El: Series crossovers consist of the opposite components one would use in a parallel crossover. While "common" crossovers are called "parallel" designs, they really are closer to a series circuit by their very nature. Some newer speakers ( and old ones like Fried and Koss ) use what is called a "series" crossover, which is really like a parallel circuit. You can see a diagram of a basic "series" crossover here at Karl's website. The terminology is quite confusing and i can understand why this baffles people. As a side note, i was doing a search and ran across Clement Perry's comments about Karl's speakers. You can read it by clicking on the link to Stereotimes CES coverage. Sean > |
Smaller drivers can produce low bass IF the cone is of a high mass design and capable of long excursion. There are several problems here though. That is, why would you want to use a smaller woofer of higher mass? This reduces both transient response and high frequency bandwidth, negating much of the benefit of using a smaller driver. On top of that, a driver with more surface area reproducing the same note at the same amplitude of a smaller driver with longer excursion will produce lower distortion. The more "throw" that the driver makes, the more distortion that it produces. On top of that, longer throw woofers produce more reflected EMF, making them harder to control. On top of that, the smaller driver will have to move more air to produce the same spl, meaning that it will be producing more distortion at all times while running into Xmax ( linear excursion capabilities ) sooner than a larger driver. Obviously, there are a lot of "on top of's" in this equation : )
In plain English, this means that if you want deep bass out of a small driver, you'll have to make several compromises in other areas to get it. The only way to get low distortion and deep bass while retaining good linearity at high volumes is to use large woofers in a large cabinet or a multitude of smaller woofers in a large cabinet.
To quote speaker designer Bill Fitzmaurice: "The bottom line for speaker cabinets is that for the goals of a small box, a reasonably efficient system, and good bass extension, you may achieve any two of those goals at the same time, but not all three". Sean
> |
Sogood51: I liked your post. The comment about ports adding quantity but lacking quality is right on the money. The analogy of a smaller woofers vs bigger woofers and the boy / man comparison was also quite good.
Drew: very nice posts that are based on both common sense and science. Those two factors rarely meet when it comes to audio and audio forums. I applaud the efforts that you put forth in your post.
Michael: Drew never said that other speakers weren't deficient in providing usable spec's. I'm quite certain that he feels that most manufacturers don't provide anywhere near the amount of info that they should about their products, especially speakers. He just said that this speaker lacked the spec's that one really needs to properly judge a speaker on paper.
Given the claims being made for this speaker and the amount of technology that supposedly went into it, one would think that the manufacturer would want to "testify" to the actual performance of the product as much as possible. Your level of involvement / personal emotions seem to be clouding your response on this one. Then again, that seems to be a common situation when it comes to products that you endorse.
Karls: I want to highlight two of your statements. Here's the first one: "It still can't break the laws of physics, but if used within its linear excursion limits, will give extraordinary performance regardless of volume level or frequency".
The only way to maintain "extraordinary performance" with this speaker contradicts the "regardless of volume level or frequency" part of that sentence. That's because as either frequency is reduced or volume is increased, excursion also escalates. As such, what you've said is that "this speaker kicks ass / remains quite linear so long as the volume is kept to a reasonable level and one doesn't expect the deepest bass". As a few posts have pointed out above, you can't maintain high spl's and / or deep bass extension with lower levels of distortion without resorting to larger drivers.
Here's the other statement that i think requires further commentary / clarification: "The issue of the extra power required by the BOMB is more of a problem for the amplifier than for the speaker. The reason for this is that there is surprisingly little energy (on a continuous basis) in the deep bass. But when it appears, it can make very heavy transient demands on the power amplifier. The BOMB has a maximum boost of 10dB at 24Hz, which equals a factor of 10 in amplifier power".
The added power requirement of the amp is most definitely a problem that the speaker shares too. That is, the 10 dB increase in power that the amp must generate has to be dissipated. ALL of that dissipation takes place within the speaker itself. I think that this is why Drew mentioned thermal compression and power handling coming into play more rapidly with this approach than when using a larger driver with less requirement for EQ.
As a side note and to somewhat respond to Drew's comments, my thoughts are that the effects of thermal compression would be somewhat reduced with this design. This is due to the design of the driver itself and the fact that it will make use of the 2" thick metal baffle acting as a heatsink for the basket of the driver. While it is true that the voice coil & coil former are doing most of the dissipation in a driver, the basket and magnet structure also enter into the picture in this area. Given that the basket is mounted to a metal structure that should be quite excellent at dispersing heat, it probably has a sizeable advantage over other designs.
Other than that, this looks to be a very well designed speaker. I remember looking at it previously and thought it was pretty nice, but also way too costly. I do understand that there is a LOT of custom machining taking place here, but $10K for a "little" speaker is still a LOT of money. Then again, $10K for a "big" speaker is still a LOT of money in my book. I do applaud your efforts and appreciate the fact that you didn't try to push some type of bass reflex design on the public in the name of "more is better" bass response.
Bluebull: Actually, if properly done, relieving the midrange driver of the longer excursions necessary for low frequency reproduction, midrange response should be improved. Given that many two ways are a compromised design, it stands to reason that the same engineers would be even more confused when adding another driver, two more crossover points and a lot more parts to the equation. Sean
>
|
Michael: Based on the information provided, it is possible to tell that someone invested a lot of time and research into these speakers. Whether or not they do everything well or suitable for a given installation is another matter. The more spec's that the manufacturer provides, the more that one can interpret just how well they might fit their specific needs and desires. Sean
>
PS... Thanks for the kind words. I hope you have a pleasant week-end too : ) |
Michael: I simply posted comments in agreement with others that had taken the time to share both their personal points of view and / or scientific data pertaining to the situation at hand. Sorry if that offends you.
Karl: Thanks for taking the time to post a response. As i mentioned, it is quite obvious to me that a lot of love and thought went into this product. I hope that this was abundantly clear in my original post.
1) I am aware that they've made great advances in terms of reducing distortion byproducts as excursion increases, but as a general rule, longer excursion still equals more distortion. Some designs are obviously better at this than others. The laws of physics still apply and we can't yet get something for nothing. As you mentioned, there are trade-offs involved in every aspect of speaker design. The end product becomes a balancing act based on what the engineer was willing to sacrifice in order to achieve their desired goals.
2) "I pointed out that while there is a 10dB boost at the amplifier, there is also nearly a 10dB loss in maximum output at the speaker".
What do the losses at the speaker involve? From what i know about such designs, the losses are incurred due to inefficiencies in power transfer below the point of resonance. The end result is a high percentage of power being dissipated as heat. As you stated, the end result might sum to a neutral response, but at the expense of much higher thermal stress.
3) According to your post here, anybody that listens above 90 dB's is "insane". Call me and dozens of other audiophiles that i know "crazy" then. Especially if you are talking about 90 dB's at 1 meter. As far as i'm concerned, spl levels should be taken and compared at the listening position, not at 1 meter. Readings taken at 1 meter are only handy for sake of sensitivity or efficiency ratings, and even then, they don't tell the whole story due to differences in dispersion patterns.
El: The original and second series 901's were a sealed design. The curve for those EQ's is different than that for the series III and all those after that.
Bose obviously had to run the drivers below resonance as the drivers were run full range. Karl is doing the same thing in principle but limiting the top end of the drivers being EQ'd and using a fancier circuit. Due to the fact that the 901 drivers resonated higher in frequency, and they were applying X amount of db's to compensate for the roll-off per octave, the total boost figure for the 901's would be much higher than Karl's design. In the long run, the use of equalization below the point of resonance is nothing new. The end result is that one can increase bass extension by appr half an octave at the expense of increased power requirements, increased power dissipation in the drivers and a lower maximum spl for the same percentage of distortion. It is really a tough balancing act to do correctly and requires very close production tolerances, both in the speaker itself and the correction circuitry being used.
As a side note, Bag End makes use of bass extension technology via a calibrated EQ curve in their subs. They chose drivers that resonated above the intended band of use and then EQ it for flat response below that point. This is a very lossy method and quite out of the ordinary, but has many advantages. Woofers and sub-woofers especially are the only drivers in most speakers were "resonance" or "break up" are considered normal and acceptable, yet most engineers / designers try to avoid that circumstance like the plague with mids and tweeters. Bag End took the high road, but in this case, the efficiency and power requirements of the system was what they were willing to sacrifice to achieve their desired goals. Sean
>
|
I had original 1st series 901's back in the 1970's. Harry Pearson stated that the 901's are the only speaker that has gotten worse with each new generation. I guess that means that he thinks that the 1st were the best. I would tend to agree.
As a side note, i've still got all of the drivers from those 901's. Those used nine 8 ohm drivers wired in a series-parallel circuit to achieve a nominal 8 ohm impedance. To be specific, they used 3 parallel banks of 3 drivers wired in series. The second series and up used nine 1 ohm drivers all wired in series. If you blew a driver with the first series, you could still use the speakers. If you blow a driver with any 901 after that, one driver takes the whole speaker down, kind of like old style Christmas tree lights. Sean
> |
Thanks for your very up-front response Karl. As fine of a speaker as you have been able to produce, your comments varify what i and several others have stated. That is, obtaining extreme low bass out of a smaller driver comes at the expense of some other area of operation. I applaud your honesty and integrity, both in your responses here and in the design attributes of your product.
As a side note, neither Mr Dartford or myself were actually making an "apples to apples" comparison between your speaker and the 901's as much as we were discussing the various manners that technology can be applied and how it is actually used in various products.
I do have one question for you though. Have you ever tried running these actively crossed? I'm curious as to what your thoughts / experiences were in this regards IF you tried that.
Other than that, i do remember you posting positively about sealed box designs in the past and contributing some excellent information / comments in various threads. I knew that i always respected your opinions and now i know why. That is, we basically agree. At least on this specific issue : ) Sean
> |
Karl: I wasn't talking about supplementing the bottom end of your design with an actively crossed sub, but about using the existing design with an active crossover and bypassing the internal series crossover. I understand that you have several different circuits ( impedance compensation, notch filters, etc... ) that would have to remain.
I also have other questions about series crossovers for you, but that is neither here nor there : ) Sean
> |
Karl: My experience with even very high grade "passive parts" placed smack dab between the amp and drivers is that they drastically reduce sound quality. One can use even a lower grade electronic crossover and get better results than if one used very simple passive crossovers of the highest quality. At least that's been my experience.
My Brother argued with me about this for quite some time. That is, until i gave him an old electronic crossover that i hadn't used in years and he tried it out. Pulling even a single cap from his tweeters made a HUGE difference in terms of sonics. Why one can get away with running a MUCH more complex circuit at line level without near as much sonic degradation is beyond me, but i'm guessing it has something to do with the current levels involved.
After studying the circuit lay-out of a series crossover, i've often wondered how much more "low frequency leakage" there is into the tweeter as compared to a standard 1st order "parallel" crossover? I've also wondered about the differences in power handling between the two. Have you ever compared the two with identical designs? Sean
>
PS... Thanks again for your taking the time to clarify your comments and respond to further comments / questions. |
