Jump to content

PHONIC PAA6 Personal Audio Analyzer


Anderton

Recommended Posts

  • Members

And now, for something completely different: Phonic's PAA6 Personal Audio Analyzer. This is a pro-oriented toolkit for anyone doing audio analysis, from showing waveforms on a scope, to checking speaker polarity, measuring sound levels, checking phase, measuring THD+noise, and...well, we'll find out during the course of this Pro Review.

 

Normally this kind of test gear is quite expensive, but despite some very pro aspects (love that color touch screen!) the PAA6 lists for $1,699.99 and the street price is around $1,200, making it affordable for serious studios, PA setups, bands looking to ring out the PA properly before a gig, and people who review gear and want to be able to say more about the specs than "It sounds pretty quiet."

 

The first attached image shows what you get in the package. Clockwise from the left, there's the:

 

* Phonic PAA6 itself.

* USB cable for connecting to a computer

* Mic stand adapters

* Padded, high-quality case for taking this puppy on the road

* Lithium-ion battery

* AC adapter

 

The second attached image shows the gorgeous touch screen. This is your gateway to accessing the various functions, and it's super-obvious: I think most people would realize that if they want to check polarity, they'd simply touch the Polarity icon.

 

I'll include as many screen shots as possible during the course of the review, but it's not particularly easy as the screen does have a certain amount of reflectiveness; if I shoot too close to the screen, you see the reflection of the camera. This reflectiveness is not a problem in use at all, it's just an issue when I'm trying to light properly and shoot.

 

Anyway, that's the basic idea of the unit. Let's look at the interfacing a little more closely.

Link to comment
Share on other sites

  • Members

There's not a lot of complication to the PAA6 connections, and so far with what I've checked out, the interface is extremely easy to navigate.

 

The first attached image shows the front of the unit, which provides access to the AC adapter jack (which I would have preferred on the rear, but no big deal), USB interface, and SD card slot (presumably for saving settings and such, I guess I'll find out soon enough). The large flange along the bottom opens up to reveal the battery compartment.

 

The second attached image shows the audio connections. From left to right, there are two XLR line inputs and the XLR signal generator output. I would like to have seen the inputs use combi XLR+1/4" phone jacks given the amount of gear in project studios that uses 1/4" phone balanced I/O, but of course, it's a simple enough matter to carry a few adapters.

 

The things that look like black crayons sticking out of the unit are the mics used to take measurements. These fold back into the case for easy carrying, or swivel out when in use. Including microphones is essential in a unit like this, as the design can compensate for any anomalies in the mic response, thus leading to more accurate measurements.

 

The third attached image shows the essence of the user interface other than the color touch screen (you can access all features and functions from this screen, which is pretty cool. It's really quite simple...again going from left to right, there's what Phonic calls a Directional Control that's like those mini-joysticks you find on laptops.

 

The next button over is the Run/Stop button, which does what you'd expect: Start running a function, like a particular test, and stop running the function.

 

The next-to-last button is the Signal Generator button, which starts and stops signal generation. Finally, there's a Power button. About the only unusual aspect regarding this is that turning on and off requires holding the button down for a couple of seconds. This helps prevent turning the unit on or off accidentally.

 

And that's it for our basic tour of the unit. Next up, let's look at its various functions, and how they work. I'm really looking forward to this - I've needed something like the PAA6 to check out gear for a long time, and I expect it will make enable quantifying the performance of various devices much more easily.

Link to comment
Share on other sites

  • Members

The Signal Generator section can be thought of as a "universal" function, because it's available in all the other functions. That makes sense; if you're using, for example, the Meter function, you might want to have a predictable signal available for metering....ditto using RTA, phase, etc.

 

Check out the first attached image, which shows the Meter function. In the lower right, you can see the signal generator parameter summary (outlined in red). The little red running guy shows whether the function is running (red means stop and green means go - what a shock!), and the green speaker indicates that the signal generator is indeed generating. 20Hz shows the frequency, and 4dBu, the level. The selected waveform (in this case, Sine) is in the lower right corner.

 

Note the little arrows next to the frequency and level parameters: They let you change the parameter values without having to go into the signal generator screen. If you choose a swept frequency range, the frequency parameter isn't editable but when the signal generator is running, it shows the frequency as it's being swept.

 

As to what exactly you can generate, refer to the second attached image, You can see the choices are sine, triangle, and square waves, as well as pink and white noise. Polarity is a special waveform when you're in polarity check mode, and sweep creates a wave with start and end frequencies. So far, so good.

 

The third attached image shows a frequency being selected. These are on standard ISO 1/6th octave intervals, which should be enough resolution to do the job.

 

However, also note the scroll bar at the side. Although the PAA6 includes a stylus for selecting details on the touch screen (for most functions, you can just touch with your finger), I found scrolling to be somewhat unresponsive. I expected the "pseudo-joystick" control in the user interface to allow going down the list, but it doesn't. If you can just touch a category and touch a value, the response is good. But when scrolling, there's room for improvement - and scrolling is something you really can't do with your finger.

 

While we're on using the stylus, I assume it's a good idea to add a clear plastic overlay on the screen to protect the touch surface, as people do with cell phones and Nintendo DS games. I happened to have an overlay for my Palm Centro (okay, it's not an iPhone - go ahead and laugh) and its fits the crucial parts of the screen. It didn't seem to affect response one way or the other. I'd be curious if Phonic recommends adding this kind of overlay.

 

Moving on to the fourth attached image, you can choose a preset gate time when you press on the interface's signal generator button or touch the speaker symbol onscreen. This is variable from 100ms to 10 seconds, but you can also set a "Continuous" mode where once you start the signal generator, it will run until you tell it to stop.

 

The fifth attached image shows one of my favorite features, the sweep generator. You can set the sweep range to step through specific intervals (e.g., 1/6 octave, 1/3 octave, 2/3 octave, 1 octave) and specify how long the signal generator should hold at each step, or enter a continuously variable sweep from a starting to an ending frequency (as shown in the screen shot). You can also set the number of repeats.

 

The continuous sweep sounds like a log rather than linear scale, which is fine with me - I hardly ever use a linear sweep (come to think of it, I can't remember the last time I did). However, if this is something users wanted, I see no reason why the on/off field next to sweep couldn't be log/lin, as you can turn the sweep on and off with the signal generator button anyway,

 

And that's pretty much the story on the signal generator. It was instructive to feed the signal into various digital audio interfaces and hear what happens if you overload the input enough to cause distortion - you can really hear the evils of foldover distortion in action! If you do this, you'll never hit the input hard again - there's a good reason why many people think it's a good idea to treat -6dB (and possibly even less) as 0dB.

 

Sweeping speakers is also fun if you want to hear what happens to the frequency response if you're room isn't well-treated. On the other hand, the Phonic PAA6 confirmed what I've always suspected: IK Multimedia's ARC really can help improve the response of less-than-ideal speakers in a less-than-ideal environment.

 

There's also a whacked-out application for the noise generator that people think is insane until they try it: Injecting some relatively high-level noise into a mix while mixing. This provides the same kind of function as listening to a CD in the car (with the road noise, engine noise, wind noise, etc.) to hear how the mix "really" sounds. What this actually does is mask the low-level sounds, so you can tell which tracks stand out, and which disappear into the background. You can set levels all the way down to -40dBu, but this technique is most useful with higher levels. Of course, remove the noise before you create your stereo master...

 

Finally, I was really happy to note that I can still hear a 16kHz tone. All those years wearing earplugs on stage paid off, and I'd advise others to do the same!

Link to comment
Share on other sites

  • Members

Well, it didn't take long for the PAA6 to prove its usefulness for troubleshooting. And I also found out about the Secret Tenth Function. Let me explain...

 

I wanted to check out the Polarity function. Remember that the PAA6 has a signal generator output as well as mics and line inputs, so you can generate a signal from the PAA6, feed it through something, and monitor its outputs with the PAA6.

 

Checking Polarity is simple. You call up the Polarity function, which automatically sets the signal generator to the polarity checker waveform - a negative-going sawtooth (this waveform makes it easy to detect phase).

 

Then you just feed the output into what you want to test (like a speaker), hold the mic up to the speaker, and see if the PAA6 display shows positive or negative.

 

I thought I'd check my system through from Cakewalk V-Studio I/O input to ADAM A7 speaker out. So I inserted the signal, read the output, and...

 

Check out the first attached image, which shows that Mr. Polarity Man (you gotta love a peace of pro audio gear with a cartoon character interface!) was not happy - polarity was reversed somewhere along the line.

 

So, I thought I'd plug directly into the speaker. The A7 has two inputs, unbalanced RCA and balanced XLR. It had been plugged into the unbalanced out, so I decided to plug the PAA6 signal generator directly into the A7's XLR in. Lo and behold, Mr. Polarity Man was happy (see the second attached image).

 

So...it must be that the speaker was okay and there was something wrong in the rest of the system, right?

 

Wrong! I tested the Cakewalk interface, and it handled polarity perfectly from in to out (remember, you don't need to use the mics to bring a signal into the PAA6, you can monitor from the line ins). After a little more testing, I found the source of the problem: the the ADAM A7's unbalanced inputs were wired out-of-phase, but the XLR ins were wired correctly.

 

This also explained something interesting. When I reviewed the speakers for EQ magazine many moons ago, I used XLR connectors but when I added the PreSonus Monitor Station - which uses 1/4" balanced jacks - I didn't bother making a 1/4" to XLR adapter but simply took half of the balanced out and plugged it into the A7's unbalanced input.

 

For some reason after doing that, the speakers didn't sound quite the same. I chalked it up to acoustics - I moved the speakers to a different place after testing - and didn't think about it any more.

 

But now, thanks to the PAA6, I know exactly what happened: I've been listening to speakers with reversed polarity!

 

BTW, in case you're thinking "a simple polarity change shouldn't make a difference," that's what I always used to think until I met someone who could tell by ear alone whether a speaker's polarity was correct or not (as long as the music contained drums and preferably a loud kick drum, which is what he used to key in to the polarity).

 

He explained that when a drum gets hit, it pushes air toward us, which is the correct polarity. When the signal polarity is reversed, with transients the speaker sucks air away from us and just doesn't sound the same. One thing I've noticed about systems with reversed polarity is you keep wanting to turn up the volume, but it doesn't seem to add the kind of presence you expect.

 

I've since done enough testing to agree with what he said. This effect seems particularly noticeable with headphones. I recently mastered a rock song where the polarity was inverted, and after correcting it, the drums had much more punch and presence.

 

Moral of the story: If you're using ADAM A7 speakers, go into the balanced XLR inputs, not the unbalanced ones!

 

Score a major win for the PAA6 (and thanks to Mr. Polarity Man, too :)). Now, about that 10th function...

Link to comment
Share on other sites

  • Members

This isn't something that Phonic touts, but you can check XLR cables by simply patching one between the input and output. If the XLR cable is good and properly wired, the display shows Positive. If the cable is a polarity switching cable (every studio should have one or two, but not every studio labels them!!), then the display shows negative.

 

If there's a short or open, then the display shows "?" because it can't get a lock on the signal generator signal to determine whether the signal has proper polarity or not.

 

Clearly, my next step is to make some more 1/4"-to-XLR adapters so I can use stereo and unbalanced connections with the PAA6. Meanwhile, I have to say it's impressing me more and more as I find useful problems for it to solve. :thu:

Link to comment
Share on other sites

  • CMS Author

I've had a PAA3 for several years and it regularly goes into my toolbox when I have a live sound gig. I've always wanted to get my hands on the PAA6 (which was announced nearly 3 years before they actually brought it out) but so far I've only played with it at trade shows.

 

You won't find anything these days thay has a frequency response far enough off flat to see on the screen . . . except for an equalizer. See how it does when you run it through the equalizer on a mixer, or an outboard EQ. As I recall, you can save the graphic plots and download them to a computer, so you'll have a simple way to put plots of EQ curves into your reviews. I also think you can overlay plots, so you should be able to plot a family of curves on a single graph. I've been looking for years for a program that emulates the old General Radio audio oscillator and tracking graphic plotter and haven't found one yet that I'm happy with. I don't think I want to spend $1200 to do this, but it's better than $10,000 for an Audio Precision or Prism analyzer. Maybe Phonic will let you keep it if you promise to mention it in every review you write. ;)

 

I've run across things where polarity gets reversed depending on which outputs or inputs you're using. Not all manufacturers are aware that it can matter. You might want to take the Adam speakers apart and see how they're wired, Maybe you can re-wire them. It's what I would do.

 

You'll probably find that you can hear a polarity inversion by listening to that sawtooth waveform. Years ago, I wrote an article about polarity and, to demonstrate that acoustic polarity inversion is actually audible (most people don't believe it) I put a polarity demonstration WAV file on line, made from a swatooth wave similar to the polarity test waveform from my NTI Mininrator, which I inverted halfway through the file. I got lots of flack about how it was a trick (it wasn't) but nearly everyone said they could hear a difference between the polarities. Most described the difference as one sounding as if it had a lower fundamental frequency. My long winded explanation had to do with Doppler shift resulting from the direction of cone motion at the leading edge of the waveform (something like your speculation about the speaker blowing or sucking). This seems, today, to have some credibility among crackpot audiophiles.

Link to comment
Share on other sites

  • Members

 

...I found scrolling to be somewhat unresponsive. I expected the "pseudo-joystick" control in the user interface to allow going down the list, but it doesn't. If you can just touch a category and touch a value, the response is good. But when scrolling, there's room for improvement - and scrolling is something you really can't do with your finger.

 

Your idea to control drop-down menus with the joystick is a good one; I might talk to them about adding it in a future firmware update. Don't personally know what it involves to make a change like that, but it's worth putting forward (couldn't be difficult to implement, I imagine).

 

As it is, you need to push the joystick down twice to activate it, and then you're given a square onscreen that you can use to select the zone/area on the screen you want to control. Then push it again to confirm. It's a bit time consuming, but it was made this way to stop you from accidently messing things up if you bump the joystick.

 

The joystick was added as a back up of sorts. Although, technically, you are able to access all functions with it, I always wonder: why would you? If your screen becomes uncalibrated, you can use the joystick to get through to the calibration screen. Otherwise, I don't know.

 

But I can't see the harm in activating the thing when you've opened a drop-down menu.

 

 

While we're on using the stylus, I assume it's a good idea to add a clear plastic overlay on the screen to protect the touch surface, as people do with cell phones and Nintendo DS games. I happened to have an overlay for my Palm Centro (okay, it's not an iPhone - go ahead and laugh) and its fits the crucial parts of the screen. It didn't seem to affect response one way or the other. I'd be curious if Phonic recommends adding this kind of overlay.

 

 

The small, plastic films used to protect screens are a great idea. They're particularly useful if you use your fingers to touch the screen, where oils and such can be transferred. It won't damage the screen right away, but over time it will add up.

 

Just a couple of thoughts on the screen protectors though (may be a bit obvious): first off, some are pretty thick. The screen cover for my PSP is about half a millimeter thick, and that would definitely reduce the sensitivity of the touch-screen. Also, there are a few I've bought in the past that have excessively sticky surfaces. Best to avoid those. I mean, if you can actually see the adhesive, it's probably no good.

Link to comment
Share on other sites

  • 2 weeks later...
  • Members

Real Time Analyzers are great, and terrible. Great, because they show you graphically just how much any untreated room needs correction...and terrible, because they show you graphically just how much any untreated room needs correction! When you call up an RTA and the display looks like a picture of the Swiss Alps, you know you're in trouble.

 

The PAA6's RTA is very full function. You can set the resolution from 1 octave if you just want a rough idea of what's going on, to 2/3 octave, 1/3 octave, and 1/6 octave. It's also possible to set the response time to 35ms, 125ms, 250ms, and 1 second (the screen shots were done with at the 125ms setting). You can also adjust the peak hold time, weighting (A, B, C, or flat), and display the sum or difference of the two channels.

 

Once you've set up your measurement parameters, you can then see the RTA results on the color touchscreen. The PAA6 handles a really wide dynamic range, so you can scroll down to see quiet signals, or up to accommodate really loud ones.

 

There's nothing like a practical example to demonstrate something, so I thought I'd use the PAA6 to test my room response with and without IK Multimedia's ARC room correction plug-in inserted (if you're not familiar with ARC, check out this review). I did a very quick test, using pink noise - I didn't put the PAA6 in a super-sweet spot, for example - but I was surprised how readily the PAA6 confirmed that the ARC really does work.

 

The first attached image shows the response without ARC engaged. It could definitely use some fixes, especially in the low end, where the response is quite uneven.

 

Now contrast that with the second attached image, which has the ARC plug-in enabled. As you can see, there's a big improvement in terms of the response - it's much more flat and even.

 

What the stills don't show is that the graphs are not static; there's always a little bit of motion in each frequency band, as the noise and measuring variations occur. When looking at the screen with ARC enabled, the display is more stable, and there's less movement.

 

Now that we know that the RTA works (and so does ARC!), let's look it in a little more detail.

Link to comment
Share on other sites

  • Members

Let's look at some more of what the RTA does.

 

One thoughtful feature is an "EQ settings" option. What this does is analyze the RTA analysis, and determine what kind of EQ changes you would need to add to compensate for anomalies in the RTA curve.

 

The attached image shows the curve that was generated when I measured the room response with ARC (for variety, I used the one-octave resolution). You'll note that the very low bass and very high treble need work, but that's because I used A-weighting. I ran the test later using the Flat response instead of A-weighted, and of course, the highs and lows were much flatter. There was a dropoff at 31Hz, but I'd expect that from small near-field monitors (I'm using the ADAM A7 monitors).

 

Anyway, taking a look at the curve and concentrating on just the midrange, you can see that the ARC did a great job of creating an even response in the midrange as very little correction is suggested.

Link to comment
Share on other sites

  • Members

You can save readings for the RTA, FFT, and RT60 functions in up to 10 memory locations inside the PAA6 (there's 100MB of memory), or to an SD card (formatted in standard FAT-16 format). I tested this function with a 1GB memory card; I assume there's the usual Windows file limit of 2GB due to the FAT formatting, but I didn't have a 2GB card to confirm this, or a 4GB card to see if it couldn't be handled. In any event, the 1GB worked fine, and that's a lot of memory anyway.

 

When you're in a function that let's you save, you simply call up a file menu. There's an on-screen keyboard for entering the file name (see first attached image). I don't know why it's not in the standard QWERTY format, but I assume this is because Phonic didn't want to have to include a zillion options for localized keyboards.

 

When you save, it saves what's on-screen and when you load, it shows that same screen, whether you're using the on-board memory or the SD card.

 

Using the SD card makes thing a little more complex, though. Files are saved as .TXT files, which the PAA6 knows how to interpret, but to a computer, it's just text. Supposedly Phonic will be coming out with software that allows you to use the computer as a "front end" for the PAA6 and presumably, it will be able to interpret the text file and create a display similar to the one on the PAA6...but I don't know when this software is supposed to be ready. Grant, any time frame for this?

 

There's a USB connector for the PAA6, and I figured I'd be able to transfer files from the SD card to the computer. Curiously, the computer recognized the SD card when inserted in the PAA6, but showed the contents of the SD card as no files and 0 bytes. Huh? This was on an XP machine, so I tried hooking up the PAA6 to a Mac and a Vista-64 computer. All of them recognized the card, but claimed there was no data on it.

 

However, if I took the SD card out of the PAA6 and put it in a card reader, or for that matter put it in my camcorder which has a USB interface for transferring photos, the files showed up. The second attached image shows the text file opened up in Windows using Notepad.

 

So...maybe the PAA6 needs some kind of driver so that the computer can communicate with the data, or I'm doing something wrong (not impossible!). Perhaps Grant could comment on this as well.

Link to comment
Share on other sites

  • Members

Okay, the SD card thing had me scratching my noggin for a few minutes (maybe longer). This is what I think the problem is - or at least what I HOPE the problem is: try formatting in FAT-32 or NTFS (though I don't know if that's possible on a card as small as 1GB). I wanted to format a card in FAT16 to test what would happen, but I honestly couldn't work out how.

 

It may be a case where the PAA6 only recognizes the FAT32 and NTFS file structures (why it lets you save/load, but not read through the computer I can't quite grasp; I'm passing that one to the engineers). If this is the case, there's probably an ammendment coming to the manual.

 

So onto the software. I have word that it'll be available before the NAMM show (Mid January next year). We're going to tie it in with a new firmware release that will include a 'waterfall analysis' function and delay time measurement (that's on the PAA6 itself). We're still finalizing a couple of features in the software and we may be asking a few people if they want to test the beta software - but that's not quite ready either.

Link to comment
Share on other sites

  • Members

Thanks, Grant. I'll try formatting in FAT-32 and see what happens. I'm used to most gear defaulting to the "lowest common denominator" format, that's why I went for regular FAT-16.

 

I'll report back with results, right now I'm traveling and don't have the PAA6 with me.

 

Good news about the software, too...January's not that far away...guess I'll need to hold on to it until then :)

Link to comment
Share on other sites

  • 2 weeks later...
  • CMS Author

Using the SD card makes thing a little more complex, though. Files are saved as .TXT files, which the PAA6 knows how to interpret, but to a computer, it's just text. Supposedly Phonic will be coming out with software that allows you to use the computer as a "front end" for the PAA6 and presumably, it will be able to interpret the text file and create a display similar to the one on the PAA6...but I don't know when this software is supposed to be ready. Grant, any time frame for this?

Disappointing that something for the computer to plot or format the data didn't come along with it. My PAA3 has a program that lets you operate it from a computer, save the data to the computer that's saved in memory, and print the display. Hopefully a similar program for the PAA6 will come along soon. I've been looking for a way to plot the frequency response of an equalizer (like when reviewing a mixer or outboard EQ). The 1/3 octave RTA in the PAA3 doesn't have good enough resolution to show the detail needed for this. Maybe the resolution of the PAA6 is good enough. Give it a try.

 

Here's what I get from the PAA3. This is a snapshot of what was on the radio when I was reading your latest on the PAA6

 

RADIORTA.jpg

Link to comment
Share on other sites

  • Moderators

Normally this kind of test gear is quite expensive, .

 

The gear that's exponentially more expensive is so because it's extremely accurate and precisely calibrated. A B&K Type 2250 handheld analyzer is $9500 because it's so precise it includes an sensor to detect when the windscreen is installed (for frequency correction) and internal documentation of each user calibration and microphone serial number so that any measurement ever taken can be traced back through an "audit trail" if needed and the measurement error corrected for (including temp and humidity).

 

A Larson-Davis 3000+ two channel real time analyzer (roughly equivalent in function to the Phonic Analyzer) is $20,000 for similar reasons. It's more accurate, precisely calibrated and very reliable.

 

These professional devices have to be as precise and as well calibrated / certified as they are because they're often used for scientific measurements where tiny differences matter, and for forensics where large sums of money are at stake due to various legal issues.

 

So the question becomes, "How accurate is the PAA6, and is that accuracy/reproducability sufficient to perform the tasks required by the typical user?"

 

Craig / Mike -

 

Does this device come with certification / calibration paperwork of any sort? What accuracy does it specify? Is there a method provided for routine calibration?

 

I guess the accuracy you need varies with the user / business, but the reproducability / drift over time would be important to everyone as I think you'd need to know what's causing changes in your monitor system or whatever it is that you're measuring.

 

Anyway, I'm curious about this device and it's specifications given the affordability of it. Even the very best microphones vary with time and use and preamps / electronics to a lesser extent; calibration extends the useful life of both. It's nice to have a lot of functions and fancy displays, but it's usually important to know the accuracy of the numbers you're looking at as well.

 

Any hint in the documentation?

 

Thanks! :wave:

 

Terry

Link to comment
Share on other sites

  • CMS Author

 

Does this device come with certification / calibration paperwork of any sort?

 

Nope, at least not the PAA3 doesn't. There are no tolerances on the published specifications. It's a tool, not a lab instrument. You take it along on a PA job or a room assessment. You don't use it to determine if you need to sue the factory for too high a noise environment or the airline for flying too close to your house.

 

That's why it only costs a few hundred bucks, I guess.

Link to comment
Share on other sites

  • Members

I think Mike has described it pretty well. The PAA6 is for general testing and ringing out of studios and PA systems, and in that respect, I have to say that it correlates really well with what I'm hearing...not sure I need more accuracy than what it has to tune a room, check performance, and uncover out of phase wiring in speakers :)

Link to comment
Share on other sites

  • 3 weeks later...
  • Members

Phonic PAA6: Oscilloscope

 

Well I must admit, it's been a while since I fired up my Tektronix Dual-Trace Scope that ruled the world a few decades ago...but now thanks to the PAA6, there's a little audio scope I can hold in the palm of my hand. Let's check it out.

 

First off, this is audio range only, so don't expect to check your microprocessor’s clock frequency. Within that constraint, however, it's very flexible. You can look at waveforms from the mic or line ins, and have a considerable range of sensitivities. For example, when measuring dB SPL, you can choose a range of 30 to 100dB SPL, 45 to 115dB SPL, or 60 to 130dB SPL. With dBu, there are four ranges spanning -85dB at the lowest end and +25dBu at the highest. dBv is similarly flexible, from -87.2dBv up to a max of +22.8dBv, again separated into four ranges. For voltage (also with four ranges) you can measure from 43.6 microvolts all the way up to 13.7V.

 

You start using the scope by setting up the range and measurement type you want, as well as the desired input. You can also set the time per horizontal division, from 0.33ms per division up to 500ms per division. Higher resolutions make it easier to zoom in on a portion of a waveform to see, for example, something like a spike.

 

Unlike a traditional analog oscilloscope, the display is quantized to a limited number of steps (hey, it’s digital!) so if you see a flat line at the top of a sine wave that's zoomed in, you won't know if it's clipping or if there's a limitation in the graphic display. However, as shown in the first attached image, the PAA6 will at least inform you if clipping is occurring on the signal coming into the unit, so at least that variable is out of the way.

 

The best part of the PAA6 oscilloscope is that it is a true two-channel triggered scope, and the trigger can come from either channel, or both channels. What this means is that with a repeating waveform, the display re-triggers based on the waveform, thus allowing it to remain stationary in the display. With a non-triggered scope, the waveform tends to "wander," making analysis difficult.

 

The second attached image shows a sine wave without triggering. If this was a video, you'd see the waveform going back and forth; the sharp vertical line indicates that the sine wave is being triggered at some arbitrary time that has no relationship to the waveform (or at least, I think that's what's happening - Grant, correct me if needed!).

 

Now look at the third attached image. In this, the sine wave is being triggered. The display is stationary, and the representation of the waveform is accurate.

 

Of course, looking at sine waves is not necessarily the most creative application for an oscilloscope, but it gets the idea across. You could use this feature to check out the waveforms of virtual synths to see what kind of violence happens to the waveform at really high frequencies, check for "ringing" on a square wave when going through a transformer, determine whether a sawtooth wave is positive- or negative-going, and the like.

 

Bottom line: The oscilloscope in the PAA6 is not a laboratory-quality instrument that's going to replace my Tektronix, but it is a convenient, easy, and fast way to check out what's happening with the various audio waveforms floating around a studio.

Link to comment
Share on other sites

  • Members

Next up: RT60 analysis. This function displays a graph of a room's reverb decay characteristics, namely, how long it takes for the signal to decay to 60dB of its original strength. But there's more to this than meets the eye; it's also possible to filter the signal (A, B, C-weighted, or flat), and look at the RT60 characteristics for any of 10 selectable one-octave frequency ranges. This is great for several applications, such as seeing how adding damping affects the high-frequency decay time.

 

As it turned out, I had a practical application for this function. There's a room in my house with a hard surface that used to be an aviary, and I wanted to try using it as an acoustic reverb chamber. When I clapped my hands in there it sounded pretty good, but I wanted an idea of what to expect in terms of reverb time.

 

To use the RT60 function, you first have to decide what you're going to use to generate the trigger signal for measurement. The PAA6's internal noise generator is one option, or set the trigger to external so it picks up the mic sound - then you can use a handclap, balloon popping, starter pistol, whatever. I opted for the handclap approach, because I was looking for a qualitative rather than quantitative measurement (i.e., which had a longer decay time). Ideally you want a really loud signal, but hey, I can clap pretty loud. The first attached image shows the Setup screen, where I'm choosing a Flat filter, the mic input, and an external trigger.

 

When you start the process by clicking on Run, the PAA6 measures the background noise (see the second attached image). Click Run again, and then it's time to generate the signal (see the third attached image).

 

After you generate the measurement, a graph appears on the display (fourth attached image). You can change the display resolution, zooming in or out so that the graph fills up most of the display. Note the sharp cutoff; this occurs at -60dB, so as you can see, the original signal wasn't all that strong in the first place...but it's good enough to get a sense of the decay time. Also, two lines appear on the display that you can trim to zero in on the numeric readings toward the right part of the display. For example, you can move the yellow line to the 60dB point and get a reading. One thing I don't understand: It seems the numeric RT60 reading is different from the reading shown on the display. For example, the display looks like it's showing about 600ms as the -60dB point, but the numeric readout is showing 1.82 seconds. This may just be a case of pilot error, or some aspect I don't understand - Grant, can you (or someone else) explain what's going on?

 

Just for grins, check out the characteristics of a typical hard-tiled bathroom (fifth attached image), which has an RT60 time of around 230ms. As you can see, there's nowhere near as much reverb as the first room I tested, but it does get across the idea that using a bathroom as an acoustic reverb chamber for short decays isn't that bad an idea!

Link to comment
Share on other sites

  • Members

This is one of the simplest PAA6 "apps" - it's basically a sound level meter that's like a "designer version" of the classic Radio Shack sound level meter. It measures dB SPL through the microphones, and dBu or dBv through the line ins. There are several measurement ranges, so if clipping occurs, you can simply switch to a less sensitive range. Conversely, if you're measuring really low levels, you can switch to a more sensitive one.

 

The first attached image shows the setup screen, where you can choose the input, measurement units, weighting, peak hold time, and response time (response time options are 35, 125, 250, and 1000m)s. The peak hold time options are 05, 1, 2, and 4 seconds. However, the display also shows the highest level that was read as a "Max" reading, so you could leave the meter unattended while it's measuring a noise source, then check back later and see the maximum level that registered. Well, at least in theory: I don't quite understand why the Peak readings can be higher than the Max reading, as shown in the second attached image; perhaps the Max reading is an average. Paging Grant... Of course, you can also reset this maximum SPL level.

 

The second screen shot also shows a few other interesting aspects. When dealing with definable tones, you'll get a readout of the frequency just above the field where you specify the measurement unit. Also notice there are Peak and Peak-to-Peak readings, and the meter has low and high limit indications to the left and right respectively.

Link to comment
Share on other sites

  • Members

 

For example, you can move the yellow line to the 60dB point and get a reading. One thing I don't understand: It seems the numeric RT60 reading is different from the reading shown on the display. For example, the display looks like it's showing about 600ms as the -60dB point, but the numeric readout is showing 1.82 seconds. This may just be a case of pilot error, or some aspect I don't understand - Grant, can you (or someone else) explain what's going on?

 

 

Somebody may have to correct me if I'm wrong (I often am), but I think the RT-60 measurement is the time it takes for the signal to decay 60 dB (for example, 1.82 seconds or 1.44 seconds in one of the pictures Craig attached). The results are usually measured using a signal 30 dB over the background noise (otherwise you may be struggling to get a decent trigger sound), so there's a bit of forumulating going on for the actual results.

 

On the PAA6, the RT-60 result is calculated from the data that lie between the red and yellow lines onscreen - which, in the case mentioned above, isn't quite a 5 dB decay. You could widen the gap between the two lines a bit to get more accurate RT-60 results.

 

Side note on the RT-60: As Craig mentioned, you can filter out particular frequencies, but you don't need to take the measurements all over again to do so. You can take the RT-60 once, then go into the set menu to choose a frequency to filter and the graph/results will be updated.

Link to comment
Share on other sites

  • Members

; perhaps the Max reading is an average.

 

I had to ask about this, but it seems the value of the maximum level is in RMS, while the peak and peak to peak are... well... I'll let a couple of diagrams say what I can't:

 

fig0101.gif

 

fig0102.gif

 

It refers to voltages here, but the theory should be the same for all measurement units that the PAA6 handles.

Link to comment
Share on other sites

  • Members

Okay, that makes sense...I figured that the max reading was peak, as the other measurements were for peaks. But if it's RMS, of course it would be lower.

 

Seems to me it would be good to have the option to have Max capture either RMS to get an average level, or peak to get the highest attained level. It's only a few lines of code, right? :)

Link to comment
Share on other sites

  • CMS Author

 

You can change the display resolution, zooming in or out so that the graph fills up most of the display. Note the sharp cutoff; this occurs at -60dB, so as you can see, the original signal wasn't all that strong in the first place...but it's good enough to get a sense of the decay time. Also, two lines appear on the display that you can trim to zero in on the numeric readings toward the right part of the display. For example, you can move the yellow line to the 60dB point and get a reading. One thing I don't understand: It seems the numeric RT60 reading is different from the reading shown on the display. For example, the display looks like it's showing about 600ms as the -60dB point, but the numeric readout is showing 1.82 seconds.

 

I'm not sure I get how it works either. Perhaps what it's doing is looking at the slope between the red and yellow lines and extrapolating the 60 dB range along the time scale based on that slope.

 

RT60 is kind of a "feel good" measurement like THD. What it doesn't tell you is how the frequencies decay with time. If you were to listen to your handclap after 1.8 seconds (assuming a real world space), it wouldn't sound the same as the initial handclap since some frequencies would decay more than others due to the uneven absorption of the reverberant space. So the best you can do with a simple measurement and display is just take an average (probably RMS) reading at the -60 dB level.

 

A waterfall display with frequency, amplitude, and time on three axes is what really can tell you what the room will sound like. A simple RT60 measurement will tell you if you're in a small room or a large room. Or, knowing (by looking around or pacing it off) what size room you're in, a long RT60 will tell you that you have a lot of reflective surfaces, a short RT60 will tell you that you have a lot of absorbent surfaces. But, as you no doubt have discovered, a room that has good high frequency absorption and poor low frequency absorption will probably have a pretty long RT60 but sound tubby, because all that reverberation time is in the low frequency range.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...