Hi Tom,
Thanks for the info. I was actually very surprised to find that large a value for the cap at 400uF+. Is it because the intrinsic impedance of the driver is somewhat low? Or is more has to do with the time-phase aligned nature of the xover?
Thanks. |
For me the Magico A3 were very impressive on a number of audiophile score-card characteristics, but they ultimately failed to really engage me.
I was wondering if this has to do with Thiel usage of first order design. I mean if you're already bought in Thiel first order philosophy, any other speakers with higher order filters will fail you in term of "emotional engagement". Personally I already made my own conclusion based on my own designs that first order results in the most musical reproduction of music, so I am sort of in the same camp as well. |
Andy - I second your motion. As car guys say, the least expensive option is the one you have. So if you have 2.4s, then upgrading resistors and caps is a huge bang for the buck.
It's just so happen that my current car is a 2004 model so I would agree :-). |
Sometimes you take it for granted what you have. After hearing a few garden varieties speakers even at pretty high price, you realize how good the CS2.4. The only thing I wish for is a bit more bass and I suspect some would agree. I guess the solution is:
1. If you like 2.4 but something better, then try to get another 7 ... that is either 2.7 or 3.7.
2. If you still want to keep the 2.4, upgrade the xover with audiophile quality caps.
3. Augment the CS2.4 with two high quality sub's - one for the left and oe for the right channel. |
Thiel used Jantzen and ERSE wire coils as equivalents.I have ERSE foils on hand and some Jantzen Wax on order.I will compare them directly, but not yet.
I definitely look forward to your impression of the two coils. |
how do Thiels sound compared to Vandersteens?
I personally have listened to a pair of Vandersteens driven by some very high end tubes amps at an audio show and I have to say they sound magical. It's something that I could never forget. I currently have a pair of CS2.4 in my living room and I have to say it's not quite what I've heard with the Van. at the show, but the amp I have is not some high end tube amps. I like the Thiel CS2.4 very much but everything considered, I would recommend the Van. since it's still in production and for most people, I think would be better. |
The more I listen to the CS2.4, the more beautiful they sound. I wish I could go back in time and hopefully I could make a difference. I've designed my own speakers but they are no equal to Thiel. |
I took a look at Crutchfield website and the Revel F208's are retailed at $5K for a pair. The CS2.4 was about $4.5K but taken into account for inflation, it would cost quite a bit more than $5K in today money so in that sense, the CS2.4 is a bit more high end compared to the F208. The F208 uses SB Acoustic aluminum drivers for the 8in. bass and 5.25in. midrange. The SB Acoustic drivers are more or less quality budget drivers. As for the tweeter, although I don't have information on it, judging by the price of the mid and bass drivers, the tweeter probably is of budget type as well. Although I have not listened to the F208, I have used similar SB Acoustic drivers in my own design. For the price they are decent, but the drivers are not in the same league as the drivers in CS2.4. Also the aluminum drivers used in the F208 have very large break up so it's almost likely that a 4th order filter must be used to suppress the break up. And as I have said before, if you're into first order design, you probably won't like 4th order filter. In my experience, the quality of the drivers are very important and will establish the limit of the final sound quality. Some of these drivers will have decent sound, but they don't quite have the inner beauty of the more exotic, more expensive drivers even with well design xovers. I have used high end drivers from Seas and ScanSpeak and some budget drivers, and after judging the sound quality vs. price, I have decided it's not worth it to use budget drivers. Here are the links for the F208 drivers if interested: https://www.madisoundspeakerstore.com/approx-5-woofers/sb-acoustics-sb15nbac30-4-5-black-aluminum-co...https://www.madisoundspeakerstore.com/approx-8-woofers/sb23nbacs45-8-8-black-aluminum-cone/ |
|
He preferred Vandersteen saying that the Vans were more musical on more recordings more often than Thiels. But he also zeroed in on what Thiel was doing, how we were approaching our work.
On the flip side, Vandersteen's were always been criticized for being dark and slow :-). He ended up having to defend his design saying that other speakers tend to have the tweeter running a bit too hot at 2-3db more than they should be. |
Hi Tom,
It was very well said. I understand the nature of the problem completely. I guess all I can say is that in a world where we have equipment makers such as DAC, preamp, amp, and last but not least speakers, and of course recording studios, each have their own ways of doing things. All I can say is how to try to fit in given all the variables.
As for how to voice a speaker, it's interesting that we have brands such as Sonus Faber, Spendor, Vandersteen who in the past tend to have a sound that is some what warm and musical, but have since become a lot more neutral in their sound lately.
In my personal design, when I design speakers with very flat frequency response, in order to pull that off, my external components have to be of fairly high quality (such as xover cap, front end electronics, cables ...) otherwise the sound will be somewhat on the harsh sound.
I have heard stories in which people have demoed Thiel speakers at the show room which sounded fabulous, but then they were very disappointed when they took the speakers home using their less than optimal equipment.
The brands which I have mentioned above whose sound has gone a lot more neutral lately may be because the quality of the
ancillary
equipment is now so much better than before, that they feel comfortable to be much neutral now without sacrificing sound quality. |
a fun journey....your results may vary....
as long as you have unlimited funds ... if you know what I mean. |
Out of curiosity with some of the discussions on speaker pricing, how much do you think the Thiel CS2.4 would be retailed for in today money (if they are built today?)
Also I know that most speaker manufacturers use CNC machines to cut the panel into shape? But in term of gluing them today, is it still human manual labors? |
Now I am looking to buy speakers used that new would be $20k- $30k and hope to pick them up for $10-$15k. I have given up the idea of buying new speakers for $10,000 or less.
It's interesting that speakers cost so much nowaday. $20K-$30 is about what an average car would cost today! My theory is that much of high end speakers still requires manual labor such as veneering, gluing, sanding and so on ... And human labor cost has inflated so much more than the government would have liked you to believe. It will be interesting to know which speakers you end up buying and how they are compared to the CS2.4. |
Chaning any of the origial drivers almost ensure the need of a new xover and since the Thiel is time-phase coherent, it's not something that can be done by just anybody. One almost have to use the original driver as replacement. |
Will I find the exact replacement I need that will bring harmony to the existing crossover? Not likely, but chances are good that I'll find something 'good enough'.
The thing is it's not like replacing a Mazda 6 with a Toyota Camry. If you have a different midrange driver, the sound balance may be too off that it would make the sound unlistenable and since our hearing is most sensitive to the midrange, the problem is exacerbated. For example, if the new midrange driver causes the midrange level to about 2-3db higher than it should be, then sound may actually hurt your hearing. Or if the new midrange driver ends up to have the region around 6-8khz to be elevated, you'll have excessive sibilance ... well I think you understand my reasons. |
I wish unsound post was a little bit more sound. |
Andy2, Really? Is that where this is going? I suggest that you either offer more constructive posts or work on your cleverness.
This thread has been taken hostage by unsound lately :-) |
Actually the Magnepan although physically composed of just one panel, it is not technically a single driver in term of phase response. Different parts of the panel will not move in complete synchronization therefore certain part of the panel will have different phase vs. some other part.
Also out of curiosity, why would you need a cross over for the Magnepan? Isn't it just a single panel? |
I wonder if Tom or someone else could comment on this. I found an image of a supposedly CS6 xover for the midrange. Based on the image, it seems the xover employs an all-pass filter for the midrange. In my experience, the all-pass dues to the amount of components, does take away some of the transparency. And seeing from the image, it does not seem like any bypass cap was used so it may help. I was just wondering if all-pass filter was widely used at Thiel for the purpose of time-phase aligning? I've read a review of the CS6 from Stereophile and the author did comment of the slight reticence of the midrange that may be related to the all-pass nework. See the excerpt below:
So far I've been digging deep into my box of superlatives. Did the CS6 have an Achilles' heel? It did, in that the midrange didn't offer quite the same degree of ultimate clarity or cleanness that so distinguished the bass or treble octaves. There was a feeling of reticence in this region, described by one visitor as a "hooded" quality, that I couldn't eliminate no matter how much I fooled around with placement. Tilting back the speaker by placing Black Diamond Racing cones under just the front did help, however, as a distinct change in timbre could be heard on the sit-down/stand-up test. But for me, at least, this was a minor problem, offset by the many things the Thiel did right. |
Hi Tom, Here is the link to the xover picture. Based on the website, it is supposed to be a CS6 xover picture. I was afraid of revealing possible intellectual property, but I guess since the website is public, it should not be an issue, at least from my stand point.
http://www.troelsgravesen.dk/vocals_files/cs6xover.jpg Thanks. |
Andy2, curious to know what you see on that board. How did you surmise an all-pass filter? An all-pass filter suppresses no frequencies, right? Seems like such a "filter" would invariably allow the driver to excite its resonant modes.
It's partly based on an opinion of someone who is fairly experienced in xover design. Also, based on my experience, it seems to me the only way to align the mid to the tweeter phase is to use an all-pass filter otherwise you have to invert the polarity (but I could be wrong). I have tried the all-pass filter in my own xover design and it takes away the transient speed and ultimately reduce the transparency. |
Hi beetlemania,
I guess I was just speculating and mostly was out of curiosity. I have not personally heard the CS6 so it would be hard for me to form any conclusion. Hopefully someone here who had listened to the CS and would offer his or her opinions on the CS6 to see how their opinions are compared to that of JA. |
As for the CS6, it's hard to tell from JA review where the "reticence" comes from. When I listen to my speakers for example, I can tell whether the "reticence" of a particular frequency comes from the lack of transparency or a dip in frequency. One could mistake a dip in frequency for "lack of transparency", and likewise, one could mistake a "lack of transparency" in the dip in frequency. You could have NO dip in frequency response but still you could still sense some type of "reticence".
Both aspects can be easily confused without having enough listening experience. As I have mentioned in my past posts, the CS2.4 does not have a dip in treble response, but as we have discussed in various posts here, due to less than ideal capacitor on the coax driver, there is some "lack in transparency" in the CS2.4 treble response but that has nothing to do with frequency response as a matter of fact I think the CS2.4 treble is fairly extended. And of course some have reported better details after upgrading the capacitors.
I wish there is a "transparency" measurement but as far as I can tell, there is no such thing. So what left is up to one's own intuition. And trying to argue about different people intuition could be a futile exercise! |
On the subject of amplifier, I swapped out my current amp for something a little less powerful. I have a spare SimAudio Moon W3 amp that is rated at 125W. Although it only has 125W, it is a fairly stout amplifier. But is it good enough to drive the CS2.4? For the most part, it's not bad, but occasionally it seems to lack a bit of dynamic as if it does not have enough current to drive the speakers and the bass lacking a bit of snap.
I think 125W is more or less the cut-off power requirement for the CS2.4. I also have a tube amp rated at 50W but it just does not have any dynamic driving the CS2.4. |
Those CS2.4 are mine. I am selling my entire system. I still have a secondary system but not quite as upscale though. I have not had any luck yet :-). I am not looking to purchase any speakers to replace the CS2.4. I probably will build my own. |
There is a sonic trait on the CS2.4 that I wonder if you agree. Maybe Tom can provide his inputs.
In speaker design the baffle is pretty important to the sound which has to do with diffraction which I won't go into detail here since it will just create an even opposing opinions. Anyway, the CS2.4 has the coax fairly lower from the top of the speakers. For comparison, the CS3.7 has the coax almost at the top with the aluminum cone on top.
When in a recording where there is a hard right or left mix, in the CS2.4, I do feel that the sound comes from the baffle which spoils the 3-D illusions somewhat. In some of my designs, I place the tweeter right close to the top of the speaker so in such recording, the image flows more from the air as opposed from the baffle.
So in the CS2.4, the high freq. sound has to traverse the entire upper baffle so you hear the reflection from the baffle and not directly from the tweeter. I suspect that Thiel may have had this in mind, therefore on the CS3.7 design, the coax is placed at the top of the baffle to create a better 3D illusions.
An example of baffle design which improves diffraction is employed by Avalon Acoustic which has the tweeter close to the top with the baffle swept back. I think Thiel curved baffle such as the CS3.7 is even better but it will probably cost more to manufacture.
|
Can’t say I’ve noticed this trait, Andy2. Some songs with the mix in one channel seem to emanate well outside the bounds of the speakers. The CS2.4 does this at least as well as the CS1.6 and Vandy 2Ce Sig II. bettlemania, I suppose it could be recording dependence. Also, although I am not a recording engineer, there are two types of hard left/right mix. One type there is still a phase relationship to the other channel so you hear the image well to the left or right of the speakers (for example at the beginning Welcome to the Machine from Pink Floyd which the CS2.4 does very well). The other type is just a pure left or right without any phase relationship to the other channel so you hear the sound basically directly from that channel speaker. I think I was referring to the latter one. If you could find some old recordings especially from the 60’s where the sound mix is basically left or right. In this regard, I was wondering if you could compare how the image "flows in the air" compared to other speaker design. As for the TAD Ref One, it has quite bigger bass drivers and midrange driver so I would expect it has more image density vs. the CS2.4. As I mentioned in my previous post, the 2.4 has a small mid so it does not sound as weighty in the lower midrange sound vs. speaker with larger mid driver such as the TAD Ref One. |
Also as we are on the subject, there is another design philosophy in which people believe that the baffle should be eliminated completely. This philosophy is adopted by Vandersteen. Part of the reason is that Vandersteen believes that the baffles modify the phase of the sound from the drivers and since he is a time/phase coherent guy, I suppose it is important to him.
|
So, Andy, even though the 2.4, like the earlier designs, does control diffraction very well, its lower tweeter creates a time discrepancy between side and top diffraction. That effect would be extremely subtle, such that I would be surprised if I could ever hear it. But, your ears are younger than mine.
Hi Tom, The subject of diffraction is very controversial and there are people who believe that you "can't hear" diffraction and people would go on and on about it without agreeing on anything. Personally, although I don't know if you can "hear it", but I think diffraction manifests itself in how well the speaker images. Of course there are other variables such as cabinet stiffness, cabinet shape, but I think diffraction plays a big role. I was wondering that the CS2.7 and CS3,7 being Thiel's latest designs, and the fact that the coax drivers were placed at the top of the baffle compared to older design, what was the main reason for that? Was it mainly to improve diffraction or was there other factors? Thanks. |
|
Anyway, since this is Thiel thread, I guess I am allowed to be a bit sentimental :-). I found this little excerpt regard to specially to the CS2.7 in which the reviewer says:
https://hometheaterhifi.com/reviews/speaker/floor-standing/thiel-cs2-7-floorstanding-speaker/
In purchasing a Thiel speaker, my sense is that you’re getting the very best effort from everyone involved, from the concept, design and engineering to fabrication, it feels like you’re buying the love and passion that any craftsman puts into a piece. When I look at the CS2.7, I could imagine staring at it in 20 years and thinking I made a wise investment. It could very well be as timeless as any speaker and it will be talked about fondly as long as we listen to music from a box pushing air.
I happen to own an old E46 M3 car which I have the same feeling. There are just to many "parts bin" products out there in either speakers or any other products. Thiel and a few that really stand out. By the way, has anyone able to hear the difference between the 2.4 vs. 27? It would be interesting to find out. |
Any 3.7 owner want to bring your speakers to my room?
Interestingly enough, that would be my next speakers but financially it's not possible for the moment. Hopefully some day. Sometimes things are just beyond one's wishes. |
That is a valid argument, but I found it is consistent in different type of instruments regardless of bass response. But anyway, as I mentioned in my previous post, this topic can bring up different opinions that people will end up having their own belief. |
With respect to unsound post regarding to baffle size, not to say which one is the most "correct", but there seems to be different design philosophies. Sonus Faber makes a few very wide baffle speakers. Thiel is somewhat in between with the curved baffle. Vandersteen basically eliminates the baffle altogether for most of its designs.
In my opinion, the wide baffle has its advantage but it also has its disadvantage. The wide baffle shifts the diffraction freq. lower, but then the problem is the low freq. image would suffer - that is the low freq. image would become "phasey" and shifting around and not stable.
I don't think you can eliminate diffraction completely. It's a matter of how to control diffraction for a given speaker design.
Just to show how the front baffle can affect the soundstage image, I found that (and most others) a small standmount monitor seems to be the best in term of "disappearing" since they have very small "sound foot print" with respect to the baffle size. Most large speakers cannot do this as well as small standmount since its baffle geometry is much larger and inevitably will get in the way of the sound front.
Anyway, baffle size, shape, and geometry will bring up different opinions and I guess different designers will have their own preferences. |
Thiel seemed to have solved that first-order crossover issue in their later designs.
Probably due to the use of coax drivers. The high freq./short wavelength xover between the mid and the tweeter is most problematic, but with coax, it's mostly not a problem. People have commented that the pre-coax Thiel also had problem with mid/tweeter integration with vertical listening axis. |
I even heard this mild hollowing-out interference a bit with the Thiel CS6 when I had it. Not nearly as bad as the Meadowlarks. I was frankly quite surprised when I got the 3.7s as to how this issue seemed totally banished. With both the 3.7s and my 2.7s, they sound remarkably even when varying listener position, either horizontally or vertically.
I have not heard the CS6 but having seen the picture of the CS6 coax, it may have been a older designer vs. the coax in the 3.7 and 2.7. I don't know. It's possible that the coax drivers in the 3.7 and 2.7 is an improvement over the older coax. |
I suppose the argument
about 1st order vs. higher filter will go on forever. Too
bad there are so few (besides Vandersteen I don't know who else)
actually make 1st speaker so people don't get to listen
and understand the actual sound so they end up instead hearing from
magazines who don't understand themselves. Of course magazines have there own agenda so who knows if they tell the truth.
I've had this little
exchange with someone who has been designing speakers as if his life
depends on it. And we are trying to “objectify” why the
difference and we both agree that the biggest difference is how the
high frequency or to be specific the treble is reproduced. With 1st
order, the treble is part of the music whereas with higher order
filter, and we actually agree that higher order generally has a cleaner,
clearer music reproduction, the treble seems to be riding on top of
the music, instead of being integrated within the music like 1st
order.
And since the treble is
where lies the most differences, what he does is that for example
with three way speakers, the xover between the mid and the tweeter,
he would use first order. The xover between the mid and the bass
which is around 400hz, he would use 2nd, and since the
wavelength at 400hz is so long, the phase does not matter much, and
of course, 2nd is much easier to implement vs. 1st.
This is consistent with speaker design in which at the low frequency
such as the bass, amplification quality is not as important and as
said above, our hearing is not that sensitive in the low frequency
due to the longer wavelength. Most sub woofers if not all using
digital amplification since using linear amplification probably does
not make much of a difference. Try using digital amp for your
tweeter :-)
Interestingly, after
listening to 1st speakers for awhile now, I seem to
attribute the sound to that of tube amplification. There is
something to 1st order sound that is similar to tubes.
And I don't think that is a coincidence either. Tubes have a way to
deal with treble better than transistors. Transistors although
always sound cleaner and clearer, compared to tubes, but they always
sound somewhat clinical and analytical compared to tubes. I won't go
into detail as to why since it may take quite a bit of spaces with
all the technical stuffs that might drive prof crazy.
This leads me to something
a bit analogous to what have said above with a slight twist. I
currently have two preamps – one is a Pass Labs X10 which is
transistor-based, and the other is a Conrad Johnson 17LS which is
tube. Both are very good in their respective domain. The X10 is
actually very smooth and warm and fluid in a transistor sort of way.
The 17LS is a touch bright and extended vs. the X10, so you would
think I would hear more “treble” on the 17LS, but that is not the
case, because with the 17LS I just hear music because the treble is
so well integrated with the entire musical range. On the X10,
although sounding a touch warmer vs the 17LS, I would hear the treble
somewhat sticking out like it is “riding on top of the music” as
I have mentioned above. BUT here is the KICKER. When I design my
speakers, I always use the X10 to fine tune my speakers because the
X10 is more neutral and it is able to tell me the strength and
weakness of my speakers better than the 17LS, and it allows me to
better optimize my speakers xover. And of course with better optimzied speakers, I can enjoy them better with my CJ 17LS
:-). So I guess it's a complementary, symbiotic thing.
A lot of time, with music
and our hearing, things can get a bit complicated and people can get
overly emotional, but there is something analogous to “sound”
that can be more easily be “objectified” and generally agreed
upon. And that is the physics of “light”. As with sound, it is
frequency dependent. For example, lower frequency light tends to be
reddish, whereas higher frequency light tends to be more bluish. And
as analogous to sound, a image that has a lot of low frequency, it
tends to be a bit less “sharp”, and likewise, an image that has a
lot of high frequency content, the image will appears sharper and
clearer.
Most people at one time or
another have bought a brand new television set. The first thing you
do when first receiving the TV is probably adjusting things like
contrast, sharpness and so on. I am using “sharpness” as an
example. So when you want more sharpness, what the TV would do is
using a high pass filter or amplifying the high frequency range (similar to the high pass filter for the
tweeter) so you get more high frequency and more bluish tonal balance
on the picture hence a sharper image, just as the sound will get more
clear if you have more treble. And of course as with sound, if the
image gets too sharp, it can be “unnatural” similar to treble
fatigue.
I am not an image
processing engineer so I don't know the detail of image filter
design, but someone told me that they actually face with something
similar in speaker design, such that if they use higher order filter,
the image can get “unnatural” albeit having more clarity. I
suppose if you were to design a spy camera to search for weapon of
mass destruction, you probably want to use higher order filter :-)
Anyway, back to speakers
design with 1st or vs. higher order filter. Interestingly
enough, if I were to run an audio recording studio, I would use
higher order speakers to monitor the recording sound since I think
higher order can tell me more about my recording sound, but when I go
home, I'll use my 1st order speaker to enjoy the music.
We human are yin and yang. Duality works.
|
|
Hi prof,
I wish I get to audition a pair of Thiel 3.7 someday ... preferably at my home :-) I actually like big speakers. |
He asked about my cabling and I told him it was standard Belden 10awg etc, and he was like "Wow, your system sure isn’t lacking in detail and amazing sound quality. Sounds like your cables are up to the task."
prof, You don't know what's you're missing :-) |
It never hurts to demo other brands on the flip-side.
I think there was a thread where prof auditioned tons of speakers but he couldn't find a reason to switch. |
I love this thread! Whenever contrasting opinions are expressed (such as on the issue of cables), there is no name-calling or attempts to shout down disagreements. Maybe if we try to discuss about power cables then see if that statement still holds :-) But joking aside, looking at the internal cables used in the 2.4, with respect to Tom and Thiel's, I was wondering if they may be a bottle neck. They look suspiciously like something from Home Depot :-). There was a post a few months back from someone I forgot (holco or something) that when he upgraded the internal cables, there was a meaningful improvement. |
However, the wire itself in the 2.4s is pretty high-end. It is 5-9s, high purity, low oxygen, long crystal, high polish 18 gauge solid in teflon jackets, precision twisted at 2.5 / inch.
Hi Tom, I was only joking about the Home Depot stuffs :-) |
Jaw drop. They sounded, thin - almost lifeless. I double checked all the cables on everything. All good.
Perhaps the old cables need to be "re-broken in" :-) |
prof 's loudspeaker audition thread is an excellent read. He covered many popular brands in detail. This information is especially helpful for those who do not have local Audio representation.
That is unless he made it all up. Like how can you prove God exists. :-) |
Second, visit the Grand Canyon
Seems like a long way to go to prove God exists though. |
That's a third for visiting the Grand Canyon.
Going through the canyon in an inflatible raft is an awesome experience.
and snake hunting lols. |
a major part of the art is prioritizing the relative value of various aspects of the whole package
That's true. Designing speakers is as much an science as much as an art. Unfortunately some put too much science over the art part. |
I have not used any high end binding posts. The most expensive I've used so far only cost me a few dollars from Madisound.com.
I guess you won't know until you try out, but for those who have used expensive posts before, do they make a difference in the sound? |
