I'm looking to build some nearfield monitors using a pair of FE103E's (possibly with a pair of tweeters as well with an active crossover in a bi-amped configuration), and it seems like the best route to go for that would be a bass reflex enclosure.
I've come up with an interesting enclosure design but need some help figuring out the best internal volume and port tuning. You can see the concept sketches I made in the attached photo.
From my drawings and measurements I came up with the following numbers (I overestimate a bit so I end up with slightly larger volumes):
Approx. FE103E volume=14.518cu.in (0.238L)
Approx. Tweeter volume=12.622cu.in (0.207L)
Port volume=25.243cu.in (0.414L) *The tube I have has an outside diameter of 2.75in and internal diameter of 2.5in and should be about 5in long for a tuning frequency of about 95Hz - this will obviously change if the tuning frequency is changed, I picked 95Hz because that's what the Fostex recommended bass reflex enclosure is tuned to.
Net area of a horizontal cross section of the enclosure=60.53sq.in (368.980sq.cm)
(for the English unit folks) Approx. Net Internal volume (in cu.in)=(60.53*internal height in inches)-(14.518+12.622+25.243)=(60.53*height)-52.383
(for the SI unit folks) Approx. Net Internal volume (in L)=(390.515*internal height in cm)/1000-(0.238+0.207+0.414)=(390.515*height)-0.859
The max. height the enclosure can be is 12in internal, so that ends up being a max. net internal volume of 637.977cu.in (11.044L). The min. height the enclosure can be is about 8.5in internal, so that means a min. net internal volume of 462.122cu.in (7.572L)
Can anyone help me come up with a good net internal volume and tuning frequency? Anyone with Martin's worksheets willing to model this little guy?
I've come up with an interesting enclosure design but need some help figuring out the best internal volume and port tuning. You can see the concept sketches I made in the attached photo.
From my drawings and measurements I came up with the following numbers (I overestimate a bit so I end up with slightly larger volumes):
Approx. FE103E volume=14.518cu.in (0.238L)
Approx. Tweeter volume=12.622cu.in (0.207L)
Port volume=25.243cu.in (0.414L) *The tube I have has an outside diameter of 2.75in and internal diameter of 2.5in and should be about 5in long for a tuning frequency of about 95Hz - this will obviously change if the tuning frequency is changed, I picked 95Hz because that's what the Fostex recommended bass reflex enclosure is tuned to.
Net area of a horizontal cross section of the enclosure=60.53sq.in (368.980sq.cm)
(for the English unit folks) Approx. Net Internal volume (in cu.in)=(60.53*internal height in inches)-(14.518+12.622+25.243)=(60.53*height)-52.383
(for the SI unit folks) Approx. Net Internal volume (in L)=(390.515*internal height in cm)/1000-(0.238+0.207+0.414)=(390.515*height)-0.859
The max. height the enclosure can be is 12in internal, so that ends up being a max. net internal volume of 637.977cu.in (11.044L). The min. height the enclosure can be is about 8.5in internal, so that means a min. net internal volume of 462.122cu.in (7.572L)
Can anyone help me come up with a good net internal volume and tuning frequency? Anyone with Martin's worksheets willing to model this little guy?
Attachments
A couple things regarding the construction of this enclosure. The idea is to cut a top and bottom out of 3/4in MDF and drill 1/2in holes through both panels so that 1/2in dowel rods can be inserted to link both panels and to provide the framing so that I can wrap thin sheets of wood around to make a nice curvy enclosure.
As far as that goes, I was wondering if that sort of construction along with some fiber fill in between the dowel rods would basically nuke any sort of standing waves inside the enclosure and how that construction would have an effect on the internal volume perceived by the FE103E?
I also wanted to place the port on the bottom, but then I'd have to raise the enclosure up off the stand/surface by at least a couple inches for it to function properly, right? I was even thinking of keeping three of the dowels (two closest to the front and the far rear one) longer to use them as legs to lift the enclosure up a few inches or so. Any reason why I shouldn't/couldn't have the port firing upward instead (besides dust or objects getting in and that can be remedied using a little piece of acoustically transparent cloth)?
Also, will the baffle diffraction step be calculated assuming the enclosure is 9in wide even though it has the 2in radius on both sides?
As far as that goes, I was wondering if that sort of construction along with some fiber fill in between the dowel rods would basically nuke any sort of standing waves inside the enclosure and how that construction would have an effect on the internal volume perceived by the FE103E?
I also wanted to place the port on the bottom, but then I'd have to raise the enclosure up off the stand/surface by at least a couple inches for it to function properly, right? I was even thinking of keeping three of the dowels (two closest to the front and the far rear one) longer to use them as legs to lift the enclosure up a few inches or so. Any reason why I shouldn't/couldn't have the port firing upward instead (besides dust or objects getting in and that can be remedied using a little piece of acoustically transparent cloth)?
Also, will the baffle diffraction step be calculated assuming the enclosure is 9in wide even though it has the 2in radius on both sides?
Hi,
if your going with a tweeter then using a fullrange driver with next
to no excursion capability as the bass/mid unit is not a good idea.
The no excursion capability is a consequence of the extended
response and relatively (for its size) high efficiency.
For active speakers the bass drivers excursion limited power
handling and the amplifiers power rating should be matched.
(To a degree, bass power handling should be ~ 20% to 30%,
depending on how much extra dynamics in the midrange)
Download WinISDPro and play with it.
Typically nearfield monitors do not have baffle step compensation.
/sreten.
if your going with a tweeter then using a fullrange driver with next
to no excursion capability as the bass/mid unit is not a good idea.
The no excursion capability is a consequence of the extended
response and relatively (for its size) high efficiency.
For active speakers the bass drivers excursion limited power
handling and the amplifiers power rating should be matched.
(To a degree, bass power handling should be ~ 20% to 30%,
depending on how much extra dynamics in the midrange)
Download WinISDPro and play with it.
Typically nearfield monitors do not have baffle step compensation.
/sreten.
Thanks for the reply sreten!
The wide range of the FE103E will allow me to experiment with a much wider crossover frequency range to find what sounds good. Also, the FE103E has the exact same sensitivity as the tweeter I want to use and the pair will also be very closely time aligned if I flush mount the tweeter on the baffle. Lastly, I really like how lifelike the FE103E makes most of my favorite recordings sound. I know the FE103E will not have much bass capability but I can always plug up the port and use an active sub. Plus, I don't listen very loud - I'd say it's around 85dB at 1m most of the time and I know the FE103Es can go a decent bit louder than that!
I was going to check out WinISDPro but got some help with the box volume and tuning from a few people over at the full range forum.
Thanks again sreten.
Most would agree with you on that point. But I still want to use the FE103E for a few reasons. The first is that they were given to me as a present and I'd like to give them a nice homesreten said:
The wide range of the FE103E will allow me to experiment with a much wider crossover frequency range to find what sounds good. Also, the FE103E has the exact same sensitivity as the tweeter I want to use and the pair will also be very closely time aligned if I flush mount the tweeter on the baffle. Lastly, I really like how lifelike the FE103E makes most of my favorite recordings sound. I know the FE103E will not have much bass capability but I can always plug up the port and use an active sub. Plus, I don't listen very loud - I'd say it's around 85dB at 1m most of the time and I know the FE103Es can go a decent bit louder than that!Coincidentally, I'll be using ~8W mono class d amps to power the FE103Es and the tweeters. They should be a fairly good match for the drivers.sreten said:
I was going to check out WinISDPro but got some help with the box volume and tuning from a few people over at the full range forum.
Well, I learned my one new thing for today.sreten said:
Thanks again sreten.
Hello BWRX,
My advice is to follow the fostex recommended bass reflex enclosures that you'll find on their web site.
As for using a tweeter, I'm considering it to match a pair of FE103 Sigma.
What I've realised is that the FE103 don't output much above 14khz, so adding a tweeter should help (although highs are not missing). In the mean time the sub is a much more urgent addition.
Anyway, the upper cut off frequency should be between 10k and 14khz. Since you'll be using a PC, active filtering is a serious option and I strongly encourage to cut the fostex at 70hz.
And what's the tweeter model you plan to use?
Bye
My advice is to follow the fostex recommended bass reflex enclosures that you'll find on their web site.
As for using a tweeter, I'm considering it to match a pair of FE103 Sigma.
What I've realised is that the FE103 don't output much above 14khz, so adding a tweeter should help (although highs are not missing). In the mean time the sub is a much more urgent addition.
Anyway, the upper cut off frequency should be between 10k and 14khz. Since you'll be using a PC, active filtering is a serious option and I strongly encourage to cut the fostex at 70hz.
And what's the tweeter model you plan to use?
Bye
Hi kepa.
After doing a little modeling with some of the programs and applets out there it seems that the Fostex recommended enclosure is a little on the small side and tuned a bit higher than it can be. An internal volume closer to 12L with a port tuned to about 90Hz would seem to be more of an optimal alignment.
A tweeter certainly isn't necessary but it would be interesting to experiment with.
The tweeter has a resonant frequency of 950Hz and a reasonably flat (+/-2dB or so) frequency response from 1.5kHz on up. The tweeter is the North D25-06S and you can find it here. They're hard to beat for the price!
Also, I won't using a PC to do the active filtering. I'll be making my own line level active crossover similar to the Linkwitz-Riley one on the ESP website.
After doing a little modeling with some of the programs and applets out there it seems that the Fostex recommended enclosure is a little on the small side and tuned a bit higher than it can be. An internal volume closer to 12L with a port tuned to about 90Hz would seem to be more of an optimal alignment.
A tweeter certainly isn't necessary but it would be interesting to experiment with.
The tweeter has a resonant frequency of 950Hz and a reasonably flat (+/-2dB or so) frequency response from 1.5kHz on up. The tweeter is the North D25-06S and you can find it here. They're hard to beat for the price!
Also, I won't using a PC to do the active filtering. I'll be making my own line level active crossover similar to the Linkwitz-Riley one on the ESP website.
Hi Tony. Your site was one of the first I came across when I started searching For the parameters I measured from my FE103s, your desing is a little on the small side coming in at around 5.5L or so. Also, how does the port tuning work with a design like yours? When you add the wooden strips in the vent doesn't that mess with the tuning? Instead of being two larger vents with a lower tuning haven't you separated them into four (or six as shown in the photos) smaller vents with a higher tuning?
For the parameters I measured from my FE103s, your desing is a little on the small side coming in at around 5.5L or so. Also, how does the port tuning work with a design like yours? When you add the wooden strips in the vent doesn't that mess with the tuning? Instead of being two larger vents with a lower tuning haven't you separated them into four (or six as shown in the photos) smaller vents with a higher tuning?
sreten said:For a number of vents they each see a smaller box volume.
Net result is the same tuning as one large vent the same length.
Ah that's the key! Thanks, that makes sense now.
Are the old Fostex FE103 drivers basically the same as the newer FE103E's? If they are I don't see how tuning such a small enclosure to 82Hz can produce good results with the 103's at all. The -3dB point will be pretty high (more than 100Hz) with a -6dB point of about 80Hz below which it falls off sharply. A larger box tuned a bit higher will lower the -3dB point significantly and still have similar cone excursion.
Great minds think alike, eh?
It's certainly serendipitous, although mine is not quite the same as yours as it will be a closed box driven by an amplifier with significant output resistance. I modelled it using WinISD (very good software). I wanted to make my box out of a composite of 3mm plywood skins with a 1" balsa core but couldn't work out how to cut the balsa blocks accurately, so I'm having to go the wasteful route of cutting loo seats of MDF and stacking them..
If you plug up the port on my design you would basically have the same enclosure as your design - although our construction methods would definitely be different. So you'll be using a tube amp to drive these then(high output impedance)? I plan to use some small class d amps which are in the works as well. It will be interesting to see what we think of these guys when we get them built.
That's not too bad, although I think it would be better to get rid of that little droop in the response by tuning it a bit higher. The T/S parameters I measured are a bit different from the ones quoted by Fostex:
Fs 104.379997
Qms 3.585399
Qes 0.561606
Qts 0.485550
I'll eventaully measure Vas whenever I get around to building a test enclosure. I've been using the quoted 6.9L in the calculations.
Fs 104.379997
Qms 3.585399
Qes 0.561606
Qts 0.485550
I'll eventaully measure Vas whenever I get around to building a test enclosure. I've been using the quoted 6.9L in the calculations.
FE103E
Hi,
Your measured parameters look like they are for a driver that has not been broken in. Fostex drivers are well know for needing up to 300 hours to break in. I would break in the driver and then re-measure.
The FE103E has very extended frequency response.
http://www.fostexinternational.com/docs/speaker_comp/pdf/fe103erev2.pdf
Most people find that a tweeter is not required.
Good luck,
Gio.
Hi,
Your measured parameters look like they are for a driver that has not been broken in. Fostex drivers are well know for needing up to 300 hours to break in. I would break in the driver and then re-measure.
The FE103E has very extended frequency response.
http://www.fostexinternational.com/docs/speaker_comp/pdf/fe103erev2.pdf
Most people find that a tweeter is not required.
Good luck,
Gio.
Hi Gio. You are quite correct in that I measured them right out of the box. They definitely have more than 300 hours of playing time by now though. I'll recheck their parameters this weekend by running them through fuzzmeasure again to see how much they've changed. Hopefully I'll have enough energy after work on Saturday to build a small box to measure their Vas as well.
Hi,
Just a few things I like to point out :
Ithink eveyone who has posted has said or implied adding a tweeter
is not a good idea, it turns a good full range driver into a poor bass/mid driver.
You will lose the transient and phase coherency which is the
main advantage of using a full range driver.
It would be fairly disheartening to build your box with a cutout
for the tweeter and then realise you'd rather have not done that.
Using TGs filter seems to me to be a very good idea.
AFAIK the basic t-amps have a fairly high bass roll-off, ~ 70Hz,
in this context this is a very good thing. What is does mean
though is your bass alignment is 5th order. (And much better
as its active than the original Wharfdale Diamond, which used
a series capacitor for a small reflex box.)
In winISD Pro Alpha you can model this by using the EQ/Filter option.
You should be able to make the box even bigger and extend
the bass a little more with a 5th order alignment. Certainly the
power handling below port frequency will be much improved.
/sreten.
Just a few things I like to point out :
Ithink eveyone who has posted has said or implied adding a tweeter
is not a good idea, it turns a good full range driver into a poor bass/mid driver.
You will lose the transient and phase coherency which is the
main advantage of using a full range driver.
It would be fairly disheartening to build your box with a cutout
for the tweeter and then realise you'd rather have not done that.
Using TGs filter seems to me to be a very good idea.
AFAIK the basic t-amps have a fairly high bass roll-off, ~ 70Hz,
in this context this is a very good thing. What is does mean
though is your bass alignment is 5th order. (And much better
as its active than the original Wharfdale Diamond, which used
a series capacitor for a small reflex box.)
In winISD Pro Alpha you can model this by using the EQ/Filter option.
You should be able to make the box even bigger and extend
the bass a little more with a 5th order alignment. Certainly the
power handling below port frequency will be much improved.
/sreten.- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.