But... What about Line Source AND Line Array Loudspeakers?

[Music] [Applause] [Music] everybody welcome back to another Tuesday step talk this week we're going to be covering something that I've had a lot of questions about you guys have been asking me about light sources and how light sources work especially considering the episodes that we did previously on comb filtering we looked at comb filtering effects of lining up a group of full range drivers and how the time delay differences within the the arrays cause huge cancellations and knocks the top end right out of them we looked at the use of multiple tweeters and the time differential between our time arrival effects from one tweeter versus the other can have the same effect it could knock the whole upper frequency ranges right out of the response that we also looked at adding a super tweeter which has again the same effects whenever you have shallow slopes and you have a high crossover core where the wave links are really short then it just takes a little bit of differential in time from one to the other for one to completely cancel the other out caused huge holes in the response and things like that so you guys kept saying what about line arrays or the bow line arrays okay line arrays is a little bit different and how we handle it so we're gonna we're gonna look at that we're gonna look at the basics and we're gonna look at some of my favorite speakers that are right here behind me which one is an actual array and the other is actually a line source the in extreme here to my right is what's considered an array it's actually a point source speaker it has a lot of advantages that you'd get from a line source and it is seven feet tall so people look at it as if it's a light source it's not really a true line source it's really it's a true point source speaker but you have to consider the wavelengths and the crossover points and how they interact and what kind of waveform they project it's like I said you're getting some of the advantages of a lion source because we're breaking down the frequency range in the Lord Regents into multiple drivers so in this design we have a planar magnetic tweeter it is offset back an inch and a half in time because that's the thickness of the baffle it's mounted on the back of the baffle which puts it physically aligned with the voice coils of the six and a half inch mid-range drivers and the tweeter being in a waveguide the waveguide increases the output in the lower frequency ranges and it allows for a much lower crossover point then if it were not within a waveguide so in this case it's crossing about 1,400 Hertz through the two mid-range drivers they're playing down to our crossover point of about a hundred and sixty Hertz or so all of it in an open baffle it's in what we call a wedge-shaped open baffle and on this type of design we do asymmetrical wings so we have a solid wing on one side and it's completely open on the other side little short wings until you get to the lowest frequency range which is a full wedge shape so again these are crossing around 160 and there's eight six and a half inch drivers covering from 160 them to 40-something Hertz depending on your breath so like I said it has some advantages a blind source when you look at the low frequency range that's being covered by eight drivers you have eight drivers sharing the load so the response is really fast when it's playing bass notes hitting those drums things like that big dynamic range you feel that impact you think wow man I think is really hitting you can just feel that bass and you walk up and you look at the speaker and the woofers are just barely moving because you have so many over sharing the load so the settling time is is much faster because you don't have one wolf or having to play highest peel level with a heavy moving mass or anything like that so not much stored energy they're really fast settling time so very much getting the advantage of a line source but with a narrow front baffle you're looking baffled it's only about eight inches wide before it starts radius in around to the side panel so imaging is fantastic transparencies fantastic speed you know across the board these speakers do a lot of really good things are they a true light sources now a true light source would mean like the speakers over here to my left this is a diamond design we originally did for serenity acoustics it is a true light source a true open baffled light source and this we have our custom new three drivers we have the BG original Neo ten drivers and we have a crossover point of about 1,400 Hertz or so between the tweeter and between the Neo tens now before I get really deep into the specifics of the speaker let's back up and grab some basics so that we kind of understand near fill far fill transitions and we can understand beaming frequency or a beaming range of a driver so what do I mean by that let's consider any driver no matter how big a diameter it is it's going to have a range where the output is going to be if it is playing a frequency range where the wavelengths are shorter than the diameter of the diaphragm so let's consider a 12 inch woofer 12 inch woofer is the wavelength of the woofer itself or the wavelength of the the distance that is the diameter of the driver that's a mouthful is about 1,100 Hertz so anything above eleven hundred Hertz in frequency range in a 12 inch woofer it's gonna start to beam that wavelength in other words the off-axis response is more omni below that wavelength and as you go above that wavelength it really starts to beam in other words output is more in a straight line versus an omni pattern because those wavelengths that it's trying to reproduce are shorter than the width of the diaphragm so when you move off axis you start losing output pretty quickly that's why when you're crossing something to a 10 inch woofer about a 1k her crossover region is usually kind of a target of that's about as high as you want to run one of those woofers otherwise it'll have no output as you move all faxes if you go to a smaller driver to a 10 inch woofer 10 inch woofer the beaming frequency or beaming wavelength is going to be around 1,300 hurts when you get smaller let's say an 8 inch woofer that frequency range is gonna be about 1600 Hertz for a 6 inch driver it's about 2200 Hertz for a 4 inch driver it'd be about 3,300 Hertz for a 2 and a half inch tall driver 2 and a half inch diameter driver that'd be memory frequency would be about 5600 Hertz for a 1 inch diameter driver like a dome tweeter that's gonna be about 1350 so consider that let's take a 2 and a half inch diameter driver anything above 50 600 Hertz is going to be projected more in a beam versus anything that's below that the output is going to begin to spread out and become more on me now the reason that pertains to a line source is when you when we design a line source and you run a group of drivers like that up in a continuous line what you're effectively doing is you're making one driver that's really long in this case the light forces over here the height of it isn't nearly 6 feet tall and the effective area of the light source were you want to stay within the range it's about 5 feet through the middle of it so anything within that range within that 5 foot range the output is going to be pretty consistent as soon as you start bringing your ear your microphone to the ends of the arrays the output and upper frequency ranges is going to start to fall off and you're gonna have an output that is very consistent within that range so long as you don't get above and below it so there's advantages there also and then it minimizes ceiling and floor reflections in all those upper frequency ranges so you're just getting an output that is continuous within that distance so how does that work and what are the distances between the drivers how important are those let's say the distances between the drivers for the stored properly need to be within about one inch if you keep it to within less than or within one inch you can somewhat maintain the same response that you would get as if it were a continuously long runner so let's say I have a tweeter and they made tweeters that were like the old Rd Series tweeters there were a continuous line and when you measured that tweeter what you got was you were hearing the output of the upper frequency ranges as it dissipated right for mr as it left the center of the driver and what happened to the output at that same frequency at the top or at the bottom of the driver was it just went past you you really didn't hear it it really didn't spread out so you don't really hear that output so much you're only hearing it at the high frequency range when you're right in front of it same goes for a continuous one of multiple drivers that are stacked that close apart it also matters that it's not a circular shape to some degree when you have a tweeter that's a circle we consider the acoustic centers of those drivers in other words we look at the distance from the middle to the middle that's we look at the voice goal of that driver versus a planar magnetic or a ribbon the voice pool itself is the actual element so when you stacked one on top of the other the voice called a voice Col basically is gonna be edge to edge so it helps improve things quite a bit when you're using that type of a driver in order to make a continuous line that way so what happens when we put a microphone out in front of this speaker and we're measuring the response and obviously we have a delay from the tweeters that are at the top versus the tweeters that were in line with versus the tweeters are at the bottom now let's take the upper frequency arrangement let's take 10 K Hertz and above remember I said the diaphragms are two and a half inches tall on these so the beaming frequency is about 5600 Hertz so everything above 50 600 Hertz is going to be projected in a straight line so in those frequency ranges as you move up with your microphone or with your ear you're basically just hearing the tweeter that you're in line with at that point so as you move up and down you're kind of skipping from one to the next and you're staying within a range to where you you have the blending of one to another and you're not hearing and the upper frequency ranges the tweeters that are at the top or the tweeters are at the bottom you're not in line with them you're outside of their beaming range so it's acting as if the same as if it were a continuous driver so you're not getting that much cancellation effect in some areas as frequency starts to decrease there are areas where there's a time differential from one set of drivers to the other and you would get a dip in the response but it's also filled in because some of the other drivers are closer to it and you're getting a less of a dip in response so you're staying within the frequency range to where there's very little cancellation on some drivers and a lot of cancellation on the others what you get is a bit of an average when you take the frequency response so what you wind up with when you measure the line source is you you get an average that's consistent as you move the microphone up and down but you get little ripples in the response across the top and you see little ripples and where you may see a 2 DB peak and then a 2 DB dip typically that's about as much as you see as you move the microphone up or down the position of those peaks and dips may swap so where you have a little bit of a dip you may see a little bit of a peak and we had a little bit of a peak you may see a little bit of a dip but it'll maintain an average response across that frequency range that's very consistent now as frequency decreases the output of those drivers really start to spread out and they act as if they're one driver so you get a coupling effect anytime you have a group of drivers to where you're you're you have one driver and you parallel it to another driver typically you gain 6 dB I like over here on this point source we have a 16 ohm driver another 16 mil driver you put those two drivers in parallel and you pick up 6 DB of output same with the low frequency drivers we have 16 own drivers here we have 2 in series 2 in series 2 in series 2 in series you parallel those four groups you're gaining as you parallel the first group in the second group you're gaining 6 DB of output as you parallel it one more time but you're doubling it again the 4 below you're gaining 6 DB of output now that is countered to some degree by the filter the crossover is controlling how much output the whole thing has the inductor that's in line with these drivers will pull the response down and I can use that to balance the output and the the impedances which are matched in this case so that it's as a balanced frequency response so with the tweeters that are all aligned here in this particular model we have 4 tweeters in sear for tweeters in series for tweeters in series for tweeters in series and then those four groups are all parallel backs we have 16 drivers in series parallel configuration so the impedance goes right back to the same that it would be with a single driver what happens to the output did we gain 60 B in gain 60 B no and yes it depends on frequency again with the microphone dead center in the array in the upper frequency range you're only gonna be getting the output of a single driver doesn't matter how many groups you parallel on a light source you're still just gonna have specially at 20k Hertz you're still only gonna be in line with one driver you'll never have any more output than you will with one driver and from there to about 10k Hertz you're gonna get a slight bit of gains but not not a lot because just like I said you're still only within one driver and then as frequency decreases you get more and more output so you're gonna gain that 12 DB of output at the lowest frequency range that the tweeters cover that's another reason why when you design a line source textbook slopes on the filters don't apply for you guys that think I'm just gonna get a line of tweeters I'm gonna get a lot of mid base drivers I'm gonna buy an electronic crossover and I'll just use it by here and I'll adjust it to where the response is is correct between the two no no it doesn't work that way those are textbook filters that are on those electronic crossovers they don't apply to a line source because you don't have a textbook slope you have a slope that you have no gain at the top end and as frequency decreases you have a lot of gain in the bottom in so the filter has to become active at the higher frequency range and basically take that frequency response and fold it over you're you're basically using a smaller capacitor at the front of the network you're folding that response over to make a flat response and then you use them the rest of the elements of your filter to control the very bottom and it can control the crossover point so it's very unconventional in the way we design a filter for a line source or a typical speaker so that's what we have to do in order to get a accurate frequency response so a little different then you get from a lot of other a lot of other types of speakers now advantages in the line source very much like the array when you're sharing the load against amongst a lot of drivers you get a lot of speed a lot of detail very little energy in this case the drivers are all playing a magnetic so they're all about the fastest driver you can you can come up with to begin list so you got lots of speed lots of quickness and then you're sharing the load amongst the whole thing it's incredible it's it's I can't even describe it how different it is versus conventional speakers and when it's playing you can literally walk right up to it and stick your ear to it and you think wow that's barely even on and you can go up and down the array with your ear to it realizing that each segment is not playing very loud at all as you get further and further away from it you are hearing more and more of the whole array so it's a it's kind of eerie how it works it's when people first walk up to it but their ear to it and back away it's usually a bit of a surprise how much each of the drivers are just sharing the load a they're not playing very loud independently but as a group playing pretty high SPL levels it's interesting how again they don't load the room like a typical speaker you have minimal floor to ceiling interactions and it's just it's different it's one of those things you have to experience in some ways it's a level of performance that you can't get from a regular speaker but it has to be designed properly mentioned earlier there's some do's and don'ts some of the don'ts if you look at some of the earlier line sources from infinity in Genesis you'd see a continuous ribbon and you and that continuous ribbon at a top end above 10 caters that just fell right off so they took little circular planar magnetic tweeters and just line them up down the side to try and bring back that top octave problem is the acoustic centers are so far apart that you got a huge cancellation effect from one to the other much like you would with the full range drivers you know because you only have a one-inch diaphragm even up to thirteen hundred and fifty Hertz or it to be as being frequency it's it's spreading out in all directions so you you're not just hearing just one driver as you move from one of the other so much you're you're catching quite a few of them all at once and there's a time differential between them and you get huge cancellation effect so lining up a group of tweeters that are a don't tweeter it's based on that far part it doesn't work at all and in the case of those speakers they're across in a really high crossover point which was up at you know 10k Hertz so as you move left or right the delay between the tweeter and the continuous tweeter of the group tweeters versus the continuous tweeter as you move left or right you could very easily create a one-inch differential and time delay without even moving very far off axis so one would be delayed in time versus the other in such a way that there would be cancellation effectors immediately especially if you start moving very far off axis left or right you get huge cancellation effects from one the other in this case with these line sources the crossover point is down around 1,400 1,500 Hertz so you're looking at a wavelength that's so long that even if you move left or right you're not far enough off access to cause a very very much phase rotation you're only looking at about a 10 or 15 degree phase rotation as you start to move off axis you know 15 20 degrees off-axis so their response doesn't change that much the wavelengths are much longer this is a good way to do it and if you look at even a lot of the line sources out there with a crossover point we get begins coming up to two and a half three K Hertz even your wavelengths are starting to get short yeah you're looking at wavelengths that are six inches or less so as soon as you start moving off axis you've got a delay in time of one set of drivers or so the other set of drivers and you start getting cancellation effects so one of the big advantages of something like this really low crossover points really long wavelengths that means the effects response is consistent in every direction so that's one of the do's versus the don'ts like I said there are a lot of light sources out there that aren't good examples of how a line source works and there are light sources out there that are pretty incredible I would love for you guys everyone out there to get to hear something like these they're a full array or a full line source that are cost no object performance levels they're not costs no object performance levels when you when you look at cost reason this is something we sell as a kit this is something we may offer as a kit we're gonna do something probably fully assembled and get those into production soon right now this is the second one that we've done another issue with doing something like this is what you build it out of the original models were made out of aluminum so the whole thing was cnc milled out of aluminum you couldn't really do it at a MDF when you cut all the holes for the drivers you have nothing left but a frame so if you took the front bath when you lifted it up on one side then there's not enough support there and just almost snap in half so it's very difficult to do something like this this particular one is done in a composite material that's super strong so it's it's still really expensive to CNC mill this whole thing out of a composite material but it works and it works really well and it's it's something that we're gonna be putting in a production I hope you guys get to hear it I know you got little bit of a plug there for some of our products but it's hard to use our products as an example and not give you a little something about it that's about it for today as far as light sources if you guys have additional questions about light sources throw them out there to me put them in the comments section I'll try to cover it and that's all for this week thanks for dropping in and see you next week

Channel: New Record Day

Views: 38,586

Rating: 4.8767653 out of 5

Keywords: audiophile, hifi, loudspeaker, fullrange driver, comb filtering, GR-Research, Danny Richie, crossover design, measurements, best loudspeakers, tube amplifier, preamplifier, diy audio, diy speakers, audiophile speakers, audiophile music, amplifier, best voice, audiophiles, disappear, high end, baffle step loss, open baffle, fast bass, subwoofer, subwoofers, line source, drivers, high power handling, low distortion, the room, speaker upgrade, line source speakers, line array speakers

Id: -GVdYAQkD8I

Length: 24min 33sec (1473 seconds)

Published: Wed Sep 25 2019