London Power ad

[-]
Kits & Books

[-]
Search the Forum








(Advanced Search)

Oscilloscopes for Audio
#1
Hi Guys

Being an analogue guy I've always had analogue oscilloscopes. My first one was a Philips 50MHz 2-ch that I bought new right from their warehouse for about $1500cdn. That was 1984 and it worked until about 2012 or so. I used a borrowed scope for a while, but it was old and temperamental, then bought a new GW Instek GOS-620 which is 20MHz 2-ch with a simple front panel. It worked for 18-months and then the vertical positioning went awry. A sine wave in the bottom 2cm of the display would look okay, but if you moved it upwards using the vertical position control, the top of the sine would distort, then flatten to zero a centimetre before the top of the screen. There was no way to accurately look at large waves or complex waves.

When I bought the Instek, it was about $560cdn and i looked at DSOs, as well. A DSO is a digital storage oscilloscope. They were initially designed and optimised for looking at digital systems where wave forms were mostly square waves and strings of pulses. None of that requires much vertical resolution, which was fortunate because at the time DSO were being born, analog-to-digital-converters (ADCs) were expensive to make. ADCs got better with time, faster, stronger - not bionic, though! - and are at the heart of 90% of all DSO still made. But now, you can get bandwidth out to 100s of GHz and memory depth unheard of a decade or so ago. 

Initially, DSOs had a latency issue, where it would take a short amount of time to build up the waveform before the scope would display it. Modern computer technology is blindingly fast and the latency is pretty much gone The only time you would have an issue is if you set the time base to be many 10s or 100s of seconds - something you could not do with an analogue scope - so, nothing to worry about for what we look at in guitar and hifi amps. I was concerned about that latency a year or so ago when I decided to go with the Instek CRT scope. I did not know what many of the DSO specs meant until the Instek partially died and I had to re-investigate scopes and figured I'd get a DSO this time.
Reply
#2
Oscilloscopes for audio - part-2

When I first started looking at DSOs, I was seeing inexpensive units in the $2-300 range. Most of these had 5.7" displays, 8-bit vertical resolution, and bandwidth typically 50-200MHz. I wanted 2-ch rather than 4-ch and did not need digital inputs for looking at low-voltage computer circuits. Most DSO have data-logging and decoding built-in, which I don't need. The tiny displays and the overall look of these scopes seemed a bit "toy-like". Then I started seeing units with 7" WVGA displays - "wide VGA". The display resolution is higher and there is more room on the display for the wave and the data around it. usually the scope tells you some of its settings onscreen, along with info about the wave itself.

Then I saw units with 8" displays and some that had VGA output connectors. For my wonky eyes these were features I could use. I also found units with 12-bit and 14-bit vertical resolution. "8-bits" is 2 raised to the power of 8, which is 256. This means there are 256 vertical steps or divisions in the vertical height of the wave, or for the limit of voltage range the wave must fit into. 12-bits is 2-to-the-12, or 4096 - 16-times as many points of resolution. 14-bits is four-times better again.

If a manufacturer is not head-lining the vertical resolution then you can safely assume it is an 8-bit machine. There is an interesting Keysight (formerly HP - Hewlett-Packard) scope that is rated at only 50MHz but has an arbitrary waveform generator (AWG) built in and software that allows the scope to display Bode plots - frequency response of an amplifier. This was pretty intriguing, but it is a $900+ machine compare to the base unit without AWG for about $630cdn. The display is only 7" and most of the features are given Keysight trade names that really don't mean anything unless you are fully immersed in the HP culture.

The 8" display units and the 12-bit+ units were by OWON. They have several ranges of DSO and at first it can be a bit mesmerising navigating through it all. Fortunately, they have one of the best websites of any test equipment manufacturer, complete with all kinds of little videos that show what their scopes can do. In fact, those videos teach you about DSOs - period - so are very useful even if you choose some other brand.

Ultimately I chose an OWON XDS2102A which is 12-bit, 100MHz, 2-ch plus trigger, 8" WVGA plus VGA output. Almost every DSO has a USB connector for storing or transferring waves, and a USB port for connection to a computer. This model has a Pictbridge output for printing and LAN for alternate connection to a computer. There is a short printed manual that explains the basics and gets you started - great for a simpleton like me - plus a CD that has a detailed manual. This unit cost about $560cdn including shipping from China by DHL. It's a fantastic bargain even without the free shipping!

Other OWON models to consider are the XDS3102A and XDS3202A which have twice the memory depth and 50% faster waveform speed. They can have the AWG, a multimeter on the rear panel and a data-logger, so they call these an "N-in-1" DSO, because it is like 2,3,4 pieces of test equipment in one box.
Reply
#3
Oscilloscopes for audio - Part-3

The OWON XDS2102A is bigger in person than the photos suggest and the 8" screen is divided into 10 vertical divisions by 16 horizontal (1cm graticule with further subdivision by ten). The unit is about 2.6kg (5.72-lbs) and feels hefty. One nice thing about a DSO is that since it is a piece of industrial equipment that is designed to measure tiny high-frequency signals, it must be fully shielded. This means its own noise-generating CPU and other circuitry cannot spray EMI into the environment. You can get the unit with a North American mains plug or a European one. I bought mine from Hilitand Instruments Store on Aliexpress.

The OWON showed up some issues with my signal generator inasmuch as the generator output is a little drifty until it has been running for quite a while. It's over twenty years old and has a frequency counter as well. The counter's two least-significant digits fluctuate constantly, reflecting either the counter's inaccuracy, the generator's instability, or both. The scope can lock the waveform on its own depending on how you set its triggering, or if you set it to catch the wave in memory.

With a music signal or guitar input, the DSO does show a small delay in its response which I was expecting. It is large enough that you could find it annoying or not? For most tests we do with the oscilloscope we are using continuous signals and the small latency is a non-issue. For example, when the generator was still drifting, the wave on the scope display would be stable but a ghost wave would appear every few seconds or so. This is very "analog-y" and you can even get the scope to show a doubled wave -an inverted wave over the noninverted wave - just as if it were a CRT scope with the trigger not set properly - you won't feel like you're missing analogue for very long - haha.

Generally, the DSO will lock the wave form right away. You can have aspects of the wave displayed, such as the peak value, RMS value, or its period (time for one cycle). There are usually "zoom" functions, as well, where you can set cursors vertically and/or horizontally and then see that part of the wave expanded in an alternate view. This is where the memory depth and vertical resolution work together to show fine detail. You should have a look at the videos on OWON's site.

One thing that the XDS2102A excels at, as I'm sure most DSO do, is showing high-frequency noise. For example, switch-mode power supplies operate at high-frequencies and produce noise output at much higher frequencies. Things get worse if secondary SMPS modules are powered from the first and "beat" frequencies are created that get into the grounds and power paths. These noise pulses feed right through any post-regulator circuitry there might be and intermodulate with the audio signal. The result is an increase in general noise - hiss - but with a an "edge". Where a regular scope may not show this, the DSO does.
Reply
#4
Oscilloscopes for audio - part-4

If you get a DSO, it will come standard with probes that are switchable from x1 to x10. The default expected setting is x10,and this will provide the maximum bandwidth for the scope-DSO combination. For example, the probes that come with the OWON XDS2102A are rated for 100MHz at x10 but only 6Mhz at x1.

Most DSOs are limited to just a 10V peak signal at the BNC input jack, so a x10 probe allows signals up to +/-100V peak to be seen. This includes any DC that might be present. If you are working with tube amps it is a good idea to get a x100 probe. Fortunately, these probes are all reasonably priced from around $15-30cdn and have bandwidths that match the oscilloscope. Some scopes will auto detect the x100 probe.

When you first set up the scope and connect the probe, you use the Utility menu to tell the scope the probe setting. This lets the scope indicate the correct volts-per-division on its display, saving you from having to multiply the reading as you would with a CRT scope.

Use the calibration oscillator output and the supplied adjustment tool to tweak the probe capacitance in x10 mode. An under-compensated probe will show spikes on the leading edge of the square wave; an over-compensated probe will have slightly down-turned leading edges; the proper "Goldylocks" setting will be just right showing a perfect square wave. To help with the alignment, you might want to move the wave vertically using the Vertical position knob so that the top of the wave lies on a graticule.

If you are trying to use the full resolution of the scope, use only one input channel at a time. The sample rate and record time are also maximised. The short printed manual will get you started but modern DSOs come with a CD with the complete manual. You can load this on to any computer with a disc drive and it has a lot more detail and easy-to-follow instructions - a lot to take in, and usually a lot of stuff you will never use. There is also software for analysing wave forms that you can capture and store on a USB key, or send via USB or LAN from the scope to the computer. This allows you to do the analysis later, elsewhere or just save the waves in a second place. You can print them from your computer or from the scope if the scope has Pictbridge output. (Pictbridge is an interface standard for digital cameras to print directly without having to download the files into a computer).
Reply
#5
Oscilloscopes for audio - part-5

One of the features of a CRT scope is that the intensity of the displayed wave form suggests that the wave is stable; the trace is being re-written over and over as the beam is swept across the screen. If there were occasional variations, these appear as "ghosts" behind the solid wave. The first DSOs did not offer this kind of information to the operator of the equipment, but new ones do.

The effect is referred to as "persistence", which is the amount of time that the phosphor glows after the electron beam has left that part of the face of the cathode-ray tube. You can exaggerate the effect by turning up the beam intensity, but this tends to obfuscate details of the wave. Later $$$ CRTs had variable persistence.

There are two ways to achieve a persistence effect with DSOs. The easiest and least expensive method uses the stored wave forms and displays the older ones in a lower intensity than the newest captures. This is accessed in a menu, where the Persist time can be set to 1, 2 or 5 seconds, infinite, or off. There is a Clear button to erase the screen and memory buffer and a new acquisition will begin. This method is used by the OWON XDS2000 and 3000-series DSOs, as well as most of Keysight's line, and many others.

The second method uses additional hardware, which of course adds cost, so these DSOs are called "DPOs" - digital phosphor oscilloscopes. They do not actually have a phosphor display, just a standard LCD; rather, they have a second data path to the display that allows both stored and live waves to be displayed. DPOs tend to be only high-bandwidth scopes.
Reply
#6
My addiction to fine lab instrumentation goes back decades, and have grown up using Tektronix 547 Hybrid Tube/SS scopes with 1A4 4-Ch plugins, a beautiful Tek R5031 Dual Beam Storage Scope (non-plug-in 7000 series instrument), having two 10uV-10V differential inputs + current probe inputs.  For years, in spite it's meager 1MHz BW (typical of their high sensitivity Diff Amps produced in their 1A7A, AM502, 7A22 and others), that scope with its phospher storage ability served as a Frequency Analyzer, driven from a Wavetek 185 Log Swp Gen, an HP3575 Gain/Phase Meter, so I was seeing dB vs Log Freq over a wide range...usually just running 20Hz-20kHz with 1 oct/Div displayed, and vertical of 5dB or 10dB/Div.  

It also worked really well in Acoustics, where I'd modulate the VCG of the Wavetek 185 to produce warbled-sine sweep, +/- 1/6 oct, and produce great detail that aligned perfectly with 1/3 oct analyzer plots, but showing all of the room details.  The Acoustic front end using a Bruel & Kjaer 1/2" 4134  condenser mic feeding a B & K 2607 Measuring Amp to drive the Y-axis of the scope with it's log DC output for 5 or 10 db/div displays (or greater as need dictates).  When I needed hard copy, I also had an HP 7045 XY plotter attached.

I still use that aging R5031 scope, but my main instrument on my present bench is a Tek 7633 Storage Scope with 7A26 2-Ch plugins.  It still allows X-Y display functions, for my Semiconductor Curve Tracer, Freq response plots & Impedance Plots for quick-and-dirty measurements.  I did invest in a LeCroy 7200A Color Digital Scope that has amazing performance, though needs to live on a Tek wide-body Scope Cart.  It's one of the few Digital scopes that offer's XY facility.  Most Digital Scopes lack that feature.

At present, the only USB-based digital scope/generator package I have is the Velleman PCSGU250.  I mostly use that connected to my laptop for generating Bode Plots with it's modest internal Sweep Generator, which lets you do Frequency Response plots out to 1MHz in Log Sweep mode, in 1/n octave steps from 1/1 thru 1/24 Oct increments.  Usable dynamic range is around 50-60db.  I buffer the front end of the Velleman with an Amber 3501a Audio Analyzer, so I have a wide range low noise differential input.  It allows annotating the plots, then exporting them as a pdf file to add to my growing database.  

I was alerted to a really nice device from Digilent ....their Analog Discovery 2.  This USB device isn't packaged as a stand-alone lab instrument...you have to do that task yourself, but what it contains is absolutely amazing, and will be my next acquisition.  It's scope function offers 2-Ch of 14-bit +/- 25V differential inputs 100MS/s, 30MHz+ BW, a 2-Ch ARB Function Gen, Network Analyzer 1Hz-10MHz, Spectrum Analyzer, Logic Analyzer, Pattern Analyzer, 2 programmable power supplies, etc.  Find this at:  https://reference.digilentinc.com/refere...nce-manual.  That package sells for under $300.  After proper packaging so it would be like another robust lab instrument on your bench, it would probably cost another $50-$150, depending on how you like your instruments to be.

When I travel to job sites, I have some Tektronix TM-500 series instruments...one being the SC504 80MHZ 2-CH SCOPE.  Tiny CRT, but it is in a 3U x 2-module-width box, so has to be small.  I normally have an AM502 Diff Amp in the package, though usually just use the spare Amber 3501a Audio Analyzer for the front end.

I've been buying top of the line used instruments for decades, way before the days of ebay.  Being a native of Los Angeles, CA, one of the gold mines years ago was the monthly TRW Radio Amateur's Techincal Swap Meet in Redondo Beach, where retired recent tecnology would hit the parking lot at pennies on the dollar for what industry paid.  That's been the story of my life....usually one or two generations behind current technology, so as wealthy industry and military unloads their 'then-current- instrumentation' to buy the new generation, it allowed average folk to buy instruments normally costing thousands of dollars for pennies on the dollar.
Reply
#7
Hey nevetslab, that's a lot of cool equipment!

Anyone in LA or San Francisco has an easy time finding great test gear and surplus parts.

I looked at the AD2 and decided against it as i do not want a computer on my work bench. Maybe if I had more space, but not now. Hearing from people who have used both PC-based test equipment versus stand-alone test gear you get a sense of the priorities of each group, and there are certain undeniable problems with the PC approach primarily with the number of boxes that must be linked together and the latency of the displayed data. Obviously, a PC can have analysis software that is far-reaching, but there is a history of lack of support for the software and the effective short useful life-time that causes.

Some of the PC-based scopes seem particularly toy-like with a very high DIY factor, where the connections to the add-on unit are on the flimsy side and the core unit is not even shielded. Fine for a uni student who is tinkering with arduino or other digital stuff, or who has access to a proper lab. I looked at these types of things over the years and computers have gotten a lot smaller and displays much thinner, so easier to accommodate on a work bench, but still too many software issues and latency. For some tests, latency is not a problem, and for low production throughput, same deal.

The weakest link in the soundcard-based scopes and analysers is the sound card itself inasmuch as its input voltage capability is very limited and not calibrated, and the 3mm TRS connnectors are not robust. Years ago, a friend of mine and I tried out the Dr.Jordan MLS but I never had much time to work with it and he never got it to work properly for what we wanted to measure. I'm sure things have improved since then, certainly sound cards have gone from 24-bit to 32-bit although there are questions as to whether the extra bits are fully useful? bandwidth is still quite limited, although there are methods of "down-sampling" high-frequencies to the bandwidth available using rather simple circuits. Again, something I've read about but not tried.

I still prefer the aesthetic of stand-alone test equipment.The DSO does what I need at the moment and just requires that I learn how to use some of its features but is otherwise a ready-to-go package.
Reply
#8
(11-03-2018, 08:04 PM)K O\Connor Wrote: Hey nevetslab, that's a lot of cool equipment!

Anyone in LA or San Francisco has an easy time finding great test gear and surplus parts.

I looked at the AD2 and decided against it as i do not want a computer on my work bench. Maybe if I had more space, but not now. Hearing from people who have used both PC-based test equipment versus stand-alone test gear you get a sense of the priorities of each group, and there are certain undeniable problems with the PC approach primarily with the number of boxes that must be linked together and the latency of the displayed data. Obviously, a PC can have analysis software that is far-reaching, but there is a history of lack of support for the software and the effective short useful life-time that causes.

Some of the PC-based scopes seem particularly toy-like with a very high DIY factor, where the connections to the add-on unit are on the flimsy side and the core unit is not even shielded. Fine for a uni student who is tinkering with arduino or other digital stuff, or who has access to a proper lab. I looked at these types of things over the years and computers have gotten a lot smaller and displays much thinner, so easier to accommodate on a work bench, but still too many software issues and latency. For some tests, latency is not a problem, and for low production throughput, same deal.

The weakest link in the soundcard-based scopes and analysers is the sound card itself inasmuch as its input voltage capability is very limited and not calibrated, and the 3mm TRS connnectors are not robust. Years ago, a friend of mine and I tried out the Dr.Jordan MLS but I never had much time to work with it and he never got it to work properly for what we wanted to measure. I'm sure things have improved since then, certainly sound cards have gone from 24-bit to 32-bit although there are questions as to whether the extra bits are fully useful? bandwidth is still quite limited, although there are methods of "down-sampling" high-frequencies to the bandwidth available using rather simple circuits. Again, something I've read about but not tried.

I still prefer the aesthetic of stand-alone test equipment.The DSO does what I need at the moment and just requires that I learn how to use some of its features but is otherwise a ready-to-go package.

All good points, K O'Conner.  

At present, I also don't have room on the bench for a computer.  When I do use that Velleman box with the laptop, it's on an adjacent bench.  I never did get started with LabView-based gear, or really set up a lot of the GPIB boxes, all of which are far more serious instruments & hardware.  For my daily dirt stuff, my basic gear is a GenRad pink noise generator, either thru a GenRad Tone Burst Generator/Khrone-Hite Filter or a B & K 1027 Sine-Random Pink thru a 1617 1/3 Octave filter to drive the back-line gear, powered speakers, etc...and use my ears & a Fluke 8060A to probe with.  And of course one of my basses to plug into what's getting serviced.  Most of the time, all that's powered up is the pink noise gen & the AC Mains power analyzer used with the 30A Variac.  

That AD2 device still intrigues me.  One of my clients who works below the water line with hydrophone arrays has started using that, though I choked a bit seeing the kluge of cheap clip leads everywhere, instead of something like we're used to using.  Still the results he and his colleagues were getting was commendable.  I can tollerate a USB cable connected to a computer, and the working side of where that USB cable is tethered would be like anything I've been using for years...

An articulate arm-based monitor and a pull-out keyboard drawer from below the bench top really takes care of putting the computer into the system.  I just haven't allocated that yet.  I'm more used to getting out of the bench chair to go sit at the desk to type in my notes and grab some more coffee from the pot next to the computer.
Reply
#9
Hi nevetslab

There is a Picoscope, I believe, that is 16-bit and a more complete box with BNCs, reasonable-ish bandwidh and priced around $1,500cdn or so. I considered that more seriously than the AD2, but again, the whole PC issue impedes my diving in.

I have some Radford and HP THD analysers and LDOs but most of what I design for hifi is below their resolution. Even the APs cannot read the THD, which makes the "if I won the lottery" prospect of buying an AP even seem pointless to some extent. (A friend has a big manufacturing concern and has a dozen APs, so I was able to have some things tested to see if their was correlation with the sims).

But day-to-day bench work really just needs a scope and sine generator and working volt meters, ammeters and ohmmeters.

I know a lot of hobbyists who only have a multimeter and that is sufficient to verify most things are working in a guitar amp. A scope is ideal for seeing that there is proper symmetry of output and for fine tuning interstage attenuators, as The Ultimate Tone, Vol.5 (TUT5) discusses. The latter can be done by ear.
Reply


Forum Jump:

[-]
Come in where it's warm!
A warm welcome to those interested in tube and hi-fi audio! Fans of Kevin O'Connor's The Ultimate Tone (TUT) book series form a part of our population. Kevin O'Connor is the creator of the popular Power Scaling methodology for amplifiers.
Please remember these three principles: respect, sharing, community.
Not familiar with The Ultimate Tone book series? See discussion topics, or click here to visit London Power/Power Press Publishing.

[-]
Forum Hosted by London Power
London Power logo