The difficulty is vastly overstated in this thread though. Also, a point to know is that every active I've played with ALL sound different. Pretty much like all components but seems to show up to a greater degree in crossovers. Like always, you pretty much get what you pay for. But I'd never go back. Removing the components from between the amp and the driver makes a big difference and unless it's messed up should be a leap better in sonics. CJ
You can build to fit the load. You have to if you want it be accurate and hit the desired frequency. But the minute you change to a new amplifer.... CJ
It sure seems difficult to someone just getting into it. Which is probably as it should be, at least to some degree. Because it truly does sound to me like anyone going into it without a fair amount of knowledge, experience, or both, is likely to be disappointed. I haven't yet found any resources online that are very helpful to someone like me who is new to this. Some say "never bi-amp with different amps." Some say "Always bi-amp with different amps." Some say "never do an active crossover." Others say "Only do an active crossover." For example. Reading about this online is truly dizzying. This thread is by far the most useful thing I've encountered so far but that's really only because I'm investing a lot of energy in reading, asking questions, and thinking things through. And I'm pretty sure I'm not even halfway to where I need to be to make a confident decision about how to do this well...
You generally terminate the filters with a lower impedance so they aren't affected as much by the amp impedance, typically some type of output potentiometer, and you can also use a trimmer to adjust the load to fit different amps. But yes, good point, they are generally built for a specific load.
That was years ago I had read the enclosure represented an average of 2/3 the cost in building a speaker the other 1/3 was for the electromechanical parts, ie: drivers, wiring, crossover etc. The cost per unit of the crossover wasn't that great (after being spread across production volume) but the amount of engineering time and energy (ie initial fixed cost) lavished on designing the crossover was the greatest acc to the article. Crossover design involves a great many of highly complex calculations was the reason given. Far more than necessary for any other subcomponent in a loudspeaker. Over the years computer technology did much to save time & labor while significantly improving design. This gets into some pretty far advanced stuff which is why I have serious doubts about the ability to improve upon in the majority of cases. I don't doubt that it has been done in some rare instances.
I use ALK crossovers because I am not a crossover design engineer, as I am given to imagine that most other audiophiles are not either. ALK crossovers are crossovers that are designed for different Klipsch Heritage speakers to be upgrades, not replacements to the original OEM factory crossovers. These designs reduce or eliminate problems that are inherent in traditional passive crossover designs. Although specifically designed for the Klipsch Heritage line, they can be successfully implemented in other speakers that require similar crossover specifications. ALK has designs that are unique and optimized for the complete Klipsch Heritage speaker line.
I would be on board with those percentages. I would also figure on the majority of that other 1/3 going toward the cost of the drivers. Other than the wire, internal crossovers are usually the smallest part of the budget. Design is one thing but building a quality crossover network involves using quality parts, which can be costly. This is something you don't see happening until you start climbing the ladder to more expensive speakers. Once you do get into the more expensive speakers, the manufacturers realize that they need quality crossover networks. You simply cannot take a loudspeaker to the next level without quality crossovers, no matter how much you spend on the cabinets or the drivers.
@Soundgarden, I noticed you were taking an interest in active systems when I brought the subject up. I have demonstrated, installed and owner several active systems and the audible benefits can be significant. One thing I should start by saying, is that apart from one system, which I bought as active only, all were designed to be easily converted from passive to active, because of this many customers and myself benefitted from the big jump of going active. What is the advantage of a passive loudspeaker? You only need to have a single amplifier and you can mix and match. It is inherently a simple system. The disadvantages of passive speakers is the frequency is divided after the power amplifier, so the crossover has to be able to handle the full power, crossover is lossy (the components heat up) and an amplifier can have a very difficult job to do as it has to drive both the crossover and the drive units, which can offer a complex load. I'm sure anyone who studied electronics at the most basic level realises that complex loads with phase shifts is not the ideal solution: in an ideal world where space, cost, complexity and the need to easily mix and match components weren't important, you simply wouldn't want to use a passive crossover. In the real world they are a necessity. The advantage of an active crossover is that this is done at line level, far more accurately and instead of using resistors to lose power, you add gain, which is obviously more efficient and loses less music in the process. A well designed active crossover can give you a flatter, tighter frequency response and the amplifier literally has to drive a frequency limited, fixed load loudspeaker. Unsurprisingly this is a simple job: you do not get the dynamic compression and a good active system will be more dynamic, detailed and controlled at any volume. The disadvantages of an active system is that it is more costly and can be a lot more complex than a passive system. An active system can also be much simpler and cheaper than high end passive set ups. You pays your money and makes your choice! The last active system I owned were Linn Akurate 242 with which I had a lot of fun with various configurations before ending with a five way active system. Bi-wiring, quin-wiring, twin mono passive quin-amping! The massive improvement came when the active cards were fitted. I preferred this set up to a more expensive amplifier driving the same speakers passively.
So there have been a number of comments to the effect of passive crossovers at the speaker should be avoided because they degrade the sound in a way that line-level crossovers between the pre-amp and power amps don't. What would be the reason for this? I understand that far less energy is bled off when signals are divided at line-level. But it sounds like something else is going on. So let's say you're able to perfectly emulate your passive speaker crossover with a line-level crossover. Why would that degrade the sound less?
Here are a few... https://www.firstwatt.com/pdf/art_diy biamp_6-24_ crossover.pdf https://www.firstwatt.com/pdf/art_coherent_xvr.pdf Active Filters https://www.audiocontrol.com/downloads/tech-papers/tech-paper-102.pdf
If you start in on quin-wiring and all that business my head will surely explode! I can imagine Triamping a three way speaker. Which in some ways is simpler cause then there’s literally no passive crossover needed in the speaker. Biamping means you still have to employ a passive crossover to separate the mid signal from the high signal. And tri amping introduces additional cost, space, and complexity issues of its own. Im gonna stick with bi amping three way speakers for my own sanity and cause my wallet is also trying to have a conversation with me. But thank you for the points not just about line level vs speaker level but also about line level active vs line level passive as concerns gain vs loss. Hadn’t thought of that yet
As @Davey mentions, you can give-up a lot of db when using resistors in passive low-pass filters. Not an issue for me, as my 12" drivers roll-off pretty low all by themselves, a 1st-order was suitable. Had I needed 12db, I likely would have needed to figure out a way to make my own coils (still on my bucket list anyhow) instead, of add some sort of buffering/amplification.
So if I can get away with 1st order and have powerful power amps then passive works fine. (Perhaps it’s even preferable?) But if I’m needing all the power I have on tap and emulating complex speaker crossovers then active is the way to go.
It also occurs to me that the ore amp itself could have an active crossover they would run off the same power supply as the pre amp. But I’m practice that would be very limiting unless it was either highly adjustable or it was custom built for a specific power amp setup.
That’s a highly dubious claim and I know of no controlled study verifying it. There may be some insertion loss but that has no bearing on SQ.
Let's say that pigs can fly... The problem is that in home audio, you can't emulate, let alone perfectly emulate your passive speaker crossover with a line-level crossover. Which is the entire reason for not doing it.
Maybe I’m not explaining it the best. I was referring to the fixed cost of initial setup and design of the crossover - not the cost of the physical elements of the crossover which is a relatively minor expense relative to the total cost of the speaker.
In post #14 I shared my first rule of designing passive crossovers. I may as well share a few more here to attempt to answer your question. Rules to Design Passive Crossovers: 1. Don't use passive crossovers 2. Don't use electrolytic capacitors 3. Don't use ferric core inductors 4. Don't use inductors made of tiny wire Most passive crossovers these days (the quality one) use film caps, as it is widely known that an electrolyte dielectric degrades the audio, typically. Electrolytics are typically used to reduce cost, at the cost of audio. Likewise, inductors with a ferrous core (iron or steel) are used to reduce cost. Unfortunately, a core with a ferrous material in it will saturate as current increased. The effect here is a change in the inductance value from desired. When the volume gets loud, crossover points and slopes start moving around. Inductors with tiny wires yield a high DCR in the inductor. Many inductors used for the woofer low-pass have hundreds of feet of thin wire. This is never a good idea because of the amount of effective resistance that results. There have been pictures posted in this thread, and the thread that spawn this one, which include violations of rules 3 and 4.
Maybe. The problem with 6db is your drivers need quite a bit of flat response beyond your crossover point, because they're still going to be audible there. So let's take a fairly typical 3k crossover point on a two-way 6.5"/1" tweeter system. With 1st order crossovers, your tweeter is only 6db down at 1.5k, so it is getting a lot of energy still and there aren't a lot of 1" tweeters that can handle being down only 6db at 1.5k. Any distortion will likely be audible, especially at higher SPL. OTOH, you could use 2nd order high-pass, then at 1.5k the tweeter is down 12db. Much better. But now your 6.5" woofer is still only 6db down at 6k, so unless your woofer rolls off nicely (forget exotic cone materials that may ring) you're going to hear that, too. And if you want a second order low-pass, you're going to pay for it with a drop in signal. I think the PLLXO is really better suited for lower frequency use. Maybe I shouldn't have encouraged it. I'm not even that good at this stuff, I'm a complete hack. I'd get your guidance from some of the engineering folks here.
That sound like a summary of what you get on any board, including this one. Still, I am glad you've got something out of this.
I don’t think you understand the difference between vertical and Horizontal bi-amping. There is nothing stupid about vertical bi-amping except when you don’t understand it. With vertical you intend to use the speaker designers crossovers. . . You gain a lot of dynamic by adding that second power supply. This is how my system is wired. This blew away a lot of other amp trials. I’ll probably never go back to a single power supply again.
When I see these quotes, it's my interpretation that they come from different topics being discussed. "Never bi-amp with different amps" comes from someone discussing passive bi-amping. "Always bi-amp with different amps" comes from someone discussing active bi-amping. Likewise for the next two quotes you've got.
No, I did understand that part. The crossover is part of the initial speaker design. Once you design it, it the design costs themselves are fixed and spread out over the entire speaker production. As a percentage of the production cost for all the speakers that are produced, the cost of the crossover design is a minimal expense. It is not so much the cost of the design, that isn't the problem. The problem is in its execution because manufacturers use cheap parts which make inferior crossovers.
