Firs do I have to tell that Mooly have already told me how to construct a power supply for his amplifier. What I am trying is to learn and understand the PSU thingy🙂
I do not learn by just copying and not understanding what is going on.
Let me start by telling what I think I may have understood.
1) A power amplifier do have a predetermined gain-multiplier, a multiplication of the input signal, that are the result of the construction and is not adjustable.
2) If the power supply's Voltage is less then (input signal x gain) you have clipping.
3) headroom do let the amplifier avoid clipping if the input signal, suddenly gets higher, due to suddenly change in input signal.
My question is first, is that right?
Second input signal x gain must have an upper "real world" limit, a limit for how high an input signal can bee?
If we take Mooly's amplifier, since that is what I am building for the moment. It have a recommended input Voltage of 45V p/p but what would 100V do (assuming all component can handle it) Would that give headroom, general higher gain or just stupid?
Please if you want, do try to explain it so even an dummy can understand it🙂
I do not learn by just copying and not understanding what is going on.
Let me start by telling what I think I may have understood.
1) A power amplifier do have a predetermined gain-multiplier, a multiplication of the input signal, that are the result of the construction and is not adjustable.
2) If the power supply's Voltage is less then (input signal x gain) you have clipping.
3) headroom do let the amplifier avoid clipping if the input signal, suddenly gets higher, due to suddenly change in input signal.
My question is first, is that right?
Second input signal x gain must have an upper "real world" limit, a limit for how high an input signal can bee?
If we take Mooly's amplifier, since that is what I am building for the moment. It have a recommended input Voltage of 45V p/p but what would 100V do (assuming all component can handle it) Would that give headroom, general higher gain or just stupid?
Please if you want, do try to explain it so even an dummy can understand it🙂
Whether your English is self-taught you should stick to clearer in the engineering sense expressions. You should also learn and understand basic electricity principles like Ohm's and Kirchhoff's laws.
Which means not to use vague descriptions like input signal and gain, but rather input voltage and voltage gain. In the context of a power amp this is important.
"It have a recommended input Voltage of 45V p/p but what would 100V do (assuming all component can handle it) Would that give headroom, general higher gain or just stupid?"
I cannot make head or tails from the question above, so "just stupid" works for me.
Which means not to use vague descriptions like input signal and gain, but rather input voltage and voltage gain. In the context of a power amp this is important.
"It have a recommended input Voltage of 45V p/p but what would 100V do (assuming all component can handle it) Would that give headroom, general higher gain or just stupid?"
I cannot make head or tails from the question above, so "just stupid" works for me.
Thanks, I will do it better, I'll try again here.
I know Ohm's law but have not heard of Kirchhoff's laws, so I'll look at that!
To make it more clear and to ensure that what I write is more correct, do we start by saying that the signal source is a CD-player. This CD-Player do have an output , measured in Volts. Lets call that our "input Voltage" to the amplifier. The amplifier do also have mains Voltage, that are transformed to 45V p/p, lets call that "supply voltage"
Am I correct in understanding that:
1) A power amplifier do have a predetermined gain, a multiplication of the input Voltage, that are the result of the construction and is not adjustable.
2) If the supply's Voltage is less then (input Voltage x gain) you have clipping.
3) headroom do let the amplifier avoid clipping if the input Voltage, suddenly gets higher.
I am also asking the following:
If we take Mooly's amplifier, since that is what I am building for the moment. It have a recommended supply Voltage of 45V p/p but what would 60V do (assuming all component can handle it) Would that give headroom, general higher gain or just stupid?
Please if you want, do try to explain it so even an dummy can understand it🙂
I know Ohm's law but have not heard of Kirchhoff's laws, so I'll look at that!
To make it more clear and to ensure that what I write is more correct, do we start by saying that the signal source is a CD-player. This CD-Player do have an output , measured in Volts. Lets call that our "input Voltage" to the amplifier. The amplifier do also have mains Voltage, that are transformed to 45V p/p, lets call that "supply voltage"
Am I correct in understanding that:
1) A power amplifier do have a predetermined gain, a multiplication of the input Voltage, that are the result of the construction and is not adjustable.
2) If the supply's Voltage is less then (input Voltage x gain) you have clipping.
3) headroom do let the amplifier avoid clipping if the input Voltage, suddenly gets higher.
I am also asking the following:
If we take Mooly's amplifier, since that is what I am building for the moment. It have a recommended supply Voltage of 45V p/p but what would 60V do (assuming all component can handle it) Would that give headroom, general higher gain or just stupid?
Please if you want, do try to explain it so even an dummy can understand it🙂
1) Yes
2) Yes
3) No
If an amp is rated at 10W but it is capable of reaching 20W for a short period of time then you can think of those extra 10W as headroom. Which means the PS voltage has to provide for that headroom but the devices may not be able to dissipate the extra power.
Not the most important parameter imho. I'd rather have a 200W with no headroom at all than a 30W amp with great headroom, but the 30W amp will clearly be cheaper and easier to build.
2) Yes
3) No
If an amp is rated at 10W but it is capable of reaching 20W for a short period of time then you can think of those extra 10W as headroom. Which means the PS voltage has to provide for that headroom but the devices may not be able to dissipate the extra power.
Not the most important parameter imho. I'd rather have a 200W with no headroom at all than a 30W amp with great headroom, but the 30W amp will clearly be cheaper and easier to build.
A dc supply voltage cannot be peak to peak, it makes no sense. You are mixing the supply voltage with the voltage at the amp output.
With that correction, yes, raising the supply voltage of an amp will increase the headroom.
Yes you are right, did use that expression because an other person did use it about the rail to rail voltage, calculated to about 32 rms. But Mooly writes +45 at the top voltage input rail, zero in the middle rail and -45V in the lower rail.
Thanks for you correcting me, it means a lot to me, without getting corrected, do I newer learn! 🙂
Your explanation makes a lot of sense but it do also confuse me because, I did often here about headroom, that you can never have to much headroom and that being more important then wattage. 🙂
Thanks for you correcting me, it means a lot to me, without getting corrected, do I newer learn! 🙂
Your explanation makes a lot of sense but it do also confuse me because, I did often here about headroom, that you can never have to much headroom and that being more important then wattage. 🙂
I don't. Seems like a very interesting amp.
Surely there is a dedicated thread and Mooly is more than capable of explaining everything concerning his amp. And doing this much better than me.
Surely there is a dedicated thread and Mooly is more than capable of explaining everything concerning his amp. And doing this much better than me.
Yes that's properly right, but my question is also not about his amp but just what extra voltage do, if an amp has an upper gain limit, about headroom and yes you saw my original question🙂
Note that the clipping point is always less than the supply voltage as perfect output devices do not exist. So you have to subtract a few volts from the supply, depending on output devices and configuration.
For instance an amp with +/-45V rails might clip at +/-40V, allowing sine wave upto about 28V rms (80V p2p). If that amp has a gain of 28, then the input rms clipping point is 1V (or 2.8V p2p).
Headroom relates the clipping point to a nominal "average" level the system's is used at - the
more headroom the fewer transient spikes will clip. If your music is already compressed/limited (much is), you won't need a great deal of headroom as spikes are limited, if you have raw recordings with a high dynamic range, it matters more.
For instance an amp with +/-45V rails might clip at +/-40V, allowing sine wave upto about 28V rms (80V p2p). If that amp has a gain of 28, then the input rms clipping point is 1V (or 2.8V p2p).
Headroom relates the clipping point to a nominal "average" level the system's is used at - the
more headroom the fewer transient spikes will clip. If your music is already compressed/limited (much is), you won't need a great deal of headroom as spikes are limited, if you have raw recordings with a high dynamic range, it matters more.
Last edited:
For the same power amplifier and speaker power you need more headroom for higher impedance speakers.
i.e. 100 watts and 4 ohm speaker requires +/- 20 volts.
100 watts and 8 ohm speaker requires +/- 28 volts.
Its just uses power formula's and ohms law.
Ohms law V=IR. Power formulas W=(VV)/R .W=IIR. W=VI.
i.e. 100 watts and 4 ohm speaker requires +/- 20 volts.
100 watts and 8 ohm speaker requires +/- 28 volts.
Its just uses power formula's and ohms law.
Ohms law V=IR. Power formulas W=(VV)/R .W=IIR. W=VI.
Assuming your speakers operate with DC voltages your formula is correct. For AC-voltages definitely not.
Find the error!
0.707 of DC if using sine wave.
So 20 volts becomes 28v
and 28 volts becomes 39v.
My calcs were correct for a square wave ! worst case scenario.
If you turn the Volume up too high, any audio amplifier will distort.
If you are building a hi-fi system, where distortion is not wanted, if you hear distortion you will Turn It Down.
In general you can not know in advance "how many volts".
But you can listen to a lot of music with not much more than 1 Watt. "Louder" stuff may need 10 Watts. LOUD with inefficient speakers we may touch 100 Watts. How many volts is that? Why would we care? Simply ensure your source, through any booster, can hit the power amp input enough to reach distortion. Not counting a few "exotics", and church-organ systems, the most I have ever seen is 2.8Vrms.
Thank you all I'll reply to each of you at once.
Imagine an amplifier where the electronic is making a 28X gain, as one wrote earlier. If you say that max input Voltage is 2.8V rms, then could you say max possible is 78.4V rms or 110.8576V peak.
No matter how much you turn up the music, you would never exceed 111V?
I have some idea of having forgotten to reply to some of you, please say if I have not replyed to one of your comments. 🙂
That is very interesting but I do not understand how you do arrive to these values. I heard a ceo of a high end brand, saying that even 10X in headroom were great and even more, even better.
Great reply that are easy to understand but I have some sort of theory, that there have to be a maximum that are calculable.
Imagine an amplifier where the electronic is making a 28X gain, as one wrote earlier. If you say that max input Voltage is 2.8V rms, then could you say max possible is 78.4V rms or 110.8576V peak.
No matter how much you turn up the music, you would never exceed 111V?
I have some idea of having forgotten to reply to some of you, please say if I have not replyed to one of your comments. 🙂
> have to be a maximum that are calculable.
Who is paying for this amplifier? Your meter? Or your ears?
Are you asking how to avoid overload when all listeners are not listening? Why? Overload rarely harms amplifiers. Sometimes overload must be avoided for other reasons. An overloaded radio transmitter will interfere with other transmitters. Governments require transmitters to be monitored. Frequently this is done with automatic level controllers (limiters) backed by overload alarms to wake-up a sleeping or remote operating engineer.
Who is paying for this amplifier? Your meter? Or your ears?
Are you asking how to avoid overload when all listeners are not listening? Why? Overload rarely harms amplifiers. Sometimes overload must be avoided for other reasons. An overloaded radio transmitter will interfere with other transmitters. Governments require transmitters to be monitored. Frequently this is done with automatic level controllers (limiters) backed by overload alarms to wake-up a sleeping or remote operating engineer.