Box design check for Dayton 12" HiFi

[jonnythan]

Box design check for Dayton 12" HiFi

I'm designing a box (most likely a sonotube) for the Dayton RS series 12" High Fidelity woofer. It will be used in a home theater application in a room that's about 14 ft x 16 ft.

I'm using WinISD exclusively for the design. I'm very much a newbie at this.

I'd like someone to take a look at what I'm planning and tell me if everything seems in order.

General parameters:
3.75 cf enclosure
22Hz tuning frequency
20.59Hz at -3 dB

Here are the WinISD graphs. I'm confused about the group delay, as I can't quite tell what it is and what it should look like.

Also, how does one determine the volume the driver takes up inside the enclosure?

Thanks!

 

[simon5]

You could try to update to WinISD Pro alpha, there's more options available.

You guesstimate the driver volume since the manufacturer don't give the volume. I'd guess 2 liters.

You could also design a slightly decreasing response curve because the room will amplify the lowest frequencies.

For example, I use a curve that starts at 80 Hz, then at 40 Hz I want it to be at -3 dB then -6 dB at 20 Hz.

Other than that, you seem to be on the right track!

 

[sreten]

Hi,

I agree with s5, a more overdamped response would be better.

The group delay is confusing, but you can compare your
design to the group delay of the same driver in a sealed box.

Group delay gives you some idea of bass transient response,
and its immediately obvious sealed boxes are much better in
this respect than vented.
But vented boxes have many advantages over sealed, not
least of all higher power handling and maximum output.

Generally try to keep GD at bay to as low a frequency as possible.

You''ll find the GD of an overdamped reflex is better than flat.

The driver has obviously been designed for ~ 2 cuft sealed.

When reflexed in ~ 4 cuft port frequency is very low, to get
some form of overdamped alignment port frequency ~ 18Hz.


🙂/sreten.

 

[jonnythan]

Perhaps something like this?

I downloaded WinISD Pro Alpha, which gives me a slightly different response curve for the same parameters. It also lets me put in a crossover, so I put in an 80Hz crossover.

I assume this is what you mean by more overdamped... I eliminated the "hump" at the low end by tuning to a lower frequency.

I'm still confused by group delay, though. I get this huge spike of 40ms around 17.5Hz. Plotting the same driver in a sealed box gives me much poorer low end (F3 is 38Hz vs 19Hz for the ported box), but the group delay is better, I suppose? I'm still not sure what the group delay *is* and what effect it has on actual sub performance, though.

Thanks again for your help and advice.

 

[jonnythan]

Here's the group delay for the two (ported vs sealed)

 

[jonnythan]

Actually I just read a paper online and then found this review:
http://www.hometheaterhifi.com/volume_12_2/polk-psw-505-subwoofer-4-2005.html

Which indicates some baseline group delay numbers as well as a good definition of what it is and how it affects the sound. It would appear that my numbers:

27Hz / 18ms
33Hz / 12ms

Are below the reviewer's "audibility thresholds."

 

[jonnythan]

Now I'm thinking about a 10" sub in a much smaller enclosure at 2cf, perhaps an 18" diameter sonotube that's 15.5 inches long, or a cube with internal dimensions of 15.75" with a pair of 5/8" dowels as braces.

F3 will be more like 21.6Hz than 19.1Hz, and I'll get about 1.5 fewer dB across the board, but my room is only 1600 cf.. that should be sufficient, do you think?

In this config group delay would have a peak of about 33ms at 20Hz dropping to 17ms at 30Hz.

If I didn't mention it, this is primarily for home theater purposes.

Thanks to anyone who actually makes it this far

 

[thalis]

In these low frequencies group delay is not a big issue, I believe. It becomes more in the mid and high frequencies.
A good freeware utility to examine box alignments together with phase - step response - impedance etc. curves ishere
Regards,
Thalis

 

[simon5]

I'd recommend 4 cu.ft internal net volume tuned with a 24" long 4" diameter pipe for the Dayton Hi-Fi with a 250W RMS amplifier.

Why changing to a 10" ?
Check for overexcursion, I think you might lose more than 2 dB or so.

 

[jonnythan]

Going to the 10" just for compactness and ease of construction.

4cf is really fairly large. 30-inch long, 18-inch diameter sonotube is manageable if I can orient it vertically, and 21.25" cubed is giant as well.

The 10" would also be a little bit cheaper.

If it's worth it, the 12" is definitely doable though.

 

[lousymusician]

Hi,

The driver has obviously been designed for ~ 2 cuft sealed.

When reflexed in ~ 4 cuft port frequency is very low, to get
some form of overdamped alignment port frequency ~ 18Hz.

🙂/sreten.

2 ft^3 sealed gives an 'optimal' Q, but 1.5 ft^3, would probably work fine for HT. The Q would be a bit higher, but the response curve is nearly identical and LF power handling would be better and I think you want that for HT.

I'd think about building a small sealed box as a trial, and then decide wheter to stay sealed or go for the big vented box. You may find that with room gain you get plenty of bass from the sealed box!

 

[jonnythan]

!!

What a great idea!!

I'll build the 2 cf sealed box, and if I find I need more bass just build a bigger one.

Thanks!

 

[tade]

i second that emotion.