PDA

View Full Version : Almost ready to start building



johnaec
09-02-2004, 10:53 AM
OK - I've collected all the drivers and am almost ready to start building a custom 4-way system. Please don't suggest different drivers - there will always be something better, but I really want to satisfy my curiosity using the following combination, and they're what I've got:
2405H
2105H or 104H-2
2021H
Sub1500

The closest JBL systems I've seen are the 4315B and 4343/4345, though obviously I won't be using the horn.

I really want to end up with a completely passive system, but due to the power hungry Sub1500's I'll also leave the bi-amp option open.

Below is a front view of what I've come up with. It's almost exactly the same size as the 4315B at about 33" x 20.5", (w/o base), but deeper at 16". This leaves me with ~ 3.5 cu. ft. for LF. Two 3" ports about 10" long will tune the box to ~ 28 hz, resulting in a system essentially flat to 30 hz.

http://mark8.org/users/johnaec/Sub1500-4way.jpg

My first question is what people think of the physcal layout? Rather than putting the main drivers in all vertical alignment I figured this arrangement would result in a closer point source. Opinions?

Second - in a .3 cu.ft. box tuned to 75 hz the JBL specs for the 2012H show essentially flat response down to 100 hz. Without the port it appears as low as I can get is 300 hz @-3db. Are there any ramification using a small port in the 2012H box? Even though the frequency plots of the Sub1500 show it essentially flat to ~ 300hz, I'd obviously like to cross over lower.

Third - crossover to the 2105H/104H2. The 4315B crosses over at 2Khz, but I'm thinking with the 2012H i might want something lower, closer to 1Khz, as in the L65A. And due to the efficiency of the 2012H it'll obviously have to be padded down.

Fourth - 'still deciding on whether the 2105H or 104H-2 will sound better. The curves Giskard posted make me wonder, though sticking with the 2105H/2405H combination would make using the HF section of the 4315B crossover mostly a no-brainer.

And last, (for now) - I'm not sure about crossing over to the 2105H at 8Khz, even though the 4315B does. It just looks like the crossover would be smoother if it were in the 6Khz range, (6.5Khz in the L65A). I've also got some 2404H-1's I can play with...

Anyway, I just wanted to get some opinion before I start putting this together. Obviously, the crossover will be the biggest challenge, but I'm in no hurry, and just want to get started in the right direction.

John

Robh3606
09-02-2004, 12:34 PM
Hello John

Those look like a nice set-up!! The only issue I have is keeping the drivers in a vertical line up. But that's me. The 4344 seem's to work just fine with the slot of to the side. I would just worry about issue found with most center channel speakers off axis. You really don't have the room unless you change the height and use a similar lay out to the 4315 with the slot and 2105 side by side above the 2012. Your idea about using the 2404 sounds like a good one if you can't get the transition right with the 104h-2. Nice to try that's for sure. Should be a real nice system when you are done.


Rob:)

Mr. Widget
09-02-2004, 02:53 PM
Looks like a fun project John!

I too would go with a vertical alignment. The drawing shown is to scale with the midpoint between the 5" mid and 2405 at 38". To achieve this with minimum baffle I would put the sub woofers in their own cabs. I have done this as high as 250Hz to good effect as long as the subs are placed very near the mid-bass drivers.

I would emulate the 4315's 8KHz crossover as it sounds better (smoother) than the L65's lower crossover. Hell you already have 2 pairs of L65s, why make more?

Widget

4313B
09-02-2004, 03:42 PM
:yes:

Robh3606
09-02-2004, 04:14 PM
Hmm Nice looking layout.

Rob:)

johnaec
09-02-2004, 04:18 PM
Well, I was hoping to keep things to a single cabinet per side, but it looks like it may be worth re-thinking...

Widget - are you saying you've run the Sub1500 up to 250 hz, or just a similar arrangement? 'Looks like an L212 setup on steroids. :)

John

Robh3606
09-02-2004, 04:24 PM
Well when I had my "poormans" L250's I had L80t3 stacked on top of Le14's in seperate boxes with a passive crossover and it actually worked better than I though it would. L212's on steroids nice:rotfl:

Rob:)

Mr. Widget
09-02-2004, 04:51 PM
Originally posted by johnaec


Widget - are you saying you've run the Sub1500 up to 250 hz, or just a similar arrangement? 'Looks like an L212 setup on steroids. :)

John

Similar setup. I wouldn't use the Sub1500 above 100Hz.

Widget


Steroids :hmm:

johnaec
09-03-2004, 07:20 AM
Originally posted by Mr. Widget
Steroids :hmm: Well, all the components except the 2105 are just a little step up from the L212...

I also wanted to thank you for the considerable time you must have spent on that drawing. I noticed you rotated the mounting holes for the Sub1500 - does this work better? And I see your preference for the offset of the drivers. I remember someone here modified their L212's this way - is it really that significant, since all the drivers are in their separate enclosed spaces anyway?

I'm really unsure now what to do. All vertical single cabs, ala 250TI, the component setup you presented, or other? My initial purpose was to stick to just two cabinets, and deal with the compromises, since they'll be in a relatively small room. I just don't want to end up with a bunch of big cabinets trying to make it ideal. I'll have to think about this for a while...

John

boputnam
09-03-2004, 07:39 AM
Originally posted by Mr. Widget
I too would go with a vertical alignment. Widget, that arrangement has me lusting for my 4313B's, same-same. They would be an interesting alternative - that is if those SUB1500 boxes ever got built... :(

Great project, John!

johnaec
09-03-2004, 08:09 AM
Originally posted by Giskard
Start with something like a 0.5 cubic foot volume tuned to ~ 80 Hz along with a 1.0 mH series inductor followed by a 3.5 ohm series resistor and a 10 ohm parallel resistor. I think that should get you good bandwidth from ~ 90 Hz to ~ 1,000 Hz at roughly 93 to 94 dB W m. Use multiple resistors to arrive at the values because you're going to need the power handling and they are going to get hot. Thanks Giskard - I'm sure I'll have plenty of crossover questions before long...

John

Mr. Widget
09-03-2004, 10:39 AM
Originally posted by johnaec
I noticed you rotated the mounting holes for the Sub1500 - does this work better? And I see your preference for the offset of the drivers. I remember someone here modified their L212's this way - is it really that significant, since all the drivers are in their separate enclosed spaces anyway?




If you notice my 2405 is different too. Rather than use your drawing, I used drawings from my driver database that I had drawn before. My mounting holes were simply rotated 45 degrees.

Offsetting the drives is important for woofers to prevent standing waves within the cabinets and it is important for mid and high frequency drivers in dealing with their interaction with the front of the baffle and cabinet edges with respect to diffraction. I have found that it is less of an issue with horns, but
can be very important with direct radiators.

Widget

Robh3606
09-03-2004, 12:20 PM
Yeah I agree on the offsets. I won't even use symetrical braces inside my cabinets. No two sections between braces on a single panel is the same dimension. Everything is offset so there are no even multiples. Might be a bit nuts but it seems to work as the cabinets seem very dead.

Rob:)

johnaec
09-03-2004, 12:24 PM
OK - I've decided how I'm going to start. For testing purposes, I'm going to make individual test boxes for each driver, like building blocks, and I'll put the Sub1500's in some 4628 cabinets for the tests. That way, I can see how significant vertical versus mixed orientation is by moving the boxes around, (assuming I get some acceptable crossovers together). Seeing as many JBL monitors got mounted horizontally in the studio, I can't think it's totally unacceptable to not have all the drivers vertically in a line. This gets back to my original intent to just have 2 cabinets as compact as possible. Either way, all the drivers end up offset from the edges, (with the exception of the subs).

If vertical versus mixed does make a significant difference after listening, I'll go from there, probably with Widget's design. It seems the main thing is to get the HF drivers nearer to ear level, but I've been listening to the much shorter L65A's, so even my "mixed" design's an improvement.

And I really appreciate all the comments! :)

John

4313B
09-03-2004, 12:33 PM
The driver orientation will be influenced by the crossover slopes. If you're going higher order slopes then you can do something like what is shown in your original drawing. Reference for instance that the 2405 in the large format Studio Monitors has an acoustic slope of 24 dB/octave. Reference that a system like the 4411 has a significantly different filter (3109) than it's Consumer counterpart the L112 (N112) since it is designed to be oriented just like your original drawing. Reference the LSR32 with it's 24 dB/octave slopes.

The idea of building test boxes is SOP.

johnaec
09-03-2004, 01:30 PM
Giskard - do you think there is merit in starting with active crossovers to get an idea what works best? I've got two I can try - a JBL DSC260A and DBX Driverack 260. They both offer multiple slopes and crossover frequencies, though both are only stereo 3-way. Obviously I'll have to use a combination to get 4-way.

My biggest concern there is that while I do have 4 Crest amps to work with, they're all significantly different models - I'd have to look into adjusting them all for the same sensitivity. I think I know someone who can help me out in that - maybe that's the way to start.

Assuming I get all the amps/crossovers adjusted to my liking, is it a fairly straightforward response translating the padding, frequencies, and slopes over to a passive interpretation?

John

4313B
09-03-2004, 05:56 PM
Well that would be cheating but it could be fun :p

boputnam
09-04-2004, 10:16 AM
Hey, John...

Dreaming of you very cool project, the 2105H or 104H-2 subenclosure came to mind.

Make sure you leave sufficient space on the baffle for this - the tube I used is 6-inch diameter, and so has a footprint larger than the 2105H or 104H-2. This impacts the spacing of the 2405, and other (L-pads, duct, etc.).

I had great success with this stuff - it's very heavy cardboard cement form. I have plenty left, I believe, stashed in the attic. You live not so far away so I'd be happy to give enough to you if you go this route.

boputnam
09-04-2004, 10:20 AM
This shows the minimum spacing required between the LE5-9 (in this case) and the 075. We swapped-in the 2405, too - identical. There was only a few mm between the UHF frame and the subenclosure.

johnaec
09-04-2004, 11:12 AM
Interesting stuff, Bo. At least for my test rig I was thinking more in terms of boxes, so I could stack them however I want. But you did remind me JBL does something similar in their cabinets - maybe I'll do that for the final versions. I'm guessing internal reflections of the 2105/LE series aren't significant enough for different dimensions?

Put some 2405's in those cabinets you posted and 'looks like you'd have a set of homemade L65's, (discounting the LE14 in one of them and white cone in the other...).

John

boputnam
09-04-2004, 12:39 PM
:rotfl:

Exactly. But, my boy preferred the "look" of the 075's, so that's how we went. We had both options. From cosmetic perspective only, I couldn't steer him differently either - I LOVE the look of those 075's!

With the help of Giskard, and input from GordonW (maybe good ol' EarlK too... :hmm:) we sourced the tube and design dimensions. As well, Giskard reminded the rear surface of the subenclosure (1/2-in ply in this case) was set purposefully non-parallel to the baffle. And, that the subenclosure was not centered on the driver, either.

I'm certain all would apply to the 2105 - being Pro versions.

(ps: Nice JBL Blue, huh?)

Ian Mackenzie
09-04-2004, 03:31 PM
Since you have a nice active crossover I would be inclined to mockup your drivers layout in stacked boxes , dial in the network and just listen to it before settling on a baffle or driver line up.

The 2105/077 thing is a bit of an issue and the 2105 is peaky up there and neither it nor the 077 is ideal in that scenario. Better a 066 or 044 crossed over lower with steep slopes. Although smoother this won't have the power handling your after to match the subs.

On home theatre your going to need some horn loading up there and controlled directivity is the in thing.

Have you though about the mini bi radial from the 4435 or the bigger 4430. Having heard Rob's system I would say its very nice and would be easier to get it right imho. Also the tried and proven 2307/2308, you will be surprise and the 2420's a quite cheap S/H

Ian

Ian

johnaec
09-04-2004, 03:48 PM
Originally posted by Ian Mackenzie
Have you though about the mini bi radial from the 4435 or the bigger 4430. I would have loved tracking down either a pair of 2342's or 2344's with 2425/6 to run above the 2012H's but never seemed to be in the right place at the right time.

For this set, I'm just going to see how it works with the drivers I have. But who knows what the future may hold, (especially after reading the JBL/TAD discussion...).

John

John
09-05-2004, 01:20 AM
Anyone seen him ???:confused:

Mr. Widget
09-06-2004, 05:11 PM
Good to hear from you!

Widget

4313B
09-06-2004, 05:26 PM
Yes! Hurry back! :yes:

Chas
09-06-2004, 05:27 PM
Earl, good to know you're alive!
:smthsail: