[abernardi]

I have a pair of Avlar Prodigy speakers, a two way small tower using the SEAS Excel 8" woofer in the title and a Heil inspired AMT.  I'm concerned about one of the woofers and fear it may be damaged.  I have a bi-amped system crossing over at 500Hz.  I've been swapping a lot of electronics in and out of my system lately and may have pushed them too far. 

The first thing I noticed was actually a few months ago in the left speaker on just one cut - "Shoeless Joe" on the "Field of Dreams" soundtrack album.  This cut is a synth heavy piece with huge low end strikes, like a Kettle Drum on steroids.  If you have it, play it and you'll see what I'm talking about.  The left speaker started emitting a crackle on one of the kettle drum hits.  I was playing it at a fairly high volume, but not ear-splitting.  At lower volumes the crackle went away, so it's not in the recording.  I was able to repeat this several times, though I've tried not to of course.

The next thing I noticed was just the other day when I tried a new amp to drive the low end, a TBI Millenia MG3.  This amp seemed to exaggerate the lowest octave (at least in my system) and I could hear both woofers huffing, I think they were reaching their extension limits.  I stopped immediately and they seem OK, but I keep thinking I'm hearing something very subtle now, a kind of dry scruffiness in voices, though I may just be paranoid.

I took out the woofer and examined it and it appears to be fine, no rips, tears, no separation from the voice coil, everything looks intact.  I lightly pushed the cone and it's smooth and quiet, same as the other woofer.  The readout on the multimeter is stable at about 6.5 ohms.  Though I did measure the resistance about half a year ago and I thought it was at about 7.5, but I might be wrong.  Can the resistance change in a speaker, or is that an indication of trouble?

I'm not sure where to go from here, any advice?

[charmerci]

First find out where the trouble is really coming from. You said that you've been changing a lot of components recently.

This will help you.

http://www.avahifi.com/index.php?option=com_content&view=article&id=88&Itemid=184

[abernardi]

Thanks Charmerci,
  Using the chart leads me to a problem with the source, but I don't think that's the whole picture.  I've determined that all the cables and electronics are working properly.  I've included a link to the problem area of the song:

https://dl.dropbox.com/u/171448/Shoeless%204.wav


  There are two drum hits with a huge reverb, the second one is the loudest and causes the most distortion.  The first one will distort too if played loud enough.  Obviously any media played back at a high enough volume will make the speaker distort, but as I wrote before, this is happening at a fairly moderate level.

  If you look at the signal you'll notice it's louder in the left channel.  I'm on a mac, so I opened it with Audacity and took a look at the left channel.  It doesn't appear to be clipping.  I noticed some flattening of some of the wave peaks, but that must already be in the source as it's peaking at about -2dB, so that leaves plenty of headroom. 




  There's an analyze function in Audacity where you can get a frequency spectrum graph.  I initially looked at it in an older version and it showed that right after the drum attack, the sound blooms hugely (the echo).  At it's peak the graph shows +22dB at 63Hz! 



  That's much more than any other part of the song.  But the signal isn't clipping.  I don't fully understand the decible scale, but I think there's a variable component, right?  I'm assuming that's what I'm seeing here. - - So, here's the weird part.  I updated to the latest version of Audacity and ran the analysis again and now it's telling me my peak level is -2.2dB at 64Hz.



 


  I went back and measured that section several times with each version and it came up with slightly different results each time, though the main difference was constant.  It has to be a preference somewhere, the lowest dB measurements are also equally different.  But I can't find it.
  Anyway, that's probably all beside the point.  Maybe my speakers simply can't handle a sound like this, maybe it's setting up some kind of resonance that only happens in these speakers. 

  Would one of you guys be willing to take a look at the wave file and maybe play it back and see if it blows out your speakers too?  ...er, I meant listen and see if you hear anything like I'm describing?

[Freo-1]

The display shows clipping with the waveform, and that is NEVER good.  Speakers do not like reproducing clipped waves.  I would expect that to be audible. 

[Letitroll98]

I was thinking the same thing Freo, looks like clipping to me.  Now generally with a clipped waveform on a two way speaker you would damage the tweeter, the clipped signal delivering a square wave to the tweeter at way more power than it was ever designed for.  A shorted output would fry the woofer, DC signal and all.  However the general over driven nature of the piece could be running the woofers into their stops, but that would be more like a click than a buzz, but might be a buzz after the unit was damaged.  If the voice coil is warped a little bit or if the glue was overheated, you could get what you describe, it's sort of an intermediary phase before the woofer craps out completely and just goes pft, pft.  You might also have a small separation where the cone meets the voicecoil.

I would run some 80-120 hz tones and measure and listen for any problems specific to the bass range rather than trying to troubleshoot with fullrange music.     

[abernardi]

The display shows clipping with the waveform.

In the display I posted, I only see one clipped wave, am I right?  I chose the part where I could find it, there were only one or two here and there.

However the general over driven nature of the piece could be running the woofers into their stops, but that would be more like a click than a buzz, but might be a buzz after the unit was damaged.

There is no buzz, I'm just getting a crackle, static, kind of a Bronx Cheer  on those peaks.  Now Audacity has a feature where you can check for actual clipping and it doesn't report any, yet I see the flat tops and bottoms of those few waveforms.  So the clipping must be in the recording itself.  I don't see how a clipped waveform in a recording that is playing back below 0dB should damage a speaker.  I could see hearing distortion in the recording, but it's not being caused by anything happening during playback, it's already there, right?

Also, I neglected to mention an important observation. This distortion happens in the other speaker if I swap channels, so it's not a damaged woofer, maybe it's something more akin to running them into there stops, as you said.

You might also have a small separation where the cone meets the voicecoil.

I would run some 80-120 hz tones and measure and listen for any problems specific to the bass range rather than trying to troubleshoot with fullrange music.

I couldn't see any separation there either, but again, I don't think it's the woofer, unless they're both damaged 

The tones is the next thing, yes, good.  THANKS.

BTW, anyone listen to the clip?

[charmerci]

Yeah, I just listened to the clip.

I'm pretty certain it's a "rattle" in the recording. (Something in the recording studio.) If that's your only reference to the problem - then there's no problem.

[abernardi]

  So you don't hear any static or crackling?  I think I know what you're talking about a rattle, it sounds like something loose around a kettle drum or part of the drum sound itself.  I heard that, but that's not the crackling I was hearing. 

  I made a sequence of sine wave tones going from 80 - 20Hz.  It sounded fine at a moderate level, then I inched up the volume and the woofers were really moving and I was afraid to go much further.  At the top level I was willing to go, I did get some vibration, but I'm not sure that was coming from the speaker or something else in the room.  I didn't play it long enough to find out.  How do you test these things without actually hurting the speaker?

  What I did find, that I didn't anticipate (but in retrospect should have) is how NOT flat the response is.  I had the sequence playing 5 sec intervals at 5Hz increments.  It started rolling off at about 65Hz fairly smoothly, at 50Hz there seemed to be a slight hump, then at 40 and 35 the room started shaking a good amount and it appeared to be louder and then it settled down at 30 and 20 was inaudible.  Time for some bass traps?

[charmerci]

  So you don't hear any static or crackling?  I think I know what you're talking about a rattle, it sounds like something loose around a kettle drum or part of the drum sound itself.  I heard that, but that's not the crackling I was hearing. 

I didn't hear any static or crackling - but I just have monitors with a 10" sub. I don't have any test equipment. I just listened closely. Sounds pretty nice and clear to me.

[neekomax]

I didn't hear any static or crackling - but I just have monitors with a 10" sub. I don't have any test equipment. I just listened closely. Sounds pretty nice and clear to me.

I also listened on my hifi system (front end is a Mac Mini, so just listened right from my browser). I didn't hear anything bad either. 

[Letitroll98]

What I did find, that I didn't anticipate (but in retrospect should have) is how NOT flat the response is.  I had the sequence playing 5 sec intervals at 5Hz increments.  It started rolling off at about 65Hz fairly smoothly, at 50Hz there seemed to be a slight hump, then at 40 and 35 the room started shaking a good amount and it appeared to be louder and then it settled down at 30 and 20 was inaudible.  Time for some bass traps?

So your room is about 15'x20' with 8' ceilings?   

[charmerci]

I just looked it up. The TBI Millenia MG3 is only 32 wpc. That tiny amp must be clipping.

"The Millenia amplifier is capable of 32 WPC into an 8 ohm load when it's paired with the linear power supply. The same power output level can be reached if 24 volts of DC power can be supplied to the amplifier. If the TBI amplifier is driven from the internal battery pack, then it is capable of 10 WPC at less than 1% distortion. The MG3 is miserly in its use of power, and has an 85% efficiency rating when run in DC mode."

If your PS is less than 24 volts, it's even less power.

[AJinFLA]

I just looked it up. The TBI Millenia MG3 is only 32 wpc. That tiny amp must be clipping.

I have a little 30w class D amp that "crackles" (very apt description) when clipping

[abernardi]

So your room is about 15'x20' with 8' ceilings?   

Hey, pretty good!  About 15'x24'x8'.  How did you get so close?  I also have 6" x 4" beams that cross the flat ceiling every 3.5' and sometimes I wonder if that's not setting up some kind of standing wave exaggerating the high end a bit.  Any thoughts on that?

I just looked it up. The TBI Millenia MG3 is only 32 wpc. That tiny amp must be clipping.

  I hadn't thought of that.  That must be it!  I'm running the TBI off 12V!  I thought the woofer had a sensitivity in the low 90's, but the specs on the SEAS website says it's 88.  The AMTs are super sensitive, over 100dB, so I have to really crank down the high end on the crossover.  My Dodd amp that's driving the AMTs is only 15wpc, but even so maybe I need more power for the low end.  Does that sound right?  I thought the active crossover would compensate for the low watts.  I think you hit it on the head charmerci, I'm going to play around a bit and report back.  THANKS for all the great feedback!

[charmerci]

  I hadn't thought of that.  That must be it!  I'm running the TBI off 12V!  I thought the woofer had a sensitivity in the low 90's, but the specs on the SEAS website says it's 88.  The AMTs are super sensitive, over 100dB, so I have to really crank down the high end on the crossover.  My Dodd amp that's driving the AMTs is only 15wpc, but even so maybe I need more power for the low end.  Does that sound right?  I thought the active crossover would compensate for the low watts.  I think you hit it on the head charmerci, I'm going to play around a bit and report back.  THANKS for all the great feedback!

I'm glad that I could help - especially in AC where there are many people who know so much more than I! I wasn't familiar with the TBI so I looked it up thinking that it was a tube amp and that maybe the distortion was coming from a bad tube. But that sample goes real low and even I wasn't going to push my system more than normal - so when I saw that tiny integrated with "35" wpc, I figured that's what was going on.

 The Dodd amp I assume is a tube amp. Tube amps usually don't distort very badly when clipping especially a good one like Dodd and sound much louder than their rated output compared to solid state.

[abernardi]

yeah, that's true.  When I was driving the speakers with a passive crossover and just the Dodd, I was getting more volume than I could handle and no distortion at all.

[Letitroll98]

Hey, pretty good!  About 15'x24'x8'.  How did you get so close?  I also have 6" x 4" beams that cross the flat ceiling every 3.5' and sometimes I wonder if that's not setting up some kind of standing wave exaggerating the high end a bit.  Any thoughts on that?

It was extraordinarily easy, the 15x20 was my first guess I didn't even dig deeper to narrow the guess down.  It was the resonant frequency you quoted.  Given that 90% of American homes are 8' ceilings, the other two dimensions are easy.  Rooms have three modes that vary in intensity, Axial, Tangential, and Oblique.  It was easy to take a 3 second guess at the Axial mode.  To answer your previous statement, yes, all rooms benefit from treatment, you have great dimensions for audio reproduction so a couple of bass traps and maybe a touch of absorption at the first reflection points would wonders and would prolly be all you need.

http://www.marktaw.com/recording/Acoustics/RoomModeStandingWaveCalcu.html

For this question, no!  Your beams are great diffusers and only help the sound.  Granted, they're not seventh root diffusers that perfectly distribute the sound between 500-3000hz, but they can't hurt. 

 

[abernardi]

Hey Letitroll,
  This is OT, but what do you think about this?  Currently I have my speakers on the long wall of the room:



  But we're thinking of switching the living room around and put the speakers on the short wall:



  Is there a general consensus on which is a better way to go?  With the short wall, the speakers are going to be closer to the walls.  Comments?

[Letitroll98]

In general, I prefer the long wall placement as it moves side wall early reflections way past the time that your ear/brain interface confuses them with the direct sound.  Also the further you can move the speakers out from the front wall the better.  To reduce bass nodes try the 1/5th - 1/3rds placement rule, e.g. place the speakers at 1/5 or 1/3 or 2/5ths of the room "length" (the short dimension in your set up).  It's not exact, but gets pretty close to the axial bass nulls and will reduce your problem bass nodes by a couple of db.  Of course to reinforce those frequencies you would use even fractions, 1/4th, 1/2, 1/6, etc.  Yes, 1/2 of the room length, try placing your speakers halfway out into the room, if only for an experiment when the spouse is away, the imaging will be revelatory.

All that being said, there's many arguments for using the short wall placement.  Smaller rooms especially when you can't get enough listening distance from the speakers and still have them well away from the front wall.  You can get a deeper soundstage (relatively) by using the short wall, especially if that allows you to move the speakers well out into the room.  This works especially well with bi-polar speakers like Maggies that don't interact with the sidewalls all that much anyway.  For box speakers you will have to pay more attention to early reflections from the sidewalls vs long wall placement.  You can see the first reflection point by using said spouse to hold a mirror on the side wall while you sit in the listening position (with a beer) and have her move up and down the length of the wall until you can see the speakers in the mirror.  Even when you see the spot, make sure to have her move back and forth a few more times just to "get it perfect", spouses just love this kind of exercise and really can't get enough of it.  After you get divorced, you can put the speakers any damn place you want.  So ends lesson one on room acoustics.

     

[neekomax]

You can see the first reflection point by using said spouse to hold a mirror on the side wall while you sit in the listening position (with a beer) and have her move up and down the length of the wall until you can see the speakers in the mirror.  Even when you see the spot, make sure to have her move back and forth a few more times just to "get it perfect", spouses just love this kind of exercise and really can't get enough of it.  After you get divorced, you can put the speakers any damn place you want.  So ends lesson one on room acoustics.

LOL Best thing I've read on the Internet all day.