[vann_d]

This thread documents the build of my 10” Dayton Quatro subwoofer. This is my first “real” build and I did it with a very minimal budget of about $150 altogether. I say “real” build because it’s the first that I actually designed. I had previously built one sealed sub for a car where enclosure size was prescribed. My woodworking skills and equipment are minimal. That said, I think the results are actually quite good considering the cost and my lack of experience. This is not, however, a sub that I would recommend a fellow DIY’er copy. The main purpose for this thread is to share my story for other first timers and in the process gather as many lessons learned as possible. Plus, documenting this will hopefully prevent me from making the same mistakes in my next DIY project!

OK, here is what I started with…

1 – 15 year old 10” Onkyo sub, ported, with plate amp in working condition and maybe $150.
The emphasis here is on "plate amp in working condition". The sub sounded terrible. I had no idea what the power output of the amp is but I figured it would be low (maybe 100 – 150 Watts).

Goal:
Using as little money as possible, create something that resembles a “real” subwoofer for music and HT.

My solution:
Buy a cheap driver and create a new design around it and my scavenged plate amp.

Ok, all I have time for now. In my next post I will reveal my embarrassing first attempt…

 

[vann_d]

So, the first decision I made was to select a driver and here is where I made my first of many mistakes. It’s not that the 10” Dayton Quatro is a bad driver, it’s that my thought process in choosing this driver was just all wrong. It went something like this…

“Hmm, how do I get by on the cheap and with minimal effort? I know! I’ve already got a sub that’s built. It’s in a 1 cu. ft. enclosure. I’ll buy a 10” driver and just change the port tuning in my existing enclosure and be done! Now, let’s go to Parts Express and see what I can get for under $100…”

I (obviously) bought the 10” Quatro. I did this without taking any time to model the driver’s behavior in the enclosure I was planning on putting it in or even understanding what some of the driver’s T/S parameters meant and how they would affect the outcome of my project.

 

[vann_d]

Now, it’s not that I didn’t know that I was supposed to do these things. It’s that I just assumed since this driver looked like pretty good quality and that there seem to be a lot of subwoofer designs in around 1 cu. ft. that I could make something work. And by “work” I meant play down to 20Hz “pretty loud”.

Anyway, I kinda learned how to use UniBox and this is what I came up with…
10” Dayton Quatro
30L Ported enclosure tuned to 35 Hz
1 4” port at 23.25 inches long

 

[Mike P.]

Do you have pics of your project?

 

[vann_d]

Pics and graphs coming soon...

 

[vann_d]

Ok, here's how the modeling turned out in my first attempt (using my existing enclosure)...

Well, the port turned out pretty difficult to implement and the result looked pretty absurd. I thought I could bend the port but the tube was so large there was no room at all. I ended up sticking half of the port outside the box. :scratch: Thankfully, I don’t have any actual photos but it went something like this…

Not very high waf is an understatement :rofl:

Anyway, I quickly scrapped this design. It actually didn't sound that bad for music but with no hp filter I was exceeding linear excursion on movies a lot. Even this poor design was much better than my original onkyo sub and the anemic 8" jbl htib sub I had used in the past...

Coming next, new enclosure design with same parts (except port)

-Vann

 

[vann_d]

OK, lessons learned so far…
1. Each driver has unique properties, is intended for a specific use, and will require unique enclosures and/or tuning to get the most out of it.
2. It is unlikely that you will be able to achieve the tuning that you want using an existing enclosure.
3. If you are going to re-use an enclosure, after checking number 1 and 2 above, make sure that the port fits!

So, after scrapping the weak initial attempt, I decided to design and build a new enclosure that can (more closely) reach my goal of having a decent HT sub on a super cheap budget.

After many attempts with iterations on enclosure size, port tuning, and port size, I ended up with these parameters:
Enclosure volume ~66 liters (2.33 cu. ft.)
Port tuning = 27 Hz using three 2” ports at 14.5” in length

This seemed to be pretty much as low as I could tune using this driver with Fs=29.6 Hz. I’m right on the limit of excursion at 100W. I may have more but I figured I could turn down the gain. This tuning seemed to get me a lot closer to being a useful HT sub than my previous effort but I knew it would still have issues with a lot of movies. I don’t have a HP filter.

Not sure how to interpret step response yet. I know this looks underdamped from my background in dynamics but I don’t know what this means for sound…probably a little sloppy.

 

[vann_d]

Here is what I came up with for the enclosure design. Outside dimensions are 12" x 12" x 42". I thought this form factor may look smaller than a cube shape.

 

[vann_d]

The build took longer than I expected. I didn’t use any fasteners and it was difficult to get everything clamped together so that edges were flush. If I would have known better I would have intentionally left overhangs that could be trimmed with the router. Instead, I ended up having to use wood filler and sanding to smooth out the ridges between panels. Not all the panels fit together perfectly anyway. Probably because I have a cheap little table saw and no experience using it.

Anyway, I lost my build photos but here is the finished product. I was trying to go for a gloss finish and I think I could have achieved it if I hung in there and spent more time sanding and re-spraying and sanding again. As it was I spent a lot of time doing this (like everyday for over a week). Some areas are nice and smooth and some are rough. I could probably wet sand the whole box and have a decent finish all around. In the end I gave up because I just didn’t think it was worth it for this project. Oh, it’s also obvious that I didn’t finish painting the end that the driver is mounted to.

Again, the plate amp was scavenged. Seems like an OK amp. It has adjustable gain and crossover frequency from 90 to 150 Hz. Phase is reverseable (0 or 180), high and low level inputs. I don't know the power output. I guess I should measure it sometime.

T-nuts were installed to mount the driver. This was also a challenge to implement because the mounting flange on the driver is small. I had to use #6 socket head cap screws. Anything bigger would not fit in the baffle and would protrude into the cutout for the driver. You can see the flanges stick out a bit as it is. To make things worse, the fastener holes in the driver flange were larger than the head of the small fasteners so I needed washers but the standard washer would barely fit. Just not a good design or intended only for wood screws into the baffle.

Internally, I ended up replacing the planned windowpane brace with one of my practice baffle cuts. I also added another brace between amp enclosure and side wall.

To flare the ports I used a round-over bit on the router. You can see on one of them I accidentally set the depth too deep. Getting the transition between pvc and mdf smooth was difficult and probably not as good as possible.

Coming next, in room response plots...

 

[Mike P.]

Looks good. Any idea what model of Onkyo that plate amp came from? Chances are it has a HP Filter.

 

[vann_d]

Thanks Mike! I've got no idea what model the sub was. It was a hand-me-down originally purchased 15 or 16 years ago. I should really sit down and do some measurements on this amp.

Seems to me the original enclosure was tuned to about 70-75 Hz. It was about 1 cu. ft. with a 3 inch port around 1 to 1-1/2 inch long. Sweet, huh?

 

[Mike P.]

It may be a good idea to measure the amp if possible. It could very well have a HPF at 35 hz. That being the case your sub won't have much low end output.

 

[vann_d]

So I went out and bought a BFD and got RoomEQ Wizard hooked up. Wow, what a value. Here I was saving my pennies for a receiver with Audessy. Uh, don’t need that anymore!

My HT is in my living room. It’s our only place for watching TV or movies or listening to music. Maybe in my next house I’ll have a basement and a dedicated theater. In the meantime, acoustics are going to be challenging. The living room is open to the kitchen, dining room, and family room. About 10,000 cu. ft. total. This sub build is the biggest part of my system so, obviously, we’re not talking huge SPL levels here but you gotta start somewhere.

Also, my RoomEQ setup is less than optimal because I don’t have a tripod to hold my Radio Shack SPL meter. I ended up stuffing the meter in between two cushions in the back of the sofa so that the meter pointed up and sat several inches proud of the cushions. It is about where the back of my head would be. I ran several sweeps in a few listening positions and averaged the results. Results within the same spot were very consistent. Surprisingly, not huge differences between positions either. Anyway, here are the before and after… 1/24th smoothing

Before

After filters

Before and after...

Not sure what's going on at 150 Hz. Isn't that a little high for a room mode? REW didn't seem to filter there.

Also, please guys (gals?) any critique on my driver/enclosure modeling, tips, etc. would be greatly appreciated. This isn't like my baby or something so I'm not going to get my feelings hurt here.

-Vann

 

[brucek]

Not sure what's going on at 150 Hz. Isn't that a little high for a room mode? REW didn't seem to filter there

I trust this is a plot of the sub alone with no mains. If so, there's no need to take any action on the response above the crossover you normally use (80Hz) until you do a plot of the sub + mains together. Then you can adjust the subs phase control to smooth the crossover region as much as possible.

Response looks fine, although it trails off fairly quick at ~ 25Hz. If you expected more from your models, check that the amp doesn't have a high pass enabled

brucek

 

[vann_d]

brucek,

Yes, the response plots are sub only, no mains.

I wasn't really expecting more low end with this design but I think I'm maxing out the excursion of the woofer at my listening levels for movies. During some scenes in LOTR or Star Wars with "rumble" my sub basically sounds like somebody is rattling a rock around in a coffee can (except a lot lower in tone). Sounds like garbage. This is why I think my amp doesn't have a HPF. I should probably set up the BFD to filter below 25Hz. It's ok to me that I'd be missing these effects for now given this modest setup. Nothing at all would sound better than that noise. Hopefully I haven't damaged the woofer.

I think I did a measurement to 2k with mains. I'll see if I can dig that up.

 

[vann_d]

Here is my in-room response to 2kHz. Somehow I ended up with a peak at 40Hz now. The 150Hz stuff I noticed with sub alone is still there but it blends in better when the mains are present. There is a pretty good null at 190Hz.

The downfall of these measurements is that in analog stereo mode (what I need to run the measurement) my cheap receiver only lets me run my mains (Klipsch RB-51) in large mode. Normally, I run them in small. So I don't really get to see the true response. To complicate things more, I don't know where the crossover is in the receiver. It is not specified in any of the documentation as far as I know.

Here is the receiver model in case anyone out there reading happens to know: JBL DCR600

1/24 octave smoothing

1/3 octave smoothing

 

[brucek]

The response is quite good.

Your crossover is likely fixed at 80Hz. It's not uncommon.

Yeah, rocks in a coffee can isn't good. You could be bottoming your driver. You're pushing too much gain from the BFD into the bandwidth below your tuning frequency - not a good idea. Look at your unequalized graph. The driver/enclosure is showing you where it wants to drop off. That's around your tuning frequency that your model shows. Don't add filters below that (as you've done). The rattling will stop.

brucek

 

[vann_d]

You're pushing too much gain from the BFD into the bandwidth below your tuning frequency - not a good idea. Look at your unequalized graph. The driver/enclosure is showing you where it wants to drop off. That's around your tuning frequency that your model shows. Don't add filters below that (as you've done). The rattling will stop.

brucek

Thanks brucek, that's great advice. I'm looking at my filter settings and I don't seem to be adding gain (see below). In fact, I'm taking it down quite a bit at 32Hz and don't have any filters below that frequency. I'm thinking I should add another filter at the port tuning to try and get rid of input to the sub below there (~26 Hz). What do you think?



Equaliser: DSP1124P
null
Cutoff 80Hz 24dB/Octave
Target level: 75.0dB
Filter 1: ON PA Fc 31.9Hz ( 32 +1 ) Gain -17.0dB BW/60 10.0
Filter 2: ON PA Fc 84.0Hz ( 80 +4 ) Gain -13.0dB BW/60 8.0
Filter 3: ON PA Fc 50.0Hz ( 50) Gain -8.0dB BW/60 4.0
Filter 4: ON PA Fc 170.0Hz ( 160 +5 ) Gain -14.0dB BW/60 4.0
Filter 5: ON PA Fc 79.2Hz ( 80 -1 ) Gain -3.0dB BW/60 2.0
Filter 6: ON PA Fc 98.0Hz ( 100 -2 ) Gain -6.0dB BW/60 2.0
Filter 7: ON PA Fc 56.5Hz ( 50 +10) Gain -6.0dB BW/60 11.0
Filter 8: ON PA Fc 46.5Hz ( 50 -7 ) Gain -2.0dB BW/60 14.0
Filter 9: ON None
Filter 10: ON PA Fc 113.8Hz ( 125 -9 ) Gain -6.0dB BW/60 3.0
Filter 11: ON PA Fc 194.0Hz ( .20k -3 ) Gain -5.0dB BW/60 1.0
Filter 12: ON None

 

[vann_d]

Here's what the predicted response plots look like for the current filter set-up and if I add a -17dB filter at 22.25 Hz and raise the gain at 32Hz to -12dB...

Maybe this will help. I'm gonna try it out.

Predicted response with old filter set...

new filter added (predicted)...

 

[brucek]

I would lose filters 6, 10 , 4, 11. They are above your crossover and are fairly meaningless.

BTW, there is a sort button on the Filters popup in REW that re-organizes the filters from low to high...

burcek