[EE] Low noise attenuator
Locked
 
|
Hi, I am curious about a general topic which mostly interests me in the
area of sensitive audio signals. What options do I have for attenuating a copper line signal while introducing a very low amount of noise? For example, if I grounded L+R channels using independent resistors would that work? OR do I need a proper high-pass or low-pass filter design? But then what about capacitors/inductors or series resistors introducing thermal noise? Open to discussion on types of noise as well :) Ryan -- http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist |
|
|
A fundamental property of absorbative attenuators (i.e. those which
dissipate the power removed from the signal) is that they introduce (1-a)kTB thermal noise in watts, where a is the attenuation factor, k is Boltzman's constant, T is kelvin temperature of the attenuator, and B is the downstream circuit bandwidth in Hz. So, for example, a 10dB attenuator (a=0.1), introduces 90% of a full equivalent of thermal noise. It is possible to attenuate reflectively (i.e., produce an impedance mismatch) and that can often circumvent this limitation although it depends on the source characteristics. There is a common practice in low noise amplifiers call "noise matching" where instead of matching the impedance of the source to that of the amplifier input, a purposeful mismatch is introduced which reduces the contribution of noise due to the real part of the amplifier's input impedance. I think the fundamental principle is the same. Sean On Mon, Feb 8, 2021 at 5:00 PM Ryan O'Connor <[hidden email]> wrote: > Hi, I am curious about a general topic which mostly interests me in the > area of sensitive audio signals. > > What options do I have for attenuating a copper line signal while > introducing a very low amount of noise? For example, if I grounded L+R > channels using independent resistors would that work? OR do I need a proper > high-pass or low-pass filter design? But then what about > capacitors/inductors or series resistors introducing thermal noise? > > Open to discussion on types of noise as well :) > > Ryan > -- > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist |
Re: [EE] Low noise attenuator
|
|
Use capacitive dividers.
On Mon, Feb 8, 2021, 4:20 PM Sean Breheny <[hidden email]> wrote: > A fundamental property of absorbative attenuators (i.e. those which > dissipate the power removed from the signal) is that they introduce > (1-a)kTB thermal noise in watts, where a is the attenuation factor, k is > Boltzman's constant, T is kelvin temperature of the attenuator, and B is > the downstream circuit bandwidth in Hz. So, for example, a 10dB attenuator > (a=0.1), introduces 90% of a full equivalent of thermal noise. > > It is possible to attenuate reflectively (i.e., produce an impedance > mismatch) and that can often circumvent this limitation although it depends > on the source characteristics. There is a common practice in low noise > amplifiers call "noise matching" where instead of matching the impedance of > the source to that of the amplifier input, a purposeful mismatch is > introduced which reduces the contribution of noise due to the real part of > the amplifier's input impedance. I think the fundamental principle is the > same. > > Sean > > > > On Mon, Feb 8, 2021 at 5:00 PM Ryan O'Connor <[hidden email]> wrote: > > > Hi, I am curious about a general topic which mostly interests me in the > > area of sensitive audio signals. > > > > What options do I have for attenuating a copper line signal while > > introducing a very low amount of noise? For example, if I grounded L+R > > channels using independent resistors would that work? OR do I need a > proper > > high-pass or low-pass filter design? But then what about > > capacitors/inductors or series resistors introducing thermal noise? > > > > Open to discussion on types of noise as well :) > > > > Ryan > > -- > > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > > View/change your membership options at > > http://mailman.mit.edu/mailman/listinfo/piclist > > > -- > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist |
|
|
David do you have any theory or practice which can explain this? I would be
interested to read more about why this helps. I'm trying to read up on Sean's topics and I'm finding it pretty complex. I have more questions than before, such as what is the relationship between impedence matching to reduce noise and impedence matching to gain maximum power? Do they happen to co-inside, or how much out of phase are they (literally can't find anything on the net about this) Cheers Ryan On Tue, 9 Feb 2021 at 13:57, David VanHorn <[hidden email]> wrote: > Use capacitive dividers. > > On Mon, Feb 8, 2021, 4:20 PM Sean Breheny <[hidden email]> wrote: > > > A fundamental property of absorbative attenuators (i.e. those which > > dissipate the power removed from the signal) is that they introduce > > (1-a)kTB thermal noise in watts, where a is the attenuation factor, k is > > Boltzman's constant, T is kelvin temperature of the attenuator, and B is > > the downstream circuit bandwidth in Hz. So, for example, a 10dB > attenuator > > (a=0.1), introduces 90% of a full equivalent of thermal noise. > > > > It is possible to attenuate reflectively (i.e., produce an impedance > > mismatch) and that can often circumvent this limitation although it > depends > > on the source characteristics. There is a common practice in low noise > > amplifiers call "noise matching" where instead of matching the impedance > of > > the source to that of the amplifier input, a purposeful mismatch is > > introduced which reduces the contribution of noise due to the real part > of > > the amplifier's input impedance. I think the fundamental principle is the > > same. > > > > Sean > > > > > > > > On Mon, Feb 8, 2021 at 5:00 PM Ryan O'Connor <[hidden email]> wrote: > > > > > Hi, I am curious about a general topic which mostly interests me in the > > > area of sensitive audio signals. > > > > > > What options do I have for attenuating a copper line signal while > > > introducing a very low amount of noise? For example, if I grounded L+R > > > channels using independent resistors would that work? OR do I need a > > proper > > > high-pass or low-pass filter design? But then what about > > > capacitors/inductors or series resistors introducing thermal noise? > > > > > > Open to discussion on types of noise as well :) > > > > > > Ryan > > > -- > > > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > > > View/change your membership options at > > > http://mailman.mit.edu/mailman/listinfo/piclist > > > > > -- > > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > > View/change your membership options at > > http://mailman.mit.edu/mailman/listinfo/piclist > > > -- > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist |
Re: [EE] Low noise attenuator
|
|
Resistors have thermal noise, capacitors have essentially no resistance
On Mon, Feb 8, 2021, 6:14 PM Ryan O'Connor <[hidden email]> wrote: > David do you have any theory or practice which can explain this? I would be > interested to read more about why this helps. I'm trying to read up on > Sean's topics and I'm finding it pretty complex. I have more questions than > before, such as what is the relationship between impedence matching to > reduce noise and impedence matching to gain maximum power? Do they happen > to co-inside, or how much out of phase are they (literally can't find > anything on the net about this) > > Cheers > Ryan > > On Tue, 9 Feb 2021 at 13:57, David VanHorn <[hidden email]> wrote: > > > Use capacitive dividers. > > > > On Mon, Feb 8, 2021, 4:20 PM Sean Breheny <[hidden email]> wrote: > > > > > A fundamental property of absorbative attenuators (i.e. those which > > > dissipate the power removed from the signal) is that they introduce > > > (1-a)kTB thermal noise in watts, where a is the attenuation factor, k > is > > > Boltzman's constant, T is kelvin temperature of the attenuator, and B > is > > > the downstream circuit bandwidth in Hz. So, for example, a 10dB > > attenuator > > > (a=0.1), introduces 90% of a full equivalent of thermal noise. > > > > > > It is possible to attenuate reflectively (i.e., produce an impedance > > > mismatch) and that can often circumvent this limitation although it > > depends > > > on the source characteristics. There is a common practice in low noise > > > amplifiers call "noise matching" where instead of matching the > impedance > > of > > > the source to that of the amplifier input, a purposeful mismatch is > > > introduced which reduces the contribution of noise due to the real part > > of > > > the amplifier's input impedance. I think the fundamental principle is > the > > > same. > > > > > > Sean > > > > > > > > > > > > On Mon, Feb 8, 2021 at 5:00 PM Ryan O'Connor <[hidden email]> > wrote: > > > > > > > Hi, I am curious about a general topic which mostly interests me in > the > > > > area of sensitive audio signals. > > > > > > > > What options do I have for attenuating a copper line signal while > > > > introducing a very low amount of noise? For example, if I grounded > L+R > > > > channels using independent resistors would that work? OR do I need a > > > proper > > > > high-pass or low-pass filter design? But then what about > > > > capacitors/inductors or series resistors introducing thermal noise? > > > > > > > > Open to discussion on types of noise as well :) > > > > > > > > Ryan > > > > -- > > > > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > > > > View/change your membership options at > > > > http://mailman.mit.edu/mailman/listinfo/piclist > > > > > > > -- > > > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > > > View/change your membership options at > > > http://mailman.mit.edu/mailman/listinfo/piclist > > > > > -- > > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > > View/change your membership options at > > http://mailman.mit.edu/mailman/listinfo/piclist > > > -- > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist |
|
|
Well, they have low resistance. But they have reactance?
On Tue, 9 Feb 2021 at 16:12, David VanHorn <[hidden email]> wrote: > Resistors have thermal noise, capacitors have essentially no resistance > > On Mon, Feb 8, 2021, 6:14 PM Ryan O'Connor <[hidden email]> wrote: > > > David do you have any theory or practice which can explain this? I would > be > > interested to read more about why this helps. I'm trying to read up on > > Sean's topics and I'm finding it pretty complex. I have more questions > than > > before, such as what is the relationship between impedence matching to > > reduce noise and impedence matching to gain maximum power? Do they happen > > to co-inside, or how much out of phase are they (literally can't find > > anything on the net about this) > > > > Cheers > > Ryan > > > > On Tue, 9 Feb 2021 at 13:57, David VanHorn <[hidden email]> wrote: > > > > > Use capacitive dividers. > > > > > > On Mon, Feb 8, 2021, 4:20 PM Sean Breheny <[hidden email]> wrote: > > > > > > > A fundamental property of absorbative attenuators (i.e. those which > > > > dissipate the power removed from the signal) is that they introduce > > > > (1-a)kTB thermal noise in watts, where a is the attenuation factor, k > > is > > > > Boltzman's constant, T is kelvin temperature of the attenuator, and B > > is > > > > the downstream circuit bandwidth in Hz. So, for example, a 10dB > > > attenuator > > > > (a=0.1), introduces 90% of a full equivalent of thermal noise. > > > > > > > > It is possible to attenuate reflectively (i.e., produce an impedance > > > > mismatch) and that can often circumvent this limitation although it > > > depends > > > > on the source characteristics. There is a common practice in low > noise > > > > amplifiers call "noise matching" where instead of matching the > > impedance > > > of > > > > the source to that of the amplifier input, a purposeful mismatch is > > > > introduced which reduces the contribution of noise due to the real > part > > > of > > > > the amplifier's input impedance. I think the fundamental principle is > > the > > > > same. > > > > > > > > Sean > > > > > > > > > > > > > > > > On Mon, Feb 8, 2021 at 5:00 PM Ryan O'Connor <[hidden email]> > > wrote: > > > > > > > > > Hi, I am curious about a general topic which mostly interests me in > > the > > > > > area of sensitive audio signals. > > > > > > > > > > What options do I have for attenuating a copper line signal while > > > > > introducing a very low amount of noise? For example, if I grounded > > L+R > > > > > channels using independent resistors would that work? OR do I need > a > > > > proper > > > > > high-pass or low-pass filter design? But then what about > > > > > capacitors/inductors or series resistors introducing thermal noise? > > > > > > > > > > Open to discussion on types of noise as well :) > > > > > > > > > > Ryan > > > > > -- > > > > > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > > > > > View/change your membership options at > > > > > http://mailman.mit.edu/mailman/listinfo/piclist > > > > > > > > > -- > > > > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > > > > View/change your membership options at > > > > http://mailman.mit.edu/mailman/listinfo/piclist > > > > > > > -- > > > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > > > View/change your membership options at > > > http://mailman.mit.edu/mailman/listinfo/piclist > > > > > -- > > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > > View/change your membership options at > > http://mailman.mit.edu/mailman/listinfo/piclist > > > -- > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist |
Re: [EE] Low noise attenuator
|
|
Yes. And you can use their reactance as a divider. Frequency response
will suffer, compensate on the input or the output. On Mon, Feb 8, 2021, 8:16 PM Ryan O'Connor <[hidden email]> wrote: > Well, they have low resistance. But they have reactance? > > On Tue, 9 Feb 2021 at 16:12, David VanHorn <[hidden email]> wrote: > > > Resistors have thermal noise, capacitors have essentially no resistance > > > > On Mon, Feb 8, 2021, 6:14 PM Ryan O'Connor <[hidden email]> wrote: > > > > > David do you have any theory or practice which can explain this? I > would > > be > > > interested to read more about why this helps. I'm trying to read up on > > > Sean's topics and I'm finding it pretty complex. I have more questions > > than > > > before, such as what is the relationship between impedence matching to > > > reduce noise and impedence matching to gain maximum power? Do they > happen > > > to co-inside, or how much out of phase are they (literally can't find > > > anything on the net about this) > > > > > > Cheers > > > Ryan > > > > > > On Tue, 9 Feb 2021 at 13:57, David VanHorn <[hidden email]> > wrote: > > > > > > > Use capacitive dividers. > > > > > > > > On Mon, Feb 8, 2021, 4:20 PM Sean Breheny <[hidden email]> wrote: > > > > > > > > > A fundamental property of absorbative attenuators (i.e. those which > > > > > dissipate the power removed from the signal) is that they introduce > > > > > (1-a)kTB thermal noise in watts, where a is the attenuation > factor, k > > > is > > > > > Boltzman's constant, T is kelvin temperature of the attenuator, > and B > > > is > > > > > the downstream circuit bandwidth in Hz. So, for example, a 10dB > > > > attenuator > > > > > (a=0.1), introduces 90% of a full equivalent of thermal noise. > > > > > > > > > > It is possible to attenuate reflectively (i.e., produce an > impedance > > > > > mismatch) and that can often circumvent this limitation although it > > > > depends > > > > > on the source characteristics. There is a common practice in low > > noise > > > > > amplifiers call "noise matching" where instead of matching the > > > impedance > > > > of > > > > > the source to that of the amplifier input, a purposeful mismatch is > > > > > introduced which reduces the contribution of noise due to the real > > part > > > > of > > > > > the amplifier's input impedance. I think the fundamental principle > is > > > the > > > > > same. > > > > > > > > > > Sean > > > > > > > > > > > > > > > > > > > > On Mon, Feb 8, 2021 at 5:00 PM Ryan O'Connor <[hidden email]> > > > wrote: > > > > > > > > > > > Hi, I am curious about a general topic which mostly interests me > in > > > the > > > > > > area of sensitive audio signals. > > > > > > > > > > > > What options do I have for attenuating a copper line signal while > > > > > > introducing a very low amount of noise? For example, if I > grounded > > > L+R > > > > > > channels using independent resistors would that work? OR do I > need > > a > > > > > proper > > > > > > high-pass or low-pass filter design? But then what about > > > > > > capacitors/inductors or series resistors introducing thermal > noise? > > > > > > > > > > > > Open to discussion on types of noise as well :) > > > > > > > > > > > > Ryan > > > > > > -- > > > > > > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > > > > > > View/change your membership options at > > > > > > http://mailman.mit.edu/mailman/listinfo/piclist > > > > > > > > > > > -- > > > > > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > > > > > View/change your membership options at > > > > > http://mailman.mit.edu/mailman/listinfo/piclist > > > > > > > > > -- > > > > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > > > > View/change your membership options at > > > > http://mailman.mit.edu/mailman/listinfo/piclist > > > > > > > -- > > > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > > > View/change your membership options at > > > http://mailman.mit.edu/mailman/listinfo/piclist > > > > > -- > > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > > View/change your membership options at > > http://mailman.mit.edu/mailman/listinfo/piclist > > > -- > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist |
|
|
But wouldn't compensation introduce noise back in anyway, nullifying any
advantage? On Tue, 9 Feb 2021 at 18:11, David VanHorn <[hidden email]> wrote: > Yes. And you can use their reactance as a divider. Frequency response > will suffer, compensate on the input or the output. > > > > On Mon, Feb 8, 2021, 8:16 PM Ryan O'Connor <[hidden email]> wrote: > > > Well, they have low resistance. But they have reactance? > > > > On Tue, 9 Feb 2021 at 16:12, David VanHorn <[hidden email]> wrote: > > > > > Resistors have thermal noise, capacitors have essentially no resistance > > > > > > On Mon, Feb 8, 2021, 6:14 PM Ryan O'Connor <[hidden email]> wrote: > > > > > > > David do you have any theory or practice which can explain this? I > > would > > > be > > > > interested to read more about why this helps. I'm trying to read up > on > > > > Sean's topics and I'm finding it pretty complex. I have more > questions > > > than > > > > before, such as what is the relationship between impedence matching > to > > > > reduce noise and impedence matching to gain maximum power? Do they > > happen > > > > to co-inside, or how much out of phase are they (literally can't find > > > > anything on the net about this) > > > > > > > > Cheers > > > > Ryan > > > > > > > > On Tue, 9 Feb 2021 at 13:57, David VanHorn <[hidden email]> > > wrote: > > > > > > > > > Use capacitive dividers. > > > > > > > > > > On Mon, Feb 8, 2021, 4:20 PM Sean Breheny <[hidden email]> > wrote: > > > > > > > > > > > A fundamental property of absorbative attenuators (i.e. those > which > > > > > > dissipate the power removed from the signal) is that they > introduce > > > > > > (1-a)kTB thermal noise in watts, where a is the attenuation > > factor, k > > > > is > > > > > > Boltzman's constant, T is kelvin temperature of the attenuator, > > and B > > > > is > > > > > > the downstream circuit bandwidth in Hz. So, for example, a 10dB > > > > > attenuator > > > > > > (a=0.1), introduces 90% of a full equivalent of thermal noise. > > > > > > > > > > > > It is possible to attenuate reflectively (i.e., produce an > > impedance > > > > > > mismatch) and that can often circumvent this limitation although > it > > > > > depends > > > > > > on the source characteristics. There is a common practice in low > > > noise > > > > > > amplifiers call "noise matching" where instead of matching the > > > > impedance > > > > > of > > > > > > the source to that of the amplifier input, a purposeful mismatch > is > > > > > > introduced which reduces the contribution of noise due to the > real > > > part > > > > > of > > > > > > the amplifier's input impedance. I think the fundamental > principle > > is > > > > the > > > > > > same. > > > > > > > > > > > > Sean > > > > > > > > > > > > > > > > > > > > > > > > On Mon, Feb 8, 2021 at 5:00 PM Ryan O'Connor <[hidden email] > > > > > > wrote: > > > > > > > > > > > > > Hi, I am curious about a general topic which mostly interests > me > > in > > > > the > > > > > > > area of sensitive audio signals. > > > > > > > > > > > > > > What options do I have for attenuating a copper line signal > while > > > > > > > introducing a very low amount of noise? For example, if I > > grounded > > > > L+R > > > > > > > channels using independent resistors would that work? OR do I > > need > > > a > > > > > > proper > > > > > > > high-pass or low-pass filter design? But then what about > > > > > > > capacitors/inductors or series resistors introducing thermal > > noise? > > > > > > > > > > > > > > Open to discussion on types of noise as well :) > > > > > > > > > > > > > > Ryan > > > > > > > -- > > > > > > > http://www.piclist.com/techref/piclist PIC/SX FAQ & list > archive > > > > > > > View/change your membership options at > > > > > > > http://mailman.mit.edu/mailman/listinfo/piclist > > > > > > > > > > > > > -- > > > > > > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > > > > > > View/change your membership options at > > > > > > http://mailman.mit.edu/mailman/listinfo/piclist > > > > > > > > > > > -- > > > > > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > > > > > View/change your membership options at > > > > > http://mailman.mit.edu/mailman/listinfo/piclist > > > > > > > > > -- > > > > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > > > > View/change your membership options at > > > > http://mailman.mit.edu/mailman/listinfo/piclist > > > > > > > -- > > > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > > > View/change your membership options at > > > http://mailman.mit.edu/mailman/listinfo/piclist > > > > > -- > > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > > View/change your membership options at > > http://mailman.mit.edu/mailman/listinfo/piclist > > > -- > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist |
«
Return to MicroControllers - PIC
|
1 view|%1 views
| Free forum by Nabble | Edit this page |