I just built a pair of D-37 cabinets from the Fostex design book. I have not yet glued them shut, just used lots of vises to attach the side wall. I have no stuffing in them at this moment and they have a SERIOUS boominess problem. I have listened to other speakers including BLH in that room without a problem (though never with a FE168EZ; this is my first build with that driver). Can I hope to fix some of this with stuffing? For that matter, where in the design ought one to play with stuffing to see whether one likes it? Below is a scan from the book showing the basic design.
Any suggestions on what to experiment with would be greatly appreciated. (The drivers have about 100 hours on them at this point...)
peter
(image removed by moderator)
Any suggestions on what to experiment with would be greatly appreciated. (The drivers have about 100 hours on them at this point...)
peter
(image removed by moderator)
Last edited by a moderator:
I built the D-58Es with the FE208e Sigma drivers. The D58 is just a bigger version of the D37. I found it very boomy when I first started, but now 15 months on, it is nearly gone. A couple of observations that may help you:
1. As the driver is breaking in, load up the compression chamber with fill (Dacron material, foam or etc) and just leave it. Eventually you will want to pull all or most of this out. But for break in, it seems to soften the boom.
2. Ensure your components are suited to your new speakers. What I mean is that some components (powerful ss amps, bright cd players, cheap cables) do not generally go well with your new speakers. You dont need to junk them, just consider what will work best. For example, I found some nice chord speaker cable works better than Cat 5.
3. I tried all manner of electrical compensation (BSC, impedence network, series resistors) but this made the boom worse. During break in, avoid these items. I do have some mild compensation now, but it would not have worked on day 1.
4. Dont load up the horn mouth with fill. This will tame the bass, but ultimately give you a muffled frequency range that you will not be happy with long term
5. Ensure you seal all gaps, including around the speaker mounting baffle. A small leak of air can give the response you are dealing with.
Ultimately, a lot of my problem was solved by 'time'. Dont waste energy, efforst and expense on solutions yet, as the character of your speaker will change. Fiddling will come later.
good luck.
1. As the driver is breaking in, load up the compression chamber with fill (Dacron material, foam or etc) and just leave it. Eventually you will want to pull all or most of this out. But for break in, it seems to soften the boom.
2. Ensure your components are suited to your new speakers. What I mean is that some components (powerful ss amps, bright cd players, cheap cables) do not generally go well with your new speakers. You dont need to junk them, just consider what will work best. For example, I found some nice chord speaker cable works better than Cat 5.
3. I tried all manner of electrical compensation (BSC, impedence network, series resistors) but this made the boom worse. During break in, avoid these items. I do have some mild compensation now, but it would not have worked on day 1.
4. Dont load up the horn mouth with fill. This will tame the bass, but ultimately give you a muffled frequency range that you will not be happy with long term
5. Ensure you seal all gaps, including around the speaker mounting baffle. A small leak of air can give the response you are dealing with.
Ultimately, a lot of my problem was solved by 'time'. Dont waste energy, efforst and expense on solutions yet, as the character of your speaker will change. Fiddling will come later.
good luck.
FWIW, the D37 was not actually designed for the current FE168ESigma, but the driver it superceeded, the FE168Sigma, which was a rather different unit & will need a bit of fiddling to make work. A few thoughts
-Try getting a bit [more?] damping into the low-pass chamber (technically, BLHs don't have a CC other than the room they are sitting in) and a some in the throat. A little goes a long way, so go easy.
-Response / behaviour does largely depend on the rest of your system, particularly the amplifier & to an extent, the electrical properties of the wire. You don't say what your amp is, so about the best I can suggest here is to make sure you've got wire of sufficiently heavy guage to keep voltage-drop to a minimum & not in itself affect the damping applied to the driver.
BTW, being a copyrighted design, technically, we're not allowed to post plans / links to them...
-Try getting a bit [more?] damping into the low-pass chamber (technically, BLHs don't have a CC other than the room they are sitting in) and a some in the throat. A little goes a long way, so go easy.
-Response / behaviour does largely depend on the rest of your system, particularly the amplifier & to an extent, the electrical properties of the wire. You don't say what your amp is, so about the best I can suggest here is to make sure you've got wire of sufficiently heavy guage to keep voltage-drop to a minimum & not in itself affect the damping applied to the driver.
BTW, being a copyrighted design, technically, we're not allowed to post plans / links to them...
I will try that.
I am using rather thick wire and am currently driving it from a revC (Mauro Penasa design as realized by TwistedPair) amp. I have others I could try (Icepower, Onyx tube amp).
I don't mean to violate rules of this board and will be happy to edit it out if I do. More generally though my use falls under the fair use right to quote from a publication in the context of a discussion (note that the cutting details are all missing; there is a whole second page with important building details...)
Peter
Thank you Adam. This seems to be echoing what Scottmoose is saying.
More generally I was looking for examples of stuffing in such a design but could find little (the Nagaoka book really doesn't seem to talk about this at all). Though I found examples where some lining is used towards the mouth end of the horn. You seem to mention that it might help but should be done judiciously as I may find that after some time I'd rather get rid of it.
Basically I want to make sure I do everything that can only be done before I glue this beast shut before that point. Or conversely, anything I do now which has a good chance of wanting to be undone can be undone...
What are the general considerations one should keep in mind when it comes to damping a BLH? There is the chamber (the volume is not very large in this design) and then maybe(?) the wide end of the horn. Anything else? (I am trying to understand the general principles here...)
peter
>>> FWIW, the D37 was not actually designed for the current FE168ESigma, but the driver it superceeded, the FE168Sigma, which was a rather different unit...
I have a pair of 168S (older style with wizzer) in boxes waiting for a home. Anybody want them?
Godzilla
I have a pair of 168S (older style with wizzer) in boxes waiting for a home. Anybody want them?
Godzilla
Another online ref to D-37
More info can be found here: http://homepage3.nifty.com/spida/eng-page9-01-16.htm
Now... how do I edit my original message to take out the image? I can't seem to find the button. I thought I was supposed to be able to edit my own messages...
peter
More info can be found here: http://homepage3.nifty.com/spida/eng-page9-01-16.htm
Now... how do I edit my original message to take out the image? I can't seem to find the button. I thought I was supposed to be able to edit my own messages...
peter
168SS 168Zigma 168EZigma...
So the original D-37 design called for 168SS. I can't seem to find much data on that driver except for this: http://www.diyloudspeakers.jp/5000html/fostex/16/fe168ss.html where you can see a scan of the data sheet. It only gives a Q0=0.23 (what type of Q is this? I am only familiar with Qes/Qts/Qms). For the 168Zigma sheet (http://www.fostexinternational.com/docs/speaker_comp/pdf/fe168sigma.pdf) gives Qes=0.4 Qts=0.37 and Qms=4.53. Finally the data sheet for 168EZigma (http://www.madisound.com/catalog/PDF/fostexdrivers/fe168ez.pdf) gives Qes=0.27 Qts=0.26 and Qms=5.328. Certainly a big difference between Zigma and EZigma. But what about the original SS?
It seems like my first order of business is to understand whether the EZigma is suitable at all. I used it because I had found a compatibility sheet at Fostex (can't find this anymore now) that listed the EZigma as a replacement for the SS. That may have been a gross oversimplification.
peter
So the original D-37 design called for 168SS. I can't seem to find much data on that driver except for this: http://www.diyloudspeakers.jp/5000html/fostex/16/fe168ss.html where you can see a scan of the data sheet. It only gives a Q0=0.23 (what type of Q is this? I am only familiar with Qes/Qts/Qms). For the 168Zigma sheet (http://www.fostexinternational.com/docs/speaker_comp/pdf/fe168sigma.pdf) gives Qes=0.4 Qts=0.37 and Qms=4.53. Finally the data sheet for 168EZigma (http://www.madisound.com/catalog/PDF/fostexdrivers/fe168ez.pdf) gives Qes=0.27 Qts=0.26 and Qms=5.328. Certainly a big difference between Zigma and EZigma. But what about the original SS?
It seems like my first order of business is to understand whether the EZigma is suitable at all. I used it because I had found a compatibility sheet at Fostex (can't find this anymore now) that listed the EZigma as a replacement for the SS. That may have been a gross oversimplification.
peter
It is not compatible. The SS is "unobtainium" now unless you score a pair on Japans ebay chances of which are slim to none. The closest thing of late to the SS is the FE166ES-R which is almost in the same "unobtainium" category.
A few suggestions if I may.
Seal the cabinet up .... properly. It is not going to work with clamps alone holding on one side.
If you need to adjust the size of the filter chamber you can only make it smaller and you can do that by adding material through the speaker cutout.
Give the drivers time to "bed" in, kind of like rings on a piston. It takes a lot of time, I have a pair of FE-208 Sigmas', the originals and they took a year at least to settle down and to my ears stopped changing after maybe a year and a half.
Play with stuffing now but be aware that you will most likely progressively keep removing same with time until you are left with very little or perhaps none at all.
I have not heard the D-37 but it is reportedly very good .... with the correct driver.
A couple of other things that I thought of overnight.
1. Scottmoose is right. The cabinet was not strictly designed for that driver, and in my case, the D-58ES was designed for the 208SS not 208E Sigma. The SS had a double magnet, so Nagaoka made the compression chamber bigger for it to fit. I think he died about 5 years ago just as these newer drivers (including yours) came on the market, so whilst they say the driver will work, it may not be the optimal cabinet design for it.
2. My sense after playing with these speakers for over a year is the compression chamber is too big. I guess that would make sense based on point 1. I have loaded mine up with 8 tennis balls just for fun!. I have also tried rocks. I cant hear a big difference so I wouldnt recommend too much time is wasted here.
3. I am also inclined to believe the design has an inherant sensitivity with a frequency somewhere around fs. A good tip I recently stumbled on is to use a 'swamping resistor'. This is simply a resistor across the speaker terminals to reduce the impedence rise at fs. It seems to work for Lowthers, and they suggest 50 ohms. I gave it a go, and it works for me!! In fact, it was far superior to a zobel which did 'boom'. More details on the Lowther America? website.
4. This is the second Nagaoka design I have built (I built the factory recommended enclosure for the 206e (which is basically a Nagaoka knock off). The sound had similar characteristics with a low boom like the one you describe, but not as prominant. So maybe this is inherant of this design?
I stress again that given time, I think you will be happy with your choice. I have just entered that zone now, 15 months on, and find my speakers extremely musical and rewarding. Persistance pays!
1. Scottmoose is right. The cabinet was not strictly designed for that driver, and in my case, the D-58ES was designed for the 208SS not 208E Sigma. The SS had a double magnet, so Nagaoka made the compression chamber bigger for it to fit. I think he died about 5 years ago just as these newer drivers (including yours) came on the market, so whilst they say the driver will work, it may not be the optimal cabinet design for it.
2. My sense after playing with these speakers for over a year is the compression chamber is too big. I guess that would make sense based on point 1. I have loaded mine up with 8 tennis balls just for fun!. I have also tried rocks. I cant hear a big difference so I wouldnt recommend too much time is wasted here.
3. I am also inclined to believe the design has an inherant sensitivity with a frequency somewhere around fs. A good tip I recently stumbled on is to use a 'swamping resistor'. This is simply a resistor across the speaker terminals to reduce the impedence rise at fs. It seems to work for Lowthers, and they suggest 50 ohms. I gave it a go, and it works for me!! In fact, it was far superior to a zobel which did 'boom'. More details on the Lowther America? website.
4. This is the second Nagaoka design I have built (I built the factory recommended enclosure for the 206e (which is basically a Nagaoka knock off). The sound had similar characteristics with a low boom like the one you describe, but not as prominant. So maybe this is inherant of this design?
I stress again that given time, I think you will be happy with your choice. I have just entered that zone now, 15 months on, and find my speakers extremely musical and rewarding. Persistance pays!
Good call.
No, the sound you describe isn't inherent to horns constructed from an expanding series of constant CSA manifolds; it's usually a case of a mismatch somewhere in the driver / cab. design or insufficient damping.
No, the sound you describe isn't inherent to horns constructed from an expanding series of constant CSA manifolds; it's usually a case of a mismatch somewhere in the driver / cab. design or insufficient damping.
- Status
- Not open for further replies.