A friend of mine, good with woodworking, had the time to build my new bass cabs recently and, last week-end I found the time to assemble everything. I’ve reporter building the bass cabs for two years so it was time, thanks Dan. I’ve used my temporary white test cabinets for over 5 years and puchased all the necessary material a long time ago so it was time. The goal was simply to double the number for drivers and raise the efficiency while keeping it to a a “reasonable” size.
The cabinets are made with 1″ baltic birch plywood and each side include 4 completely sealed volumes. Each volume has a dedicated passive radiator that can be externally tuned. Basically, it’s like having 8 small subwoofer organised in 2 groups. I could, if wanted, tune each passive radiator to different frequency and may eventually try it. My first experimentations had me tune them all to around 24 hz, for now. My previous tests regarding passive radiators had me prefer tuning them as low as possible, when you have the possibilities to do active correction. I use to tune my previous cabs under 20hz but with this 24hz tuning in this setup I get some impressive outputs from 20 hz to 370hz.
Inside the cabinets I went with a mix of 2″ carbon foam and bubble wrap (a la sonus faber and goldmund). Those are placed behind the driver, between the driver and the passive radiator. The final volume for each separate volume is around 60-63 liters. I didn’t unroll the bubble wrap since I got rolls that were exactly the perfect height and already tightly turned so, I simply jammed one roll per cavity in a slighlty asymetric positioning. The old “battery test” tell me they are now slightly under damped. I can add damping later but the frequency response is really smooth so I think it’s doing a good job.
The drivers are 8 visaton tiw250xs, it’s a unique woofer/subwoofer with a very light cone (49g) but with very good excursion and that can take massive corrections without breaking a sweat. The output is clean enough to be used well over 500hz, closer to 1 khz in some applications but that’s not my goal here. yet, this give me lots of possibilities for the crossover point considering that my largest horns are 320hz concrete JMLC.
Doubling the number of drivers per side not only changed the low end response, it also changed the sound dispersion and efficiency of the low mid. They are way more powerful, sound a lot larger and fuller in this configuration. Of course it was expected by doubling and arranging the drivers in a quad configuration and I’m really happy about the results. Now it sound like a large driver with tons of weight in the 100-400 hz while keeping the possibility of shaking the house in the 20-50 hz. It’s like having a subwoofer and a efficient woofer in the same cabinets. I understand that it’s not a small solution but there is no free ride when you want basse and efficiency, the result won’t be small.
My final crossover point is to be determined. With the previous configuration I enjoyed more having the jmlc horns crossed as low as possible . This was using the large horns like a wideband driver. I went with a selection of motor and diaphragm to get a usable 370 hz and up. With the increasing of the surface area, now 4x 10″ per side, it’s quite possible that moving my crossover point higher be beneficial. It’s on the list of things to experiment.
I still have to build the horns rack so I can move the larger horns over the smaller ones, having them in a vertical axis. I also need to build a set of passive radiator hiding grills or metal sheets, the css apr10 are quite ugly because they are externaly tunned. I’m not sure what would look best, for now they will stay visible, like on the included pictures, until I decide on the way to go.
I have a pair of beyma cp21 in the mail that could potentially replace the super tweeters I’m using right now. I will post results when I have the time of comparing the two solutions.
In short, the systeme right now is :
- a pair of bass cabs using visaton tiw250xs and css apr10 passive radiator, custom design with walnuts finish. All internal wiring is done with belden 5t00up double runs. The speaker cables are made with double runs of the same belden cables and a set of furutech spades and a set of banana plugs so each run get a good contact patch. Crossover point from 370hz to 500hz.
- the mids are matched jbl 2440 with radian 1245-16 diaphragms and JMLC 320hz concrete horns. The back cover are removed from the motors (kind of a dipolar configuration) and the speaker cables, made of belden 9402) are attached directly to the diaphragms. Crossover point from 1.2khz to 6khz (the larger horns start to beam a lot over 5khz)
- the highs are beyma TPL-150 with customs JMLC horns in the front and the rear so they are dipolar. The speaker cables are custom made too but with silver plated copper and teflon protection and vampire termination. They do 1.2khz and up but preferably over 2khz.
- the super tweeters are vintage diatone horn tweeters with unknow specs but around 101 db/w and 8 ohms. I simply put a 1.5 uf caps on those connected to the tpl-150 so the crossover point is really high at around 15khz. They may change for beyma cp21 if the 15khz and up is better on the cp21.
It’s imortant to note that my back wall is quite damped as si the room. I largely prefer a damped environment to be closer to the recording.
A DEQX hpd3 control everything. My actual crossover points right now are 370 hz, 5 khz 15 khz. I also have quick programmation for 370 hz, 3.5 khz, 15 khz. I’m not decided between the two since they can be both really good, both horns do the 2khz to 5khz really well so it’s mostly about off axis response and technically the modified tpl-150 should be better but I mostly listen on axis and the larger horns are really energic. I can switch to 4 predetermined configuration on the deqx so I can switch bettwen different configuration depending if I’m on axis or off axis with the remote control.
The amplifiers are :
- for the bass, bryston 7b3 mono blocs (my charges are 8 ohms) so I get a nice 675 watts per channel.
- for the mids, custom made 7189 single ended and triode configured. Output are james 6123hs with a dead silent power supply. It’s crazy over built (audio note transformer, large polystyrene coupling capacitors …) and deliver a “massive” 3.5 watts per side. No global feedback is used here.
- for the highs it’s a pair of custom made magnequest SETH PP 2A3 amplifiers using nikel core and only quality parts … delivering a big 10 watts for each side. No global feedback is used here.
I would estimate the efficiency of each section around :
- 95db/w/1m for the bass cabs
- 112 db/w/1m for the 320hz JMLC horns
- 102 db/w/1m for the JMLC TPL-150 and the super tweeters
My listening position is about 10 feet from the speakers.
The preliminary results are incredible. I will try to optimize more but it’s already fantastic results. I leave you with some pictures and will update later for my tests regarding the low end integration frequency.
I still have lots of experimentation to do like :
- trying to remove the back horns on the tpl-150
- maybe doubling the number of tpl-150 since I have 4 wood horns to play with
- changing the super tweeters to beyma cp21 so it can match the efficiency if doubling the tpl-150
- crossover points for the bass cabs
- adding stuffing to add damping of the bass cabs
- tuning each passive radiators to a different frequency (ex : 20hz, 30hz, 40hz, 50hz) since all my tiw250xs are in different volumes.
- building a metal support to hold the incredibly heavy 320hz horns made of concrete over the tweeters so I can get a vertical disposition