hearing
And---in much of the world of audio, and especially in recording studios, practically everyone is deficient above about---wait for it---4 kHz! This revelation is vigorously suppressed by various organizations but the truth is emerging despite this. What is the safe maximum level to avoid non-age-related deterioration? Apparently about 85dB. That's not all that loud!
Now that SMPTE is actually collaborating with AES after years of acting autonomously, we can anticipate some significant reduction in cinema audio levels arising from recent studies.
And we have the hilarity of a major and now very rich industry figure like Iovine, in a preposterously self-aggrandizing piece in Rolling Stone, declare that despite his profound hearing loss, he can hear what's wrong with something a block and a half away.
And so it goes.
Brad
What happens a lot is that IM distortion at high levels of otherwise-inaudible high frequencies will generate difference-frequency IM that is audible by HF-hearing-deficient subjects (no less that Brad Plunkett was complaining about this to me the other day). The type of nonlinearity that produces the diff frequency is the same that produces even-order harmonic distortion, the kind that people think of as generally innocuous and even benign. I agree with Putzeys and Keith Howard that it is not either one.
And---in much of the world of audio, and especially in recording studios, practically everyone is deficient above about---wait for it---4 kHz! This revelation is vigorously suppressed by various organizations but the truth is emerging despite this. What is the safe maximum level to avoid non-age-related deterioration? Apparently about 85dB. That's not all that loud!
Now that SMPTE is actually collaborating with AES after years of acting autonomously, we can anticipate some significant reduction in cinema audio levels arising from recent studies.
And we have the hilarity of a major and now very rich industry figure like Iovine, in a preposterously self-aggrandizing piece in Rolling Stone, declare that despite his profound hearing loss, he can hear what's wrong with something a block and a half away.
And so it goes.
Brad
Brad,
I know one recording engineer who is always wearing foam ear plugs most of the time to protect his hearing, it is his tool he makes money with and protects them. I have listened to him mixing and was shocked at how low level he worked at but he actually has a bunch of Gold records to his credit so I would say he seems to understand his craft.
I know one recording engineer who is always wearing foam ear plugs most of the time to protect his hearing, it is his tool he makes money with and protects them. I have listened to him mixing and was shocked at how low level he worked at but he actually has a bunch of Gold records to his credit so I would say he seems to understand his craft.
I agree. The waterfall diagram for my speakers is just fine uo to about +1 dB of nomial input voltage, and goes haywire rather fast, after about +2 dB. On the other hand, +22 dB works out to 35.6 Vrms, or roughly 158 Watts. That is listed as the absolute top short term input voltage, since their working spec is 100W of contiuous input. By that time, it should be producing (92+20) 112 dB SPL at 1 m, far too loud for any sustained listening, but good for dynamic peaks here abd there.
Hence its power rating at 100W continuous and 150W peak programme material. More han enough for me.
Curously enough, the waterfall diagram for my H/K PA 2400 power amp is exceptionally similar in shape and form, but at about 1-2 dB above the speaker. It is very curiosly specified, on different markets with the same grid voltage in one country it's specified as 170W/8 and in other countries as 200W/8. In theri manual, H/k specify it as 200W/8, and as 170W/8 as the lowest considered as "minimum". So, it will produce next to a perfect waterfall diagram up to 22 dBW, and will start to have problems at just above 23 dBW, literall 0.5 dB above that.
I can't wait to see what will happen when I refresh it and replace it's native 12,000uF/70V caps with larger, say 15,000 uF or better caps. The nominal line voltage is +/- 59.4V, which I suspect is the real limiting factor. It has no problems pumping out peaks of just above 700W into 2 Ohm loads and uses four pairs of 2SC3281/2SA1302 from Toshiba per side.
I can't wait to see what will happen when I refresh it and replace it's native 12,000uF/70V caps with larger, say 15,000 uF or better caps. The nominal line voltage is +/- 59.4V, which I suspect is the real limiting factor. It has no problems pumping out peaks of just above 700W into 2 Ohm loads and uses four pairs of 2SC3281/2SA1302 from Toshiba per side.
Last edited:
About damping of speakers, it is near impossible to cure many resonances in a satisfactory way by any electronic medecine, contrary to the response curve that you can correct by various ways.
Let me take an example. The bass speaker of a two way enclosure with a cross-over of, let says, 1000Hz.
Ok, there is all the chances that 1000Hz is near the point where the membrane begin to fractionate, but, till there, it is working nicely in piston.
The hightest damping factor of a normal amp is far to be enough. We can think to use some active servo ?
The result will be a signal send to the speaker with slew rate too brutal to let him staying in this piston area. I mean, it will excite the membrane's resonances, you know, the resonances witch give a colored sound, with a character coming from the speaker material itself.
Let me take an example. The bass speaker of a two way enclosure with a cross-over of, let says, 1000Hz.
Ok, there is all the chances that 1000Hz is near the point where the membrane begin to fractionate, but, till there, it is working nicely in piston.
The hightest damping factor of a normal amp is far to be enough. We can think to use some active servo ?
The result will be a signal send to the speaker with slew rate too brutal to let him staying in this piston area. I mean, it will excite the membrane's resonances, you know, the resonances witch give a colored sound, with a character coming from the speaker material itself.
Last edited:
@Kindhornman, bcarso
What you guys are saying sounds almost like the Son Audax sales pitch went. It's been 13 years, but if memory serves they said something much along those lines, that to get the best of a titanum tweeter one had to have its own resonance either below its crossover or above its nominal cutoff frequency. And it had to be fairly "clean", or we would get a shrill sound. I don't know if I still have their sales literature, I'll dig some tomorrow morning, to refresh my own memory.
But the point is, it is not shrill, or odd in any way, and it does ambience like a dream. At the time, in 2002, this was their second best high frequency driver, right ofter their somewhat odd "Tiger Eye" yellow tweeter, which was franky a letdown for me, but it seems everybody else as well, very few ever made it to the market. I do remember that at the time, top of the line tweeters from other better known sources varied in prices from about 45 to about 85 Euros, this one cost €98. A few years after that, their prices escalated, no idea why, but they went nuts crazy.Today, for current equivalents you can expect to pay €150-300 per piece.
What you guys are saying sounds almost like the Son Audax sales pitch went. It's been 13 years, but if memory serves they said something much along those lines, that to get the best of a titanum tweeter one had to have its own resonance either below its crossover or above its nominal cutoff frequency. And it had to be fairly "clean", or we would get a shrill sound. I don't know if I still have their sales literature, I'll dig some tomorrow morning, to refresh my own memory.
But the point is, it is not shrill, or odd in any way, and it does ambience like a dream. At the time, in 2002, this was their second best high frequency driver, right ofter their somewhat odd "Tiger Eye" yellow tweeter, which was franky a letdown for me, but it seems everybody else as well, very few ever made it to the market. I do remember that at the time, top of the line tweeters from other better known sources varied in prices from about 45 to about 85 Euros, this one cost €98. A few years after that, their prices escalated, no idea why, but they went nuts crazy.Today, for current equivalents you can expect to pay €150-300 per piece.
Last edited:
If you had a speaker which does not break up and had no other resonances except the one fundamental...... you add some long decay times when you introduce a crossover. If you do a water fall plot on a cross over with a resistive load, you see the corners of the filer having longer decay times.
So, it would be best to not cross-over where the ear is sensitive (midrange) and use wide band drivers.
THx-RNMarsh
So, it would be best to not cross-over where the ear is sensitive (midrange) and use wide band drivers.
THx-RNMarsh
Last edited:
Reason why i use compensation networks to flatten the impedance curve of the driver ? They damp the speakers a lot.
You are right. It was very easy to "voice" the crossover of my big enclosures in the studio (700Hz) while it took years to fine tune the one in my home (1500Hz) while the horn driver was the same.
I mean .... any LCR crossover -- active or passive will have its own resonances freq with its own decay time. That is added to what ever the speaker does independantly. I did this test with MLLSA back when it was still DOS OS. Might be able to do a SIM of it now? You dont just get a freq response for no neg side. The time delay of the LCR network will be added as well but you have to look for it in the correct domain (water-fall plot).
It is very hard to hide time and freq amplitude variations and decays in the mid range....... 700Hz is in the middle of voice range. I love compression drivers but they never went low enough and the drivers below the midrange didnt do such a good job up to and thru the mid range. It was too frustrating for me even though I loved the clarity of good compression drivers (with conical horn) an octave away from the crossover was great. At least with JBL drivers of the time. But even now, you just cant get great clarity in the mid range with almost any driver if there is an LCR in the midrange. I used 6db/oct crossover to midrange horn and shattered the diaphram according to JBL. They repaired/replaced it for free. The slope was too gradual and the excursion was too great..... yet if I went higher in freq crossover with the single Cap the low-mid and bass driver sounded worse.
THx-RNMarsh
It is very hard to hide time and freq amplitude variations and decays in the mid range....... 700Hz is in the middle of voice range. I love compression drivers but they never went low enough and the drivers below the midrange didnt do such a good job up to and thru the mid range. It was too frustrating for me even though I loved the clarity of good compression drivers (with conical horn) an octave away from the crossover was great. At least with JBL drivers of the time. But even now, you just cant get great clarity in the mid range with almost any driver if there is an LCR in the midrange. I used 6db/oct crossover to midrange horn and shattered the diaphram according to JBL. They repaired/replaced it for free. The slope was too gradual and the excursion was too great..... yet if I went higher in freq crossover with the single Cap the low-mid and bass driver sounded worse.
THx-RNMarsh
Last edited:
Richard,
You have those who don't want any crossover and go for a full range driver but I must say I have never seen any device that I think has a chance of actually doing full range or anything close to it so that it out. Then we have two way systems and choosing the crossover point is also difficult if the bottom to mid is covered by a large format driver like a 15" as Christophe has just pointed out once above the pistonic range things start to fall apart in so many ways. With smaller bookshelf and stand mounted speakers things are a bit simpler if you aren't trying to play pipe organ music. I really see large format speakers as working best as a three-way system, you can use each device to its best advantage that way but then the network really does have to get it right.
I really don't understand why more speaker companies don't include a simple resonance compensation network with there drivers or at least some helpful hints on how to do that, it makes things better for the speaker and the driving amplifier if they are all working with a flat impedance curve.
You have those who don't want any crossover and go for a full range driver but I must say I have never seen any device that I think has a chance of actually doing full range or anything close to it so that it out. Then we have two way systems and choosing the crossover point is also difficult if the bottom to mid is covered by a large format driver like a 15" as Christophe has just pointed out once above the pistonic range things start to fall apart in so many ways. With smaller bookshelf and stand mounted speakers things are a bit simpler if you aren't trying to play pipe organ music. I really see large format speakers as working best as a three-way system, you can use each device to its best advantage that way but then the network really does have to get it right.
I really don't understand why more speaker companies don't include a simple resonance compensation network with there drivers or at least some helpful hints on how to do that, it makes things better for the speaker and the driving amplifier if they are all working with a flat impedance curve.
When we talk about waterfall of speakers, we are talking about >300 ms of resonances, Richard.
Richard,
As you know there is no free lunch in any of this, you have to choose your poison well and take the best you can. Until someone comes up with a way to change physics and make everything linear and phase linear at the same time there are no perfect solutions. I assume I am looking at a port resonance at the bottom end.
As you know there is no free lunch in any of this, you have to choose your poison well and take the best you can. Until someone comes up with a way to change physics and make everything linear and phase linear at the same time there are no perfect solutions. I assume I am looking at a port resonance at the bottom end.
Last edited:
I never found drivers of any type if crossed over in the critical midrange to work good enough for me....partly because of the issues I just brought up. However, there are electrostatics and there are planar (some narrow strips up to 7 feet long) which can be crossed over at 200-300Hz.... the beginning of the voice range. They have great potential to do it all well, IMO.
BUT,if I could find a compression driver that could be crossed over below 300Hz,I would include it in my short list.
There are a few cone/dome drivers which can cover the midrange well and leave the audibly problematic crossover regions out of the midrange.
A couple other problems are minimized in audible significance is matching the dispersion characteristics of two drivers (say a cone and a compression/horn) at the cross-over region - and -
Mismatched distortion characteristic series...... which really makes a voice sound strange when crossed over in the voice range.
THx-RNMarsh
Last edited:
All of which you guys mentioned is why I believe a three way is optimal, although also considerably harder to get right.
Richard's point on XO points is well taken, albeit made in vain. I say that because in all my life I've never seen or heard any nominally 3 way speaker have its lower XO point, bass to mid, below about 450 Hz, and its upper XO point fares somewhat better, as there are two strategies. The usual one is to have the XO placed somewhere between say 2 kHz and 8 kHz at best. Between thes etwo, I'd prefer the 8 kHz, but in practice that may not be so good. The other strategy is to have a two way design with a super tweeter, in the mnner of the old B&W DM 2, with a Celestion dome driver taking over at somewhere around 18 kHz, if memory serves. As ever, both approaches have their pros and cons.
In my case, the lower XO point is at 800 Hz, since my bass drivers was recommended to be crossed over at 1 kHz or lower. My upper XO is at 3.5 kHz, even if my midrange cone was said to be able to be linear up to 7 kHz, and the tweeter was supposed to be used no lower than 2 kHz, and that only with a 12 dB/oct filter.
Anyway, I see in the current JBL catalog (which may be different in Europe than in USA) that their better moels use two bass units up to 2*10" sizes, a cone midrange and a horn loaded tweeter. They have their own sexy names for each of the above, and unfortunately, I haven't heard any of them - yet. My wife's Ti600 pair of floorstanders uses two 6.5" bass drivers, 1 4" midrange cone, and 1 1" titanium dome, which is quite reasonable, no obvious vices. All drivers are Son Audax made, as evident when you take off the JBL sticker. Overall, pleasing sound, but not really exciting.
Richard's point on XO points is well taken, albeit made in vain. I say that because in all my life I've never seen or heard any nominally 3 way speaker have its lower XO point, bass to mid, below about 450 Hz, and its upper XO point fares somewhat better, as there are two strategies. The usual one is to have the XO placed somewhere between say 2 kHz and 8 kHz at best. Between thes etwo, I'd prefer the 8 kHz, but in practice that may not be so good. The other strategy is to have a two way design with a super tweeter, in the mnner of the old B&W DM 2, with a Celestion dome driver taking over at somewhere around 18 kHz, if memory serves. As ever, both approaches have their pros and cons.
In my case, the lower XO point is at 800 Hz, since my bass drivers was recommended to be crossed over at 1 kHz or lower. My upper XO is at 3.5 kHz, even if my midrange cone was said to be able to be linear up to 7 kHz, and the tweeter was supposed to be used no lower than 2 kHz, and that only with a 12 dB/oct filter.
Anyway, I see in the current JBL catalog (which may be different in Europe than in USA) that their better moels use two bass units up to 2*10" sizes, a cone midrange and a horn loaded tweeter. They have their own sexy names for each of the above, and unfortunately, I haven't heard any of them - yet. My wife's Ti600 pair of floorstanders uses two 6.5" bass drivers, 1 4" midrange cone, and 1 1" titanium dome, which is quite reasonable, no obvious vices. All drivers are Son Audax made, as evident when you take off the JBL sticker. Overall, pleasing sound, but not really exciting.
Yes, and we should be able to get a couple hundred Hz lower XO by now would be perfect IMO.For compression drivers -- a lot of excellent work was posted over at the Loudspeaker forum by Welter (6thMay 2012) see Multi-way > "high Freq Compression driver Evaluation".
THx-RNMarsh
- Status
- Not open for further replies.