I have another thread going where people are chiming in on what my first diy project should be... feel free to chime in there 
However, I put too many questions in there and want to extract one out.
How in the heck do you calculate correct enclosure volume and what are the important speaker parameters that determine this?
I understand that WinISD is a favorite, but for the life of me I can't model these speakers. WinISD likes to complain about inconsistencies w/ the "Q" type params.
Qes(.63) Qms(3.91) Qts(0.52) Fs(56) Vas(.16 ft^3).
If I go w/ "sorta close enough" values by letting WinISD calculate Qms (which doesn't look that close to me), then I still can't get volumes that match this part's express list nor the volume used in this project
And the values on that list from part's express is freaking me out. There are some 6.5" speakers that should use a smaller enclosure than some 4" speakers!? crazy.
Also, not sure if other params are important (other than the Q* ones) when doing the modeling.
Lastly, do people usually talk liters or ft^3 for volumes?
So, help the noob get past this initial hurdle, please.
-Silly.
However, I put too many questions in there and want to extract one out.
How in the heck do you calculate correct enclosure volume and what are the important speaker parameters that determine this?
I understand that WinISD is a favorite, but for the life of me I can't model these speakers. WinISD likes to complain about inconsistencies w/ the "Q" type params.
Qes(.63) Qms(3.91) Qts(0.52) Fs(56) Vas(.16 ft^3).
If I go w/ "sorta close enough" values by letting WinISD calculate Qms (which doesn't look that close to me), then I still can't get volumes that match this part's express list nor the volume used in this project
And the values on that list from part's express is freaking me out. There are some 6.5" speakers that should use a smaller enclosure than some 4" speakers!? crazy.
Also, not sure if other params are important (other than the Q* ones) when doing the modeling.
Lastly, do people usually talk liters or ft^3 for volumes?
So, help the noob get past this initial hurdle, please.
-Silly.
I like a box calculating program to find the parameters for itself if I don't have them all, but only if it ask's my permission first, or I ask it to.
Anyway, I use litres but most programs will offer you whichever you feel more comfortable with.
There is a rule of thumb that suggests a driver will be happiest in a closed box that is about a third of the drivers Vas. I think AR was responsible for this finding a few decades back IIRC.
In such a box, the system's resonance will be at a frequency of twice that of the driver in free air. The system Q will be twice that of the Driver's Qts (in free air).
From this you can work backwards from what you want to achieve, to what you want to find in a prospective driver.
If all this doesn't work, just change the box size to suit. Closed boxes are not known for being too critical.
Anyway, I use litres but most programs will offer you whichever you feel more comfortable with.
There is a rule of thumb that suggests a driver will be happiest in a closed box that is about a third of the drivers Vas. I think AR was responsible for this finding a few decades back IIRC.
In such a box, the system's resonance will be at a frequency of twice that of the driver in free air. The system Q will be twice that of the Driver's Qts (in free air).
From this you can work backwards from what you want to achieve, to what you want to find in a prospective driver.
If all this doesn't work, just change the box size to suit. Closed boxes are not known for being too critical.
Hi,
Do you want a sealed box speaker or a vented box speaker?
When working with vented boxes this does not follow:
For a vented box this is wrong:
Do you want a sealed box speaker or a vented box speaker?
When working with vented boxes this does not follow:
The optimum box volume depends on the driver's Qts. for Vas = 3times vented box volume, the required Qts is 0.29 I also suspect it is not true for sealed boxes either.
For a vented box this is wrong:
The Fb= 1.6times Fs when Vas=3times Vb and Q=0.29
I was talking only of closed boxes, not vented (sorry if I wasn't clear).
From an early version of the LDC, (IIRC) I found Vb =
Vas
-------------
./ /Qtc\2...\
| |----| - 1|
.\ \Qts/..../
and fc (IIRC) =
./Qtc\
|----| x fs
.\Qts/
(Ignore the dots).
It has been a while since I used these formulas and if I have made a mistake, I will stand corrected.
From an early version of the LDC, (IIRC) I found Vb =
Vas
-------------
./ /Qtc\2...\
| |----| - 1|
.\ \Qts/..../
and fc (IIRC) =
./Qtc\
|----| x fs
.\Qts/
(Ignore the dots).
It has been a while since I used these formulas and if I have made a mistake, I will stand corrected.
Hi Indm,
you made it completely clear you were discussing a closed box, that's why I removed the word in the quote.
But for comparison I gave the equivalent numbers for a vented box to show the option and to show that the volume and frequency rules vary with Qts. They are not fixed.
BTW your formulae match your description.
you made it completely clear you were discussing a closed box, that's why I removed the word in the quote.
But for comparison I gave the equivalent numbers for a vented box to show the option and to show that the volume and frequency rules vary with Qts. They are not fixed.
BTW your formulae match your description.
Sillyboy,
I don't believe a closed box is necessary for good sound even though my post sounded like that, I just chose to talk about closed boxes.
I haven't used WinISD, if you like it will you let us know please?
AndrewT,
that's cool. I was just being careful, hate to think I was giving out bad formulas.
BTW I checked to be sure
http://www.arcavia.com/kyle/Equations/Closed.html
I don't believe a closed box is necessary for good sound even though my post sounded like that, I just chose to talk about closed boxes.
I haven't used WinISD, if you like it will you let us know please?
AndrewT,
that's cool. I was just being careful, hate to think I was giving out bad formulas.
BTW I checked to be sure
http://www.arcavia.com/kyle/Equations/Closed.html
Sillyboy said:
Hi,
The values are inconsistent, e.g. Qes needs to be 0.60 to be consistent.
Also note for a sealed box you are usually looking at a final Q of 0.6 to 0.9.
For a reflex design before adding the port you are usually looking at a Q of
0.4 to 0.6 if ther box was sealed, so reflex boxes are always bigger.
/sreten.I was sorta enjoying those formulas from http://www.arcavia.com/kyle/Equations/Closed.html
Vb = Vas / ((Qs/Qt)^2-1)
Only problem is that I have no idea what value I am supposed to use for Qs.
Any hints?
Vb = Vas / ((Qs/Qt)^2-1)
Only problem is that I have no idea what value I am supposed to use for Qs.
Any hints?
<sigh>
See, this is exactly why enclosure volume calculation is just brutal for a n00b. Zaph uses the Hi-Vi W5 in his project 19 .
If you take stats from http://www.solen.ca/hiv.htm and choose of low'ish Qtc of .8:
Vb = Vas / ((Qtc/Qts)^2-1)
Vb = 9L / ((.8/.49)^2-1)
Vb = 5.4L or 0.19 ft^3
Compare that too that stats that Zaph used for this woofer (I assume measured values? No idea)
Vb = 7.27 / ((.8/.376)^2-1)
Vb = 2.06L or 0.07 ft^3
Okay, pretty big difference there... and then you continue reading Zaph's plans, and he actually uses an enclosure sizes of 0.25 ft^3. 4 times bigger than theoretical optimum.
So what's a n00b supposed to take away from this!?
See, this is exactly why enclosure volume calculation is just brutal for a n00b. Zaph uses the Hi-Vi W5 in his project 19 .
If you take stats from http://www.solen.ca/hiv.htm and choose of low'ish Qtc of .8:
Vb = Vas / ((Qtc/Qts)^2-1)
Vb = 9L / ((.8/.49)^2-1)
Vb = 5.4L or 0.19 ft^3
Compare that too that stats that Zaph used for this woofer (I assume measured values? No idea)
Vb = 7.27 / ((.8/.376)^2-1)
Vb = 2.06L or 0.07 ft^3
Okay, pretty big difference there... and then you continue reading Zaph's plans, and he actually uses an enclosure sizes of 0.25 ft^3. 4 times bigger than theoretical optimum.
So what's a n00b supposed to take away from this!?
Sillyboy said:
First, use actual measured T/S parameters rather than the often inaccurate manufactuer's specs.
Actual measured T/S parameters
With that, I came up with this:
Vented response in .25 box with 53 hz tuning
Vented is really optimum for this woofer, considering it's low Fs and teeny actual Vas measurement. Note that my crossover sims and measurements were taken sealed for simplicity.
The first thing a noob often learns to do is use a box modeling program. The problem is, what these programs often recommend as optimum is far from it. Optimum is not always the lowest anechoic F3. Optimum is more accurately a custom tailored low end that takes into consideration usage, the room size, typical speaker location, a woofer's low end distortion profile and finally, actual measurements. I don't want to make things more complicated... but they are, and box modeling is the easy part.
Hope this helps, and I hope I didn't scare you.
Totally makes sense. So, I was struggling with two things:
- trusting manufacturers stats
- forgetting that Zaph's recommendation was for a *ported* box, which are always bigger than sealed.
So the moral here is to only use drivers for which Zaph has measures t/s Or maybe learn to do that myself (http://www.bcae1.com/spboxad3.htm?)
Thanks for everyone's input. I now feel like I have at least *some* handle on enclosure volumes (or as Zaph calls it... "the easy part")
- trusting manufacturers stats
- forgetting that Zaph's recommendation was for a *ported* box, which are always bigger than sealed.
So the moral here is to only use drivers for which Zaph has measures t/s
Or maybe learn to do that myself (http://www.bcae1.com/spboxad3.htm?)Thanks for everyone's input. I now feel like I have at least *some* handle on enclosure volumes (or as Zaph calls it... "the easy part"
)So, after a long day of work, I got home... and got solidly confused again
Attached a screen shot of a comparison between Hi-Vi W5 and Hi-Vi B3S. I stole my t/s params from Zaph's measured data.
I just can't wrap my head around what this... *means*.
I have fixed Qts on both models at 1.08. By my eyes, it looks like:
- the w5 wants a really, really, really small enclosure (good luck getting a 5" woofer in a 0.03 ft^3 enclosure)
- okay, apparently silly Qts for the W5. but still, .1 cu ft gives, a Qts of .703 which seems oddly small.
- uuuhhhh, the 3" woofer has better/lower bass response?!
- the B3S seems hard constrained to no lower than Qtc = 1.01 at infinite volume. What's the deal here? On the other hand, its easy to
- I can't wrap my mind around how the differece t in db (efficiency) plays into this chart, or more importantly the actual bass response. It seems that in reality the 5" should be "lifted up" on the chart (to compensate for its extra db)... which would also affectively makeit look like its moving towards lower bass response. I have no idea what the correct apples to apples comparison is.
- I actually keep wonder if I put in one of the parameters wrong, cuz there is just so much that is ... well, 'weird' to me in this comparison. I mean it looks like f3 for the 5" is 125hz, where its more like 88hz for the 3". no way I can wrap my head around that.
So... thanks again for helping the n00b understand volumes and such.
-Silly
Attached a screen shot of a comparison between Hi-Vi W5 and Hi-Vi B3S. I stole my t/s params from Zaph's measured data.
I just can't wrap my head around what this... *means*.
I have fixed Qts on both models at 1.08. By my eyes, it looks like:
- the w5 wants a really, really, really small enclosure (good luck getting a 5" woofer in a 0.03 ft^3 enclosure)
- okay, apparently silly Qts for the W5. but still, .1 cu ft gives, a Qts of .703 which seems oddly small.
- uuuhhhh, the 3" woofer has better/lower bass response?!
- the B3S seems hard constrained to no lower than Qtc = 1.01 at infinite volume. What's the deal here? On the other hand, its easy to
- I can't wrap my mind around how the differece t in db (efficiency) plays into this chart, or more importantly the actual bass response. It seems that in reality the 5" should be "lifted up" on the chart (to compensate for its extra db)... which would also affectively makeit look like its moving towards lower bass response. I have no idea what the correct apples to apples comparison is.
- I actually keep wonder if I put in one of the parameters wrong, cuz there is just so much that is ... well, 'weird' to me in this comparison. I mean it looks like f3 for the 5" is 125hz, where its more like 88hz for the 3". no way I can wrap my head around that.
So... thanks again for helping the n00b understand volumes and such.
-Silly
Attachments
Hi,
the W5 :
Qts = 0.38, Vas =7.3L, Fs = 62Hz.
On inspecting Vas and Qts I can immediately say the driver is
suitable for sealed box loading in the range ~ 1.5L to ~ 5L and
suitable for reflex loading in the range ~5L to ~10L, though
generally speaking using a box volume any bigger than Vas is
not optimum, the extra volume above Vas doesn't give you much.
For a full range speaker reflexing is the way to go, but for use with a
subwoofer and size being a critical issue sealed is the only way to go.
/sreten.
the W5 :
Qts = 0.38, Vas =7.3L, Fs = 62Hz.
On inspecting Vas and Qts I can immediately say the driver is
suitable for sealed box loading in the range ~ 1.5L to ~ 5L and
suitable for reflex loading in the range ~5L to ~10L, though
generally speaking using a box volume any bigger than Vas is
not optimum, the extra volume above Vas doesn't give you much.
For a full range speaker reflexing is the way to go, but for use with a
subwoofer and size being a critical issue sealed is the only way to go.
/sreten.Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.