Holmz
Major Contributor
I don’t think that the speed of a higher frequency is higher in meters/second.
But I also have not been to the embassy in Cuba.
-
Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Slew Rate
- Thread starter DonH56
- Start date
But I also have not been to the embassy in Cuba.
Not sure why I am bothering, but your post gets scored a D-minus for that glaringly cavernous insight into technical illiteracy.
Slew rate and small-signal bandwidth are independent variables and if an amplifier has a small-signal bandwidth of 300kHz or 50MHz, that has nothing to do with the noise measured in bandwidth that is relevant to audio. The 300kHz amplifier could be orders of magnitude noisier than the 50MHz amplifier.
Last edited:
OP
- Thread Starter
- #83
Not sure either. These latest posts were spawned by a discussion about power amplifier bandwidth and large-signal slew rate, not small-signal response, and large-signal bandwidth is relevant to slew rate although slew-enhancement techniques could modify the effective bandwidth (small and large). Greater slew requires higher current which increases (shot) noise. Small-signal bandwidth and noise was not a consideration in this instance, and the independence of small-signal bandwidth and large-signal slew rate also depends upon the amplifier topology and circuit design; they are not always independent.
A higher power bandwidth is a consequence of a higher slew rate, not a requirement for it. So when you wrote:
It was fair to assume that you were talking about small-signal bandwidth. If by "bandwidth" you were instead referring to power rather than small-signal bandwidth, then your statement, I am afraid, isn't any less nonsensical. How the hell does a higher power bandwidth increase noise?
You are in a quite a muddle here.
In any amplifier topology, slew rate and small-signal bandwidth can be modified independently of each other.
Seems like you didn't understand what he wrote. If bandwidth is low, then that is a limiter to slew rate. As Don wrote in his OP, Max Slew Rate of a sine wave = 2*pi*F*A. A is the amplitude. F is the frequency. If you keep "A" constant, then F sets the max slew rate. If an audio amplifier filters everything above 30 kHz, it will force a lower limit on slew rate than one that goes to 90 kHz.
Of course, reverse is also true that slew rate sets the (undistorted) limit to how much of the bandwidth can be used.
So while one parameter is not derived from the other, in the case of audio where the bandwidth is quite low, it certainly places a limit on slew rate. This is why high-slew-rate amplifiers tend to have very large bandwidth. Something that is not desirable but is the way to get high slew rates for the marketing material.
OP
- Thread Starter
- #87
Chicken and egg. If the goal is a high slew rate, I need to design for higher bandwidth, and/or add circuitry for slew enhancement. If I need very wide bandwidth, then (small-signal) slew rate is consequently high as well. Most of my career was analog ICs for things like radar systems where very wide bandwidth and very high slew rate were both required for pulse integrity, as well as very low noise and excellent overload recovery because the dynamic range was huge. I do not have much experience with high-power amplifiers, audio or RF/mW/mmW. I was focused on baseband through X to W band stuff but on the receiver side or initial signal generation before the high-power stages. Mostly data converters (ADCs/DACs) and associated circuits.
A higher power bandwidth is a consequence of a higher slew rate, not a requirement for it. So when you wrote:
The discussion was centered on and driven by a post a page or three back when a member questioned if the slew rate of an older Crown amplifier was adequate to meet its power specs. It is...
Higher bandwidth means greater noise bandwidth so all else equal (i.e. same noise floor in nV/rtHz) the circuit with greater bandwidth will have higher noise. That does not assume any filtering so your comment re. the audio band is correct (for bandlimited noise). Except for some hobby things my work focused on wideband circuits so I tend to think more about wideband (multi-GHz) systems than narrowband or bandlimited design; filtering was usually applied later.
I think it more likely we are simply defining things differently, or thinking about things from a different viewpoint and background. For transmitters, higher power bandwidth means a host of circuit trades to support that bandwidth, including things like greater bias current in the driving and output stages, so the noise floor tends to rise with increasing output power specifications. A 10 kW amplifier has a higher noise floor than a 1 kW amplifier, both due to the extra driving stage(s) and higher output current (higher shot noise). Again for radar/pulse applications, higher slew helps since the noise bursts around zero-crossings are shorter, potentially making integrated noise lower despite the additional shot noise from higher bias, but there are many design trades.
In a switching or transitioning dif pair, noise peaks during signal zero crossings when it is in the small-signal region, and the net noise is a function of the small-signal noise and time in the small-signal region which is determined by the slew rate which is related to the bandwidth. That is how I tend to think of things, for a single stage, and at a lower level circuit-wise than say an entire audio amplifier.
Ah, I do not usually distinguish large- and small-signal slew rate, since both are important for most of the circuits I designed (not audio). You can define the slew rate in the small-signal region as well as large-signal output slew. I designed data converters where small and large signal bandwidth and slew rate were both specified and fairly tightly coupled for low-level class A stages. The entire signal range (small to large) had to be linear over a very wide bandwidth, and slew and bandwidth was set by the differential pairs' design. These were low-level circuits where bandwidth and slew rate were specified and tightly coupled.
It would be good of you to write an article delving into small- and large-signal characteristics with respect to bandwidth and slew rate explaining how they are decoupled and why. That is not something this thread attempted to address at all, and I have no desire to write such an article at this time -- other things going on. I could say that I do understand the basis for your argument but am quite sure you would not believe me anyway.
OP
- Thread Starter
- #88
OK, this post (quoted from above) is a hot mess, but I am leaving it for posterity to exemplify a couple of bad examples:
1. Given normal daytime waking hours, it is never a good idea to post at 3 am (up late with younger son, celebrating his 31st birthday and viewing through the telescope on the only clear night in a while). What makes perfect sense at 3 am reads mostly like nonsense at 10 am. Bad idea whether in college or retired.
2. Engaging a person who starts off with the unshakeable premise that you are a completely unschooled, unskilled, idiotic, ignoramus is a waste of time for everyone. The 'net makes it easy to dismiss a person's decades of experience and knowledge based upon one misunderstood, poorly-worded post.
Most amplifier designs decouple large- and small-signal response including slew rate. The application that started my career did (could) not, leading to a complex analysis of how small-signal noise relates to and impacts large-signal performance (and vice-versa) that is well beyond the scope of this thread. For the specific question regarding the Crown's slew rate spec, the answer was provided several pages ago, and I see no point in extending things in this thread.
As a long-time (trumpet-playing) friend quoted: "Enjoy your day, unless you've made other plans." I need to return to lurking and not posting.
p.s. Amir @amirm, and others, thank you for the support.
1. Given normal daytime waking hours, it is never a good idea to post at 3 am (up late with younger son, celebrating his 31st birthday and viewing through the telescope on the only clear night in a while). What makes perfect sense at 3 am reads mostly like nonsense at 10 am. Bad idea whether in college or retired.
2. Engaging a person who starts off with the unshakeable premise that you are a completely unschooled, unskilled, idiotic, ignoramus is a waste of time for everyone. The 'net makes it easy to dismiss a person's decades of experience and knowledge based upon one misunderstood, poorly-worded post.
Most amplifier designs decouple large- and small-signal response including slew rate. The application that started my career did (could) not, leading to a complex analysis of how small-signal noise relates to and impacts large-signal performance (and vice-versa) that is well beyond the scope of this thread. For the specific question regarding the Crown's slew rate spec, the answer was provided several pages ago, and I see no point in extending things in this thread.
As a long-time (trumpet-playing) friend quoted: "Enjoy your day, unless you've made other plans." I need to return to lurking and not posting.
p.s. Amir @amirm, and others, thank you for the support.
Has the cause of that mysterious hearing damage problem there ever been found?
Holmz
Major Contributor
Similar threads
