I have been studying the fostex 4inchers for awhile, i have come down to a choice between the 4 drivers.
the fostex FE103E,FE107E,FE126E,FE127E.
after looking at various implimentations, i am still undecided.
i have seen the recommended enclosures and have decided against backloaded horns because of the complexity involved in building and tuning them. i have been looking at voight pipes too.
there are various articles online discribing the implimentation of the FE103 but none have compared them to the rest of the range.
i see it from a point that the FE103E is the worst off in the group with the bumpiest response graph. it shows 10db dips at 2 critical frequencies.
the FE107E is much better in the response but its magnet is supersmall. nearly half the size of the FE103E. therefore, i doubt the ability to produce good dynamics and detail.
the FE126E seems to be the best contender but it has a extremely low Qts of 0.25 and has rather bumpy high frequency response.
the FE127E also seems to be a good choice but again, its magnet is supersmall, nearly half of the FE103E.
i was actually thinking of the FE126E in a TQWT but it seems i won`t get good bass since the Qts is low.
i am very much a bass lover listening to things like dire straits and pink floyd. what is a good driver and a suitable enclosure to go with it?
Thank you for reading this long thread!
the fostex FE103E,FE107E,FE126E,FE127E.
after looking at various implimentations, i am still undecided.
i have seen the recommended enclosures and have decided against backloaded horns because of the complexity involved in building and tuning them. i have been looking at voight pipes too.
there are various articles online discribing the implimentation of the FE103 but none have compared them to the rest of the range.
i see it from a point that the FE103E is the worst off in the group with the bumpiest response graph. it shows 10db dips at 2 critical frequencies.
the FE107E is much better in the response but its magnet is supersmall. nearly half the size of the FE103E. therefore, i doubt the ability to produce good dynamics and detail.
the FE126E seems to be the best contender but it has a extremely low Qts of 0.25 and has rather bumpy high frequency response.
the FE127E also seems to be a good choice but again, its magnet is supersmall, nearly half of the FE103E.
i was actually thinking of the FE126E in a TQWT but it seems i won`t get good bass since the Qts is low.
i am very much a bass lover listening to things like dire straits and pink floyd. what is a good driver and a suitable enclosure to go with it?
Thank you for reading this long thread!
If you're thinking of going single driver then you have to pretty much accept the fact that you'll probably have to build a sophisticated box for it to live in. If you want strong bass you need both an Fs below 35 Hz and at least a .35 Qts, and those characteristics aren't often seen in high efficiency wide range drivers. Choose what's most important to you and then decide on the driver that will give it to you, rather than deciding on a driver first and then expecting it to do something that it was not designed for. None of the drivers you mention are capable of good bass without a very good horn loaded enclosure. They would be OK used along with a sub.
I can't help you choose one of these. My experience has been with the FE103A (40 yr old alnico version of the 103), the Radio Shack 40-1197, the FE108 Sigma, and a pr of FE108Esigmas (which are just getting broken in enuff to listen too).
A pr of FE103s in a Tim Forman-style ML-TQWT is where i'd go.
dave
A pr of FE103s in a Tim Forman-style ML-TQWT is where i'd go.
dave
If you are going to cross to a sub woofer, my advice is to use a sealed box for the four-inch. I would also advise using a second order high pass filter on the four-inch and (assuming your sub woofer driver is more or less flat to 400 Hz) a fourth order low pass filter on the sub. And as long as you are going for a sub, I would suggest considering a servo or MFB subwoofer configuration.
With a crossover in the 100 Hz or so region, this generally allows you to drive the four-inch to its power max. Although using different drivers of different manufacture, I have played such systems loudly enough to be bad for your hearing.
Best of luck,
Mark
With a crossover in the 100 Hz or so region, this generally allows you to drive the four-inch to its power max. Although using different drivers of different manufacture, I have played such systems loudly enough to be bad for your hearing.
Best of luck,
Mark
MarkMcK said:
I'm confused are you suggesting crossing the four incher at 100hz? I don't think any of these drivers will really go flat to 100hz in a sealed enclosure. Wouldn't something more like 200hz be appropriate? I am still learning about all of this, but I was also under the impression that crossing over that close to the resonate frequency would cause problems (I believe these drivers have a resonate frequency somewhere around 70-80hz). I'm certainly not trying to argue with you, rather I'm merely trying to educate myself, so please set me straight on anything I've misunderstood. Also if you wanted to elaborate on the crossover you would recomend. I'm all ears.
Joe
There are a number of variables to consider in determining crossover frequency and slopes. My recomendations consider two variables. One variable is holding the four-inch to its linear excursion region. The second is the actual acoustic slope of the driver output. Properly sized (or compensated for by a split pole crossover filter), a sealed box provides an easy to work with and constant acoustic slope.
If you tune the box for lower frequencies and port or backwave horn load or any of the other tricks you can play to extend apparent low frequency performance, you produce a non constant slope. Slope starts off slow and then becomes very steep. In addition, all of the non sealed techniques complicate things in the regions where the driver acoustically crosses over to the box resonance or the backwave horn loaded output and so on.
If you are going to use the four-inch full-range in the low end, then horn loading may extend the bass response. If you are going to cross to a sub woofer anyway, then this extra bass and all the transient and acoustic center shift complications inherent in such designs get in the way of a smooth sub to four-inch crossover.
Sealed boxes roll off 12 db per octace below Fs. Add twelve more electrical and you have a 24 db per octave slope. If your woofer is not purposely mass and inductance loaded, it will probably have response several octaves above cross, and so a fourth order electrical is required to match. The fostex are capable of flat output in a sealed box to around 100 Hz. If you want to go small box and have a rise in the acoustic output at resonance, then you can split the poles of the electrical filter to compensate for this. First pole is set to a higher frequency. I talk a little more about this in my Tang Band modification series for the W4-1052SA driver.
Also, I have seen several posts saying that you have to cross a subwoofer at 80 Hz. This is not always the case. Instead, the higher the slope the higher the cut-off frequency for the sub woofer before there is any image contamination. The 80 Hz figure was set with second order slopes in mind. This was a common crossover value for subwoofer plate amps. More of these amps are becoming available with fourth order low pass filters. A good design change. Keep the sub low pass slope to 24 db per octave acoustic output and you can cross at an even higher frequency.
I have crossed as high as 300 Hz using a 36 db low pass acoustic output slope for the first 50 db down and have no problems with imaging or woofer placement.
Lastly, delays or acoustic center shifts are an unavoidable consequence of finite bandwidth. Any driver at the limit of its bandwidth will shift and any crossover filter will cause shift. The point, however, is not to try to avoid them (because you cannot), but rather to include them in an integrated design. The goal is not to avoid this or that problem at all cost, but to produce an acoustic signal that is as close as possible to a reproduction of the input signal.
Although this post is getting much too long already, I have only touched upon integrated design. Still, assuming there are not undocumented problems with the drivers or enclosures you are using, this should be enough information to get you close, to get you started.
Good designing and good building,
Mark
If you tune the box for lower frequencies and port or backwave horn load or any of the other tricks you can play to extend apparent low frequency performance, you produce a non constant slope. Slope starts off slow and then becomes very steep. In addition, all of the non sealed techniques complicate things in the regions where the driver acoustically crosses over to the box resonance or the backwave horn loaded output and so on.
If you are going to use the four-inch full-range in the low end, then horn loading may extend the bass response. If you are going to cross to a sub woofer anyway, then this extra bass and all the transient and acoustic center shift complications inherent in such designs get in the way of a smooth sub to four-inch crossover.
Sealed boxes roll off 12 db per octace below Fs. Add twelve more electrical and you have a 24 db per octave slope. If your woofer is not purposely mass and inductance loaded, it will probably have response several octaves above cross, and so a fourth order electrical is required to match. The fostex are capable of flat output in a sealed box to around 100 Hz. If you want to go small box and have a rise in the acoustic output at resonance, then you can split the poles of the electrical filter to compensate for this. First pole is set to a higher frequency. I talk a little more about this in my Tang Band modification series for the W4-1052SA driver.
Also, I have seen several posts saying that you have to cross a subwoofer at 80 Hz. This is not always the case. Instead, the higher the slope the higher the cut-off frequency for the sub woofer before there is any image contamination. The 80 Hz figure was set with second order slopes in mind. This was a common crossover value for subwoofer plate amps. More of these amps are becoming available with fourth order low pass filters. A good design change. Keep the sub low pass slope to 24 db per octave acoustic output and you can cross at an even higher frequency.
I have crossed as high as 300 Hz using a 36 db low pass acoustic output slope for the first 50 db down and have no problems with imaging or woofer placement.
Lastly, delays or acoustic center shifts are an unavoidable consequence of finite bandwidth. Any driver at the limit of its bandwidth will shift and any crossover filter will cause shift. The point, however, is not to try to avoid them (because you cannot), but rather to include them in an integrated design. The goal is not to avoid this or that problem at all cost, but to produce an acoustic signal that is as close as possible to a reproduction of the input signal.
Although this post is getting much too long already, I have only touched upon integrated design. Still, assuming there are not undocumented problems with the drivers or enclosures you are using, this should be enough information to get you close, to get you started.
Good designing and good building,
Mark
MarkMcK said:
Ok, so I understand clearly the reasoning in using a sealed enclosure for the fostex. As for the crossing over on a steep slope for the bottom end, say you're using a pro-sound woofer for the bottom end which has a smooth response up to 2kHz, then would the steep crossover only give you freedom in placement of the woofer? Or would this still be necesary even in the fostex was right next to the woofer?
As for the crossover on the fostex, a passive 12dB crossover is alright even if it is within less than one octave of the Fs of the driver? Finally I am not sure what you mean by inegrating your "acoustice center shifts" (is this the same thing as a "phase shift") into your design. Thanks so much for the explanations.
Joe
You will need to match acoustic slopes to successfully cross the woofer and four-inch. This is about acoustic summing and is necessary regardless of woofer to four-inch separation.
The Fs that is important is the system resonance. That means as mounted in the enclosure. If you tune the sealed box to say 70 Hz but are rolling in an electrical -3 at 100 or 150 or whatever, then you have made a split pole cross and you will have to match this varying slope with the woofer low pass filter.
You want to design your box so that you can use or at least account for the high pass slope it provides. What is important is the acoustic F3 and the subsequent slope. (In the case of a L/R crossover, then we are looking at F6, but otherwise the same.)
Phase is a special case term that, while perfectly reasonable in particular applications, can cause confusion when used in other situations. We, working in acoustics and in transient domains have been saddled with the concept of phase for historical reasons. I would rather not use the term phase anymore. Still, in general terms, what can be described as phase shift is the same as a shift in acoustic center or origin.
At least for a sealed enclosure, the mechanical resonance and slope are analogues of an electrical filter. That means phase shift and acoustic center are interchangeable in most cases. It also means that the phenomena you see in box simulation programs is the same phenomena you create with a filter. In other words, if the box tuning is done correctly, there are no negative consequences to using this mechanical filter as an adjunct to an electrical filter. The combined performance would be identical to a higher order electrical filter feeding an acoustically perfect transducer.
In the real world, there are all sorts of things that go wrong to complicate our work, but there is no inherent reason to fear Fs in a sealed system.
Mark
The Fs that is important is the system resonance. That means as mounted in the enclosure. If you tune the sealed box to say 70 Hz but are rolling in an electrical -3 at 100 or 150 or whatever, then you have made a split pole cross and you will have to match this varying slope with the woofer low pass filter.
You want to design your box so that you can use or at least account for the high pass slope it provides. What is important is the acoustic F3 and the subsequent slope. (In the case of a L/R crossover, then we are looking at F6, but otherwise the same.)
Phase is a special case term that, while perfectly reasonable in particular applications, can cause confusion when used in other situations. We, working in acoustics and in transient domains have been saddled with the concept of phase for historical reasons. I would rather not use the term phase anymore. Still, in general terms, what can be described as phase shift is the same as a shift in acoustic center or origin.
At least for a sealed enclosure, the mechanical resonance and slope are analogues of an electrical filter. That means phase shift and acoustic center are interchangeable in most cases. It also means that the phenomena you see in box simulation programs is the same phenomena you create with a filter. In other words, if the box tuning is done correctly, there are no negative consequences to using this mechanical filter as an adjunct to an electrical filter. The combined performance would be identical to a higher order electrical filter feeding an acoustically perfect transducer.
In the real world, there are all sorts of things that go wrong to complicate our work, but there is no inherent reason to fear Fs in a sealed system.
Mark
Mark
thanks so much for your patience in explaining all of this... I am going to build a two way speaker using a fostex 108e sigma, mated to an Emenience Delta-12lf. Obviously I've still got a lot to learn, but I think I am starting to understand the basics. I'll probably start with the fostex running open with just a inductor on the Emenience to start off with, adding an l-pad as needed. I know this goes against what you said about matching acoustic slopes, but this was suggested to me by someone as a means to fill in for the 6dB drop occuring due to step baffle step. Anyways I'll start there and then maybe try a 12dB crossover on the fostex where the sealed enclosure starts dropping off ,which looks to be -3dB at 180Hz, and then cross the Emenience 24dB an octave.... we'll see. I'm looking foward to experimenting, I just wish crossover parts were cheaper
.
thanks again
Joe
thanks so much for your patience in explaining all of this... I am going to build a two way speaker using a fostex 108e sigma, mated to an Emenience Delta-12lf. Obviously I've still got a lot to learn, but I think I am starting to understand the basics. I'll probably start with the fostex running open with just a inductor on the Emenience to start off with, adding an l-pad as needed. I know this goes against what you said about matching acoustic slopes, but this was suggested to me by someone as a means to fill in for the 6dB drop occuring due to step baffle step. Anyways I'll start there and then maybe try a 12dB crossover on the fostex where the sealed enclosure starts dropping off ,which looks to be -3dB at 180Hz, and then cross the Emenience 24dB an octave.... we'll see. I'm looking foward to experimenting, I just wish crossover parts were cheaper
.thanks again
Joe
Actually Mark while I still have your attention maybe I could get a quick bit of crossover advise....(sorry hacknet I hope you don't mind that I've completly hijacked your thread). I've got some ceramic enclosures that I made around 2L which I figure will give me around 1.5L internal volume when you figure in the driver and foam on the interior walls. Using the fostex 108e sigma this should give me a -3dB of 180Hz. Would you suggest rolling the Emenience in with a first order crossover at 180Hz and letting the fostex run open (in hopes of filling out the bottom end of the fostex which will be down due to baffle step losses), crossing the Emenience second order at 180Hz and letting the fostex run open(matching them accustically while still leaveing the fostex with nothing in the signal path), or would you suggest crossing the fostex second order at 180Hz and the eminence fourth order at 180Hz(perhaps making things easier for the fostex on the bottom end therefore cleanning up the top end). Any thoughts, certainly I plan on doing some experimenting, but I can not really afford to buy all of these crossover parts at once so I thought I would see if you had any suggestions.
Thanks in advance
Joe
Thanks in advance
Joe
I suggest you measure the Fostex performance in the box. You might not be down -3 at 180 Hz. Just speaking from experience.
Next, baffle step phenomena shelves, you drop and then flatten. In a two liter box, the Fostex response may be rising at it approaches Fs, even free field.
Where do you believe your baffle step loss will begin? If it is near the crossover frequency, then you might want to add a first order electrical to the point that the bs shelves. This will give you a third order slope below 180 (assuming you are actually F3 at 180).
Again, just based on experience, the Eminience will have a slightly rising response with frequency. This requires more filtering to achieve a comp cross. Because of cost, I never passively speaker level cross below 300 Hz. Plate amps are becoming really inexpensive. I advise doing a cost comparision before committing to passive speaker level.
If you can account for the actual acoustic performance of the drivers, either passive or active crossover, this frequency will work really well. There is great potential for this combination.
Best,
Mark
Next, baffle step phenomena shelves, you drop and then flatten. In a two liter box, the Fostex response may be rising at it approaches Fs, even free field.
Where do you believe your baffle step loss will begin? If it is near the crossover frequency, then you might want to add a first order electrical to the point that the bs shelves. This will give you a third order slope below 180 (assuming you are actually F3 at 180).
Again, just based on experience, the Eminience will have a slightly rising response with frequency. This requires more filtering to achieve a comp cross. Because of cost, I never passively speaker level cross below 300 Hz. Plate amps are becoming really inexpensive. I advise doing a cost comparision before committing to passive speaker level.
If you can account for the actual acoustic performance of the drivers, either passive or active crossover, this frequency will work really well. There is great potential for this combination.
Best,
Mark
I made a pair of speakers for my office using the FE107E. They are ML-TL's. They go down to 65 Hz, but of course you can't fill a cathedral with them. I have a Peerless 10" XLS sub that I sometimes use with them after everyone else goes home. I am quite pleased with the speakers, particularly considering the small cost. My boss has a pair of Quads in the adjacent office. He paid $1 grand for them. He fancies himself an audiophile. He told me recently that my system sounds better than his. It visibly hurt him to say it.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=36305&highlight=locust
http://www.diyaudio.com/forums/showthread.php?s=&threadid=36305&highlight=locust
MarkMcK said:
Well, I just got my ceramic enclosures out of the kiln. They are 2.3 liters and measure about 6" across the baffle which will be 3/4" oak rounded over attached with liquid nails. If I understand baffle step correctly the 6dB drop occurs where the one wave length is equal to the baffle width, so ignoring the round over it would be 2.2kHz? Not very close to the crossover.... Maybe drop everything above 2.2kHz -6dB (assuming 2.2kHz).
I have to admit before I really started this project I just figured in my ingnorance that I could get away with an inductor on the Emeniece, try out a couple different l-pads and call it a crossover. I hadn't figured for the cost of a couple of 2cnd order crossovers at 180hz...
ouch! I make my living through pottery, and I get by, but I'm not sure I can swing that right now.... the plate amp looks comparable in price. What about pushing the cross-over up higher? It would be cheaper and certainly allow for higher power handling (not that I need it), I don't know.... maybe there is no cheap way out of it.You are correct about the Emenience it is around 94dB at 200Hz and then slowly rises up to around 97db, I don't have the chart in front of me, but that is pretty close. I'll put the baffles on the ceramic enclosures this weekend and see if anybody around here is able to test them for me. I think I might know someone who can. Thanks again for your patience and knowledge, Mark.
David- I followed your locust journeys with some interest...however, I've already got my ceramic sealed enclosures, and I think I'll try and see this through assuming it doesn't end up breaking the bank.
Joe
Josephjcole said:
No that is not correct. The simple answer is BSf3=4560/(baffle width in inches), but it is really more complex than that.
http://www.t-linespeakers.org/tech/bafflestep/index.html
dave
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.