Quick and Dirty Surrounds

Figured i'd post some pics of my recent surround build.

They were a low budget, low effort affair. I just needed something to replace my ugly rear channel satallites.

I looked through Zaph's speaker reviews for a small driver that was suitable for full range operation. The Fountek FR88EX rated very well and was recommended by several people on this forum. Unfortunately, the price had just jumped significantly on those drivers and they were out of stock from Madisound. I chose the Aurasound NS3 as a substitute because it was in stock and Zaph mentioned it *could* be run without a filter and i'm lazy

I wanted the box as small as possible. I chose a sealed design 6"Wx5"Dx7"H
that will produce a Qtc of 0.74 according to WinISD and result in an F3 around 95Hz. This was larger than i wanted to be perfectly honest, but i went ahead anyway.

Because i thought i may change drivers in the future and i was going to play around with different approaches to veneering the front, i went with a removable baffle.

I had some time to listen to them before i started on the veneer and they were a big improvement over the tiny old satalites they replaced. I think i have some air leaks after the veneer work because the baffle does not sit flush anymore. i haven't listened to them and will probably disassemble and do some sanding to get them fitting better before i have a finished product.

Edit - these are nothing special, but i figured i'd share since i like seeing picks of other people's builds, regardless of the complexity or level of quality...

initial box assembly. 1/2" MDF... i usually use 3/4 but i figured that was overkill for these little guys. it was a bit harder to assemble with the thinner panels if you ask me. i used wood glue and brads... i usually use glue and screws.

i've seen a trend in Zaph designs to route the back edge of the baffle to allow better airflow around the back of the driver. this time i remembered NOT route the areas where the screws would hold in the drivers.

also needed to route a deep channel on one side to give the side-mounted speaker terminals clearance.

some terminals and wiring...

a little veneer clamping... i wasn't thinking that these boxes were only 1/2"... i tightened the clamps a bit too much. i heard the strangest sound the bottom of one of the boxes started buckling and splitting.

i ended up filling the crack with glue and clamping it sort of back into position. then i used a hand plane to level off the outside of the bulge. final result was good enough for these little guys. definitely a bummer tho.

also had to veneer the edges of the baffles separately since they would be removable. i'm going to try making round-over baffles and veneering them next but i'm not sure how that's gonna go so i wanted to start off with something a bit more sure-fire.

i'm using paper-backed ash veneer that i had left over from my previous projects.

after veneering, i stained these guys to match the rest of my room. they didn't stain as even as i hoped. i think perhaps due to my uneven prep-sanding. but they're not too bad. after the stain dried i hit them with 4 or 5 coats of shellac . i ordered a series of Abralon sanding pads and i'm going to try my luck at sanding them to a nice semi-gloss.

I kind of butchered the sanding job with the abralon pads. Too aggressive, especially around the edges. I didn't really like the way these looked in their assigned location anyway - they sort of stick out. I'm probably going to redesign and rebuild... maybe vented.

Anyway, i recently got an Omnimic and wanted to see how these guys performed relative to the Sony Satallites (maybe a 1" or 2" driver in them, i don't know) so i played around this morning trying to take measurements. I was equally interested in the speaker performance and learning to use the Omnimic.

Attached are some curves. They're not very good. And right as i was writing the graph title in excel, "IPhone Input", it occurred to me that these results are garbage anyway. I thought i was being clever by ripping the Omnimic CD in iTunes and putting the tracks on my iPhone for ease of use. I forgot that the tracks would get compressed, completely negating their usefulness So i'm atttaching these more as a refernce point for when i have some time to hook up a CD player and retest.

I've done very little with the Omnimic so far. I tested a sub with a plate amp, which was simple. And i tested my surrounds using an extra receiver i have laying around - just using one channel and one speaker. I agree, it would really be a pain to test response on a per-speaker basis in my main HT room using this method for 5.1.

Your post reminded me that i saw a reference to a downloadable DVD with test tones on it in the Omnimic thread. I don't have the time to read through the DVD description thread right now. I'm not really clear if it would even work with Omnimic given the fact the Omnimic system is expecting specific sweep durations, etc... Thought i would post this link in case you haven't seen it yet:

http://www.avsforum.com/avs-vb/showthread.php?t=742969

Just retested the Sony Satallites and my Aura 3" with Omnimic, this time using the CD as the source (instead of iPhone, like an ass).

I played around with the Omnimic software a bit more and it does more than i thought - lets you load multiple saved curves and adjust their display, then save out a graph with a legend. This is a pretty fundamental piece of functionality when testing speakers, so i'm very glad it's there.

Anyway, it looks like my 3" Aurasound sealed boxes are a total bust. They aren't really any better than the Sony's, which are about 1/4 the size. DIY fail

There are a few things i can think of that could improve things.
1) A vented design would give me better low end, but the boxes would be bigger still
2) I didn't round-over the baffle edges
3) There may be some air leaks in the Aura box.

Overall, i'm very disappointed with my project. But i learned a good amount about finishing techniques and working with 1/2" MDF - delicate stuff. New box design coming in the near future.

The FR curve is 1/12 Octave smoothing with 10 sample averaging. The HD curves are 1/12 Octave smoothing also. It's interesting, the SPL curve (Black) in the HD plots does make the Aura look a little better than the Sony if you interpret that as the FR data. But in the FR image with them side-by-side, they both look equally poor.

thanks bwaslo - that is an excellent suggestion.

i just retested at a significantly louder level. things look a little better for my Aurasound in the retest - still a bit of a mess tho. There is some benefit in the low end (50-200hz) with my DIY, which is what i expected.

Interesting that there are dips at ~125Hz and ~200Hz for both speakers. Any chance that is a room node? I think the reflecting wall in that test room is about 8 feet away and the back wall about 1.5 feet away. Microphone was 2 or 3 feet away from the speakers pointed directly at them.

The Aurasound does fall off pretty bad above 6K... i guess i really should have made this a 2way. Maybe i'll start looking into tweeter selection and crossover design.

i also wanted to mention that when i stated in an earlier post that my DIY speakers sounded much better than the Sony's ... i was actually using a different Sony satallite that sounded like it might have been blown. Since then i found some other Sonys laying around that are in much better shape - those are the ones used in these measurements.

Some followup work... i ended up building a vented enclosure but never reported on it. I believe it was 0.142cuft tuned to 50hz with 2 vents 3.9" long (i think it was 1" outer diameter pvc?). With 4" lengths, i had to put a 90deg kink via miter saw and pvc cement. That may have comprimised performance - i was too lazy to research the impact of that approach.

I built one vented enclosure and tested it against one of the completed sealed ones. Results were underwhealming. They tested very similarly. I was hoping for a bump in the mid-upper bass region, but didn't see anything. There's a tiny response bump just under 50hz for the vented curve... maybe miscalculations on my part (or missed-cuts) lead to a tuning slightly below the 50hz target. It's irrelevant anyway tho because the response is about 20db down at that point.

I'm rather embarassed about the build quality of these enclosures. I think both the sealed and the vented were leaky.

Attached is an enclosure comparison pic and measurements

just for fun i tried running the vented enclosure with no stuffing, polyfill (as typically used in a sealed enclosure) and then the general vented box foam lining.

I can barely remember doing this (was 4 months ago), but i don't think i changed the volume on the receiver between runs. So i can't account for the relatively large gain over most of the frequency range that the foam-padding run exhibits. I was so disgusted by the whole project that i didn't dig any further into it.

During some cleanup last month i threw out the vented enclosure. The sealed cabinets are sitting in my garage. They're too out of place for the locations i had picked for them but i couldn't bring myself to throw them out. One day i'll probably rebuild a fancy (curved maybe?) sealed enclosure and maybe then i'll get some use out of the Aura's as surrounds or pc speakers. I'm currently using my ugly little sony satallite surrounds because they're so tiny that they fit the space well. Can't win 'em all i guess.