X-over questions

**JonMarsh** · 18 September 2001, 11:16 Tuesday

The picture for the crossover, natrually, is the one that doesn't display. Doesn't it always work that way?

It's common, with the advent of computer optimizers, to use as few a components as you can get away with in a commerical design, as long as you get fairly close to the target acoustical function. Even a DIY kit (for sale) is a commerical product.

Many kit designs I've seen don't use a zobel. Without a zobel, the changing output impedance of the driver interacts with the filter impedance, and you don't get the electrical crossover slope you think you're getting. Sometimes that's OK, because the full electrical slope isn't needed, because of the existing acoustical roll off. Using natural roll offs at the top of a driver's pass band is a questionable way to go, though, if you're seeking clarity.

The zobel network has another possible function, it is often critical to implementing baffle step correction, when adjusted in concert with the nominal filter component values.

Many also don't include baffle step compensation (for the change in driver loading vs frequency as the baffle size goes from being a "launch" platform at some wavelengths, to minimal effect at others). This typically occurs in the region of 500-800 Hz. This affects how the speaker can be used and how it sounds considering room placement. Without baffle step compensation, the upper midrange will be lifted by 3-4 dB. This can be compensated by putting the speaker against the wall. This makes for other problems, if your room or the speaker isn't really designed for boundary loading. Also, other reflections from wall boundarys and furniture will typically severely compromise imaging. Imaging, of course, isn't important to everyone. But, the early reflections that destroy imaging also create comb filtering in the frequency response, which is also audible, and undesirable.

What kit did you get, Andrew, and what is the intended application? Computer? Bedroom? Rear channels? Can you get the crossover schematic properly referenced? And what tweeter is used?

Best regards,

Jon

Earth First!
_______________________________
We'll screw up the other planets later....

**AndrewM** · 18 September 2001, 11:50 Tuesday

Well I fixed the pic-link (damn case sensitive servers).

Anyways, the kit is an Aria 7 TLR. I managed to pick it up as a "demo" from somebody in Canada (I love exchange rates), got all the drivers and an assembled x-over (using Solen caps/inductors with Theta bypass) for VERY cheap.

So the kit uses the Focal TLR tweeter and the 7k4411 woofer in an MTM. They'll be used as the mains in my setup for now.

Right now I'm playing around with the enclosure design and I'm leaning towards going with a stand mounted setup as opposed to the tower design (Orca has plans for both, but with no x-over change). So despite the kit being designed by a well known person I think it was more of a quick-n-dirty design rather than one that was built and tweeked to death. And considering the components that are used I'd like to get the most out of it.

I've been thinking for awhile about buying something like LspCAD Pro/MLS (I've got enough computers laying around to use), but I wonder if I would use it enough to justify the expense (not that I've ever let that stop me). One of my biggest reasons for going with it is that I'd like this simple 2-way MTM to grow into a 3-way MTMW setup, but I think that will require more accurate measuring/tweaking ability than I can get out of using an SPL meter and test tones.

Andrew

**JonMarsh** · 18 September 2001, 15:27 Tuesday

Hi Andrew,

Well, that's a nice expensive tweeter you've got there, and some pretty fancy seven's, too.

Looking at the crossover, it's an electrical 2nd order low pass on the woofer (probably counting on part of the woofer roll off to contribute another pole, making it equivalent to a 3rd order acoustic.

The Audiom tweeter has always been something of an enigma to me- I haven't ever seen the qualities on paper which would justify the somewhat hideous price they charge for it compared with the Td120dx2, one of my favorite all time tweeters.

Now, examining the impedance curve would lead one to think that the Audiom tweeter doesn't use ferro fluid damping. But Focal claims it is Ferro fluid cooled. Efficiency is quite high- rated at 95 dB, even higher than the Td120dx2.

However, though the nominal resonant frequency is at 1 kHz, the response begins rolling off below 3 kHz. Puzzling, and annoying.

Now, the crossover shown looks like an electrical 3rd order at about 3.5 - 3 kHz. I could plug the numbers into LSPCAD and figure out what it is, nominally. There's no zobel for tweeter resonance, and there's no zobel for HF impedance rise. I wouldn't do that with a tweeter this expensive- there'll be a slight kick up in response at the LF resonance, and there may be some interaction due to HF rise. Attenuation is only via a series resistor- which forms a voltage divider with the tweeter impedance.

Your words, cheap and dirty, are how I would describe this crossover. Glad I didn't have to use them first; occasionally I like to fool myself into thinking I'm more of a gentleman.

My own studies and calculations show that dual sevens are generally not a good option for a full range speaker- compared with a high performance 8, they usually won't have the same bass extension, though efficiency and mid band power handling will work in their favor. Cost is not so good, of course.

If you want to use these without modifying the crossover much, consider this idea.

Ideally you need to deal with the crossover not having baffle step compensation. One way would be to run these high pass at about 125 Hz, in a smallish box, with a series 1st order electrical crossover; net high pass crossover with driver roll off would be 3rd order, particularly away from boundaries. Have a bass cabinet with comparable nominal efficiency (a slight problem- mostly one of cost), with a first order roll off starting at 125 Hz. That will fill in some of the baffle step loss, but you'll need to have LF drivers clean to 700 Hz or so. Also, efficiency will need to be in the 93 dB range overall for the bass- not as easy as it sounds. (this, BTW, is what the real X1 SLAMM does). Eton 11's, or dual Eton or Focal 8's come to mind. The 8K5412 is fairly smooth. The 13V7511 comes close to having the efficiency you need with just one driver. If the 13V7511 is boundary loaded near the floor, which gives it some extra effective output, that might work. If you biamped, then efficiency matching goes away as an issue.

Just some ideas.

BTW, do you know anyone nearby that could measure them for you with a good setup, using LAUD or CLIO or LMS? If so, I could look at the crossover design for you. I could measure them, but I'm in Northern CA; I have a few different test boxes, including one for dual 7's, but I understand you might not feel comfortable packing off a set of drivers to someone you've never met to make some measurements.

If you can get some measurements done, if you want to play with the software yourself, you could at least use better measurement data then you can get with an RS meter. (not very good!). I use CLIO with my vintage HP mic preamp and Bruel and Kjar 4133 mic.

I think you can wind up with a pretty nice system when you're finished. If you don't want to totally scratch build an enclosure, the Woodstyle 123 REV would work well with these drivers; you can get those from Madisound for a reasonable cost; just plan on re-inforcing the heck out of them, like my M8 project. I build a set of MTM's for a friend using Eton 7" midwoofers and the Td120Dx2 that way. Platform panels for the mid woofers, lots of internal oak bracing, and a tweeter back brace/subenclosure- I don't believe in exposing tweeters to the woofer back pressure.

Best regards,

Jon

Earth First!
_______________________________
We'll screw up the other planets later....

**AndrewM** · 18 September 2001, 16:34 Tuesday

The Audiom tweeter has always been something of an enigma to me- I haven't ever seen the qualities on paper which would justify the somewhat hideous price they charge for it compared with the Td120dx2, one of my favorite all time tweeters.

Well as to the tweeter, I managed to get the whole kit for under $700US...this was mostly due to the favorable exchange rate between US and Canadian money. So almost every comment I've read about the TLR was that it does indeed sound better than the TC120tdx2, but as you said, most people don't hear the almost $400 difference (over the TC120 series), but I luckily got around that.

Your words, cheap and dirty, are how I would describe this crossover. Glad I didn't have to use them first; occasionally I like to fool myself into thinking I'm more of a gentleman.

Well, this has been a common complaint about these designs from what I've read...they work, but are far from optimal.

My own studies and calculations show that dual sevens are generally not a good option for a full range speaker- compared with a high performance 8, they usually won't have the same bass extension, though efficiency and mid band power handling will work in their favor. Cost is not so good, of course.

I'm not to worried about the absolute in bass response with this setup, I have a new Tempest sealed sub going which I'm pretty happy with and extends high enough to match the Focal 7's. At least until I can move the whole system to a 3-way.

I'm kind of holding off on trying out the 3-way setup as I'm waiting for the new Focal line to appear and more importantly I'm moving in the next month. It looks like Focal is dumping the V-line of woofers (no more 11v or 13v, but still a 10v), but they are adding a 13" Kevlar and a 13" W cone, however none of them have the effeciency of the 13V (the above 13"'ers are 91db). I also need to try and keep the enclosure fairly small (no X1 clone for me

). So I could look at doing a dual 10" (or 11") driver (fairly expensive option though, and a big enclosure), go with one of the 13" drivers and hope for enough "floor gain", or pad down the whole MTM section a couple db. But I'd need to invest in a good measuring system for all of that (and no, I haven't heard about anybody near me who has a measuring setup).

I think I'll end up just building the whole setup as spec'd in the "quick-n-dirty" design, and then invest in a measuring/design package, and then go to town with it. I nice big 13" kevlar woofer sitting under the two kevlar 7"'s would look pretty kewl (and properly design sound pretty damn good also).

BTW, you can see the new Focal lineup (but no specs) at

http://www.focal.tm.fr/fr/compo/index.htm

And prices from a German distributor (doesn't mean much though).

http://www.speaker-en-co.nl/Focal-u.htm

Thanks for the info Jon,

Regards
Andrew

**JonMarsh** · 19 September 2001, 00:41 Wednesday

I've seen the page for the new Focal drivers, like many others, I'm wondering when they'll publish the specs and when distributors over here will have info and parts to sell (can't build much with just vapor ware).

Focal rotates their models a lot more often than some other manufacturers, like, say, Eton. I hope the 8V5411 is similar to the current 8V model, at least as far as having a clean midrange goes.

You got a steal deal on your Aria 7 kit, as I'm sure you realize; you'd normally pay more than that for just the Audiom tweeters. I expect the Audioms will sound very nice in an optimized system; will they sound five times as nice as a TC120dx2? Hmmmmm, that's a little hard to imagine. Will they sound better than Accuton C23's? That's even harder to imagine- given the measured performance- and the Accutons are half the price. But also less efficient. For a system that you want to play loud without problems with voicecoil heating, you can't beat the big Focal tweeters.

With the price of Focal and Eton woofers, you really *do* have to think twice before considering a dual woofer cabinet.

It sounds to me like you're on the verge of getting sucked into serious speaker tweaking and development. That's how it goes, you know. For a budget approach, LSPCAD with their MLS software, plus one of the decent inexpensive electret mics would get you off the ground nicely. A key to being able to do some useful measurements in your home is MLS capability, as I imagine you realize. That, plus a decent package for crossover optimization. The pro version of LSPCAD is neat because of the filter emulation function, but you can do a heck of a lot without that.

You've got some good parts, and I think you're going to have a lot of fun getting the most out of them, if you can get a few tools to help with the process. But watch out, that's how you get hooked on doing this. There's always one more system you want to build....

Regards,

Jon

Earth First!
_______________________________
We'll screw up the other planets later....

**AndrewM** · 19 September 2001, 10:33 Wednesday

Focal rotates their models a lot more often than some other manufacturers, like, say, Eton. I hope the 8V5411 is similar to the current 8V model, at least as far as having a clean midrange goes.

Any theories on why they do this? It makes it a pain in the butt for those of us still building kits, or looking to build older design's. Of course you fool yourself into thinking the specs are close enough, etc that you end up spending more money to get a non-optimimzed system.

I'm also waiting to see the general reaction on the W-coned drivers, they're promising a lot like having all the benefits of kevlar but without the nasty break-up's higher in the range, but none of the new line is out yet, except for a couple drivers at Zalytron. So maybe you'll have a couple of options when it comes to the 8"'er.

You've got some good parts, and I think you're going to have a lot of fun getting the most out of them, if you can get a few tools to help with the process. But watch out, that's how you get hooked on doing this. There's always one more system you want to build....

That's what I'm afraid of, pretty soon I'll have a high end stereo system in every room of my house, then my friends house's, etc. My girlfriend will probably roll her eyes even further back in her head.

Andrew

**JonMarsh** · 19 September 2001, 15:43 Wednesday

That's what I'm afraid of, pretty soon I'll have a high end stereo system in every room of my house, then my friends house's, etc. My girlfriend will probably roll her eyes even further back in her head.

Geez, Andrew- what have you been doing, taking a guided tour of my home? Actaully, that's a bit of an exaggeration- what I have is a collection of somewhat vintage quasi high end systems with DIY stuff scattered around- like my bedroom system, which is a WMTMW using ScanSpeak Kevlar 8's, MB Quart Mids and tweeters, with a Denon Digital preamp and Denon monoblocks. And that dates from the late 80's. The problem is that the rate of building and change is picking up, not slowing down! In the last year I've bought seven power amps, built two electronic crossovers, bought another one, and designed three speaker systems. I've got to get this problem under control! Just as soon as I get the CLIOWIN software, and finish the M8 and M10 MkII. And the new subs with Blueprint drivers and Styke drivers.

Fortunately I get to exercise some of my design and development mania on ThomasW's projects, like the Whisper Klones. Now he's working on a dipole system, using the GB ribbons, and Eton midwoofers- this should be fun, and a little less of a woodworking puzzle than the Whisper's were. I leave the strange woodworking problems to him, I just concentrate on drivers and crossovers, and genearl configuration.

Seriously, Focal *is* starting to annoy me; even the tweeters are all changing, and the Td120dx2 hasn't been out that long- plus there's a new Audiom model, which looks like it doesn't have the waveguides. Why are we going from a tx2 to a TD5? Obi Wan, what does it all mean?

Now, how close is the 8V5411 going to be to the 8V4412? I'm hoping that they've just taken the polyglas cone and put it on the larger magnet assembly, as used in the 8K5411- they share the same VC diameter, on an anodized aluminum former. The 8V4412 is cleaner up to 2.5 kHz than their other 8's, including the kevlar. We'll see. It's almost like they're the Detroit of high end driver manufacturers- they're certainly R&D driven- which I don't mind in one sense, but there's something I like about being able to get the same Eton drivers over such a long period of time, for example.

Enough ranting.

Keep us posted on how your project goes once you get into it. My sympathies to your girlfriend. Maybe we should start a forum section for significant others of audioholics and the DIY dependent. Mine might be interested!

Best regards,

Jon

Earth First!
_______________________________
We'll screw up the other planets later....

**AndrewM** · 19 September 2001, 16:34 Wednesday

Seriously, Focal *is* starting to annoy me; even the tweeters are all changing, and the Td120dx2 hasn't been out that long- plus there's a new Audiom model, which looks like it doesn't have the waveguides. Why are we going from a tx2 to a TD5? Obi Wan, what does it all mean?

Now, how close is the 8V5411 going to be to the 8V4412? I'm hoping that they've just taken the polyglas cone and put it on the larger magnet assembly, as used in the 8K5411- they share the same VC diameter, on an anodized aluminum former. The 8V4412 is cleaner up to 2.5 kHz than their other 8's, including the kevlar. We'll see. It's almost like they're the Detroit of high end driver manufacturers- they're certainly R&D driven- which I don't mind in one sense, but there's something I like about being able to get the same Eton drivers over such a long period of time, for example.

I'm very interested in what the W-cone will do for people, E-speakers has been selling the automotive versions for awhile now, and with good results. Zalytron has 3 of them listed (5, 6 and 10 inch models), and from the spec sheets it looks very nice, the 5" driver is flat almost to 10Khz, the 6" to 3Khz and the 10" to 1Khz. What is interesting is that all the measurements have a little wobble (for lack of a better technical term) at 1500Khz, I wonder if that was a problem in their measuring setup. All the while keeping the sonic benefits of a composite cone. What concerns me is that none of the drivers listed with Zalytron are listed anywhere on the Focal site or with any of the overseas Focal distributors (who already are listing the new lineup for sale).

Keep us posted on how your project goes once you get into it. My sympathies to your girlfriend. Maybe we should start a forum section for significant others of audioholics and the DIY dependent. Mine might be interested!

I'm lucky in that my SO is as much a music/movie lover as I am, so she can see the benefits to the whole thing. It's just when I explain a project to her I get those "you need to be locked up" looks. Like my recent Tempest sub, it's sealed and big, like 32" tall x22" x22", when I carried it into the living room and set it down to get the old sub out of the way I got that look, then when I pulled out the old sub (less than half the size literally) I really got the look. My saving grace was when I pluged it all in, moved the sub back out of the way and threw in some of her favorite music and cranked it up and that look faded away. And was replaced with praise after throwing a few of her favorite bass heavy movies on.

Back to the topic at hand, I was going over some of the other Focal drivers and stopped at the 10V6411, it has a 93.1db sensitivity and drops down pretty nicely (f3-32Hz), but needs a huge box (not a big deal if I need just one), and is much cheaper than going the other routes (dual 10 or 11 drivers).

So to ask a loaded question, how close do you think somebody could get using say your above example to do a 3-way? Using textbook values in an x-over and rudimentary measuring?

Andrew

**JonMarsh** · 20 September 2001, 00:58 Thursday

I'm lucky in that my SO is as much a music/movie lover as I am, so she can see the benefits to the whole thing. It's just when I explain a project to her I get those "you need to be locked up" looks. Like my recent Tempest sub, it's sealed and big, like 32" tall x22" x22", when I carried it into the living room and set it down to get the old sub out of the way I got that look, then when I pulled out the old sub (less than half the size literally) I really got the look. My saving grace was when I pluged it all in, moved the sub back out of the way and threw in some of her favorite music and cranked it up and that look faded away. And was replaced with praise after throwing a few of her favorite bass heavy movies on.

I can see this in my mind's eye like it was on HDTV! It sounds like you're a good match for each other. My lady friend came out with me to Colorado this summer, so she got to watch ThomasW and I build the Aerial Stryke sub and horse it around, plus measure and listen to it at ungodly levels... then she had to endure being dragged up to my friend's at Ayre Acoustics in Boulder; more listening sessions, but some watching, too, with their new DVD player. My friend is tamed a bit since he's got a wife and two kids, but still hardly mainstream.

The 10V6411 looks pretty promising for what you describe. Have you tried plugging the parameters into something a little more reasonably sized, like 65-70 liters? Qts is a little on the high side for a reflex, but a sealed box with Q of 0.577 would probably work quite nicely for you- with your Tempest's, you won't need deep bass extension from the 10's.

Do you have Unibox? It's a free Excel spreadsheet that's good for low fequency driver/enclosure modeling. I don't have the URL on this system, but it may be linked from one of the lists off Madisound.

Like a number of other Focal drivers, it does have a curious impedance glitch which could be their test baffle or may be a cone mode. Still, it's high enough it shouldn't be a problem in the system you propose.

If you're interseted in this idea, I could model the basic system and crossovers and post or send it to you. You'd get pretty close using that plus the crossover's you already have- it would be a darn site better than a lot of commerical systems, just not fully dialed in.

I can plug in the driver parameters for inductance, efficiency, and so forth, and do a basic model without measured data. The basic version of LSPCAD is pretty inexpensive, so if you're considering going down this road, that, perhaps, should be your first investment. You'll probably wind up needing to experiment with things like whether the woofer works better floor loaded or off the floor (comparable to the upper woofer in an X1 or Maxx), but you can figure some of that stuff out just by listening. At about $125 each, they aren't inexpensive, but considering the alternatives, that's pretty reasonable.

Regards,

Jon

Earth First!
_______________________________
We'll screw up the other planets later....

**AndrewM** · 20 September 2001, 10:23 Thursday

John,

I would certainly be appreciative if you can plug in some numbers and at least get me a basic x-over system, that would get me into the ballpark and hold me over till I can pick up some type of measuring/simulation type software.

To move onto other potential headaches how did you implement the 2nd tweeters into your X1 setup? This always looked like an interesting thing to try. Is this just as simple as throwing in a 1st or 2nd (or higher) x-over on the tweeter at a fairly high range (I think you had mentioned something like 9Khz once).

Also, I don't know how often you get to check out the Madisound board, but since this came up a few times the specs on the new Focal stuff is on the site, but doesn't look to have a link to it...but somebody found it, here's the link to the mid-bass units

http://www.focal.tm.fr/fr/compo/midbass/index.htm

You can navigate the with the top bar (tweeters, midrange, midbass, woofers), all the specs are there and it's all in English.

Thanks again,
Andrew

**JonMarsh** · 20 September 2001, 10:53 Thursday

Andrew,

Thanks for the link! But those frogs make me crazy! The new 8V5411 specs, if they're to be believed, are rather bizarre, compared with the 8V4211. Both are 40 mm VC, according to the table data, though the bullets for the 8V4211 say 25 mm voice coil. The one's I've got here clearly DO have a 40 mm VC. Now, here's where it gets strange. The 5411 has the next larger size magnet, like the Polykevlar driver, as I speculated. It has a *shorter* voice coil, 13.5 mm instead of about 16 mm. Gap is the same size, so Xmax is reduced from 5.25 to 3.5 mm. Efficiency is up, of course, to 90.3 dB from 89.7 dB.

Then, to ice the cake, the rated Qts is 0.51, compared with the Qts for the 8V4211 of 0.3. What were they thinking!!?!? It just doesn't add up that they could have a shorter voice coil, a larger magnet/higher BL product, and a higher Q.
The key is in the compliance- much lower, and the Fs is rated at 50 Hz with a Qms of 7.9.

Aaarrrrrggggghhhhhhhh!

Well, if this one system turns out well with the 4211's, I'll probably buy a few more pairs of the 8V4211, or maybe just standardize on the 800/37 Eton, and to heck with the cost difference.

Back to your situaion, I'll run the numbers this weekend and let you know how it looks- I can simulate with the Orca crossover, and nominal characteristics for the drivers (radiating size, impedance and VC inductance, etc.) It should get close. I'll post here to let you know how it looks.

Best regards,

Jon

Earth First!
_______________________________
We'll screw up the other planets later....

**AndrewM** · 21 September 2001, 12:52 Friday

Back to your situaion, I'll run the numbers this weekend and let you know how it looks- I can simulate with the Orca crossover, and nominal characteristics for the drivers (radiating size, impedance and VC inductance, etc.) It should get close. I'll post here to let you know how it looks.

Thanks Jon, I certainly appreciate it.

I found a copy of unibox and have been playing around with it and I found something that may raise an issue, the 10V woofer has a pitiful 3.5mm Xmax, which means that (according the Unibox) even at 50 watts of power I've exceeded that x-max figure at like 80Hz (for both sealed and vented alignments). Now granted 50 watts into that speaker is going to be around 109-110db, but it looks like I'll be able to overdrive these things pretty easily if I try and run them at full range or set to "Large" from my processor (which has just a 80Hz bass management). Any thoughts of how bad of a problem this might be? Whichever processor I decide to go with the next time I upgrade is going to have a bit more bass management built into it but for now...

Also on the power response tweeter, I dug up the info you posted awhile ago, but I didn't see anything on how the x-over was handled. I assume it's just another network thrown in with a high x-over point of like 9Khz, so in reality you're creating a 4-way speaker. Or have I over-simplified it?

Andrew

**JonMarsh** · 21 September 2001, 14:35 Friday

Your point regarding Xmax is important- it's probably how they got the efficiency, but it is a potential limit. Now, the question is, how often will you be likely to hit close to 109 dB or more in the 80 Hz region? For one speaker? (two is 111 dB, approximately). Yes, that's an upper limit, and you'd ideally like to have some headroom over it- that's why Wilson's bigger speakers look like the Max or X1. Also, remember that those are anechoic numbers, and you do get room boundary re-inforcment- up to 6 dB. I think you might be fairly satisfied with the results regardless.

The other alternative would be to bite the bullet and get drivers with a larger Xmax, 3-4 dB or so less efficiency, and parallel on each side.

Well, now Focal has updated the site again, and the English section leads to the new drivers. I don't like some of the design tradeoffs they've made in them. The 10W6411 has a good Xmax, but the Qts is rather high for ported systems; would work well in a largish sealed box, say with a box Q of 0.577.

Your best bet might be the new 10V6412, which reports an Xmax of 8mm, efficincy of 92.68 dB. Note all the old specs are gone, and the images don't display anymore.

Regarding the auxiliary tweeters, 1" dome tweeters start narrowing their dispersion above 9-10 kHz. The axial response usually stays flat, but the room reverberent power response drops. Using a rear firing "fill" tweeter fills in the "air" in the far field response. In a sense you're right, it becomes a four way system, but not on axis- just for the room reverberent reponse. Obviously, room placement and walls become a factor, but in general I think it works OK, and I like it subjectively.

On a measurment basis, you can see the difference doing 1/3 octave RTA type measurements at two to three meters with or without the Aux tweeter. The RTA is a good measure of the room response. The MLS is a good measure of the early arrival axial response.

If you try to make a conventional tweeter measure flat at 2 - 3 meters with an RTA, you'll wind up with a "hot" axial response, which you won't enjoy.

Regards,

Jon

Earth First!
_______________________________
We'll screw up the other planets later....

**AndrewM** · 21 September 2001, 16:06 Friday

That's a good point about the pair of speakers and the room gain, if I stick with a ported design and tune it as low as I can (without losing some SPL) I can still hit some pretty outrageous volumes without over-driving the speaker.

So I think for now I'll stick with the original plan of the older 10V (providing they're still available), and if I find I'm not happy with the results, well hey I always wanted to try out those 8w monoblock tube amps for $99...and I'd already have a nice pair of woofers for a nice 3-way system, just need to pick up maybe a nice set of ribbon tweeters and then a couple of midrange drivers... (this is how this whole speaker designing thing starts isn't it?).

Andrew

**JonMarsh** · 23 September 2001, 17:16 Sunday

Hi Andrew,

I've run some numbers, using data from drivers I've got and tested that are similar to what you would be using. I've used the Aria 7 crossover values, making no changes in the tweeter circuit(net 3), and only modifying the "mid" circuit (net 2) to add the necessary LF zobel (for driver LF resonance) and the high pass capacitor. For the woofer circuit (net 1) there is also an LF resonance control zobel, and the voice coil inductance zobel, plus the LP inductor. These LF resonance zobels are *mandatory*, because with out them you get major league interaction with the crossover network, and peaks in the response. This is what makes the passive crossover for a Wilson X1 a rather expensive proposition.

Below are Net 1 and Net 2, plus the calculated summed response and impedance; midrange polarity is inverted. This plotted response is not totally relevant, but indicative of what should roughly happen. The measured data I used to rough this out is for Eton 370 7" drivers and Focal Td120dx2, with some adjustments to efficiency to mimic your drivers more closely. LF driver is a Focal 8, again with some adjustments.

This should get you in the ballpark.

Regards,

Jon

Earth First!
_______________________________
We'll screw up the other planets later....

**AndrewM** · 23 September 2001, 23:36 Sunday

Thanks a bunch Jon, it is much appreciated. I have a question about C2111 in the Net 2 Zobel, 1000uf? Are the components in the Zobel important quality wise? Madisound has a 1000uf Bennic Elctrolytic for $8 (cheaper to get less tolerance from other sources), I would assume this is good enough.

And lastly about the inductors on Net 1, the 6.5mH can that slip up a bit in DCR? The sledgehammer core inductors can't even manage .1Ohm at 6.5mH, but it looks like around .25 is do-able. I'm not to worried about power saturation since the woofer won't be able to handle it anyways. Also, in the LCR circuit is the DCR less important?

Thanks again Jon, I really appreciate it.

Andrew

X-over questions

X-over questions

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment