn't take any losses into consideration, this VA-rating would, of course, have to be higher in a real world application, but this is just to be sure i got it right.

Please correct me if im wrong.

Thanks in advance.

[nightanole]

um did you account for the %50 efficiency or what ever is for your amp? If its class ab chip amp its going to need 300 watts per channel.

[metalsculptor]

Originally Posted by Tonge

Let's assume i have build an amplifier that runs on +/- 35V supply rails and puts out +/- 24,7V RMS, to the connected load.

In this case, a 4R speaker is connected as load, and for simplicity, let's say this speaker is purely resistive (it is not, i know).
At max. output this would result in the speaker drawing: 24,7V/4R = ~6,2A RMS, or 24,7V * 6,2A = ~150W, from each of the power supply rails, through the amplifier of course.

Good attempt, few problems there is usually a voltage drop across the output stage due to drive requirements, 4V is not unusual so there will be only 31V peak available then the supply will sag 5 to 10% under load.
The power draw is 50% + and 50 % - meaning that each rail only sees load 50 % of the time making the per rail load half what was calculated.

Then there is amplifier efficiency, but if you work in current rather than voltage and add 100 mA or so for bias then the result will be close.

For a sine wave into a 4 ohm resistive load with 10% power supply sag at full load and 4V drop in the output devices at the onset of clipping

25*1.414*.9 = 31.8V under load - 4V = 27.8V peak/ 1.414 = 19.7V RMS
/4ohm = 4.9 A RMS + 100 mA bias - 5 A * 35V unloaded supply voltage = 175W.

The unloaded DC rail voltage is used in the power calculation because that is the voltage before losses.
Also I only calculated assuming one supply rail because the load is a 50/50 on each rail so effectively 100% one one rail

This approach is simplistic but close enough for most use. If this is an application with 100% duty then transformer utilisation factor needs to be factored in, for a single phase capacitor input filter with bridge rectifier it is
somewhere between 1.5 and 1.8 depending on the transformer leakage inductance and the filter capacitor size
300VA would do this easy.

FWIW music is unlikely to present a 100% duty 175W load to the power supply.

[metalsculptor]

Forgot a few things like the voltage drop in the bridge rectifier (roughly 0.8V per rail) and the wattage shouldn't be calculated on the unloaded voltage, rather the loaded voltage because the transformer is rated for it's loaded voltage or it should be
Also a 25V 0 25V transformer might produce 35V loaded because the maker has accounted for the sag under load and raised the no load voltage accordingly

Hope that makes sense.

[Mooly]

Have a look at this,
The Signal Transfer Company: Power Output

[metalsculptor]

The example the link is not representative of a typical audio amplifier, to achieve saturation the base of the output transistor needs to be driven above the output supply rail voltage which is achievable with boosted driver voltage but that is not what the OP plans on doing. The other reason to keep the output stages out of saturation is that the gain falls off rapidly near saturation.

Sketch the VAS and driver stages of a typical amplifier with all the relevant voltage drops included and you will see what I mean.