The dScope III is an amazingly capable new measurement system by PrismSound. In a relatively small enclosure, its size is such that a laptop computer fits nicely on top of it making for a very compact computer controlled measurement setup. It is a serious upgrade of the previous dScope II that, if memory serves me correctly, was in the form of a card that mounted inside a PC. Connection to the host computer is via a USB cable. Requirements for the host PC are Windows 98, ME, 2000 or XP with USB a Pentium 200 or faster processor and a minimum of 24 MB of RAM. These are relatively modest requirements and, as usual, a faster processor and more RAM are recommended.
Test and measurement
Windows compatible; hardware interface; signal analyzer; FFT analyzer; continuous time analyzer; signal generator; multiple waveform styles; programmable
Prism Sound at 973-983-9577,
All signal generation and analysis in the dScope III is done in the digital domain and on a two-channel basis. The selected generator output signal is simultaneously available at the analog outputs via D/A converters, and at the digital outputs. Unique to the dScope III is that the two generator outputs can be independently chosen allowing for different waveforms, amplitude, and frequencies. A wide choice of waveforms are available including sine, square, ramp, noise, and complex waveforms such as bursts, pulses, MLS, and twin multitones. Additionally, there is capability for user-defined waveforms.
Arrangement of the signal measurement and analysis section consists of three functions. These are the signal analyzer, continuous time analyzer (CTA), and the FFT analyzer (FFTA). The signal analyzer selects between analog and digital inputs and measures the RMS amplitude and frequency of the incoming signals along with the phase difference between the two incoming signals. Amplitude references along with impedance reference for power measurements are set in the signal analyzer window. The CTA provides for the usual measurements of THD+N, bandpass, bandreject, crosstalk, IM distortion and more. It operates continuously on both Analyzer channels. On entering the CTA, the input signal is filtered according to the selection of high-pass, low-pass and weighting filters. According to the selected function, the signal is then passed to the peak-detector either directly, or via a band pass or band reject filter, or via the SMPTE IMD demodulator.
Operating by capturing a buffer of audio samples, the FFT analyzer then processes these to display the results as Scope (amplitude vs. time) and/or FFT (amplitude vs. frequency) traces. FFT size can range up to 256K points with a broad selection of window functions including industry standard and high-performance proprietary types. Of interest, it is possible to measure up to 40 simultaneous FFT detectors (FFTD). Each of these can perform different measurements on the input data, with different filter settings if desired. A bin summation process essentially emulates many of the functions of the CTA allowing such interesting measurements as summing the harmonics of the input signal for a THD reading without noise or measuring a particular signal harmonic.
For the plotting of curves in the trace window, the sweep setup panel allows for a wide variety of parameters to be used as sweep sources in addition to the usual frequency or time. In like manner, a wide variety of parameters can be plotted on the vertical axis with up to four results plotted simultaneously. Upper and lower limits can be applied to traces with various alarm actions being initiated if the limits are exceeded.
Two pairs of assignable monitor output BNC connectors are provided, one monitoring the signal generator, the other the signal analyzer. These can be set to monitor a wide range of functions, including CTA residual (e.g. THD+N) or digital carriers. An assignable loudspeaker and headphone feed can also be assigned to the analyzer or generator monitor functions.
A major feature of the new dScope III is that it is programmable using Microsoft Visual Basic scripting (VB Script) language. This allows for creating automatic testing sequences with limits on tested measurements.
Scripts allow for passing data between Window applications. For example an automation script could first perform a pre-arranged series of tests and then log the results in an Access database. Scripts are easy to write in the dScope III’s Script Editor. dScope III variables can be “dragged” from a hierarchical tree of options and when needed, allowed named constants are prompted. Alternatively, VB Scripts can be written using any text editor.
The digital measurement capabilities of the dScope III are extensive indeed. In addition to the measurement of the audio properties of incoming digital signals, the instrument can measure and display source and cable-induced jitter components, sample-rate, carrier waveform (eye diagram), and amplitude. Further, it can simulate lossy cables and/or jittered sources by the digital generator, routing of the demodulated jitter signal to the FFT analyzer for viewing of the jitter spectrum and full support for channel status generation and analysis.
The user interface is said to be carefully designed to be simple and intuitive whether the job in hand is basic or complex – more on this shortly. A configurable toolbar lets you choose which shortcut tools you have available at any time. A configurable “user buttonbar” lets you run user scripts or load configurations at the touch of a button. The dScope III desktop has several “pages,” which can be quickly switched – this eases space problems on the desktop and allows different measurement modules to be set up on different pages for quick page swapping.
Calibration of the dScope III is done in software and requires no hardware adjustment as calibration coefficients are stored in non-volatile memories in their respective modules – a handy feature.
Indeed, an impressive array of features for a formidable piece of measuring gear. I suggest checking out PrismSound’s web site at www.prismsound.com to more fully check out the dScope III features and to examine a good array of screen shots of various dScope III software windows.
I installed the dScope III software on three different computers. The first was my main desktop machine, a 600 MHz P3 with 256 MB of RAM. The second was on my laptop, a 450 MHz P3 with 128 MB RAM. My intention was to use the laptop to run the dScope III in my lab to measure various things. The third machine was a much older laptop with a 133 MHz P2 processor and 32 MB of RAM. The software ran OK on this older machine but I never ran the hardware as this machine lacked a USB connection. After some considerable initial fooling around, I began to get the feel for the dScope III. Although it is similar in lots of ways to the Audio Precision equipment that I am very familiar with, the way that the dScope III does things is enough different that I feel that a considerable learning curve is required (at least for me) to get real good at running it.
I set out to measure some common things I do all the time like frequency responses, THD+N vs. power output of power amplifiers, and such. Right off, I found that it did not seem possible to plot distortion vs. measured power on the horizontal axis as I can with my Audio Precision System Two Cascade. I corresponded with PrismSound about this and was informed that there hadn’t been any particular call for this capability and therefore wasn’t included in the software. However, being the programmable machine that it is, a script was written for me to try out. Unfortunately, it was written in a newer version of the software than the V1.0 released version and wouldn’t run with my setup. This is not to say that ultimately, this test and many other ones can be made to run on dScope III.
The standard way of the dScope III software coming up at start up is shown in Figure 1 for page 1 of the five pages of the interface. Page two has a full sized trace window, page three has the sweep setup and settling windows, and page four has digital related windows, input and output channel and bit characteristics, and carrier input and output properties. One FFT detector window was added here to show what one of those looks like. This is at 1024 by 768 resolution. A word on screen size and resolution. I found that a 1024 by 768 or greater resolution was best for viewing the basic module windows. With lower resolution, like in my laptop’s 800 by 600 pixels, one spends too much time using the scroll bars and reducing modules to clear clutter. Of course, the ability to “page out” one’s desired windows to other pages help in this.
Figure 2 is an illustration of how one can have multiple plots in the trace window with different horizontal and vertical quantities being shown. In this figure, we have the fundamental and harmonic distortion residue of a small lab amplifier being shown both as time and frequency displays. An illustration of the multiple FFT measurements is shown in Figure 3. Here, we have the spectrum of a 5 kHz square wave being shown in the main trace window. Along side are three FFT reading windows that I dragged and dropped from individual FFT detector windows. After I got the expanded meter reading windows, I reduced the detector windows. In this particular example, we have the total harmonic distortion, second, and third harmonic values in the three reading windows. Pretty slick!
Figure 4 is a plot of the frequency response of the lab amplifier as a function of output loading for open circuit, 4 ohm and 8 ohm loads. Needless to say, a subsonic peak occurs in this design for loads above 4 ohms. Next, I set up the digital output for a full scale 1 kHz sine wave audio signal on the unbalanced S/PDIF output and fed it into a Crystal CDB4328 D/A converter development board. I then set in a jitter signal of 80 ns at a 500 Hz frequency.
Setting up the trace window for an FFT of the audio output of the D/A, I overlaid the unjittered and jittered traces as shown in Figure 5. The effect of this type and amount of jitter was to add sidebands of plus and minus 500 Hz around the main audio signal of 1 kHz. Note that the trace window legend has notes about the traces that can be entered there. Lastly, to illustrate the use of one of the supplied user bar programs, I invoked the multitone program, and ran the analog back-to-back (ties the generator to the analyzer internally) test. The results in the trace window, Figure 6, indicate that the internal resolution of the dScope III could resolve distortion components on the order of several parts per million. For the digital back-to-back test, the resolution, of course, is much greater with a distortion noise floor of greater than 170 dB.
I did have a few things that I felt weren’t so wonderful about the dScope III.
Due to its small size, there are no dual banana plug inputs on the generator outputs and analyzer inputs. However, neither does the small Audio Precision ATS-2. This was a minor inconvenience for me as I do use dual banana plug leads a lot for connecting to the analyzer inputs of instruments. I had to make up short dual banana to XLR adapters to use some of my usual test leads.
Also, starting the program sometimes would fail and the unit’s power switch would have to be turned off and then on again. This happened with the two computers that I ran the hardware on.
When a configuration file is loaded, it comes up with the measurement going. Further, if a user toolbar program is initiated, it starts the scripted measurement regardless as to whether the values are appropriate for the device being measured. I didn’t find a way to alter this.
And in trying to set the parameters for a trace in the trace window, say to be a plus and minus value for a time waveform, the dialog box for entering values would not allow for minus values! Bumbling along and putting in something like 0 and plus values and then moving the trace down with the trace window controls ended up by having minus values in the dialog box.
The dScope III is a great piece of measurement gear considering its attractive price – compared to some of the major competition and its extensive capabilities, it ought to find its way onto many measurement lab benches and production lines. Being very interested in audio measurements, I wish I had one long enough and the time to really learn all of its ins and outs.