AVS › AVS Forum › Audio › Receivers, Amps, and Processors › "Official" Audyssey thread (FAQ in post #1)
New Posts  All Forums:Forum Nav:

"Official" Audyssey thread (FAQ in post #1) - Page 1972

post #59131 of 62251
1/18/13 at 1:05pm
Quote:
Originally Posted by djoberg View Post

So, in light of Keith and Feri's last posts, I'm going to give my last calibration with results and what I've done since talking to the CS rep at Onkyo. In the last calibration I set the gain to about 2 o'clock, ran the full 2EQ calibration and the sub's level came in at +11.5dB. It obviously sounded imbalanced with too much bass. I then turned the trim level down to about +7.5dB and it sounded better, but it still had too much bass. This morning I called Onkyo and right after making my post on here I turned the trim level down to +3.5dB and the sub's gain down to 12 o'clock. I've only listened to some music (CDs), but it sounds EXCELLENT! There is a seamless blending of the sub and the mains ,center, & surrounds, with just the right amount of bass. Granted, I'm going by my ears in forming this conclusion.

Did my turning down the trim level and sub gain negate my last calibration (in other words, did my actions negate the goal of Audyssey, which is to "achieve 85dB SPL in the LFE channel)? And just as important, am I in danger of experiencing the same thing that Keith did (see post above) when he had distortion during loud bass scenes, which turned out to be clipping that eventually ruined his sub? Of course, in asking this last question it seems that Keith did not adjust his trim levels (he left them on +10dB), so *maybe* if he had lowered them the clipping, distortion and ruining of his sub wouldn't have taken place.

I'm betraying my ignorance in asking these questions (I've been told countless times that "ignorance is not stupidity, so feel free to ask questions to your heart's content"), but I really want to know the truth so I can act on it. Keith and Feri seem to in disagreement on some issues and I want to be able to digest what they're both saying and try to arrive at the truth...and then apply it.

 

Yes, Feri and I disagree on this. But I am the one who ran his AVR at +10dB and destroyed a sub because of it. Following Feri's line of thinking +10dB ought to be fine, but I know from experience it isn't. I can only tell you what I believe to be true - the sub trim should ideally be well clear of the maximum 'end stops'.

Reply
Reply

Gear mentioned in this thread:

post #59132 of 62251
1/18/13 at 1:13pm
Quote:
Originally Posted by kbarnes701 View Post

We'll have to agree to disagree on this, Feri. I think the case for staying within the +/-3.5dB range (approximately) is overwhelming, for the reasons I gave (and which Selden added to).

Sorry for that. I see you are still sticking to the idea that the trims have a centroid/ neutral position. How about this analogy: the ABC that starts from A and ends with Z has it's central position between the letters "M" and "N". True, but has no practical meaning. cool.gif
Reply
Reply
post #59133 of 62251
1/18/13 at 1:16pm
Quote:
Originally Posted by kbarnes701 View Post

Yes, Feri and I disagree on this. But I am the one who ran his AVR at +10dB and destroyed a sub because of it. Following Feri's line of thinking +10dB ought to be fine, but I know from experience it isn't. I can only tell you what I believe to be true - the sub trim should ideally be well clear of the maximum 'end stops'.

Keith, did that poor little sub at +10 dB trim level deliver an SPL of 85 dBs at the MLP?
Reply
Reply
post #59134 of 62251
1/18/13 at 2:43pm
Feri -- I think you are missing the point on the extreme positive trim. The issue is not that the subwoofer was damaged due to clipping because it was TOO LOUD. As you noted, the calibrated volume is the same regardless of the combination of receiver trim + sub gain. The hypothesis is that the SW pre-out signal itself was clipping due to the extreme voltage of the signal output when the sub trim was that high.

Somebody way back in this thread actually measured the sub pre-out of his receiver and measured this happening. I'll see if I can dig up the posts. It doesn't imply that it will *always* happen with an extreme positive trim, but the phenomenon was measured to be real.

Other than the general idea that having the trim "centered" gives you plenty of room to adjust up/down to taste, the argument against the extreme negative trim AFAIK is purely just about the fact that it might generate insufficient voltage to wake a sub's auto-on circuit when playing at lower volumes. I don't think anyone has posited that an extreme negative trim can be harmful.
Reply
Reply
post #59135 of 62251
1/18/13 at 2:49pm
Here, found it. It appears to me that the issue derives from the fact that a subwoofer's input will have a specific range of voltage sensitivity, and the combination of a high sub trim, loud volume, and the massive LFE in a modern action flick can generate a voltage level at the SW pre-out which can cause the sub's amp to clip. So your discussion of 85dB is somewhat of a red herring, that's not the root of the issue.

Quote:
Originally Posted by [unique] View Post


Hi Jm!


Today i connected my denons 2809 SW output to an oscilloscope, and ran the movie "the hurt locker", using the first bomb scene.


In order to test the subwoofer clipping problems with especially dyn eq enabled i noticed a few things.


1) Using my settings of -12db sw and -16db ref im getting +/- 1.5V in the bomb scene. This range is _just_ ok for my bk xls200 amp, im guessing it has around +/-2V input range. I will test the limits later this week by ear to check which input range i can use. (update: input range of sub is +/- 700 mV, thus even in this scene it was clipping
)

2) This is the more amazing one, i checked if the denon would clip.

I put SW trim on 0DB, LFE on 0db (default), and reference volume.

Dyn eq had no influence using these reference levels, as Chris always has explained, but the SW output clipped! Scope was reading +/- 6V and the sine wave was getting distorted.

This is pretty amazing since this shows that running reference volume using all trims on 0db will cause clipping using this movie.

I might consider posting my findings with more info/photos, since this seems to be a design flaw in the denon. Or are we not supposted to run reference volume at home?


Either way, when setting SW trim on -12DB you can run this movie without clipping in that scene, the levels go to +/- 5V then. So a subwoofer input amp should be able to handle +/- 5V to run on reference. Other movies are maybe even louder, so +/-5V should be a minimum for any subwoofer amp manufacturer to handle.


Either way, i fully agree with your statement,to avoid clipping in the output and inputs, put the SW trims as low as possible in the receiver, especially on a denon 2809.

Edited by batpig - 1/18/13 at 2:54pm
Reply
Reply
post #59136 of 62251
1/18/13 at 2:49pm
Quote:
Originally Posted by mogorf View Post

Quote:
Originally Posted by kbarnes701 View Post

We'll have to agree to disagree on this, Feri. I think the case for staying within the +/-3.5dB range (approximately) is overwhelming, for the reasons I gave (and which Selden added to).

Sorry for that. I see you are still sticking to the idea that the trims have a centroid/ neutral position. How about this analogy: the ABC that starts from A and ends with Z has it's central position between the letters "M" and "N". True, but has no practical meaning. cool.gif

 

I hear you Feri. I just don't agree with you.

Reply
Reply
post #59137 of 62251
1/18/13 at 2:54pm
Quote:
Originally Posted by mogorf View Post

Quote:
Originally Posted by kbarnes701 View Post

Yes, Feri and I disagree on this. But I am the one who ran his AVR at +10dB and destroyed a sub because of it. Following Feri's line of thinking +10dB ought to be fine, but I know from experience it isn't. I can only tell you what I believe to be true - the sub trim should ideally be well clear of the maximum 'end stops'.

Keith, did that poor little sub at +10 dB trim level deliver an SPL of 85 dBs at the MLP?

 

Of course. The SPL is not affected by whether you input the correct voltage though a high trim level or a high sub gain.  It makes no difference, from a voltage POV whether the trim is set to +10db and the sub gain set to, say, 9 o'clock or whether the trim is set to -10dB and the sub gain set to, say, 4 o'clock. The output will be the same. But I know from experience that a trim of +10dB clipped the input of the sub amp when it was trying to hit 85dB and thus ultimately destroyed the sub. For this reason, I believe your advice is, on this occasion, dangerous, which is why I have addressed it several times already.

 

I remain true to my earlier remark that the best solution is a trim somewhere between approx +/- 3.5dB. You can keep hammering away indefinitely but you will not persuade me otherwise :)

Reply
Reply
post #59138 of 62251
1/18/13 at 3:03pm
Quote:
Originally Posted by kbarnes701 View Post

Yes, Feri and I disagree on this. But I am the one who ran his AVR at +10dB and destroyed a sub because of it. Following Feri's line of thinking +10dB ought to be fine, but I know from experience it isn't. I can only tell you what I believe to be true - the sub trim should ideally be well clear of the maximum 'end stops'.

I hear you Keith and believe me, I'm concerned. But do you think it will make a difference (concerning the *clipping* issue) that I turned the trim level down to +3.5dB (and the gain is now at 12 o'clock)?

I've got company coming in a half hour for overnight, so after they leave tomorrow I hope to reconnect my Pioneer and do a calibration to see if the sub is the culprit. If it isn't, then I'll try Feri's suggestion of doing a reset on the Onkyo and the sub. This is all frustrating, because I really do love the sound of the Onkyo and of course I've spent a good 4-5 hours becoming acquainted with it and going through all the various settings.
Reply
Reply
post #59139 of 62251
1/18/13 at 3:03pm
Quote:
Originally Posted by batpig View Post

Feri -- I think you are missing the point on the extreme positive trim. The issue is not that the subwoofer was damaged due to clipping because it was TOO LOUD. As you noted, the calibrated volume is the same regardless of the combination of receiver trim + sub gain. The hypothesis is that the SW pre-out signal itself was clipping due to the extreme voltage of the signal output when the sub trim was that high.

 

 

 

That is exactly correct. Sending a huge signal voltage to the sub, which it needed because its gain was set way too low, caused the input on the sub to clip and this destroyed it after a while. The SPL is a red herring. This is why it is dangerous to run a very high trim level. As it makes no difference to the SPL whether you run a trim at the extremes and a sub gain at the extremes, but as it does give the possibility of damaging the sub, the sensible thing to do is to stay within a broad central range of trim.  I don't really understand why Feri feels the need to challenge this so vigorously. If his contention is that it doesn't matter where you set the trims so long as they are not maxed out, then clearly there is no problem in setting them +/- 3.5dB - and thus avoiding any possible damage caused by clipping the sub's input.

 

Quote:
Somebody way back in this thread actually measured the sub pre-out of his receiver and measured this happening. I'll see if I can dig up the posts. It doesn't imply that it will *always* happen with an extreme positive trim, but the phenomenon was measured to be real.

Other than the general idea that having the trim "centered" gives you plenty of room to adjust up/down to taste, the argument against the extreme negative trim AFAIK is purely just about the fact that it might generate insufficient voltage to wake a sub's auto-on circuit when playing at lower volumes. I don't think anyone has posited that an extreme negative trim can be harmful.

 

Agreed.

Reply
Reply
post #59140 of 62251
1/18/13 at 3:08pm
Quote:
Originally Posted by djoberg View Post

Quote:
Originally Posted by kbarnes701 View Post

Yes, Feri and I disagree on this. But I am the one who ran his AVR at +10dB and destroyed a sub because of it. Following Feri's line of thinking +10dB ought to be fine, but I know from experience it isn't. I can only tell you what I believe to be true - the sub trim should ideally be well clear of the maximum 'end stops'.

I hear you Keith and believe me, I'm concerned. But do you think it will make a difference (concerning the *clipping* issue) that I turned the trim level down to +3.5dB (and the gain is now at 12 o'clock)?
 

 

 

I think it isn’t the issue in your case. In your case, something is wrong and what we need to is to discover what - the AVR or the sub. My money is on the AVR.  It's a pity you don't have any measuring gear because it would be simple to test the sub output voltage on the AVR and see if it is in spec or not.

 

Quote:
I've got company coming in a half hour for overnight, so after they leave tomorrow I hope to reconnect my Pioneer and do a calibration to see if the sub is the culprit. If it isn't, then I'll try Feri's suggestion of doing a reset on the Onkyo and the sub. This is all frustrating, because I really do love the sound of the Onkyo and of course I've spent a good 4-5 hours becoming acquainted with it and going through all the various settings.

 

I think it is sensible to test with the Pioneer. I bet the trims are acceptable when you do.

Reply
Reply
post #59141 of 62251
1/18/13 at 3:18pm
Quote:
Originally Posted by batpig View Post

Feri -- I think you are missing the point on the extreme positive trim. The issue is not that the subwoofer was damaged due to clipping because it was TOO LOUD. As you noted, the calibrated volume is the same regardless of the combination of receiver trim + sub gain. The hypothesis is that the SW pre-out signal itself was clipping due to the extreme voltage of the signal output when the sub trim was that high.

First off, what does "extreme positive trim" mean? Isn't the +/- 12 dB trim well defined and safe to use? Secondly, no way I could ever accept this kind of reasoning. Sorry bp, good try but not true. Hypothesis remains hypothesis in this case. Why? Because a SW pre-out signal can not clip, an AVR can not deliver "extreme" voltages even at the higher end of the trim allowance. If you think so, you are degrading the abilities of the R&D engineers of the AVR makers. Let's not do that. cool.gif
Quote:
Somebody way back in this thread actually measured the sub pre-out of his receiver and measured this happening. I'll see if I can dig up the posts. It doesn't imply that it will *always* happen with an extreme positive trim, but the phenomenon was measured to be real.
Surely it will turn out to be nothing more than an urban legend. Remember, there is no sucha thing as "extreme positive trim", just max. positive trim! smile.gif
Quote:
Other than the general idea that having the trim "centered" gives you plenty of room to adjust up/down to taste, the argument against the extreme negative trim AFAIK is purely just about the fact that it might generate insufficient voltage to wake a sub's auto-on circuit when playing at lower volumes. I don't think anyone has posited that an extreme negative trim can be harmful.

Can we forget this centered trim idea? See my analogy with the ABC! smile.gif Meanwhile, adjusting up/down to taste is not in accordance with the intended use when it comes to film material that has been recorded against known standards. Talking about music or games is another strory, as you also know, better than me. smile.gif
Reply
Reply
post #59142 of 62251
1/18/13 at 3:45pm
Quote:
Originally Posted by mogorf View Post

First off, what does "extreme positive trim" mean? Isn't the +/- 12 dB trim well defined and safe to use?

I don't know where the specific level is that presents a "threat", nor do I claim that such a "threat" will universally be present when running a large positive sub trim. All I'm providing is the basic reasoning for the potential issue.

Quote:
Secondly, no way I could ever accept this kind of reasoning. Sorry bp, good try but not true. Hypothesis remains hypothesis in this case. Why? Because a SW pre-out signal can not clip, an AVR can not deliver "extreme" voltages even at the higher end of the trim allowance. If you think so, you are degrading the abilities of the R&D engineers of the AVR makers. Let's not do that. cool.gif

So let me get this straight -- your counter argument is, "it can't happen because AVR makers can't f*ck up?" We are officially straining the limits of even your credibility when it comes to defending the manufacturers. And, in this case, hypothesis was VALIDATED because it was a measured phenomenon, not a theoretical.

Furthermore, this doesn't have to imply a "mistake" on the part of the AVR manufacturer; rather, it's a MISMATCH between the voltage of the Sub pre-out compared to what the sub itself is designed to accept. The AVR could be designed perfectly well but it still may present a signal which doesn't "play nice" with a specific subwoofer. Or do you think AVR manufacturers have tested their receivers with every possible subwoofer out there? rolleyes.gif

Quote:
Surely it will turn out to be nothing more than an urban legend. Remember, there is no sucha thing as "extreme positive trim", just max. positive trim! smile.gif

Hmmm, I bet you wish you hadn't said that since I was able to dig up the precise post which demonstrated the actual measurements of the phenomenon. I guess it's easier to ignore evidence that doesn't fit your viewpoint and dismiss it as "urban legend", even when I quoted the relevant post.

Quote:
Meanwhile, adjusting up/down to taste is not in accordance with the intended use when it comes to film material that has been recorded against known standards.

WTF are you talking about? People shouldn't adjust their subwoofer volume to taste? You are going off the deep end now.
Reply
Reply
post #59143 of 62251
1/18/13 at 3:50pm
Quote:
Originally Posted by batpig View Post

Here, found it. It appears to me that the issue derives from the fact that a subwoofer's input will have a specific range of voltage sensitivity, and the combination of a high sub trim, loud volume, and the massive LFE in a modern action flick can generate a voltage level at the SW pre-out which can cause the sub's amp to clip. So your discussion of 85dB is somewhat of a red herring, that's not the root of the issue.

Heck with it bp, this guy is plain stupid (sorry). He is changing sub channel trims at will,...from-12 dB to 0 dB and watching an oscilloscope to see how clipping occurs. I told you, it's nothing but an urban legend. Forget this post, please. biggrin.gif
Reply
Reply
post #59144 of 62251
1/18/13 at 3:57pm
Changing a parameter and then using an instrument to measure the effects of that change.... How stupid! rolleyes.gif
Reply
Reply
post #59145 of 62251
1/18/13 at 4:04pm
Quote:
Originally Posted by batpig View Post

Changing a parameter and then using an instrument to measure the effects of that change.... How stupid! rolleyes.gif

Let's move on! cool.gif
Reply
Reply
post #59146 of 62251
1/18/13 at 4:22pm
Feri, I have to agree with Keith and BP. I don't think you quite get what they're saying here.

In professional sound reinforcement, there's a practice called gain matching, i.e. you need to match the gain structure in your equipment chain to produce the best results. If you really do want to read up more to understand it, here's a handy link for you:
http://www.hometheatershack.com/forums/home-theater-receivers-processors-amps/35677-gain-structure-home-theater-getting-most-pro-audio-equipment-your-system.html#axzz2INJnZDY3

In pro and consumer equipment, different components and manufacturers have different acceptable ranges for the signal inputs and outputs. Exceed this range and you will produce clipped signals in the chain. As outlined in the article above, the ideal way to ensure that all your equipment is setup correctly as far as gain structure goes, is to actually measure the output and input ranges of all the components in your chain, but most folks aren't going to be doing that. In lieu of measuring the limits of the input and output ranges in each piece of equipment (specifically for the subwoofer in this case), the safer route is to adjust the subwoofer's gain so that the avr's trim settings are centered, and not too close to the extreme ends of the range.

Feri, you seem to have ignored BP's post quoting another poster who TESTED AND MEASURED the signal chain in his avr and showed that the output range on some settings exceeded the input range on his subwoofer. In addition, it appears that the avr's sub output was clipping at Reference level AS HE MEASURED THE WAVEFORM DISTORTING USING AN OSCILLOSCOPE ON THE OUTPUT. The simple fact is, some equipment can handle the range, some can't. To be safe without actually measuring everything, set the sub's gain so the avr's trim is near the center of its range.

To the poster having issues with the extreme positive trim settings: your previous Pioneer didn't have Audyssey room correction did it? It is quite possible that your subwoofer is actually in a less than optimal position in your room and might be in a partial null, requiring boosted levels based on what the Audyssey mic is reading. It's of course, possible that the Audyssey mic may be having problems with the lower octaves too, but have you experimented with something like a subwoofer crawl? If the Audyssey mic is measuring a partial nullification of the subwoofer in the MLP and other measurement locations, it will cause the trims to be set at the extreme positive end to try to compensate.

This is one of the reasons why measurement equipment can help. With measuring equipment, you can find the best locations for your sub(s) prior to applying room correction.


Max
Reply
Reply
post #59147 of 62251
1/18/13 at 6:10pm
I can't for the life of me figure out why my center channel wont get picked up by the Audessey mic. I am using it on a tripod with my receiver Onkyo TX-NR809. I have tried 15+ times with a number of different positions. It is a Phase Technology V6. Any ideas?
Reply
Reply
post #59148 of 62251
1/18/13 at 7:01pm
Quote:
Originally Posted by djbluemax1 View Post

To the poster having issues with the extreme positive trim settings: your previous Pioneer didn't have Audyssey room correction did it? It is quite possible that your subwoofer is actually in a less than optimal position in your room and might be in a partial null, requiring boosted levels based on what the Audyssey mic is reading. It's of course, possible that the Audyssey mic may be having problems with the lower octaves too, but have you experimented with something like a subwoofer crawl? If the Audyssey mic is measuring a partial nullification of the subwoofer in the MLP and other measurement locations, it will cause the trims to be set at the extreme positive end to try to compensate.

Max

Thanks Max for all the helpful info. My Pioneer AVR employs its own calibration system called MCACC and it had no problem with the sub during the calibration. I had my gain on my SVS sub set at 12 o'clock and if memory serves me it resulted in a -1 or -2dB trim level, so all was good. That begs the question, "If my Pioneer calibration mic had no problems with my sub in the position it's still in (in my front corner flanking my left front speaker), why would Onkyo's Audyssey have trouble with it? To frame the question in a different way to better suit your hypothesis, "Could there be a problem with Audyssey because of a "partial null" when there wasn't a problem with MCACC because of a "partial null"?"

Well, I've got to get back to our company but I thought I'd "jump in" and ask that quick question.
Reply
Reply
post #59149 of 62251
1/18/13 at 7:06pm
Quote:
Originally Posted by bmjones66 View Post

I can't for the life of me figure out why my center channel wont get picked up by the Audessey mic. I am using it on a tripod with my receiver Onkyo TX-NR809. I have tried 15+ times with a number of different positions. It is a Phase Technology V6. Any ideas?

Can you give more details:

- Does the speaker actually make sound when Audyssey pings it?
- Have you tried any basic troubleshooting, like swapping speakers to see if the problem "follows" the speaker or sticks to that channel?
- is Audyssey able to to complete and save successfully with the other speakers?
Reply
Reply
post #59150 of 62251
1/18/13 at 7:10pm
Quote:
Originally Posted by batpig View Post

Can you give more details:

- Does the speaker actually make sound when Audyssey pings it?
- Have you tried any basic troubleshooting, like swapping speakers to see if the problem "follows" the speaker or sticks to that channel?
- is Audyssey able to to complete and save successfully with the other speakers?

1. Yes the center does make noise just audibly lower.
2. I have not thought of this, i will swap one of the center and a bookshelf to see.
3. No it is not able to complete. It gives an Error on the center and gives me the option to retry or cancel.
Reply
Reply
post #59151 of 62251
1/18/13 at 7:27pm
Quote:
Originally Posted by bmjones66 View Post

1. Yes the center does make noise just audibly lower.
2. I have not thought of this, i will swap one of the center and a bookshelf to see.
3. No it is not able to complete. It gives an Error on the center and gives me the option to retry or cancel.

Ok It definitely has something todo with the speaker. It Errors out on the channel it is on. It doesnt just say "No", it says "Error". So it hears it, but something goes on.
Reply
Reply
post #59152 of 62251
1/18/13 at 10:21pm
Quote:
Originally Posted by djbluemax1 View Post

Feri, I have to agree with Keith and BP. I don't think you quite get what they're saying here.

In professional sound reinforcement, there's a practice called gain matching, i.e. you need to match the gain structure in your equipment chain to produce the best results. If you really do want to read up more to understand it, here's a handy link for you:
http://www.hometheatershack.com/forums/home-theater-receivers-processors-amps/35677-gain-structure-home-theater-getting-most-pro-audio-equipment-your-system.html#axzz2INJnZDY3

In pro and consumer equipment, different components and manufacturers have different acceptable ranges for the signal inputs and outputs. Exceed this range and you will produce clipped signals in the chain. As outlined in the article above, the ideal way to ensure that all your equipment is setup correctly as far as gain structure goes, is to actually measure the output and input ranges of all the components in your chain, but most folks aren't going to be doing that. In lieu of measuring the limits of the input and output ranges in each piece of equipment (specifically for the subwoofer in this case), the safer route is to adjust the subwoofer's gain so that the avr's trim settings are centered, and not too close to the extreme ends of the range.

Feri, you seem to have ignored BP's post quoting another poster who TESTED AND MEASURED the signal chain in his avr and showed that the output range on some settings exceeded the input range on his subwoofer. In addition, it appears that the avr's sub output was clipping at Reference level AS HE MEASURED THE WAVEFORM DISTORTING USING AN OSCILLOSCOPE ON THE OUTPUT. The simple fact is, some equipment can handle the range, some can't. To be safe without actually measuring everything, set the sub's gain so the avr's trim is near the center of its range.

To the poster having issues with the extreme positive trim settings: your previous Pioneer didn't have Audyssey room correction did it? It is quite possible that your subwoofer is actually in a less than optimal position in your room and might be in a partial null, requiring boosted levels based on what the Audyssey mic is reading. It's of course, possible that the Audyssey mic may be having problems with the lower octaves too, but have you experimented with something like a subwoofer crawl? If the Audyssey mic is measuring a partial nullification of the subwoofer in the MLP and other measurement locations, it will cause the trims to be set at the extreme positive end to try to compensate.

This is one of the reasons why measurement equipment can help. With measuring equipment, you can find the best locations for your sub(s) prior to applying room correction.


Max

Thanks Max, I will read through the article linked in your post. Meantime, on another note, why do you think gain matching is something that is discussed on Internet forums only, while it seem that manufacurers are not avail of this phenomenon? Why do you think that a lonely poster with an oscilloscope in hand trying to prove something even he himself does not understand to the ful extent will make the whole audio industry collapse due to his wrong conclusion that have never been proved. I'm always sceptical here on AVS when I read such posts that are not backed up with scientific reasong behind, yet presented as fact to be acceptalbe worldwide.
Reply
Reply
post #59153 of 62251
1/18/13 at 11:38pm
Quote:
Originally Posted by mogorf View Post


Thanks Max, I will read through the article linked in your post. Meantime, on another note, why do you think gain matching is something that is discussed on Internet forums only, while it seem that manufacurers are not avail of this phenomenon? Why do you think that a lonely poster with an oscilloscope in hand trying to prove something even he himself does not understand to the ful extent will make the whole audio industry collapse due to his wrong conclusion that have never been proved. I'm always sceptical here on AVS when I read such posts that are not backed up with scientific reasong behind, yet presented as fact to be acceptalbe worldwide.
Discussed on internet forums only? I've discussed it more with audio professionals in the real world than on the internet. Just because the majority of laymen are clueless about something doesn't mean it doesn't exist. Heck, all the regular Joe's believed the earth was flat for the longest time too.

The manufacturers ARE aware of this phenomenon. Unfortunately, there aren't a whole lot of standards set in stone ESPECIALLY for consumer level equipment, which is why you have some amplifiers with a 32db gain (like Emotiva and some with a 24db gain and others with different gain structures, just as you have some equipment that will output a 0-1.0V signal on a line out, whilst some others may output up to 1.5V and still others even higher, or lower. Some inputs have a wider input range, but if the equipment hooked up to that input never outputs the maximum range, you'll never get the maximum volume out of it unless you adjust the gain. With other equipment, if the output range of the equipment upstream is higher than the input range of the next piece of equipment, then the signal ends up clipping.

As I said, this is all common knowledge in the professional audio industry. Just because you haven't heard of it doesn't mean no one knows about it. As far as the manufacturers go, because as mentioned, there is no set standard for consumer equipment, some manufacturers may produce equipment with ranges that aren't as common as others. It's up to the owner to discover any incompatibilities. Most consumer equipment is built around a more common spec, but not all. Before I bought my XPA-1 amps, I checked Emotiva's specs for the RCA and XLR input ranges, and checked my Onkyo 5008 for its pre-out signal range to make sure that they're compatible. Sure, I could just assume that they should be and fork out the money and plug and pray like most folks, but since I know about this, it would have been dumb for me NOT to check the specs first.

As for scientific reason, all the scientific reason is there, it's just whether people choose to take advantage of it, or choose to ignore it for their own agendas. I still like Markus's signature.:
"In science, contrary evidence causes one to question a theory. In religion, contrary evidence causes one to question the evidence." - Floyd Toole


Max
Reply
Reply
post #59154 of 62251
1/19/13 at 3:23am
Quote:
Originally Posted by mogorf View Post

 
First off, what does "extreme positive trim" mean? Isn't the +/- 12 dB trim well defined and safe to use? Secondly, no way I could ever accept this kind of reasoning. Sorry bp, good try but not true. Hypothesis remains hypothesis in this case. Why? Because a SW pre-out signal can not clip, an AVR can not deliver "extreme" voltages even at the higher end of the trim allowance. If you think so, you are degrading the abilities of the R&D engineers of the AVR makers. Let's not do that. cool.gif
 

 

It wasn't hypothesis that destroyed my sub. It was clipping the input for 3 hours.  Not was it hypothesis in the measured result Batpig quotes above.  

Reply
Reply
post #59155 of 62251
1/19/13 at 3:32am
Quote:
Originally Posted by mogorf View Post

 I'm always sceptical here on AVS when I read such posts that are not backed up with scientific reasoning behind, yet presented as fact to be acceptalbe worldwide.

:)  You mean like this: "A SW pre-out signal can not clip, an AVR can not deliver "extreme" voltages even at the higher end of the trim allowance."

Reply
Reply
post #59156 of 62251
1/19/13 at 3:47am
Quote:
Originally Posted by djbluemax1 View Post


In pro and consumer equipment, different components and manufacturers have different acceptable ranges for the signal inputs and outputs. Exceed this range and you will produce clipped signals in the chain. As outlined in the article above, the ideal way to ensure that all your equipment is setup correctly as far as gain structure goes, is to actually measure the output and input ranges of all the components in your chain, but most folks aren't going to be doing that. In lieu of measuring the limits of the input and output ranges in each piece of equipment (specifically for the subwoofer in this case), the safer route is to adjust the subwoofer's gain so that the avr's trim settings are centered, and not too close to the extreme ends of the range.

 

 

Max, with your consent, I will add a slightly modified version of that paragraph to the FAQ where setting the trims to, ideally, +/- 3.5d is discussed. Currently we just say that this is a good idea but we don't mention the safety aspect at all. I will add it as a technical note, along with the link to the article you also quoted. The current discussion has reminded me of the potential danger to the sub if its input is clipped for some time (as my own was back in the day before I understood these things).

Reply
Reply
post #59157 of 62251
1/19/13 at 4:34am
Quote:
Originally Posted by kbarnes701 View Post

It wasn't hypothesis that destroyed my sub. It was clipping the input for 3 hours.  Not was it hypothesis in the measured result Batpig quotes above.  

Keith my friend, clipping of a sinusoidal signal means it's being transformed into a square wave. Feed this into a speaker and the speaker will play it. It will sound distorted, but the speaker does not care, it just plays what is fed into it. Feed a speaker with a distorted signal or a pure signal makes no difference for the speaker. You can't distroy a speaker with a clipped signal because the peak-to-peak voltage does not change. A speaker can only be destroyed by mechanical or thermal overload. A clipped signal will cause neither of these two cases.

Please be careful with amending the FAQ. If you don't beleive me (as I see you don't), then why not ask around or study electronics a bit more.

Take care. smile.gif
Reply
Reply
post #59158 of 62251
1/19/13 at 4:41am
What I learned, and it may be wrong, is that reproducing a square wave generates higher order distortion at high frequencies thus making it possible for the tweeter to be overloaded and burn out. That's one of the reasons for active loudspeakers being superior, the distortion is only fed to the driver powered by the clipping amp.
Reply
Reply
post #59159 of 62251
1/19/13 at 5:37am
Quote:
Originally Posted by mogorf View Post

Quote:
Originally Posted by kbarnes701 View Post

It wasn't hypothesis that destroyed my sub. It was clipping the input for 3 hours.  Not was it hypothesis in the measured result Batpig quotes above.  

Keith my friend, clipping of a sinusoidal signal means it's being transformed into a square wave. Feed this into a speaker and the speaker will play it. It will sound distorted, but the speaker does not care, it just plays what is fed into it. Feed a speaker with a distorted signal or a pure signal makes no difference for the speaker. You can't distroy a speaker with a clipped signal because the peak-to-peak voltage does not change. A speaker can only be destroyed by mechanical or thermal overload. A clipped signal will cause neither of these two cases.
 

 

 

That contradicts everything I have ever read about clipping. I am not especially knowledgeable about electronics, but the sites I read are written by people who are. On this occasion therefore, Feri, I prefer to go with the majority view on this, especially after destroying a subwoofer already. My understanding was that feeding the speaker a clipped signal would cause the voice coils to drastically overheat, thus eventually destroying them.

 

This is becoming OT for this thread and I don't especially wish to discuss it - as I said before, we will have to agree to disagree on this one, each one of us happy that we are right :)

 

EDIT: Even discounting the possibility of damage for a moment, it still makes no sense to feed the speaker with a clipped signal. Why would anyone want to risk that - from the distortion angle, let alone the  possibility of damage. Your argument seems to be that it is "alright' to feed the speaker a clipped signal, which seems strange to me.

 

Quote:
Please be careful with amending the FAQ. If you don't beleive me (as I see you don't), then why not ask around or study electronics a bit more.

Take care. smile.gif

 

The FAQ has always contained the advice to keep the speaker trims, ideally, in the +/- 3.5dB range - you have never commented on it before.

 

I have now amended the FAQ to take account of this issue and to elaborate on it. As there is without question no harm done by keeping the trims within +/- 3.5dB and as there is at the very least some argument about possible damage to speakers caused by using extreme trim levels, it seems most sensible to err on the side of caution. Nobody will damage his sub by using trim levels of +/-3.5dB so that is the best advice the FAQ can give.

 

You say that it makes no difference where the trims are set so you too must agree that +/-3.5dB is good ;)

 

The updated FAQ advice is here:

 

f)3.    How do I set the controls on my subwoofer before running MultEQ?


Edited by kbarnes701 - 1/19/13 at 5:43am
Reply
Reply
post #59160 of 62251
1/19/13 at 5:40am

Feri - if you are so confident that your theory of clipping is correct, perhaps you would conduct a small experiment for us that will settle it conclusively; feed your speakers a high level square wave for an hour or so and report back please ;)

Reply
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Receivers, Amps, and Processors

Gear mentioned in this thread:

AVS › AVS Forum › Audio › Receivers, Amps, and Processors › "Official" Audyssey thread (FAQ in post #1)