Although I can see some points within that generalization that are true (taming room modes and getting good bass) overall I would disagree. In real life the engineer and the listener are doing very different tasks and usually end up with very different rooms suited to those tasks. The engineer is taking a bunch of close mic'd tracks that are very one dimensional and usually quite different sounding from each other and trying to bring them together into a cohesive sound with a lot of EQ, dynamics processing and artificial echoes and reverb. Most seem to prefer a fairly dead room to do this in so they can tell whether the spaciousness they're hearing is actually down on tape or not. A listener doesn't need to know this and I think he can in fact enjoy the mix better if he steps back from the "microscope" so to speak. I'd make the analogy to a chef and a diner. The chef needs to carefully weigh and measure his ingredients as he mixes. The diner will not enjoy the meal more by setting himself up with the same tools as the chef so he can tryi to measure them as well.
In the end it's all a matter of preference, but I'd bet money that if you go over to one of the recording forums and take a quick survey the large majority of engineers do NOT listen to music at home on near field monitors in a replica of a control room. They take a lot of pride in being able to make something in the control room that is somewhat abstract, something that will "translate" well as they put it when it's played in a real room. They make their mix knowing that the final step has yet to be done. I think of it kind of like preparing something for projection. It's not the reel of 35mm film we're interested in when we go to a movie, it's the image that comes to life when we see it on the screen. The waveform on an album is like that reel of film to me. You could listen to it on near perfect headphones and have an exact reproduction of what's on the disc but I don't find that as satisfying as hearing it projected into room.
poptart , control rooms are not built to be acoustically dead. And if a recording engineer needs to have a dead room to confirm that his recording is dry enough, headphones are used. Control rooms are constructed in such a way that they provide a reflection free zone, i.e. ITDG of 20 ms. This derives from observations by Beranek who did studies on what parameters make up a good sounding concert hall. Take a look at typical control rooms:
http://www.rpginc.com/proaudio/projects/professional_recording_studios.htm
To make a instrument sound near with the sensation of being in a large hall, the ITDG needs to be big. Bigger than that 2 speakers in a normal small room provide. That results in the need of delaying and absorbing early reflections in a normal listening room so they don't cover the room information on the recording. This requirement is valid for control rooms and high quality listening rooms at home alike. A drawback is that you loose spaciousness and envelopment. But that's why multichannel was invented.
Best, Markus
P.S. And please don't say NS-10 when talking about good "monitor" speakers, say Klein+Hummel, ME-Geithain or Genelec.
http://www.rpginc.com/proaudio/projects/professional_recording_studios.htm
To make a instrument sound near with the sensation of being in a large hall, the ITDG needs to be big. Bigger than that 2 speakers in a normal small room provide. That results in the need of delaying and absorbing early reflections in a normal listening room so they don't cover the room information on the recording. This requirement is valid for control rooms and high quality listening rooms at home alike. A drawback is that you loose spaciousness and envelopment. But that's why multichannel was invented.
Best, Markus
P.S. And please don't say NS-10 when talking about good "monitor" speakers, say Klein+Hummel, ME-Geithain or Genelec.
Any sentence that starts with a generalization like this is wishful thinking because there is no standard control room, no standard monitor speaker, no standard mixing technique, standard set of engineer's ears (people in the music industry tend to have more damage at a given age than usual) or anything else standard about this in the real world. Wouldn't it be great if this were true and sound reproduction could be solved by a simple formula like "just do what the pros do" but there is no "what the pros do". It's a free for all out there. I've been in control rooms smaller than my bedroom with bark on the walls (cork?) and nothing more acoustically sophisticated than racks of gear behind the engineer. Somehow this studio managed to create hit records despite it's lack of a RFZ.
The nasty sound of ns10s was exactly my point. I think we'd probably all be surprised to find out how many of our albums were mixed on junk like that and if you really think the holy grail of audio is hearing it just like the engineer did you need to get a pair. Radiohead is a very famous band right? They've sold umpteen millions of copies of their albums. Their much lauded producer Nigel Godrich mixed every album on ns10s.
Here's a quote from another very famous engineer/producer, Steve Albini:
(if you don't listen to rock you probably haven't heard of this guy but he's recorded everyone from the John Spencer Blues Explosion to Nirvana, Cheap Trick to the Pixies. List of work here http://en.wikipedia.org/wiki/List_of_Steve_Albini's_recording_projects)
Sorry for my late answer as I was on a business trip for the last week.
Thank you for all your interesting contributions, although they are in my opinion slightly off topic now
.
Intentionally the idea of constructing an elliptic horn for the tweeter was to reduce early reflections from the floor and to homogenize the directivity on the horizontal axis. This would mean something like a nearfield speaker.
I didn't want to reduce all reflections in my listening (from acoustic point of view a completely unperfect) room. The side and wall reflections are still wanted, as they increase the impression of a wider room. Although I once heared a 4 way DSP controlled speaker system in a reflection free room - and I was extremely impressed of the natural sound with many details I never heard before on that CD (LTJ Bukem).
So from your experience does the construction of an elliptic or what soever like horn make sense? Will bundling of the tweeter sound make any positive effect on the sound reproduction under at home listening conditions?
Thank you for all your interesting contributions, although they are in my opinion slightly off topic now
.Intentionally the idea of constructing an elliptic horn for the tweeter was to reduce early reflections from the floor and to homogenize the directivity on the horizontal axis. This would mean something like a nearfield speaker.
I didn't want to reduce all reflections in my listening (from acoustic point of view a completely unperfect) room. The side and wall reflections are still wanted, as they increase the impression of a wider room. Although I once heared a 4 way DSP controlled speaker system in a reflection free room - and I was extremely impressed of the natural sound with many details I never heard before on that CD (LTJ Bukem).
So from your experience does the construction of an elliptic or what soever like horn make sense? Will bundling of the tweeter sound make any positive effect on the sound reproduction under at home listening conditions?
Aoxomox said:
I think that the answer to this question is an unqualified "Yes". But the question that is more relavent is how far to go. To what extent does a small horn work when compared to a larger one? I make waveguide from 17" down to 10" in mouth size - all the same coverage angle. There are similarities and there are differences, but in gnerela bigger is better. There is nothing acoustic (not size and cost of course) about the smaller waveguides that is better. So to what extent do you want to shrink the device until it is "aestetically" pleasing and within your budget. ALL waveguides are an improvement, bigger ones improve more. Is there a sweet spot? Wish that I knew!
I will say that I have found a 90 degree device as being the optimum and can point to some rationale for this. Wider than this and the diffraction rises, narrower than this and the device gets too long.
Aoxomox, sorry for being off-topic again. I promise that it's my last post on that topic here.
poptart, we're talking at cross purposes. I'm talking about high quality sound reproduction based on psychoacoustic observations. The fact that an album sold a lot of copies is not correlated to its sound quality. I believe one can enjoy music even with the standard stereo of a Golf I.
Albini is right that it's possible to "hear through" a room. But you need a lot of experience that you simply don't get when listening to reproduced sound at home. By the way, his control room doesn't look very arbitrary to me: http://electrical.com/StudioA/controlroomA.php
You'll find this "LEDE inspired" type of control rooms everywhere. I would call it a standard.
So the goal still is to have a setup at home that is capable of delivering what is on the record without giving wrong cues (like too early reflections from the listening room). Wrong cues will make information on a recording dissapear. Yes, there are recordings that don't carry such detailed information (lots of them). And yes, you can enjoy music with one speaker dead and the drivers of the second one near dead (lots of fun). That doesn't change the fact that there are recordings out there that deliver more than just the notes.
Just an interesting sidenote on mastering: http://www.chicagomasteringservice.com/loudness.html
The whole discussion started with my statement that I don't consider the floor reflection as being very important. This reflection is there all the time in our everyday live. It's more like a constant because we tend not to be more or less tall on different days. However side wall and ceiling reflections change strongly with every room we visit. Maybe we can discuss that topic in another thread.
Best, Markus
poptart, we're talking at cross purposes. I'm talking about high quality sound reproduction based on psychoacoustic observations. The fact that an album sold a lot of copies is not correlated to its sound quality. I believe one can enjoy music even with the standard stereo of a Golf I.
Albini is right that it's possible to "hear through" a room. But you need a lot of experience that you simply don't get when listening to reproduced sound at home. By the way, his control room doesn't look very arbitrary to me: http://electrical.com/StudioA/controlroomA.php
You'll find this "LEDE inspired" type of control rooms everywhere. I would call it a standard.
So the goal still is to have a setup at home that is capable of delivering what is on the record without giving wrong cues (like too early reflections from the listening room). Wrong cues will make information on a recording dissapear. Yes, there are recordings that don't carry such detailed information (lots of them). And yes, you can enjoy music with one speaker dead and the drivers of the second one near dead (lots of fun). That doesn't change the fact that there are recordings out there that deliver more than just the notes.
Just an interesting sidenote on mastering: http://www.chicagomasteringservice.com/loudness.html
The whole discussion started with my statement that I don't consider the floor reflection as being very important. This reflection is there all the time in our everyday live. It's more like a constant because we tend not to be more or less tall on different days. However side wall and ceiling reflections change strongly with every room we visit. Maybe we can discuss that topic in another thread.
Best, Markus
markus76 said:
Markus - we should move this topic, agreed.
You argument is plausible and I would agree that the floor reflection is the least troublesome from experince, but I don't think that there is any solid data to substantiate your belief and I think that it is easy enough to get rid of the floor reflection so I do it. But if your saying that the floor reflection is desirable and should be left alone, then I would have to say that I'd need to see some supporting data as your comments are all based on your experinces and some "logical" arguments, but no solid data.
As you can see from the measurements a nominal 5-5.5 inch driver crossing over in the 3kHz. region to a 90degree waveguide around 78mm. in diameter can give a smooth constant directivity characteristic above 3kHz.
If you want to extend the constant directivity frequency downwards you need a bigger waveguide, that in a three way will encompass the mid driver for best results.
If you lower the crossover to much below 2kHz. then the linear dome excursion becomes an issue, and the Visaton tweeters I have seen the specs for do not feature large linear excursions.
In the end constant directivity is a good thing and the lower in frequency you go arguably the better, but the bigger the speaker, and as we know for most people the,"faf", ( female acceptance factor), trumps all other considerations.
rcw.
If you want to extend the constant directivity frequency downwards you need a bigger waveguide, that in a three way will encompass the mid driver for best results.
If you lower the crossover to much below 2kHz. then the linear dome excursion becomes an issue, and the Visaton tweeters I have seen the specs for do not feature large linear excursions.
In the end constant directivity is a good thing and the lower in frequency you go arguably the better, but the bigger the speaker, and as we know for most people the,"faf", ( female acceptance factor), trumps all other considerations.
rcw.
I would agree with all of this. Which is why my next design will be a smaller speaker system not a larger one. People seem interested in *talking* about large systems with premium performance, but they always seem to compromise in their actual purchases.
I am going to use a B&C DE10 and an 8" woofer so the system will be quite small. It should also be relatively inexpensive as these drivers cost a lot less than the others and there is less cabinet material, etc. The crssover will have to be scalled back in complexity to keep the cost down (this is a concern). As a kit it may be arround $300. The waveguide may be elliptical.
I am going to use a B&C DE10 and an 8" woofer so the system will be quite small. It should also be relatively inexpensive as these drivers cost a lot less than the others and there is less cabinet material, etc. The crssover will have to be scalled back in complexity to keep the cost down (this is a concern). As a kit it may be arround $300. The waveguide may be elliptical.
Interesting idea, but I doubt that its practicle. What will more than likely happpen is that the crossover point will just move up. It has to with the DE10. BUT, if I find that the 8" woofer is too wide at the crossover point to match the waveguide then there are two ways to solve this. A waveguide on the woofer or a wider waveguide on the tweeter. Of the two the wider tweeter waveguide would be the more effective I would think. But thats waht design is all about - concept -> test -> modify -> test and hopefully your done
gedlee said:
Compression drivers work better on waveguides than domes, because compression drivers are designed to be mated to horns or waveguides.
To be specific, there's a hideously complex phase plug that the diaphragm of the compression driver is mated to. Here's a pic of one; you can see it provides a smooth transition from the dome to the throat:
An externally hosted image should be here but it was not working when we last tested it.
That's how we can get directivity control from a compression driver with a three inch diaphragm and a one-inch throat; it's the phase plug. Without the phase plug we'd be SOL.
Having said that, I've long been curious about ring radiators, because they have a built in wavguide, that goes right through the diaphragm. My hunch is that a ring radiator may be a good candidate for a waveguide.
Not saying that it's a replacement for a compression driver, but it MIGHT be a step up from a dome.
In other words, perhaps a ring radiator is a candidate for projects where directivity is required, but there isn't enough space or budget for a compression driver.
Based on this line of thought, I found some data on ring radiators on waveguides.
In the pic below, I've compared the following:
1 - Gedlee Summa. The speakers that are my reference, and which have an uncompromising 90 degree waveguide. Note the smooth and consistent polar response.
2 - BMS 4540ND compression driver on an XT1086 waveguide. The XT1086 isn't perfect, but it's darn good. It has a weird step in the throat (check out the pic) which keeps it from being as good as it could be. But check out the vertical plots! They're excellent. (The step in the throat only exists in the horizontal axis, not the vertical.)
3 - BMS 4540ND on MCM h65. Same waveguide as above, but on a $12 waveguide. This is the same one used by Zaph for his famous waveguide speaker. Note that the BMS doesn't work well, likely due to the abrupt change in coverage angle. (There's a sharp and abrupt transition at the throat of the WG.)
4 - "ring radiator" on MCM H65. Here's the interesting one. The same MCM waveguide with a "ring radiator" has impeccable performance. This is a $12 waveguide and a $40 tweeter, and it's polar response is almost state of the art. This kind of performance would be remarkable with a compression driver, but with a conventional tweeter it's unheard-of.
There are a couple of other reasons that a ring radiator is an intriguing candidate for a waveguide. A compression driver uses a substantial diaphragm, and due to it's mass, it's upper frequency limit is relatively low. That's why uber-expensive compression drivers like the TAD2001 use a beryllium diaphragm. But the diaphragm of a ring radiator is miniscule, and it's top end is close to 30khz. I am only aware of one compression driver with output that can approach it - the BMS 4540ND. Not coincidentally, it's also a ring radiator.
An externally hosted image should be here but it was not working when we last tested it.
The source of my data is here:
http://www.aeronet.com.au/waveguide.htm
http://www.htguide.com/forum/showthread.php4?t=20386
An externally hosted image should be here but it was not working when we last tested it.
This is the EV SX500+ which they claim to have controlled the dispersion down to 500hz using a waveguide on the woofer. Having never seen this speaker in person, I have no idea how big or deep that portion is, but you might find something similar could be done for the 8" woofer in order to offer pattern control down to 1000-1500hz.
I'm surprised this isn't more common. Back when I was scheming on a way to clone a Summa, my idea was to use a 12" wooofer instead of a 15", and put the twelve in a waveguide. You get a few advantages with this:
1 - Allows you to use a smaller woofer
2 - The waveguide will increase gain from 700hz and up, right where you need the extra SPL to mate with the compression driver
3 - There aren't many 15in woofers that can play to 1khz, but there are dozens of 12in woofers that will
You could take this to an extreme and use an eight inch woofer too.
1 - Allows you to use a smaller woofer
2 - The waveguide will increase gain from 700hz and up, right where you need the extra SPL to mate with the compression driver
3 - There aren't many 15in woofers that can play to 1khz, but there are dozens of 12in woofers that will
You could take this to an extreme and use an eight inch woofer too.
An externally hosted image should be here but it was not working when we last tested it.
Patrick Bateman said:
Compression drivers work better on waveguides than domes, because compression drivers are designed to be mated to horns or waveguides.
To be specific, there's a hideously complex phase plug that the diaphragm of the compression driver is mated to. Here's a pic of one; you can see it provides a smooth transition from the dome to the throat:
An externally hosted image should be here but it was not working when we last tested it.
That's how we can get directivity control from a compression driver with a three inch diaphragm and a one-inch throat; it's the phase plug. Without the phase plug we'd be SOL.
Having said that, I've long been curious about ring radiators, because they have a built in wavguide, that goes right through the diaphragm. My hunch is that a ring radiator may be a good candidate for a waveguide.
Not saying that it's a replacement for a compression driver, but it MIGHT be a step up from a dome.
In other words, perhaps a ring radiator is a candidate for projects where directivity is required, but there isn't enough space or budget for a compression driver.
Based on this line of thought, I found some data on ring radiators on waveguides.
In the pic below, I've compared the following:
1 - Gedlee Summa. The speakers that are my reference, and which have an uncompromising 90 degree waveguide. Note the smooth and consistent polar response.
2 - BMS 4540ND compression driver on an XT1086 waveguide. The XT1086 isn't perfect, but it's darn good. It has a weird step in the throat (check out the pic) which keeps it from being as good as it could be. But check out the vertical plots! They're excellent. (The step in the throat only exists in the horizontal axis, not the vertical.)
3 - BMS 4540ND on MCM h65. Same waveguide as above, but on a $12 waveguide. This is the same one used by Zaph for his famous waveguide speaker. Note that the BMS doesn't work well, likely due to the abrupt change in coverage angle. (There's a sharp and abrupt transition at the throat of the WG.)
4 - "ring radiator" on MCM H65. Here's the interesting one. The same MCM waveguide with a "ring radiator" has impeccable performance. This is a $12 waveguide and a $40 tweeter, and it's polar response is almost state of the art. This kind of performance would be remarkable with a compression driver, but with a conventional tweeter it's unheard-of.
There are a couple of other reasons that a ring radiator is an intriguing candidate for a waveguide. A compression driver uses a substantial diaphragm, and due to it's mass, it's upper frequency limit is relatively low. That's why uber-expensive compression drivers like the TAD2001 use a beryllium diaphragm. But the diaphragm of a ring radiator is miniscule, and it's top end is close to 30khz. I am only aware of one compression driver with output that can approach it - the BMS 4540ND. Not coincidentally, it's also a ring radiator.
The source of my data is here:
http://www.aeronet.com.au/waveguide.htm
http://www.htguide.com/forum/showthread.php4?t=20386
Very interesting stuff here. Thanks for this.
I have some parts express 12" waveguides I was going to try putting to use as a DJ speaker, but maybe I will try some ring radiators with it as a pet project instead. It sounds like that JBL part you commented on in the horn vs waveguide thread would be a better bet for the DJ speaker...
Thanks again.
-Tony
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.