I've designed and built several power amps from scratch but they were all well under 100W. I'm thinking of moving up to higher powers and am wondering if it's possible to economize a bit by using a power supply transformer that's rated somewhat below the continuous maximum power draw, and still be safe in practice.
For example, let's say I design a 300W Class AB amp that draws 500W at full power. Is it acceptable to use a transformer that can take a continuous 400W load?
Regarding the transformer rating, I mean something I design and build myself after considering copper and core losses, Irms/Idc, etc. and can actually handle a 400W DC load continuously without overheating.
For example, let's say I design a 300W Class AB amp that draws 500W at full power. Is it acceptable to use a transformer that can take a continuous 400W load?
Regarding the transformer rating, I mean something I design and build myself after considering copper and core losses, Irms/Idc, etc. and can actually handle a 400W DC load continuously without overheating.
Thanks for the reply. I should have mentioned that that the amplifier - which is hypothetical at present - would be a general-purpose tool, occasionally used for PA and music. It would never be used to blast out a continuous sine wave at full power, except perhaps for brief tests on rare occasions. Even that is highly unlikely.
By the same token, I take it that a 300W SMPS would be sufficient to power a 300W Class D with 85% efficiency. I'm leaving a wider safety margin here because perhaps SMPSes are less tolerant to overloads.
Is there any commonly used rule-of-thumb for estimating such PS requirements?
By the same token, I take it that a 300W SMPS would be sufficient to power a 300W Class D with 85% efficiency. I'm leaving a wider safety margin here because perhaps SMPSes are less tolerant to overloads.
Is there any commonly used rule-of-thumb for estimating such PS requirements?
For domestic duty you will find that a transformer of xVA will power a total maximum power of xWatts to 50% of xWatts
i.e. a 400VA transformer will give satisfactory performance that is also "safe" on domestic listening duty for a total maximum output power of 200W to 400W
For PA duty the transformer should probably be >2times the maximum total power output. i.e. that 400VA transformer should be good enough for upto 200W of maximum output.
The class D amplifiers are a different load because depending the music program and speakers, some part of the energy delivered to speakers (and not transformed in sound or heat) are recovered by the amplifier to the filtering capacitors. This is the reason (at least one) of power supply pumping. Some reference here.
As I see it, it is a question about if you run out of voltage or current to the amplifier.
Normally you have a fixed voltage from the power supply being connected to the amplifier. When you reach the voltage limit, a class AB amplifier starts "clipping" (the voltage ceiling is reached) as it is well known. Class D amplifiers may have a sort of soft-clipping function that causes less stress on the loudspeakers.
Limitation by voltage is the normal way.
If you have a supply voltage to the amplifier that allows the amplifier to draw more current from the power supply than the power supply can provide, the power supply voltage will drop. Like if you use a too low loudspeaker impedance. Then again, your amplifier will "clip" the music signal but in a less controlled manner. Short temporary current overloads from transients in the music can be compensated by large buffer capacitors on the power rail(s).
When you overload a power supply you may blow the fuse (not a major problem), may harm the rectifier bridge (worse) or thermally overload the transformer (even worse). Thus, if you intend to save on the transformer size then consider using a cheap digital thermometer for monitoring of the transformer temperature.
My limited experience with monolithic class D amplifiers is that a protection circuit in the amplifier is activated and causes a partly disruption of the amplifier. This may leave an unpredictable noise in the loudspeaker.
I have often used insufficient power supplies for initial test of amplifiers (typically a variable laboratory supply with the current limiter set at a low level: If there is a problem with the amplifier in initial test, I do not want a lot of energy available to destroy the amplifier). Just be prepared for that at a certain moment you reach the power supply current limit when the sound level is increased.
You specify power levels of 300W-500W. Without trying to sound like your mother, the worst damage may be to your hearing if used at home. For PA use (damaging the hearing of others) I will recommend you not to design for uncontrolled distortion because uncontrolled distortion is very unpleasant in front of a powerful PA loudspeaker and may result in the public leaving or at least complaining.
Normally you have a fixed voltage from the power supply being connected to the amplifier. When you reach the voltage limit, a class AB amplifier starts "clipping" (the voltage ceiling is reached) as it is well known. Class D amplifiers may have a sort of soft-clipping function that causes less stress on the loudspeakers.
Limitation by voltage is the normal way.
If you have a supply voltage to the amplifier that allows the amplifier to draw more current from the power supply than the power supply can provide, the power supply voltage will drop. Like if you use a too low loudspeaker impedance. Then again, your amplifier will "clip" the music signal but in a less controlled manner. Short temporary current overloads from transients in the music can be compensated by large buffer capacitors on the power rail(s).
When you overload a power supply you may blow the fuse (not a major problem), may harm the rectifier bridge (worse) or thermally overload the transformer (even worse). Thus, if you intend to save on the transformer size then consider using a cheap digital thermometer for monitoring of the transformer temperature.
My limited experience with monolithic class D amplifiers is that a protection circuit in the amplifier is activated and causes a partly disruption of the amplifier. This may leave an unpredictable noise in the loudspeaker.
I have often used insufficient power supplies for initial test of amplifiers (typically a variable laboratory supply with the current limiter set at a low level: If there is a problem with the amplifier in initial test, I do not want a lot of energy available to destroy the amplifier). Just be prepared for that at a certain moment you reach the power supply current limit when the sound level is increased.
You specify power levels of 300W-500W. Without trying to sound like your mother, the worst damage may be to your hearing if used at home. For PA use (damaging the hearing of others) I will recommend you not to design for uncontrolled distortion because uncontrolled distortion is very unpleasant in front of a powerful PA loudspeaker and may result in the public leaving or at least complaining.
One difference would be that a 50Hz Xformer will take a lot of short-term overload and "only" sag in voltage. a SMPS may shut down if overloaded for even a short time. All depends on the construction of the SMPS and whether it is designed for audio.
Hope it help
TroelsM
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.