Member
Joined 2009
Paid Member
I have been thinking about a Fonken for an HT set up. Today, I realized that my HT set up will very likely incorporate a sub, which means I don't need to stretch the LF performance of the main speakers. In fact I don't really want them interfering with each other do I.
I'm new to this game so my thoughts may be off-track - here goes:
Venting allows me to get more LF with some tradeoffs. Most of the tradeoffs are ok, but I am lead to understand that a possible issue is that the cone is left 'unloaded' at low frequencies where excessive signal would drive the cone hard against it's xmas and produce distortion in the mid frequencies. This could be a scenario with an HT set up if I'm not there to supervise !
So I go to thinking that maybe an FE127e sealed box would be a better approach as it keeps the cone better loaded at LF and the HT sub fills in for the quick fall off at 100Hz or so?
I understand that one may have chosen a different driver for such an enclosure type, but the driver is already purchased.
Perhaps 100Hz is too high a cut off for the sub in that it may be too directional and not 'disappear' as effectively as the more traditional 80Hz cut off ?
I'm not sure if the resistive nature of the Fonken ports provides more effective loading at LF than a 'regular' BR box and perhaps represents a compromise that is the best of both worlds ?
I'd welcome opinions on the above ?
p.s. I'm still looking for a free excel based vented box calculation spreadsheet that shows me the frequency response of the speaker whilst allowing me to specify the box volume.
p.p.s. I try to be diligent in doing a 'search' before I start a new thread, but I find the search engine digs up too many results without a good match to my search expression, am I the only one ?
I'm new to this game so my thoughts may be off-track - here goes:
Venting allows me to get more LF with some tradeoffs. Most of the tradeoffs are ok, but I am lead to understand that a possible issue is that the cone is left 'unloaded' at low frequencies where excessive signal would drive the cone hard against it's xmas and produce distortion in the mid frequencies. This could be a scenario with an HT set up if I'm not there to supervise !
So I go to thinking that maybe an FE127e sealed box would be a better approach as it keeps the cone better loaded at LF and the HT sub fills in for the quick fall off at 100Hz or so?
I understand that one may have chosen a different driver for such an enclosure type, but the driver is already purchased.
Perhaps 100Hz is too high a cut off for the sub in that it may be too directional and not 'disappear' as effectively as the more traditional 80Hz cut off ?
I'm not sure if the resistive nature of the Fonken ports provides more effective loading at LF than a 'regular' BR box and perhaps represents a compromise that is the best of both worlds ?
I'd welcome opinions on the above ?
p.s. I'm still looking for a free excel based vented box calculation spreadsheet that shows me the frequency response of the speaker whilst allowing me to specify the box volume.
p.p.s. I try to be diligent in doing a 'search' before I start a new thread, but I find the search engine digs up too many results without a good match to my search expression, am I the only one ?
I have the same problem with the "search" function. I do my best to add words to get it narrowed down, but I still fail at it quite a bit.
Bigun said:
Well, you could certainly find no shortage of suggestions for different drivers - including Fostex (i.e. the FE167), but you could certainly do worse.
that could depend on your room, but quite possibly so - one reason why a lot of folks recommend (at least) 2 powered woofers. I've had plate amps with XO's as high as 120 or more - "sub" is a bit of a misnomer at these frequencies
I think this has been discussed in another thread, but the short answer to this one is "yes, in fact quite effective controlled loading" - based on some quickie experiments with a small FE127 design, I'd tend to say - "not much of a compromise, actually..."
Hi,
http://home20.inet.tele.dk/kou/ubdwnld.html
As long as your HT set up does high pass filtering (set to small)
for the speakers venting can be used to increase power handling
and maximum levels around the cut-off, low frequency unloading
will not matter with the hp filter. Fostex driver excursion needs
all the help it can get, Qts is usually too low anyway for sealed.
🙂/sreten.
http://home20.inet.tele.dk/kou/ubdwnld.html
As long as your HT set up does high pass filtering (set to small)
for the speakers venting can be used to increase power handling
and maximum levels around the cut-off, low frequency unloading
will not matter with the hp filter. Fostex driver excursion needs
all the help it can get, Qts is usually too low anyway for sealed.
🙂/sreten.
Member
Joined 2009
Paid Member
Thanks guys.
One thing these box calculators exclude is the affect of the 'Onken resistance' in the performance of the system. Given that the Base Reflex calculation is based on a Helmholtz oscillator it should be a simple matter of adding a resistor (R) to the equivalent circuit diagram and comparing the results with R=0 and R>0 to see what happens to the frequency response and phase information. If I was one of those people who had a spice simulator and knew how to use it we might get some quick data.... of course it's not that simple to determine the correct value of R for a given vent configuration but at least it would reveal the general behaviour ?
Call me lazy, but I haven't played with Spice for 20 years and so I'd have to start from scratch as well as find a free version for my iMac. Perhaps somebody out there would like to pick this up? OR maybe it's just wasting time when I could be experimenting - I heard that some mail hs arrived at home for me today - hopefully my Fostex drivers, so one less excuse for building something !
One thing these box calculators exclude is the affect of the 'Onken resistance' in the performance of the system. Given that the Base Reflex calculation is based on a Helmholtz oscillator it should be a simple matter of adding a resistor (R) to the equivalent circuit diagram and comparing the results with R=0 and R>0 to see what happens to the frequency response and phase information. If I was one of those people who had a spice simulator and knew how to use it we might get some quick data.... of course it's not that simple to determine the correct value of R for a given vent configuration but at least it would reveal the general behaviour ?
Call me lazy, but I haven't played with Spice for 20 years and so I'd have to start from scratch as well as find a free version for my iMac. Perhaps somebody out there would like to pick this up? OR maybe it's just wasting time when I could be experimenting - I heard that some mail hs arrived at home for me today - hopefully my Fostex drivers, so one less excuse for building something !
Gareth:
to follow up on my final point on post #3 above:
http://www.diyaudio.com/forums/showthread.php?postid=1719202#post1719202
so in other words, sometimes the math only answers part of the question, and only at specific pressure levels - the highly dynamic nature of musical signals can mean that the enclosure's tuning might only be "perfect" at a particular volume level
to follow up on my final point on post #3 above:
http://www.diyaudio.com/forums/showthread.php?postid=1719202#post1719202
so in other words, sometimes the math only answers part of the question, and only at specific pressure levels - the highly dynamic nature of musical signals can mean that the enclosure's tuning might only be "perfect" at a particular volume level
Member
Joined 2009
Paid Member
Hi Chris,
so all these graphs and plots and calculators and discussions about ripples and curves and bumps... gets us maybe only half way there
but on a practical note, I see the danger of going around in cirles here, the key things I need to take away are:
a) there's enough benefit in providing vents that I should do so
b) the vents will extend the LF enough to avoid placing more difficult constraints on the sub, will improve the power level and is probably more suitable for my chosen driver
c) the Onken style vent is the optimal method for venting in this case and I can employ some HF filtering (either via the receiver or more directly) if high LF levels cause problems
d) unless following a proven design, don't be shy to the idea of doing a bit of post-build 'tuning' on the ports
d) the music will speak louder than the maths
And having opened the box but 20 minutes ago !....
so all these graphs and plots and calculators and discussions about ripples and curves and bumps... gets us maybe only half way there
but on a practical note, I see the danger of going around in cirles here, the key things I need to take away are:
a) there's enough benefit in providing vents that I should do so
b) the vents will extend the LF enough to avoid placing more difficult constraints on the sub, will improve the power level and is probably more suitable for my chosen driver
c) the Onken style vent is the optimal method for venting in this case and I can employ some HF filtering (either via the receiver or more directly) if high LF levels cause problems
d) unless following a proven design, don't be shy to the idea of doing a bit of post-build 'tuning' on the ports
d) the music will speak louder than the maths
And having opened the box but 20 minutes ago !....
Attachments
Hi Bigun, in terms of simple, easy newbie-friendly tools, there are lots. have you tried WinISD or Unibox?
The search function works flawlessly for me -- are you choosing "Show results as posts" rather than the default of "Show results as threads"? Within a very long thread, there's also a "Search within this thread" at the bottom of the page, right-hand side.
The search function works flawlessly for me -- are you choosing "Show results as posts" rather than the default of "Show results as threads"? Within a very long thread, there's also a "Search within this thread" at the bottom of the page, right-hand side.
Member
Joined 2009
Paid Member
HI,
That's a very handy tip on the search feature, I believe that will answer to the problem.
Couldn't resist playing with the calculations. I also found something helpful at :http://micka.de/org/en/index.php#ideal
I extracted the dimensions from published data for the planet10 Fonken and mFonken which use the same drivers I have. Looking at the Fonken, with Vb=13L, port length of roughly 26cm and an effective port equiv. diameter of 7.8cm I find that the frequency response looks very close indeed to the recommended vented box frequency response for this driver using the calculator on that web site I just quoted. This makes sense, confirming that the Fonken is set up like a BR.
Then I used the data for the mFonken, Vb=4.5L, port length roughly 16cm and a port equivalent area of roughly 4.3cm. I tried varying the port dimensions to convince myself that this was the optimal in terms of flatness of response and I think it is. So again, a nicely designed box set up like a BR.
However, with the smaller box of the mFonken there are limitations and the frequency response doesn't offer a 3dB point much lower than a sealed box. I'm not convinced that the venting of this box provides much benefit in real life. I haven't read much about what people think of the sound from the mFonken. It looks a very tight box with this driver. I do read many good things about the Fonken, but with that design, the box volume really allows this thing to perform.
Here's a plot showing my attempt at modelling the mFonken (and what I learn is that for small HT system I may not want to vent after all, or if I do then the box volume should be made large enough - which I think I have space to accommodate)
That's a very handy tip on the search feature, I believe that will answer to the problem.
Couldn't resist playing with the calculations. I also found something helpful at :http://micka.de/org/en/index.php#ideal
I extracted the dimensions from published data for the planet10 Fonken and mFonken which use the same drivers I have. Looking at the Fonken, with Vb=13L, port length of roughly 26cm and an effective port equiv. diameter of 7.8cm I find that the frequency response looks very close indeed to the recommended vented box frequency response for this driver using the calculator on that web site I just quoted. This makes sense, confirming that the Fonken is set up like a BR.
Then I used the data for the mFonken, Vb=4.5L, port length roughly 16cm and a port equivalent area of roughly 4.3cm. I tried varying the port dimensions to convince myself that this was the optimal in terms of flatness of response and I think it is. So again, a nicely designed box set up like a BR.
However, with the smaller box of the mFonken there are limitations and the frequency response doesn't offer a 3dB point much lower than a sealed box. I'm not convinced that the venting of this box provides much benefit in real life. I haven't read much about what people think of the sound from the mFonken. It looks a very tight box with this driver. I do read many good things about the Fonken, but with that design, the box volume really allows this thing to perform.
Here's a plot showing my attempt at modelling the mFonken (and what I learn is that for small HT system I may not want to vent after all, or if I do then the box volume should be made large enough - which I think I have space to accommodate)
Attachments
Gareth - just for kicks, you could build a pair of the mFonkens* ( a far simpler build than the Prime), and report back on how what you hear in the real world relates to what the modeling shows in cyberspace.
You could find it's not exactly "set up like a(n average) BR". Is there any software capable of modeling with the type of highly dynamic and spectrally complex signals that we actually listen to?
* if you do, please note that the driver bracing is important, and the box is small enough that lining with damping material is probably easier if you install the back panel last.
As for not reading much about what people think of the sound from the mFonken, certainly as compared to the "Prime" - that's likely as there have been few, if any, of the small ones yet built (other than our own the 3 or 4 pairs), while at last count there are probably over a dozen pairs of the Fonkens by DIYers. Why not be the first report your impressions on the little guys?
You could find it's not exactly "set up like a(n average) BR". Is there any software capable of modeling with the type of highly dynamic and spectrally complex signals that we actually listen to?
* if you do, please note that the driver bracing is important, and the box is small enough that lining with damping material is probably easier if you install the back panel last.
As for not reading much about what people think of the sound from the mFonken, certainly as compared to the "Prime" - that's likely as there have been few, if any, of the small ones yet built (other than our own the 3 or 4 pairs), while at last count there are probably over a dozen pairs of the Fonkens by DIYers. Why not be the first report your impressions on the little guys?
Member
Joined 2009
Paid Member
Hi,
I like the idea of doing some experiments but of course in reality my HT set up does allow me to accept a much taller box (e.g. floorstander) with a larger volume than the mFonken.
I'm not that confident that my ears are very expert - actually I was thinking it would be handy to have a microphone and spectrum analyzer software for my iMac so I do something more scientific, like watching what happens if I make adjustments. But I'm getting ahead of myself with all this armchair DIY - first step is buy some wood (perhaps start with something that won't be the final box) and get something in place to listen to
The modeling really doesn't do it justice I would agree.
You mention good bracing and padding - I had a question about shallow enclosures (if I go floorstander or shallower) and the risk I read somewhere about sound reflecting off the back of box where it's so near to the cone. Do you think a shallow box requires some special considerations regarding the bracing behind the speaker and if so is it as simple as ensuring the back of the box has some kind of surface that is angled to the front of the box to direct reflections away from the cone (see attached) ?
I like the idea of doing some experiments but of course in reality my HT set up does allow me to accept a much taller box (e.g. floorstander) with a larger volume than the mFonken.
I'm not that confident that my ears are very expert - actually I was thinking it would be handy to have a microphone and spectrum analyzer software for my iMac so I do something more scientific, like watching what happens if I make adjustments. But I'm getting ahead of myself with all this armchair DIY - first step is buy some wood (perhaps start with something that won't be the final box) and get something in place to listen to
The modeling really doesn't do it justice I would agree.
You mention good bracing and padding - I had a question about shallow enclosures (if I go floorstander or shallower) and the risk I read somewhere about sound reflecting off the back of box where it's so near to the cone. Do you think a shallow box requires some special considerations regarding the bracing behind the speaker and if so is it as simple as ensuring the back of the box has some kind of surface that is angled to the front of the box to direct reflections away from the cone (see attached) ?
Attachments
Gareth:
Maybe someday I'll take a lesson on how to archive links to previous posts or photos for posting, but for now I'll need to rely on others.
FYI, there is a small CSA floorstanding member of the Fonken family that would well be worth considering for a video system. It definitely has more extension into the lower midbass area than the "milli", and with the support of powered woofer(s), I'd expect quite satisfactory results for most musical and modest HT systems, particularly with carefully selected bass management control in the latter.
http://homepage.mac.com/tlinespeakers/FAL/box-plans/fonken-tune-FE127-FS-Mk2.pdf
http://www.planet10-hifi.com/fonken.html
Hopefully it won't take long for Dave to respond by posting a couple of ( or links to ) pictures of this design.
Regarding the bracing, note on most of the drawings on the P10 website, including the in first of above links, that the bracing generally runs vertically in the longer dimension of the enclosure, and they usually include a number of holes.
In fact the 2 smallest 2 of this group are the only ones in which the bracing is not extended to contact at least 3 of the enclosure panels (in the case of the Prime, the vertical brace is full height, and couples 4 of the surfaces), with a U-shaped driver cut-out.
An idea for wide/ shallow cabinets that's been percolating in the back of my mind for several years, and I've just never got around to, is to create a "stepped cone" by stacking a series of circular plywood cutouts, with the smallest approx the diameter of the magnet, and in contact with it. It would certainly serve as a driver brace, as well as possibly help with radial dispersion of the rear wave energy to the damped side and top/bottom panels. This may have been done before, or be similar to what you envision, but I don't recall seeing any reports of such.
Maybe someday I'll take a lesson on how to archive links to previous posts or photos for posting, but for now I'll need to rely on others.
FYI, there is a small CSA floorstanding member of the Fonken family that would well be worth considering for a video system. It definitely has more extension into the lower midbass area than the "milli", and with the support of powered woofer(s), I'd expect quite satisfactory results for most musical and modest HT systems, particularly with carefully selected bass management control in the latter.
http://homepage.mac.com/tlinespeakers/FAL/box-plans/fonken-tune-FE127-FS-Mk2.pdf
http://www.planet10-hifi.com/fonken.html
Hopefully it won't take long for Dave to respond by posting a couple of ( or links to ) pictures of this design.
Regarding the bracing, note on most of the drawings on the P10 website, including the in first of above links, that the bracing generally runs vertically in the longer dimension of the enclosure, and they usually include a number of holes.
In fact the 2 smallest 2 of this group are the only ones in which the bracing is not extended to contact at least 3 of the enclosure panels (in the case of the Prime, the vertical brace is full height, and couples 4 of the surfaces), with a U-shaped driver cut-out.
An idea for wide/ shallow cabinets that's been percolating in the back of my mind for several years, and I've just never got around to, is to create a "stepped cone" by stacking a series of circular plywood cutouts, with the smallest approx the diameter of the magnet, and in contact with it. It would certainly serve as a driver brace, as well as possibly help with radial dispersion of the rear wave energy to the damped side and top/bottom panels. This may have been done before, or be similar to what you envision, but I don't recall seeing any reports of such.
Member
Joined 2009
Paid Member
forgive my ignorance, but what is 'CSA' ?
my plan is for typical 5.1 set up where each satellite is a single driver and a sub rather than a woofer, meaning I want a single separate sub which operates at low enough freq. as to not be directional. The HT rear surrounds have to mount on an a ceiling bracket due to the configuration of my basement - the viewing area doesn't neatly end at a rear wall and so I'll probably hang them horizontal to avoid creating lumps on my forehead
So perhaps all the speakers will be the same design, the rears and centre mounted horizontally, the left/right mounted vertically. All 3 front speakers will be mounted on the wall, or just off it via suitable brackets.
Hence the speakers should have an LF roll-off that will work well with a sub. I haven't researched this yet but I gather that this means the cross-over frequency between the sub and the other speakers will be somewhere around 100Hz (i.e. both down at -3dB) and the sub designed so that it's flat from 80Hz down as far as can be reasonably achieved (subject to the room effects) ??
If this is right (?) and the calculated freq. response I posted above for the mFonken is also correct (?) then they are actually good for the job. But I think I can afford to go larger (taller for vertical orientation) and I believe something around 8Lites would be close to ideal. What I hope to have time for this weekend is to shamelessly take a p10 design and stretch/squeeze it to suit my size constraints, run a calculation to determine the port size and ... go buy a router!
I will employ bracing as per p10 suggestions, my little diagram was just to look at the idea of something specially shaped near the back of the driver.
my plan is for typical 5.1 set up where each satellite is a single driver and a sub rather than a woofer, meaning I want a single separate sub which operates at low enough freq. as to not be directional. The HT rear surrounds have to mount on an a ceiling bracket due to the configuration of my basement - the viewing area doesn't neatly end at a rear wall and so I'll probably hang them horizontal to avoid creating lumps on my forehead
So perhaps all the speakers will be the same design, the rears and centre mounted horizontally, the left/right mounted vertically. All 3 front speakers will be mounted on the wall, or just off it via suitable brackets.
Hence the speakers should have an LF roll-off that will work well with a sub. I haven't researched this yet but I gather that this means the cross-over frequency between the sub and the other speakers will be somewhere around 100Hz (i.e. both down at -3dB) and the sub designed so that it's flat from 80Hz down as far as can be reasonably achieved (subject to the room effects) ??
If this is right (?) and the calculated freq. response I posted above for the mFonken is also correct (?) then they are actually good for the job. But I think I can afford to go larger (taller for vertical orientation) and I believe something around 8Lites would be close to ideal. What I hope to have time for this weekend is to shamelessly take a p10 design and stretch/squeeze it to suit my size constraints, run a calculation to determine the port size and ... go buy a router!
I will employ bracing as per p10 suggestions, my little diagram was just to look at the idea of something specially shaped near the back of the driver.
in this context, CSA = Cross Sectional Area
There's certainly no shame in borrowing some of your design inspirations, we all do it from time to time, but unless you want to build your learning curve into this project (and the attendant multiple iterations of actual enclosure builds), you might want to check with Dave off-line regarding revised dimensions for vent tuning.
He's got numerous more plans already mapped out that aren't posted on the site, and might already have one that would work for your application.
FWIW, some rough calcs on net enclosure volumes are approx:
Fonken Prime / GR Fonken & Floorstander- approx 13 litres
mFonken ("milli") 4.5 litres
so the 127 can definitely work in a range of volumes with appropriate tuning.
There's certainly no shame in borrowing some of your design inspirations, we all do it from time to time, but unless you want to build your learning curve into this project (and the attendant multiple iterations of actual enclosure builds), you might want to check with Dave off-line regarding revised dimensions for vent tuning.
He's got numerous more plans already mapped out that aren't posted on the site, and might already have one that would work for your application.
FWIW, some rough calcs on net enclosure volumes are approx:
Fonken Prime / GR Fonken & Floorstander- approx 13 litres
mFonken ("milli") 4.5 litres
so the 127 can definitely work in a range of volumes with appropriate tuning.
- Status
- Not open for further replies.