- Joined
- Jan 14, 2011
- Messages
- 74,912
- Location
- Gillette, Wyo.
- Tagline
- Halfbiass...Electron Herder and Backass Woof
[TABLE="class: main"]
[TR]
[TD="class: maintd"][TABLE="width: 100%"]
[TR]
[TD]
Audio.TST February 2013
Hello Leland,
I spent the first 15 years of my career working for a certain well-known oscilloscope company in Beaverton Oregon. As a young and fledgling design engineer, I rarely got to experience direct contact with customers. When I did, it was often under unusual and sometimes cryptic circumstances. About three years into my career, I vividly remember receiving a telephone call (no e-mail back then) from a customer in New Mexico that was having a problem with some oscilloscope time-bases. He would not directly answer some of my questions regarding his application, and it quickly turned into a game of 20 questions. I finally suggested he send the problem units back to the factory for analysis and repair. He said he couldn’t do that, and remained evasive. After 20 more questions I was finally able to deduce that he was working with test equipment that had been buried in a tunnel near an underground nuclear explosion. Those were fairly common in that day, and they provided a lucr ative chunk of business for my employer. To make a long story short, this customer couldn’t send his units back because they were radioactive! In the end I could not help this customer, but I will never forget the joy of learning about his unusual application.
It has been my observation that larger corporations tend to treat their design engineers as technical automatons, people who are supposed to pull rabbits out of a hat often under near impossible deadlines. Direct customer contact is infrequent, indeed something to be avoided because everyone knows that engineers tend to talk too much. Trade show attendance was only a dream, usually reserved only for the very highest level engineers. In 1977 when I was given a promotion and the opportunity to manage a new group of engineers tasked with the design of the company’s first entry into the world of audio test and measurement, I realized something had to change regarding this institutionalized policy of isolating design engineers from customers. I attended my first AES (Audio Engineering Society) convention in the fall of 1977 at the Waldorf Astoria in New York City. I made many new friends and contacts, and returned with the powerful belief that direct customer contact and understanding of his typical applications (aka “use casesâ€) was essential to the success of any good product. My group went on to develop the SG505 low distortion oscillator (in 1979) and the AA501 fully automatic distortion analyzer (in 1980) which were considered significant advancements at that time.
In late 1982 after we developed programmable versions of these two products, senior management let my team know it was to be disbanded; and that there would be no further development in the audio test and measurement market. (There was an old saying with this company that “profit is proportional to bandwidthâ€â€¦therefore audio test and measurement products could never provide an attractive business opportunity.) Having had the most rewarding phase of our careers up to that point, four of us continued to meet clandestinely to discuss possibilities. We ultimately resigned from the company during the summer of 1984, and Audio Precision was officially incorporated on October 28, 1984. We began Audio Precision with the premise that the customer MUST be well served with quality products, at the right price, with the correct levels of performance, and with timely and efficient service. To this day I still strongly believe in that founding premise.
As always, we want to hear from you, particularly if your application involves new technology or new applications for existing technology.
Bruce Hofer, Chairman & Co-Founder, Audio Precision
[h=1]Output: Bode Plots in APx500[/h]Q: Can the APx500 software make Bode plots?
A: A Bode plot is a graph of the transfer function of a system showing the magnitude and phase plotted on a logarithmic frequency axis. In a Bode plot, usually the phase is plotted on a separate graph above the magnitude graph, but sometimes magnitude is plotted above phase, or they can also be shown on one graph with dual vertical axes.
In APx, the transfer function of a system (a Device Under Test (DUT) or a circuit) can be measured using one of the frequency response measurements (Frequency Response, Stepped Frequency Sweep, Continuous Sweep, Acoustic Response or Multitone Analyzer). Of these, the easiest and most convenient to use is Continuous Sweep. Although it has 16 primary results, the only two of interest for a Bode plot are Gain (which is the transfer function magnitude) and Phase. Note that below the Phase graph there is a Mode control which can be used to select “relative to Ch1â€, etc. For a Bode phase plot, this control should be set to “Input-to-output, wrappedâ€, as shown below:
As a simple example, in the attached project file, a Continuous Sweep measurement has been used to measure the frequency response (or transfer function) of an analog A-weighting filter. Note that all the primary results except Gain and Phase have been deleted, and the Gain result has been renamed to “Gain (Transfer Function Magnitude)â€. These results are the components of a Bode plot.
For those users who need to see them plotted one above the other, the project file contains a report using the APx custom reporting feature to plot the phase graph directly above the magnitude graph.
Downloads:
[h=1]Sound Advice: APx Burst Waveform Utility[/h]The new APx Burst Waveform Utility creates sine burst waveforms and then saves them and/or uploads them into the APx generator. A burst waveform contains a specified number of cycles of a waveform at full level, followed by a period of cycles at a lower level. The transition from full to low level happens instantly at the zero crossing of the waveform. The low level may be set at any level between 0 and 100%. This utility does not create shaped bursts, where the transition from high to low level occurs gradually over time.
[h=5]Fig 1 Burst waveform, burst on for 1 cycle with a total interval of 3 cycles. Low level is 10% (-20 dB).[/h]Burst waveforms can be useful in testing a number of properties, including power supply transient response and reserve, meter ballistics, and compressor/limiter behavior.
The utility provides control over the Burst Parameters, the Pattern, and the Wav File Parameters. The frequency setting is limited to frequencies that are synchronous at the selected sample rate. 192 kHz is set as the default rate in order to provide the greatest resolution. This can be reduced if you need to save .wav files at a lower sampling rate.
[h=5]Fig 2 APx Burst Waveform Utility (larger view).[/h]Wave file length may be set to minimum length, specified samples, or specified seconds. Minimum length will produce the shortest synchronous clip that is at least 256 samples long. The APx500 generator will seamlessly loop this clip until stopped. When the clip is loaded into Measurement Recorder, the Sweep Time will get set to 1 second, because the file is so short. After loading, you may reset the length of the acquisition to a longer length of time.
To use the utility, open it, set the desired parameters, and then click either “Upload to APx†or “Save .wav fileâ€. “Upload to APx†loads the waveform file directly into the generator for the current APx measurement. If APx500 is not running, or its version does not match the utility’s, then this button will be disabled.
“Save .wav file†will prompt for a file location to save to disk. The file that is created may be played with any .wav file player, or uploaded later to the APx generator. It may also be imported into AP2700.
Downloads:
[h=1]Test Results: AP News & Events[/h]Events:
[/TD]
[TD="width: 10"] [/TD]
[TD="class: contents"]Notes from the Test Bench
Output: Bode Plots in APx500
Sound Advice:
Test Results: AP News and Events
[/TD]
[/TR]
[/TABLE]
[/TD]
[/TR]
[/TABLE]
[TR]
[TD="class: maintd"][TABLE="width: 100%"]
[TR]
[TD]
Audio.TST February 2013
Hello Leland,
I spent the first 15 years of my career working for a certain well-known oscilloscope company in Beaverton Oregon. As a young and fledgling design engineer, I rarely got to experience direct contact with customers. When I did, it was often under unusual and sometimes cryptic circumstances. About three years into my career, I vividly remember receiving a telephone call (no e-mail back then) from a customer in New Mexico that was having a problem with some oscilloscope time-bases. He would not directly answer some of my questions regarding his application, and it quickly turned into a game of 20 questions. I finally suggested he send the problem units back to the factory for analysis and repair. He said he couldn’t do that, and remained evasive. After 20 more questions I was finally able to deduce that he was working with test equipment that had been buried in a tunnel near an underground nuclear explosion. Those were fairly common in that day, and they provided a lucr ative chunk of business for my employer. To make a long story short, this customer couldn’t send his units back because they were radioactive! In the end I could not help this customer, but I will never forget the joy of learning about his unusual application.
It has been my observation that larger corporations tend to treat their design engineers as technical automatons, people who are supposed to pull rabbits out of a hat often under near impossible deadlines. Direct customer contact is infrequent, indeed something to be avoided because everyone knows that engineers tend to talk too much. Trade show attendance was only a dream, usually reserved only for the very highest level engineers. In 1977 when I was given a promotion and the opportunity to manage a new group of engineers tasked with the design of the company’s first entry into the world of audio test and measurement, I realized something had to change regarding this institutionalized policy of isolating design engineers from customers. I attended my first AES (Audio Engineering Society) convention in the fall of 1977 at the Waldorf Astoria in New York City. I made many new friends and contacts, and returned with the powerful belief that direct customer contact and understanding of his typical applications (aka “use casesâ€) was essential to the success of any good product. My group went on to develop the SG505 low distortion oscillator (in 1979) and the AA501 fully automatic distortion analyzer (in 1980) which were considered significant advancements at that time.
In late 1982 after we developed programmable versions of these two products, senior management let my team know it was to be disbanded; and that there would be no further development in the audio test and measurement market. (There was an old saying with this company that “profit is proportional to bandwidthâ€â€¦therefore audio test and measurement products could never provide an attractive business opportunity.) Having had the most rewarding phase of our careers up to that point, four of us continued to meet clandestinely to discuss possibilities. We ultimately resigned from the company during the summer of 1984, and Audio Precision was officially incorporated on October 28, 1984. We began Audio Precision with the premise that the customer MUST be well served with quality products, at the right price, with the correct levels of performance, and with timely and efficient service. To this day I still strongly believe in that founding premise.
As always, we want to hear from you, particularly if your application involves new technology or new applications for existing technology.
Bruce Hofer, Chairman & Co-Founder, Audio Precision
[h=1]Output: Bode Plots in APx500[/h]Q: Can the APx500 software make Bode plots?
A: A Bode plot is a graph of the transfer function of a system showing the magnitude and phase plotted on a logarithmic frequency axis. In a Bode plot, usually the phase is plotted on a separate graph above the magnitude graph, but sometimes magnitude is plotted above phase, or they can also be shown on one graph with dual vertical axes.
In APx, the transfer function of a system (a Device Under Test (DUT) or a circuit) can be measured using one of the frequency response measurements (Frequency Response, Stepped Frequency Sweep, Continuous Sweep, Acoustic Response or Multitone Analyzer). Of these, the easiest and most convenient to use is Continuous Sweep. Although it has 16 primary results, the only two of interest for a Bode plot are Gain (which is the transfer function magnitude) and Phase. Note that below the Phase graph there is a Mode control which can be used to select “relative to Ch1â€, etc. For a Bode phase plot, this control should be set to “Input-to-output, wrappedâ€, as shown below:
As a simple example, in the attached project file, a Continuous Sweep measurement has been used to measure the frequency response (or transfer function) of an analog A-weighting filter. Note that all the primary results except Gain and Phase have been deleted, and the Gain result has been renamed to “Gain (Transfer Function Magnitude)â€. These results are the components of a Bode plot.
For those users who need to see them plotted one above the other, the project file contains a report using the APx custom reporting feature to plot the phase graph directly above the magnitude graph.
Downloads:
[h=1]Sound Advice: APx Burst Waveform Utility[/h]The new APx Burst Waveform Utility creates sine burst waveforms and then saves them and/or uploads them into the APx generator. A burst waveform contains a specified number of cycles of a waveform at full level, followed by a period of cycles at a lower level. The transition from full to low level happens instantly at the zero crossing of the waveform. The low level may be set at any level between 0 and 100%. This utility does not create shaped bursts, where the transition from high to low level occurs gradually over time.
[h=5]Fig 1 Burst waveform, burst on for 1 cycle with a total interval of 3 cycles. Low level is 10% (-20 dB).[/h]Burst waveforms can be useful in testing a number of properties, including power supply transient response and reserve, meter ballistics, and compressor/limiter behavior.
The utility provides control over the Burst Parameters, the Pattern, and the Wav File Parameters. The frequency setting is limited to frequencies that are synchronous at the selected sample rate. 192 kHz is set as the default rate in order to provide the greatest resolution. This can be reduced if you need to save .wav files at a lower sampling rate.
[h=5]Fig 2 APx Burst Waveform Utility (larger view).[/h]Wave file length may be set to minimum length, specified samples, or specified seconds. Minimum length will produce the shortest synchronous clip that is at least 256 samples long. The APx500 generator will seamlessly loop this clip until stopped. When the clip is loaded into Measurement Recorder, the Sweep Time will get set to 1 second, because the file is so short. After loading, you may reset the length of the acquisition to a longer length of time.
To use the utility, open it, set the desired parameters, and then click either “Upload to APx†or “Save .wav fileâ€. “Upload to APx†loads the waveform file directly into the generator for the current APx measurement. If APx500 is not running, or its version does not match the utility’s, then this button will be disabled.
“Save .wav file†will prompt for a file location to save to disk. The file that is created may be played with any .wav file player, or uploaded later to the APx generator. It may also be imported into AP2700.
Downloads:
[h=1]Test Results: AP News & Events[/h]Events:
[/TD]
[TD="width: 10"] [/TD]
[TD="class: contents"]Notes from the Test Bench
Output: Bode Plots in APx500
Sound Advice:
Test Results: AP News and Events
[/TD]
[/TR]
[/TABLE]
[/TD]
[/TR]
[/TABLE]
