Trying to design the world's simplest sub

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I have built a cable with a voltage divider in the middle, to halve the signal going to the miniDSP from the main amp, and this greatly reduced the hissing, down to a level where it was not really noticable, but I could still hear some if I really put my head down by the driver and if I set the gain to maximum there was some buzzing coming through. So, I thought, 'It's a sub, I don't need the treble.' and added a passive filter (2nd order, f3 at 2K) to roll off all the treble. That did the trick, so now it is working acceptably, and when there is no signal it is silent. The filter also seems to kill some pinging and popping that I got on startup and shutdown, so that was good too.

I still don't know why the meter in the miniDSP software reads low compared to what I expect, but if I measure the voltages in and out, the output from the miniDSP is down only 1dB on the input, which is close enough. The voltage divider also means that the 0.9V output range from the miniDSP should match neatly to the 0.5V input on the BK amp.

So, I just have to make sure the extra transformer and PSU and the little filter fit in with the amp in the space, and I can build the thing.
 
Last edited:
Finally gluing stuff together

An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.


Right, I have finally got to the point of gluing the first panels up. I have to start with the little chamber for the electronics. which now has to go all the way to the top of the cabinet, which means putting that panel on the back and the side first.

I'm wishing I had stuck with the more flexible cable I started with, but after I had problems with hum, one of the things I tried was using thicker shielded cables for the signal. They are a pain to route correctly, I just hope I never need to take any of this stuff out, as it might be hard to put it all back through the hole for the heatsink. Part of me is wishing I had just made a seperate box for the electronics, but then I hope it will look neater all in one. The large washer embedded in the side is where the miniDSP volume control pot sits. The four holes are just for ventilation, as the PSU and miniDSP are up the top, and might bake in the heat from the amp otherwise.

Hoping the next weekend, will see this assembled!
 
Right, I've finally got it all glued up and filled with the insulation material!

An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.


I have not got any clamps, so I've been using metal brackets internally to hold each panel to the others and gravity for the top one. When the glue dries, I shall finally get to have some fun with it.
 
Just picked up on this thread now - very interesting as I have the 15" Dayton in a 150 litre sealed enclosure (built before I got engaged, let alone married). I've been using the sub in two different houses over the last two years and cannot fault it. Deep clean extension with way more dynamic range than I need. I'm using it with a 120W class-d power amp and relying entirely on Audyssey processing in my receiver to balance it out.

I'm allowed to keep the sub in our new house but have conceded to a lick of duck egg blue paint to keep things aesthetically sympathetic......

I'm sure you'll love the Dayton in your sealed box. Keen to know how it turns out.
 
Seems I'm late to the party, but this
http://www.tek4.com/ForSale/Genelec1029Rig/InsightintoSubwoofers.pdf
should be of interest.

In particular, the Fielder/Benjamin reference at the very end. I actually saw their talk-the basic point was if you can't generate high SPL at low frequencies then don't bother trying to reproduce those frequencies as you will not hear/feel them anyway. I don't remember the numbers, but the 86 dB at 20 Hz mentioned earlier I'm pretty sure is inadequate.

On another note, maybe I missed something, but I saw no mention of the positioning of the sub. That could make more difference than the details of the cabinet design, actually. Be sure to experiment with this. Use reciprocity-put the sub where your head(s) would be, and then stick the measurement mic at possible sub locations. And then play around-with a friend, we were pretty surprised that moving the sub a couple of inches made a small but noticeable difference.

Was the pattern of the bracing panel holes and radiusing done according to the textbook? :p
 
Right got the driver in and have been having fun all day long.

An externally hosted image should be here but it was not working when we last tested it.


Here is a photo with my son, for scale. He's got his new toy too, a Nintendo 3DS.

I have not been doing a completely serious assessment today, just playing favourites and watching Star Wars blu rays that I got for Christmas. First impressions are great, I have set the sub up to match the levels of the left and right speakers just going by the published sensitivities and gain of the amps (+5 dB gain to the sub in the preamp stage) and then adding another +1 as I don't go so loud, then applying a Linkwitz transform in the miniDSP to lower Fsc from 33 Hz to 20 Hz, with a 24 dB/octave high pass at 10 Hz to protect the drivers. This results in a gain which rises from about 40Hz linearly up to 4dB at 15 Hz and then drops off again, although it is xtill +2 dB at 10 Hz.

The crossover is set to 60 Hz, which seems as low as I can go before the left and right speakers will start to fall off (-3dB at 55 Hz) and this may be something I play with. At 60 Hz there is no way I ever sense the sub as the source of anything, it all sounds perfectly tied to the left, right and centre channels.

The quality of the bass seems great, there is not a hint of colouration so far (although full volume tests will have to wait for now) and it sounds just like the rest of the system, just lower. There is not really any separate character to the bass at all.

I'll be playing some more with it over the holidays but I don't want to rush into any judgements as I have hardly run it in yet.

The only annoying thing is that the volume control pot is not working. Maybe I have pulled the wires during assembly. Anyway, I'm not taking it apart just to fix that!
 
Seems I'm late to the party, but this
http://www.tek4.com/ForSale/Genelec1029Rig/InsightintoSubwoofers.pdf
should be of interest.

In particular, the Fielder/Benjamin reference at the very end. I actually saw their talk-the basic point was if you can't generate high SPL at low frequencies then don't bother trying to reproduce those frequencies as you will not hear/feel them anyway. I don't remember the numbers, but the 86 dB at 20 Hz mentioned earlier I'm pretty sure is inadequate.

As I initially intended to work with 80 litres size, but ended up building 137 litres, I realise I never put up any models relating to the final size. so here they are.

An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.


So, the maximum SPL at 20 Hz is about 101 dB, and 104 dB at 30 Hz, if I run with the Linkwitz transform set to 25Hz. If I didn't use any transform I could have the response shown below.

An externally hosted image should be here but it was not working when we last tested it.


So, this is before any room effects, and given my listening distance is about 2.5 m, it does sound loud. The exact value of the knee frequency to equalise to is something I plan to play with.

On another note, maybe I missed something, but I saw no mention of the positioning of the sub. That could make more difference than the details of the cabinet design, actually. Be sure to experiment with this. Use reciprocity-put the sub where your head(s) would be, and then stick the measurement mic at possible sub locations. And then play around-with a friend, we were pretty surprised that moving the sub a couple of inches made a small but noticeable difference.

It's next to the main speakers at present, just so I can try the whole range of crossovers to see what sounds best, with no delay effects. This is quite a convenient spot so it may not move unless there is a problem.

Was the pattern of the bracing panel holes and radiusing done according to the textbook? :p

The bracing was just one at the top to hold the top layer of insulation, and one behind the driver, with holes to let it breathe. The whole thing is asymmetric (including horizontally, the driver is 15 mm closer to the front right side than the left) and I was keen to make all the bracing also asymmetric to spread out any resonances as much as possible. I spent a day knocking on all the panels and adding more wood braces to the inside wherever the note sounded lower than 150 Hz, until it was all giving high notes. So, emprical-scientific rather than theoretical-scientific.
 
Cool!
Please post an impedance plot? thanks.

I suspect all plots are with filters in winisd?

I do not have an impedance plot, and I'd need to take the amp out to make the measurement. I'll have to remember to do this if I need to take the amp out for some reason in future.

The modelled impedance looks like this

An externally hosted image should be here but it was not working when we last tested it.


In post #69 above the first three figures have a Linkwitz transform to lower F3 from 33 to 25 Hz, the last has no LT applied. All the figures have 4th order Butterworth high pass filters set at 10 Hz. Should have put this in, sorry.
 
Well, after a few days of watching Star Wars on blu-rays, the bass on tap is just what I wanted. The floor (3/4" thick wooden planks, 60 years old) shakes as if a truck is driving past during some of the louder bits. I've been running with the Linkwitz transform in the minidsp set to 25Hz (the -3 dB point should be 33 Hz although I can't measure it until I buy a mic that is calibrated for these frequencies) or even turned off, and it still shakes the floor massively. The lowest rumblings do seem to be getting the driver moving visibly, but it looks as if I've only used 6-8 mm of travel so far. I've got the volume control working on the minidsp again (cables were too short and got broken during assembly) so my wife can dial it back if she wants to, and put some nylon feet about 3 mm thick and 6 cm diameter on, so the jobs to do still are to round off all the other edges apart from the front, which I did already, then paint it, and put a cloth gille on the front. I'm going to make a big hoop of mdf with 8 pegs and sockets about 1/2" larger in radius than the driver.

I did start this by wondering how much bass would be 'enough', and I definitely have enough for this room. It's so perfect for films, it is just effortless. I really love the fact that the dynamics seem to be preserved, and the compression with the 6" drivers of my old front speakers at low frequencies is gone. My kids also got 'How to Train Your Dragon' on bluray for Xmas, and this turns out to have some wonderful bits for the sub to show off. Basically, whenever a dragon does anything much, there are wonderful sound effects.

I've listened to a bit of music too, and it seems that this is showing up some of the rap and pop songs that have an overly equalised bassline in the mix. Presumably this is to make it sound great on the average car stereo? The Stones and Pink Floyd are coming out just fine though, as is Daft Punk. I have not yet had a quiet enough house to do some serious stereo listening, but I hope to get round to it at the weekend. I'm trying the crossover between 60 and 90 Hz, but can't detect much difference so far. I have really enjoyed building this so I'd like to build myself some small two ways to work as rear speakers next. I think at this point, I'll have to buy a decent mic, and start measuring.

I think at this point I should give thanks to everyone whose pages I read and who commented at the start of this thread, when I had little idea what I was going to do! Without all the helpful suggestions I would probably have built something much less fun. Cheers guys!
 
Just picked up on this thread now - very interesting as I have the 15" Dayton in a 150 litre sealed enclosure (built before I got engaged, let alone married). I've been using the sub in two different houses over the last two years and cannot fault it. Deep clean extension with way more dynamic range than I need. I'm using it with a 120W class-d power amp and relying entirely on Audyssey processing in my receiver to balance it out.

I'm allowed to keep the sub in our new house but have conceded to a lick of duck egg blue paint to keep things aesthetically sympathetic......

I'm sure you'll love the Dayton in your sealed box. Keen to know how it turns out.

I'm actually thinking whether to paint mine to match the floor (light grey) or the wall which is also duck egg blue, wondering which will make it less conspicuous under my desk. Did you ever post any pictures up of your sub, Sharpi?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.