I'm tinkering with ideas for a subwoofer to compliment my tangband computer speakers. Parts Express has a JBL/Vifa buyout 4.5" woofer for $7 that I'm looking at using in a ported isobarac enclosure.
My original plan was to use the 5.25" Dayton drivers in an isobarac, but I like how this Vifa models, and it costs half as much.
I realize that I'm not going to get room shaking bass out of these tiny drivers, but I'm not trying to. It's just a computer system after all. I modeled this driver in WinISD, and with a 10 watt input, it shows an output SPL of approximately 92dB in the frequency band that I'll be using it. Unfortunately, WinISD (at least my version) doesn't show cone excursion, so I can't tell if this output is actually possible.
Here's a WinISD plot with a 10 watt input. I've got the crosshairs at the -3dB point relative to the output below 100hz:
Is there a simple way to calculate cone excursion given t/s and enclosure parameters at a specific frequency?
My original plan was to use the 5.25" Dayton drivers in an isobarac, but I like how this Vifa models, and it costs half as much.
I realize that I'm not going to get room shaking bass out of these tiny drivers, but I'm not trying to. It's just a computer system after all. I modeled this driver in WinISD, and with a 10 watt input, it shows an output SPL of approximately 92dB in the frequency band that I'll be using it. Unfortunately, WinISD (at least my version) doesn't show cone excursion, so I can't tell if this output is actually possible.
Here's a WinISD plot with a 10 watt input. I've got the crosshairs at the -3dB point relative to the output below 100hz:
An externally hosted image should be here but it was not working when we last tested it.
Is there a simple way to calculate cone excursion given t/s and enclosure parameters at a specific frequency?
Unibox will estimate this if you have enough of the parameters to enter. You can find it at www.danbbs.dk/~ko/ubdwnld.htm
Jay
Jay
Below is the Parts Express 299-495, made by Vifa. On sale for $6. I guess that is the one you mean.
It is in an 11 liter ported enclosure, ported at 45 Hz. It is run at 6.5 volts, (10 watts for a 4 ohm speaker). From the curve you posted, I assumed that is pretty close to what you are using.
Parts Express said the outside diameter of the basket was 5", so I assumed the cone was 4" exactly. I think I am probably pretty close.
The green line is the output, the purple line is the cone excursion. As you can see, the Vifa's cone excursion of +/- 3.5mm can indeed put out the bass you are looking for, at least down to 40 Hz or so.
By the way, if anyone else wants to try this, the BL is probably 6.38, according to Bullock and White's BoxModel. I estimated the Sd as .00785 sq m, or 78.5 sq cm.
I used BoxModel to find the Bl factor, then used Subwoofer Simulator, a program written by our won F4ier, to illustrate the cone excursion.
http://www.geocities.com/f4ier/speaker.htm
It is in an 11 liter ported enclosure, ported at 45 Hz. It is run at 6.5 volts, (10 watts for a 4 ohm speaker). From the curve you posted, I assumed that is pretty close to what you are using.
Parts Express said the outside diameter of the basket was 5", so I assumed the cone was 4" exactly. I think I am probably pretty close.
The green line is the output, the purple line is the cone excursion. As you can see, the Vifa's cone excursion of +/- 3.5mm can indeed put out the bass you are looking for, at least down to 40 Hz or so.
By the way, if anyone else wants to try this, the BL is probably 6.38, according to Bullock and White's BoxModel. I estimated the Sd as .00785 sq m, or 78.5 sq cm.
I used BoxModel to find the Bl factor, then used Subwoofer Simulator, a program written by our won F4ier, to illustrate the cone excursion.
http://www.geocities.com/f4ier/speaker.htm
Attachments
I didn't do isobaric, just a single one to get the idea.
This is for 10 watts input, (actually 10.5 watts at 6.5 volts input).
Instead of estimating the vifa "4.5 in" cone area, I went to Vifa's website and looked the Sd of the other 4 1/2 inchers.
It was 58 sq cm, not 78.5 sq cm. Speaker must have an oversized rim. Oops.
So that changes the Sd to 58, and the Bl to 4.71 Tm, (as calculated by BoxModel).
Here is the revised simulation-it's pretty close to the last one.
Here are the other Thiele-small parameters as stated on the Parts Express website:
PE 299-495
Power handling: 35 watts RMS/70 watts max *
VCdia: 1" *
L1: .5 mH *
Znom: 4 ohms *
Re: 2.95 ohms *
Frequency range: 65-5,000 Hz *
Fs: 65 Hz *
SPL: 88dB 2.83V/1m *
Vas: .16 cu. ft. *
Qms: 1.44 *
Qes: .34 *
Qts: .27 *
Xmax: 3.5 mm *
Dimensions: Overall Diameter: 5",
Cutout Diameter: 4",
Mounting Depth: 2-3/8".
This is for 10 watts input, (actually 10.5 watts at 6.5 volts input).
Instead of estimating the vifa "4.5 in" cone area, I went to Vifa's website and looked the Sd of the other 4 1/2 inchers.
It was 58 sq cm, not 78.5 sq cm. Speaker must have an oversized rim. Oops.
So that changes the Sd to 58, and the Bl to 4.71 Tm, (as calculated by BoxModel).
Here is the revised simulation-it's pretty close to the last one.
Here are the other Thiele-small parameters as stated on the Parts Express website:
PE 299-495
Power handling: 35 watts RMS/70 watts max *
VCdia: 1" *
L1: .5 mH *
Znom: 4 ohms *
Re: 2.95 ohms *
Frequency range: 65-5,000 Hz *
Fs: 65 Hz *
SPL: 88dB 2.83V/1m *
Vas: .16 cu. ft. *
Qms: 1.44 *
Qes: .34 *
Qts: .27 *
Xmax: 3.5 mm *
Dimensions: Overall Diameter: 5",
Cutout Diameter: 4",
Mounting Depth: 2-3/8".
Attachments
The output curve is for 6.5 volts into 4 ohms, or 10.5 watts.
So it would be 87 dB for one watt. Which is actually a bit high, since it really is an 85 dB/watt speaker. But Jim just seems to be interested if the 4.5 woofer can put out the goods around 40 Hz, and the answer seems to be "yes". (Okay, maybe just a little into the nonlinear range).
So it would be 87 dB for one watt. Which is actually a bit high, since it really is an 85 dB/watt speaker. But Jim just seems to be interested if the 4.5 woofer can put out the goods around 40 Hz, and the answer seems to be "yes". (Okay, maybe just a little into the nonlinear range).
I don't recommend isobaric layout. Two drivers placed toghether in smaller than ideal boxes [not isobaric] would still have about the same bass eficiency and 6dB higher SPL capability compared to isobaric.
Resulting efficiency in midbass and midrange should be also 3dB higher and less cone excursion for the same SPL should be expected due to mutual loading
Resulting efficiency in midbass and midrange should be also 3dB higher and less cone excursion for the same SPL should be expected due to mutual loading
In theory, two drivers [non isobaric] placed in boxes of volume V liters each [placed together] have 6dB/W higher efficiency than two drivers [isobaric] placed in a box of volume V/2 liters
But with two independent drivers you may shrink the boxes until you lose these extra 6dB of efficiency at low frequencies and then use active equalisation to compensate for it. You will have the same efficiency of isobaric at LF and higher efficiency at higher frequencies, and more important, that system will have a SPL capability at least 6dB higher than isobaric [this is the most valuable property of using two independent drivers and may compensate for the extra space required]
You may also try to compensate for the LF rolloff of the shrunk box by loading the sub placing it in a corner of the room, this will make the sub behave as if the box were bigger
Simulation programs calculate box volumes assuming 2*pi spatial load [half space] but in a corner you have 0.5*pi spatial load and thus you get the same frequency response with a smaller box and less cone excursion than expected
But with two independent drivers you may shrink the boxes until you lose these extra 6dB of efficiency at low frequencies and then use active equalisation to compensate for it. You will have the same efficiency of isobaric at LF and higher efficiency at higher frequencies, and more important, that system will have a SPL capability at least 6dB higher than isobaric [this is the most valuable property of using two independent drivers and may compensate for the extra space required]
You may also try to compensate for the LF rolloff of the shrunk box by loading the sub placing it in a corner of the room, this will make the sub behave as if the box were bigger
Simulation programs calculate box volumes assuming 2*pi spatial load [half space] but in a corner you have 0.5*pi spatial load and thus you get the same frequency response with a smaller box and less cone excursion than expected
Re: Vifa in 11L box
Looking at the group delay can be useful too. It shows a huge delay at about 43Hz, but much smaller delays at adjacent frequencies. This means that there will be a lot of ringing at 43Hz, and combined with the peak in sensitivity it could result in a very "boomy" sound. Damping the box with stuffing should improve this, but it will of course reduce the sensitivity gains from using a large box.
Sub Sim lets you model the addition of stuffing material, but it's a bit awkward. AFAICT the "Ql" setting lets you change the reactance (mechanical springiness) of the air in the box, and the "Qa" setting controls its mechanical conductance.
I've found by amateur measurements on different speakers that adding a small to moderate amount of loose stuffing can be modelled by setting the Ql between about 5 and 3. I haven't noticed any drop in overall sensitivity in the midrange even when I added lots of stuffing (so I guess that Qa is largely unaffected), but impedance measurements were consistent with a Ql of about 1.9.
11L seems way too big, how about a 2L box?
CM
Looking at the group delay can be useful too. It shows a huge delay at about 43Hz, but much smaller delays at adjacent frequencies. This means that there will be a lot of ringing at 43Hz, and combined with the peak in sensitivity it could result in a very "boomy" sound. Damping the box with stuffing should improve this, but it will of course reduce the sensitivity gains from using a large box.
Sub Sim lets you model the addition of stuffing material, but it's a bit awkward. AFAICT the "Ql" setting lets you change the reactance (mechanical springiness) of the air in the box, and the "Qa" setting controls its mechanical conductance.
I've found by amateur measurements on different speakers that adding a small to moderate amount of loose stuffing can be modelled by setting the Ql between about 5 and 3. I haven't noticed any drop in overall sensitivity in the midrange even when I added lots of stuffing (so I guess that Qa is largely unaffected), but impedance measurements were consistent with a Ql of about 1.9.
11L seems way too big, how about a 2L box?
CM
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.