Academy
Why we prefer lower order crossovers
1,202 views
In this episode:
02:00 - Why do we use passive crossovers?
04:56 - How can you 'hear' a passive crossover?
06:39 - What does a crossover consist of?
08:35 - What is first order?
11:24 - Why do we prefer lower order crossovers?
Follow Us
► YouTube https://www.youtube.com/user/DynaudioOfficial?sub_confirmation=1
► Facebook https://www.facebook.com/DynaudioGroup
► Twitter https://twitter.com/dynaudio
► Instagram https://instagram.com/dynaudio/
► LinkedIn https://www.linkedin.com/company/dynaudio-as
Find a Dynaudio dealer http://www.dynaudio.com/find-dealer/
Visit http://www.dynaudio.com
View transcript
- Welcome to Ask The Expert. My name is Christopher and I'm your host. With me today, I have Alex Newman, who's a acoustician in our R&D department. Alex, can you talk a little bit about what it is that you do at Dynaudio? - Yeah, I'll try and put it briefly. I'm an acoustic engineer and my main job is to tune the speakers and the sound of the speakers. And the task involves a lot of things from the start to the beginning of the project. And from the start, it's coordinating between other departments, mechanical engineers, drivers... and choosing what we're going to start with and what's best for this project. And that takes a lot of preparation. And after that, we start building up prototypes, measuring, listening... And I think the majority of the time we spend listening to our loud speakers, testing if something doesn't sound right, we go back to different departments or to whatever we were doing and figure that out. And then, it's different for active and passive designs. So for a passive speaker, we'll do a lot of hand electronics and measuring the speaker and soldering and then listening to that and changing... - And lots of hands on work. - Yeah, hands on work. And for active, it's a lot more digital processing and playing with tools to improve the speaker that way. - Cool. The topic for today, Alex, is passive crossovers. And we're gonna dive into what they are and how they work and why we use them and all that stuff. So I think that we should start out with just the basic question: Why are they there and what's their purpose? - Yeah, so the passive crossover is there to split the frequency regions of a tweeter and a woofer. So the woofer plays bass and a tweeter plays high frequencies, treble. When you have more three-way system, then you have a mid-range. Then you need another electrical crossover to split that again. And the reason we do that is because, otherwise they will share the same frequency range and we can get distortions and lots of other problems. And I think, the most important for the tweeter is If we don't have a crossover on the tweeter it can blow-up, it can burn up. It can be lots of problems. And the woofer again, it's trying to knit that with the tweeter. - So it's making sure that you take the musical signal and actual divide it to go to the right part of the speaker, right? - Yeah, so it sounds natural. We want to design the crossover so that it can't be heard. And then after that, so the speaker can't be heard at all. - That sounds like a tricky challenge. - It is a very tricky challenge but that's what we pride ourselves in trying to do. - How do you approach something like trying to... Because it's a pretty central part of the speaker, right? - Yes. - And so, trying to eliminate it, if you understand me correctly, how do you go about that? - So a lot of the time, we start off by measuring the loud speaker and measuring the frequency response and the time response. And looking for something that's close to ideal of what we want. And when we get that, we get a good knit as we call it. - A good what? - A knit. - A knit? - So how they blend together. - Ah, yeah. - So if we were to do a tone that went all the way from the base to the treble, we wouldn't hear something strange. It would just be natural going from base to treble. And that's the part where the crossover comes into play. It really needs to be spot on. But sometimes, listening tests change that. So we measure it, and it might look brilliant on paper from how we measured it. But then we go listen to it and we can hear the crossover. And it sounds strange. Maybe there's too much mid-range, or there's not enough depth. And so, from then on, we start changing components to improve those parts. We take it back and measure it. And a lot of time, when we measure it after listening to it, It sometimes measures better, than it did originally. Just from us listening to it and tuning it again. - There's one thing that stuck with me. You said something about, um, you know, it has sound... And you said you could hear it. And I'm like, how do you hear crossover? And I'm a novice, so to me it sounds strange that you can hear it. And I don't even know if you can explain what it's that, it's probably expertise too - I think the most obvious way is when in frequency response, the region where they cross over sometimes seems emphasized or it's lacking or... - Okay. - That's the most obvious. That's the easiest to hear. You can stray away here, maybe it's a bit distorted. But the tricky part, and I think the most important part is listening for the time response. So, the speakers are different sizes and they also move at different rates. But when they got close together, they're supposed to move at the same time. Together. And then when that gets to you, it's supposed to be a summation of both drivers. - Yeah. - And when that's out, when the timing is out, you can hear that in certain things like snare drums or anything transient a lot of the time. And it sounds unnatural. - And that's the things that you're then listening for? - Yeah. - And trying to figure out, okay so there's an emphasized frequency area here so now you actually know where you have to go in and fiddle around with the crossover, right? - Yeah, exactly. So we listen to that. We kind of know, we have good guesses of what components will change what and how it will effect the rest of the speaker. - 'Cause it always, change one thing, it changes another. - Yeah, yeah, yeah. - So, we have an idea. But sometimes it takes some iterations to discover what we want. - You were talking about, you know, certain things that you can change and parts of the passive crossover. Can you walk me through what a passive crossover actually consists of? - Yeah, so passive crossover is made up of multiple electrical components such as resistors, capacitors, and inductors. And all of these components electrically change the signal that's coming in to it before it goes to the drivers. And we use inductors and capacitors to roll off the frequencies we don't want. So we use a capacitor in series on a tweeter to roll off the low end so that we don't have loads of bass going in to the tweeter. And then, for the woofer, we use and inductor in series to roll off the high end. And that's how we manage it. The resistors we use to balance a lot of the time. So, it's just padding, changing the levels of the tweeter. - Okay. - There is also other places. Crossovers can be very complex. - Yeah. - The one I just described is a simple, first order, no fuss crossover. But there can be tens, tens of components. I think the aim a lot of the time is to reduce the amount of components for use. Each component adds more coloration. - Yeah. - And sometimes it also can effect the way we hear the music, even. - And I guess every time you add something, what you're actually doing is tampering it a little bit with the music segment, right. Like you said, you're coloring it, changing what was recorded in the studio, just a tiny little bit. But, I mean, it's all something that we can hear when we're listening afterwards in our living rooms, right? - Yeah. - And the next thing you said was it's a first order, and I guess you're talking about typology now, right? - That's correct, yes. - And I've heard about first order, second order, third order, fourth order... - It can keep going. - And keep going, okay - Impassive maybe not. - So, is that just about all of the components that you put in? - It's to do with how you arrange the components. And how many you put in. So the one I described was just an inductor on the woofer and a capacitor on the tweeter. That's first order. That gives you a slow roll off. So the bass doesn't sharply end from the tweeter so it slowly goes. When you add another inductor to that circuit in parallel, it gets a bit complicated, but when you add it, we get a steeper slope. - Okay. - So the cutoff is faster and we use that to control the frequency response of the tweeter. And we can also play with the shape it makes. - The shape? - The shape of the frequency response. - Okay. - And that's... by changing values of each component. The less components, the better group delay we have which means the better timing. - Okay. - And we like to have really good timing at Dynaudio. We aim for that. But sometimes it's necessary to have second order where it works better for that driver in that enclosure and that speaker. And depending on what we're aiming for, sometimes we compromise and go for a second order because it sounds better. - And I guess, as with everything, you were talking about earlier that choices means that there's things that you don't choose and it's a compromise from all of the different parts that goes into a loud speaker. - Yeah. - And it's not just that we focus especially on the drive units or especially on the crossover. It really is the whole nine yards, right? - I think a lot of it changes how we work with the crossover. - Yeah. - Feed is a really good example. - Feed? - Feed on a floor standing loud speaker changes our perception a lot. So we might, say you were listening to it without the feed, the mid-range might sound really present or gritty for lack of a better word. And once you add the feed, it cleans up. But if we were to tune that loud speaker without the feed, we would have been changing the crossover. Ah, this is wrong, trying to fix it. And we would have done it incorrectly because once you have the feed, it sounds fine. - It's a lot of work. - Yeah, so we need to consider all the factors and something we try do is have all our reference parts and all the actual parts we're building a loudspeaker with to tune it with. - Before you go into that process, right? - So that we know when start tweaking, we know that we're making the right decisions. - Okay, um... Talking about usually preferring first order, is that just a general rule? Is that because we just, it's just what we prefer? Or does it give us an advantage? - It gives you an advantage in terms of the timing response of the loudspeaker. It has less effect on the way it moves. And that, in essence, can give you better imaging, better depth. But sometimes the disadvantages of first orders is that the power handling of the loudspeaker is compromised or we get too much low frequency in the tweeter for example. And also, sometimes, the whole back to hearing the crossover, sometimes that's influenced by having a first order. So sometimes we step it up to a second. Just to improve the sound or to improve the timing. Sometimes we have a better timing response if we use a second order. - And timing response, you mentioned that was for the imaging. So it's really, if that's not really in place in a good place, it becomes more difficult to pinpoint instruments, voices, in the sound image, correct? - Yeah, that's correct. So, a lot the time we focus on imaging and depth and width and all that kind of stuff. Rather than focusing on so much the frequency response because that part we can balance. That part's easier to balance. The depth is harder to gain and sometimes lower order crossovers gain more depth or better imaging. But, like I said, there can be times where it's compromised. And so, maybe we want a more precise image where, say, the vocals are right in the middle. Sometimes a second order can achieve that. - Okay, and it's really about what you're trying to achieve, right? And then it's choosing stuff and then not choosing other stuff? - Yep. - To try and meet that goal. - Yeah, that's very true. But I think the main thought it revolves around is the drivers. And so, a lot of the new products we're coming out with are using similar drives. - Okay. - Similar technologies within them with some changes. And that means the drivers are quite close in the way they act. So that it end up being, we have used similar typologies for those drivers, because they sound the best for that specific driver. - And from what you've been telling me, it sounds like if you have something that's already, it shares technology, it shares a lot of design basics, that's going to be easier for the crossover to deal with. And it's going to be easier for you to design. - It's easier to start with. - Yeah. - It's a good tool to be able to use, to give yourself a direction, where to go. Sometimes you can start something and go, where should we go with this? - You were talking about new products and I can hear that they just started a sound demo of the new Confidence that we're previewing. So I think that we should try and wrap up. And the final thing I want to ask you is, it might be tough, what is the single most important thing in your mind in a passive crossover? - In a passive crossover? Quality of components. - Yeah. Perfect. Thank you so much for joining us Alex. - Yeah, no worries. - It was a pleasure. And I won't take any more of your time so you need to head out and experience the high end show, right? - Yeah, need to look around a bit more. - Thank you so much for watching and keep the questions coming.