Fostex FE167E TL project

Lensi

Member
2005-04-29 11:43 pm
NC
hi all,
I've just finished a year-long project to build my first full-range speaker. Until I fired them up for the first time I had never even heard a full-range, single-driver speaker. I wouldn't have even attempted this project without the valuable information provided by Martin King at www.quarter-wave.com. Although I didn't copy his plans, I did use his Mathcad applet. Thanks, Martin! And thanks to all the posters at diy forum, among others who convinced me to give this a shot.


FYI, I'm using a Fostex FE167E. Internal wiring is DHLabs T-14. Binding posts are Cardas unplated copper. The cabinets are internally braced with dowel rods. Cabinets are 44" high, driver is 37.5" up from the floor. Acousta-Stuff inside, as well as 1/2" and 1" acoustic foam.

Lensi

[IMGDEAD]http://home.earthlink.net/~raflynn/Boxen/small_jpgs/mini_CRW_3111b.jpg[/IMGDEAD]
 

Lensi

Member
2005-04-29 11:43 pm
NC
The enclosure is a straight MLTL. I designed a BSC and impedance correc tion before I finished the cabinets, but the bass was so profound that I had to take out the BSC entirely. I'm tweaking the impedance correction by ear. So far , I'm getting the best mid-high frequency reproduction using a parallel combination of 6.8uF, 5.6uF, 4.3uF and a film and foil .47uF. In series with that is 13 Ohms of resistance.


Lensi
 
Lensi said:
In series with that is 13 Ohms of resistance.
Lensi

Greets!

Yikes! 3-4 ohms + BSC I can see, but 13 with a medium Qt driver?! This would push the base effective Qt up to ~0.86! More 'boom tube' than ML-TL WRT SQ if not tuned super low (like ~25Hz) and stuffed to a fare-thee-well (at least 1.0lb/ft^3).

Oh well, different strokes..........., or is this a typo?

GM
 

Lensi

Member
2005-04-29 11:43 pm
NC
GM,

I've done some more testing/tweaking. The over-abundance of bass was due to room lift. The BSC's I had tried were 1.8mH with various resistors, 6 Ohm, 5 Ohm, 4 Ohm and lastly 3.3. Still too much bass boom, but at the time I hadn't done any testing on room modes.
The impedance correction that I was using (16.7uF + 13 Ohm) was serving to remove some of the room resonance creeping up into the vocal range, giving singers a chesty resonance. The impedance correction helped remove some of that, but didn't solve the underlying problem. I have just repositioned my listening chair to solve the room mode issue, and have made a subsequent change in the impedance correction, using now 6.8uF + 13 Ohms (as well as re-installing the bsc : 1.8mH, 5 Ohm) I also changed the port tube (I have three lengths), using the longest of the three, about 4". This has smoothed out the bass response, making it more articulate with being boomy. I'll keep you posted as I move toward a permanent configuration.

Lensi
 

Lensi

Member
2005-04-29 11:43 pm
NC
hi Weidok,

I have a Cayin TA-30 (www.cayin.de) that I've modified. I removed the preamp stage and feedback circuit. I also replaced all electrolytic capacitors with Black Gates, all film caps with AudioCap Thetas. I replaced the stock bridge rectifier with a home-built bridge using Ixys soft recovery diodes.

Now regarding my previous post....


I spent most of last night zeroing in on a final BSC and impedance correction for the speakers. After repositioning the speakers and the listening chair to reduce the effects of room modes, I settled on a BSC of 1,8mH + 5 Ohms. The impedance correction circuit is now 6,8uF+4,3uF+0.47uF Film & Foil with an 11 Ohm resistor. This combination produces the most strikingly realistic vocals I've heard.

I guess the moto is: you have to accurately identify the problem before you can create a solution. The values above are quite in line with the values calculated during the design phase of this speaker. I knew something wasn't quite right when I was testing impedance correction with values of 16-17uF and resistances of 13-15 Ohms, but I just didn't think about room modes as the source of the sonic problems.

I'm frankly blown away by the sound this Fostex MLTL produces. Cavernous soundstaging, beautiful timbre and excellent dynamics.


Lensi
 
Looks good. Have you heard any higher end Fostex drivers to compare it to? I'm looking at the Fostex range because they're not excessively expensive and they're pretty damn efficient, which is important to me because I want to use a hot-rodded SI T-Amp to drive them. I've pretty much settled on the FE167E. The FE206E looks tempting but it's not quite as efficient and the upper end of the frequency response doesn't look quite as good either. Have you measured what the f3 of the finished speaker is yet?
 
JWFokker said:
I've pretty much settled on the FE167E.

The 167 is a great all-rounder, working in many simplier boxes... the xx6s tend to need back-loaded horns to get the best out of them... actually, i haven't met a Fostex i haven't liked yet. The FE108ES is my favorite so far (but i'd really like to play with the FE168ES).

dave
 

Lensi

Member
2005-04-29 11:43 pm
NC
hi JWFokker

I haven't hear any other Fostex drivers. I made the decision to build for the FE167E based on Qts and other driver specs, as well as how they measured in the mathcad simulations. I felt that the FE167 E probably offers the best compromise of extension both at the low and high end of the frequency range. Depending on port tube length, the f3 calculates to 42-46 Hz. I plan to do some actual measurements from the listening position with a friend's dB meter. I'll post that info as soon as I have it.

Lensi
 

Lensi

Member
2005-04-29 11:43 pm
NC
FYI,

Here's how Martin King's Mathcad applet predicted the bass response of my speaker with a 2" and a 3.25" port. I originally was planning a port length in the 1.75"-2.25" range because I assumed the baffle step loss would be greater than what it apparently is. In-room response leads me toward a port length in the 3"-4" range. The images are screen captures from the Mathcad applet.

Lensi

[IMGDEAD]http://home.earthlink.net/~raflynn/Boxen/small_jpgs/2%20inch%20port_sm%20.jpg[/IMGDEAD]
[IMGDEAD]http://home.earthlink.net/~raflynn/Boxen/small_jpgs/3.25%20inch%20port_sm.jpg[/IMGDEAD]
 

weidok

Member
2004-01-27 8:24 pm
dkm
kenev said:
Also consider a long break-in period for the driver, before you end up on a compensation circuit. I own also a pair of FE167E mounted on a TQWP and I found out that it took a couple of hundred hours for the driver to break-in. Now, I play them without any compensation circuit.

Hi kenev

I have a pair of FE 167 E playing in a ML TL..
Just want to try them in a TQWT ??
Do you have drawings and pictuters from your speakers