| Home |
| Search |
| Today's Posts |
|
#41
|
|||
|
|||
Chris Hornbeck wrote:
On Mon, 29 Aug 2005 23:09:38 GMT, SSJVCmag wrote: I haven;t seen this many inventive out-there curses since reading the Captain Haddock collection. **He told me to go to Hell and to stay there. I'm profoundly grateful. But not yet dead. Very funny stuff overall. There'll probably be a fan website someday. Don't give me ideas ! ;-) Graham |
|
#42
|
|||
|
|||
|
Harry Houdini wrote:
Would you care to go into a little more detail on how EMI triggers slew rate limiting. Very interesting. IOW EMI can trigger slew rate limiting. Whenever I play that recording of Carmen on EMI, the thing starts mistracking like mad as soon as Rise Stevens starts singing about her friend Lillas Pastia. Clearly a case of mechanical slew limiting. --scott -- "C'est un Nagra. C'est suisse, et tres, tres precis." |
|
#43
|
|||
|
|||
|
"nap" Slew rate is = dV/dT. Most definately affects the 'rise time ' specification. ** Bull****. You are a total ****ing ass - nap. **** the hell off. poor thing. ** **** off - you anencephalic turd. ............. Phil |
|
#44
|
|||
|
|||
|
Graham Stevenson ** YOU are a mental retard and an autistic moron. YOU are a monstrous liar and raving lunatic. YOU are a notorious usenet stalker and harasser. .......... Phil |
|
#45
|
|||
|
|||
|
"martin griffith" Risetime=digital slew rate= audio ** My god - it just gets worse and worse. RAP = Congenital ****wit's burial ground ? ........... Phil |
|
#46
|
|||
|
|||
|
"Harry Houdini" Cheers! - everyone - for taking the time to shed some light on the subject of slew rates. I have a *much* better understanding of what's going on. ** No you do not - you PITA troll. ............ Phil |
|
#47
|
|||
|
|||
|
Graham Stevenson Don't give me ideas ! ** YOU never had a single idea in you life you did not steal from someone. YOU are a fake, a liar and a damn charlatan. YOU are a piece of sub human garbage. ............ Phil |
|
#48
|
|||
|
|||
|
|
|
#49
|
|||
|
|||
|
"SSJVCmag" ** Mike Rivers would like you to come over and suck his penis like you usually do. Leave your false teeth at home. ............ Phil |
|
#50
|
|||
|
|||
|
On Tue, 30 Aug 2005 03:10:01 GMT, SSJVCmag
wrote: Let's move on now to the Hall Of Living Arms. This was original shown (and believed to be a cinematic trick) in the original French version of BEAUTY AND THE BEAST. Too, too great (sorry; don't know how to say it in French). The 1946 Cocteau original sets a standard for both beauty and technical trickery still seldom approached. This summer's _War of the Worlds_ was, comparatively, pathetic. An accessible and too-too-fun approach is to study the finale "ascension" of the beauty-beast pair. Run it in reverse to understand the dynamics. Beautiful, beautiful work. And, no computers needed apply. Chris Hornbeck |
|
#51
|
|||
|
|||
|
Scott Dorsey wrote:
Harry Houdini wrote: Would you care to go into a little more detail on how EMI triggers slew rate limiting. Very interesting. IOW EMI can trigger slew rate limiting. Whenever I play that recording of Carmen on EMI, the thing starts mistracking like mad as soon as Rise Stevens starts singing about her friend Lillas Pastia. Clearly a case of mechanical slew limiting. Clearly, you've been taking way too much EMI. Back off a little... -- ha |
|
#52
|
|||
|
|||
|
Pooh Bear wrote:
Phil Allison wrote: Graham Stevenon, ** YOU are a mental retard and an autistic moron. And you're an ignorant arse who hates being shown up as wrong. Now shut up and go learn something about slew rate. When y'all's brains develop fully you won't be responding to Silly Assilon. The Troll lives by responses. Let it not live. -- ha |
|
#53
|
|||
|
|||
|
|
|
#54
|
|||
|
|||
|
"Pooh Bear" wrote in message
... That 'overload' could be provoked by applying a step impulse to a gain stage too. In such a case the amplifier output will slew towards the required voltage before settling. This illustrates nicely that square wave testing of audio equipment is not entirely sensible. Any test signal should not exceed the bandwidth of the device under test. Why not? If the device stands a chance of being exposed to wideband signals, a square wave is a great way to find out whether it will react sensibly (bandwidth-limited at all levels) or badly (slew-limited at high levels). Peace, Paul |
|
#55
|
|||
|
|||
|
Arny Krueger wrote: "Harry Houdini" wrote in message Arnie, Would you care to go into a little more detail on how EMI triggers slew rate limiting. Very interesting. IOW EMI can trigger slew rate limiting. The maximum slew rate of a signal is 2*pi*signal frequency*signal amplitude. What is the steepest slope of the response of a baseband limited transfer function to a step function input? Can't remember for sure. Bob -- "Things should be described as simply as possible, but no simpler." A. Einstein |
|
#56
|
|||
|
|||
|
"Paul Stamler" This illustrates nicely that square wave testing of audio equipment is not entirely sensible. Any test signal should not exceed the bandwidth of the device under test. Why not? ** You were just given the reason - ****HEAD !! If the device stands a chance of being exposed to wideband signals, ** Beware of ****wits that begins sentences with unexplained " if " conditions. a square wave is a great way to find out whether it will react sensibly (bandwidth-limited at all levels) or badly (slew-limited at high levels). ** Just as "sensible" to use sine waves of any frequency & amplitude. Just as mad as well. ............ Phil |
|
#57
|
|||
|
|||
|
Pooh Bear wrote: If you care to do the calculus it's possible to demonstrate that the fastest change in signal for a sinewave is at the zero-crossing point and equates to 2 x pi x freq x V pk. Yes, but what is it for a signal that is a band limited step function instead of a sinusoid? Bob -- "Things should be described as simply as possible, but no simpler." A. Einstein |
|
#58
|
|||
|
|||
|
On Tue, 30 Aug 2005 00:29:27 -0700, Bob Cain wrote:
Arny Krueger wrote: "Harry Houdini" wrote in message Arnie, Would you care to go into a little more detail on how EMI triggers slew rate limiting. Very interesting. IOW EMI can trigger slew rate limiting. The maximum slew rate of a signal is 2*pi*signal frequency*signal amplitude. What is the steepest slope of the response of a baseband limited transfer function to a step function input? Can't remember for sure. Bob It doesn't have one. A band limited circuit has a step response slope that is proportional to amplitude - hence it is theoretically unlimited. A slew rate limited circuit, on the other hand, is limited. d |
|
#59
|
|||
|
|||
|
"Bob Cain" If you care to do the calculus it's possible to demonstrate that the fastest change in signal for a sinewave is at the zero-crossing point and equates to 2 x pi x freq x V pk. Yes, but what is it for a signal that is a band limited step function instead of a sinusoid? ** That would depend on the rise time of the band limited step - ie time expired to 63% of final value. For a simple 6 dB/oct roll off, BW = 0.35 / Tr or Tr = 0.35 / BW So, for a circuit that is - 3 dB at 50 kHz, the Tr = 7.0 uS. So, if the step is say 10 volt, 6.3 volts of it is done in 7.0 uS at a rate of 0.9 V/uS. ............ Phil |
|
#60
|
|||
|
|||
|
"Phil Allison" wrote in message
Graham Stevenson ** YOU are a mental retard and an autistic moron. YOU are a monstrous liar and raving lunatic. YOU are a notorious usenet stalker and harasser. He's a heck of a nice guy compared to Phildo. ;-) |
|
#61
|
|||
|
|||
|
"Pooh Bear" wrote
in message This illustrates nicely that square wave testing of audio equipment is not entirely sensible. I agree with that, but on different grounds. Any test signal should not exceed the bandwidth of the device under test. Huh? If a signal exists in the typical use then it is fair game to use it or something like it for testing. Good equipment typically has bandwidth limiting in or around the input stage to make it difficult or impossible to overload. |
|
#62
|
|||
|
|||
|
"Arny Krueger" "Phil Allison" Graham Stevenson ** YOU are a mental retard and an autistic moron. YOU are a monstrous liar and raving lunatic. YOU are a notorious usenet stalker and harasser. He's a heck of a nice guy compared to Phildo. ;-) ** So is nearly anyone - short of " Chemical Ali ". ......... Phil |
|
#63
|
|||
|
|||
|
"Arny Krueger" This illustrates nicely that square wave testing of audio equipment is not entirely sensible. I agree with that, but on different grounds. Any test signal should not exceed the bandwidth of the device under test. Huh? If a signal exists in the typical use then it is fair game to use it or something like it for testing. ** Where do infinite band square wave comes from - Arny ?? Out the arse of some audio Satanic Demon ??? Get real for Christ's sake. ............ Phil |
|
#64
|
|||
|
|||
|
On Tue, 30 Aug 2005 21:56:23 +1000, Phil Allison wrote:
"Arny Krueger" This illustrates nicely that square wave testing of audio equipment is not entirely sensible. I agree with that, but on different grounds. Any test signal should not exceed the bandwidth of the device under test. Huh? If a signal exists in the typical use then it is fair game to use it or something like it for testing. ** Where do infinite band square wave comes from - Arny ?? Out the arse of some audio Satanic Demon ??? Get real for Christ's sake. ........... Phil How do you get from a signal that exceeds the bandwidth of the device under test to an infinite band square wave? Just interested. d |
|
#65
|
|||
|
|||
|
Paul Stamler wrote:
"Pooh Bear" wrote in message ... That 'overload' could be provoked by applying a step impulse to a gain stage too. In such a case the amplifier output will slew towards the required voltage before settling. This illustrates nicely that square wave testing of audio equipment is not entirely sensible. Any test signal should not exceed the bandwidth of the device under test. Why not? If the device stands a chance of being exposed to wideband signals, a square wave is a great way to find out whether it will react sensibly (bandwidth-limited at all levels) or badly (slew-limited at high levels). Actually, you ask the relevant question I had just been musing over. Namely what actual inputs might a gain stage see ( 'worst' case ) ? I doubt that any moving coil mic would present a challenge but I'd be interested to know what kind of impulse response a condensor mic might provide when say miking a snare drum. Any ideas anyone ? Of course once the signal's been through the record - reproduce train it'll be limited to the system bandwidth anyway but I'd be curious to know about how it starts off. Anyone have the kit to digitize such a signal and perform a Fourier analysis ? Graham |
|
#66
|
|||
|
|||
|
Bob Cain wrote: Pooh Bear wrote: If you care to do the calculus it's possible to demonstrate that the fastest change in signal for a sinewave is at the zero-crossing point and equates to 2 x pi x freq x V pk. Yes, but what is it for a signal that is a band limited step function instead of a sinusoid? I believe that in rough terms it approximates to a full scale sinusoid at the highest frequency. The important caveat being 'band limited' of course. Graham |
|
#67
|
|||
|
|||
|
Don Pearce wrote:
On Tue, 30 Aug 2005 00:29:27 -0700, Bob Cain wrote: Arny Krueger wrote: "Harry Houdini" wrote in message Arnie, Would you care to go into a little more detail on how EMI triggers slew rate limiting. Very interesting. IOW EMI can trigger slew rate limiting. The maximum slew rate of a signal is 2*pi*signal frequency*signal amplitude. What is the steepest slope of the response of a baseband limited transfer function to a step function input? Can't remember for sure. Bob It doesn't have one. A band limited circuit has a step response slope that is proportional to amplitude - hence it is theoretically unlimited. A slew rate limited circuit, on the other hand, is limited. Logically, the amplitude would have to be within system dynamic range for any test to be meaningful. In the same way as we use sinewave testing. Graham |
|
#68
|
|||
|
|||
|
On Tue, 30 Aug 2005 14:17:57 +0100, Pooh Bear wrote:
Don Pearce wrote: On Tue, 30 Aug 2005 00:29:27 -0700, Bob Cain wrote: Arny Krueger wrote: "Harry Houdini" wrote in message Arnie, Would you care to go into a little more detail on how EMI triggers slew rate limiting. Very interesting. IOW EMI can trigger slew rate limiting. The maximum slew rate of a signal is 2*pi*signal frequency*signal amplitude. What is the steepest slope of the response of a baseband limited transfer function to a step function input? Can't remember for sure. Bob It doesn't have one. A band limited circuit has a step response slope that is proportional to amplitude - hence it is theoretically unlimited. A slew rate limited circuit, on the other hand, is limited. Logically, the amplitude would have to be within system dynamic range for any test to be meaningful. In the same way as we use sinewave testing. Graham Right. Of course all these rise time limitations are to do with charging capacitors through resistances. Slopes are related to how quickly the caps can charge. In the case of slew rate limited slopes, you have a constant current source - in the typical amplifier this will be the long tail pair on the input - charging the dominant pole capacitor. This makes for a very simple calculation to make sure that no in-band signal is ever going to saturate that situation. Given a high enough slew rate (lots of current in the input pair, and as small a dominant pole cap as possible) and band-limiting of the amplifier signal, it is possible to ensure that slew rate limiting never happens in practise. d |
|
#69
|
|||
|
|||
|
|
|
#70
|
|||
|
|||
|
"Pooh Bear" wrote
in message Don Pearce wrote: On Tue, 30 Aug 2005 00:29:27 -0700, Bob Cain wrote: Arny Krueger wrote: "Harry Houdini" wrote in message Arnie, Would you care to go into a little more detail on how EMI triggers slew rate limiting. Very interesting. IOW EMI can trigger slew rate limiting. The maximum slew rate of a signal is 2*pi*signal frequency*signal amplitude. What is the steepest slope of the response of a baseband limited transfer function to a step function input? Can't remember for sure. Bob It doesn't have one. A band limited circuit has a step response slope that is proportional to amplitude - hence it is theoretically unlimited. A slew rate limited circuit, on the other hand, is limited. Logically, the amplitude would have to be within system dynamic range for any test to be meaningful. In the same way as we use sinewave testing. How does one find that a system has a fault if all the tests you do are constrained from demonstrating that fault? |
|
#71
|
|||
|
|||
|
"Pooh Bear" wrote
in message Anyone have the kit to digitize such a signal and perform a Fourier analysis ? Having digitized a lot of stuff with a high-bandwidth system (well high bandwidth compared to regular studio gear), the bandwidth champs aren't snare drums, but stuff like keychains, triangles, tambourines, etc. Here's some pretty pictures: http://www.pcabx.com/technical/sample_rates/index.htm Here's somebody else's pretty pictures: http://www.cco.caltech.edu/~boyk/spectra/spectra.htm |
|
#72
|
|||
|
|||
|
"Arny Krueger" How does one find that a system has a fault if all the tests you do are constrained from demonstrating that fault? ** This is the exact *dumb***** error Arny the Asshole is besotted with. Amusingly - it is the exact same "error " used constantly by snake oil merchants - the error of applying an unrealistic & unjustified criteria in testing. Any device of any kind will, of course, fail some test - if you make it tough enough. Proves nothing of any value to anyone. It is the *mark* of an good engineer that to devise tests that are indicative of the needed performance in the intended application - not too easy and neither so tough that passing is next to impossible. Fools set up impossible criteria and then pass judgements on how badly the test subjects all fail. Fools. .......... Phil |
|
#73
|
|||
|
|||
|
On Wed, 31 Aug 2005 00:57:58 +1000, Phil Allison wrote:
"Arny Krueger" How does one find that a system has a fault if all the tests you do are constrained from demonstrating that fault? ** This is the exact *dumb***** error Arny the Asshole is besotted with. Amusingly - it is the exact same "error " used constantly by snake oil merchants - the error of applying an unrealistic & unjustified criteria in testing. Any device of any kind will, of course, fail some test - if you make it tough enough. Proves nothing of any value to anyone. It is the *mark* of an good engineer that to devise tests that are indicative of the needed performance in the intended application - not too easy and neither so tough that passing is next to impossible. Fools set up impossible criteria and then pass judgements on how badly the test subjects all fail. Fools. ......... Phil So clearly the ultrasonic stability of an amplifier is a matter of no interest to a design engineer. He can happily limit all his stimuli and measurements to 20kHz Nice one, Phil - don't forget to fuse your tweeters! Hee hee hee! d |
|
#74
|
|||
|
|||
|
"Don Pearce" Phil Allison wrote: "Arny Krueger" How does one find that a system has a fault if all the tests you do are constrained from demonstrating that fault? ** This is the exact *dumb***** error Arny the Asshole is besotted with. Amusingly - it is the exact same "error " used constantly by snake oil merchants - the error of applying an unrealistic & unjustified criteria in testing. Any device of any kind will, of course, fail some test - if you make it tough enough. Proves nothing of any value to anyone. It is the *mark* of an good engineer that to devise tests that are indicative of the needed performance in the intended application - not too easy and neither so tough that passing is next to impossible. Fools set up impossible criteria and then pass judgements on how badly the test subjects all fail. Fools. So clearly the ultrasonic stability of an amplifier is a matter of no interest to a design engineer. ** No such idiotic conclusion follows at all from my post. You are a demeted, know nothing ass - Don Pearce. Kindly go **** yourself - again . ** Hey Don - RF and audio are not the same !!! Reflections occur at impedance discontinuities. ** Hey Don - RF and audio are not the same !!! You don't need even a millimetre of cable. ** Hey Don - RF and audio are not the same !!! Any time an impedance changes, power is scattered ** Hey Don - RF and audio are not the same !!! - some goes forwards into the load, some goes backwards into the source. ** Hey Don - RF and audio are not the same !!! The bigger the discontinuity (rom 300 to 100000 ohms, for instance) the more of the available power is scattered backwards. ** Hey Don - RF and audio are not the same !!! That is the basis of Scattering Parameters ** Hey Don - RF and audio are not the same !!! - a standard method of specifying matching, terminating and reflecting. ** Hey Don - RF and audio are not the same !!! Pearce Consulting http://www.pearce.uk.com = a bloody RF web site !!! Don Pearce !! YOU are one, PITA scatterbrained ****ing IDIOT !!!! .............. Phil ........... Phil |
|
#75
|
|||
|
|||
|
On Wed, 31 Aug 2005 01:13:01 +1000, Phil Allison wrote:
"Don Pearce" Phil Allison wrote: "Arny Krueger" How does one find that a system has a fault if all the tests you do are constrained from demonstrating that fault? ** This is the exact *dumb***** error Arny the Asshole is besotted with. Amusingly - it is the exact same "error " used constantly by snake oil merchants - the error of applying an unrealistic & unjustified criteria in testing. Any device of any kind will, of course, fail some test - if you make it tough enough. Proves nothing of any value to anyone. It is the *mark* of an good engineer that to devise tests that are indicative of the needed performance in the intended application - not too easy and neither so tough that passing is next to impossible. Fools set up impossible criteria and then pass judgements on how badly the test subjects all fail. Fools. So clearly the ultrasonic stability of an amplifier is a matter of no interest to a design engineer. ** No such idiotic conclusion follows at all from my post. You are a demeted, know nothing ass - Don Pearce. Kindly go **** yourself - again . ** Hey Don - RF and audio are not the same !!! Reflections occur at impedance discontinuities. ** Hey Don - RF and audio are not the same !!! You don't need even a millimetre of cable. ** Hey Don - RF and audio are not the same !!! Any time an impedance changes, power is scattered ** Hey Don - RF and audio are not the same !!! - some goes forwards into the load, some goes backwards into the source. ** Hey Don - RF and audio are not the same !!! The bigger the discontinuity (rom 300 to 100000 ohms, for instance) the more of the available power is scattered backwards. ** Hey Don - RF and audio are not the same !!! That is the basis of Scattering Parameters ** Hey Don - RF and audio are not the same !!! - a standard method of specifying matching, terminating and reflecting. ** Hey Don - RF and audio are not the same !!! Pearce Consulting http://www.pearce.uk.com = a bloody RF web site !!! Don Pearce !! YOU are one, PITA scatterbrained ****ing IDIOT !!!! ............. Phil .......... Phil Well, that one certainly expolded with far more ordnance than I put in. What went bang up your end, Phil? d |
|
#76
|
|||
|
|||
|
Don Pearce wrote:
Nice one, Phil - don't forget to fuse your tweeters! Hee hee hee! No need - he has fused synapses. -- ha |
|
#77
|
|||
|
|||
|
Don Pearce wrote: On Tue, 30 Aug 2005 00:29:27 -0700, Bob Cain wrote: Arny Krueger wrote: The maximum slew rate of a signal is 2*pi*signal frequency*signal amplitude. What is the steepest slope of the response of a baseband limited transfer function to a step function input? Can't remember for sure. It doesn't have one. A band limited circuit has a step response slope that is proportional to amplitude - hence it is theoretically unlimited. A slew rate limited circuit, on the other hand, is limited. Right, so the slew rate requirement for transient inputs can't be estimated by the max slope of steady state, HF sinusoids. Isn't it is higher than the latter would indicate? Bob -- "Things should be described as simply as possible, but no simpler." A. Einstein |
|
#78
|
|||
|
|||
|
Pooh Bear wrote: Anyone have the kit to digitize such a signal and perform a Fourier analysis ? A time domain analysis would be better, seeking the pair of samples with the largest difference between them. Without taking into account transducer properties, for a digital to analog signal that would be rail to rail in about 22.7 microseconds (at 44.1 kHz.) On second thought and from some quick sketches, I think that the zero crossing slope of a signal at Nyquist (Arny's answer) is greater than any transient slope could be. No time for a real analysis. Anyone? Bob -- "Things should be described as simply as possible, but no simpler." A. Einstein |
|
#79
|
|||
|
|||
|
Phil Allison wrote: "Bob Cain" If you care to do the calculus it's possible to demonstrate that the fastest change in signal for a sinewave is at the zero-crossing point and equates to 2 x pi x freq x V pk. Yes, but what is it for a signal that is a band limited step function instead of a sinusoid? ** That would depend on the rise time of the band limited step - ie time expired to 63% of final value. For a simple 6 dB/oct roll off, BW = 0.35 / Tr or Tr = 0.35 / BW So, for a circuit that is - 3 dB at 50 kHz, What would you calculate for a D/A system with a very high order reconstruction rolloff at 44.1 kHz? Let's say a brick wall for grins. Without looking deeply (enough, probably) I'm back to thinking that the max for a sampled system is pi * samplerate * Vpp * 1E-6 (V/uS) which for a 44.1 kHz, 10 Vpp system would be 1.39 V/uS. Bob -- "Things should be described as simply as possible, but no simpler." A. Einstein |
|
#80
|
|||
|
|||
|
"Bob Cain" wrote in message
Pooh Bear wrote: Anyone have the kit to digitize such a signal and perform a Fourier analysis ? A time domain analysis would be better, seeking the pair of samples with the largest difference between them. Without taking into account transducer properties, for a digital to analog signal that would be rail to rail in about 22.7 microseconds (at 44.1 kHz.) On second thought and from some quick sketches, I think that the zero crossing slope of a signal at Nyquist (Arny's answer) is greater than any transient slope could be. No time for a real analysis. Anyone? The rate at zero crossing, F=Nyquist-delta, peak signal = FS defines maximum slewing for a signal from the digital domain. |
Linear Mode
