diyAudio (
-   Subwoofers (
-   -   Tiny 12 litre subwoofer making 15Hz and 600ccm? (

ChocoHolic 30th December 2003 07:04 AM

Tiny 12 litre subwoofer making 15Hz and 600ccm?
2 Attachment(s)
Some days ago I stated that I had no Audio projects since 10 years.
I simply had forgotten my subwoofer !!!
About 5 years ago I designed a funny small sub woofer, which is
still satisfying my ears.....
Only around 12 litres (netto), but impressive 15Hz and a linear volume displacement of around 600ccm (max. 1400ccm !).
I will put some more details about the construction in my next posting.

ChocoHolic 30th December 2003 10:18 AM

2 Attachment(s)
What happens if we design a closed box and simply make it to small?
The resonance frequency will be high, the damping poor.
First let's have a look to the resonance frequency. Closed cabins give
a nice high pass filter with 2nd order (please don't try this approach with vented designs!!!).
In an active solution you can easily compensate this by giving more gain to the lower frequencies
with a 2nd order characteristic reverse to the box.
I used standard OP amps for this (N5534A: yes old, but OK.....).
The box has 11 small Dynaudio Woofers (ESOTEC 15W-75) and volume of
around 12litres. This results in a resonance frequency of about 155Hz.
The pentagon shape avoids undesired resonances.
I used this type of speaker because they have 75mm coils, which makes them rugged enough
in thermal and the driving force is given to the membran well distributed.
Of course one drawback: The required power at low frequencies is getting high, very high.
Let's estimate for 30Hz, here this tiny box is already giving 28db less than nominal.
The proper compensation will give 625 times more power to the speaker.
WOW! OK, I really tried this concept in an excessive way.
Currently I am using a traditional analog amp, which can deliver around 1.5kW
to the 11 parallel speakers.
For long term power I put a limiter. When running the box at 100W continuosly
for 3 hours, the coils temperature stabilizes at 80C. The entire box is warm.
In fact the speakers still have some margin, but going to limit would only bring around
2db more average output.....
Please note that in normal music the ratio of short term power and long term
average is around 10:1...20:1 in this frequency range. So the 1.5kW amp matches
quite good with the 100W average limit.

Damping: Here we have a direct influence of the resistance in the electrical circuit.
normaly this is dominated by the speaker. The output impedance of the amp + wiring typically
is 50....200 less than the speakers impedance.
Well, let's give a negative output-impedance to the amp!!!!!
With this we can adjust the damping to any value we need.
For low frequencies this is easy ( for high frequencies up to now not a good idea......).
You simply need to measure the output current and feed it back positive.
Please note. The negative output impedance must be less than the speakers impedance,
otherwise the system ist getting instable!

In the last 3 years several new and especially for sub woofer applications designed speakers
came to the market. So this posting here may encourage you to do your own experiments.
My personal feeling is that a less excessive design might be more fortunate with respect to
the required power. Setting up your own design will require some good fundamental know how about
speakers, filters, feedback control and amplifiers.... Not the right thing for newcomers.
Of course I can give some more detailed hints, if you really want to set up a subwoofer with this

Have fun

P.S. Is there anybody who has done something similar? Which experiences did you make?

ChocoHolic 31st December 2003 11:20 AM

No reply at all???
I would have expected at least that somebody would propose a switch mode amp for such projects! I really the onlyone here who likes good bass but not big boxes?

JCoffey 31st December 2003 01:29 PM

I love big bass, and small boxes. I also love Dynaudio.(Thats why I have them in my car) but what I shy away from is really high cost, and 11(!) ESOTECS?
Its a beutiful project, and very impressive results. Its just a bit too much out of my range of knowlage to even begin to make suggestions, or comments.

ThorstenL 31st December 2003 03:20 PM



Originally posted by JCoffey
I love big bass, and small boxes. I also love Dynaudio.(Thats why I have them in my car) but what I shy away from is really high cost, and 11(!) ESOTECS?
Its a beutiful project, and very impressive results. Its just a bit too much out of my range of knowlage to even begin to make suggestions, or comments.

Well, things of that sort are quite common in Germany, where this principle is called URPS (Under Resosance Principle Subwoofer). I still have also one of these on the drawing board, using a simple square pillar and 8 Drivers. The ones that caught my eye are the beautifully made 8" unit from Apex Jr.

These cost $ 70 in pairs, so 8pcs will set you back $ 280. The Volume dispacement should be similar to that quoted above, around 600cm^3++.

The Column looks like it would have to be probably around 12" X 12" X 18" made in 1" MDF. This gives 25 liter volume gross, probably around 16 Liter once the drivers static displacement is accounted for. This gives around 2 liter volume per driver and thus as single driver in halve space produces 90db/2.83V/1m @ 100Hz with a resonance at around 150Hz. At 50Hz the SPL is down 14db and at 25Hz it is down 28db.

Using a sensible 150W RMS power limit per driver we would like an Amplifier with 1K2W output Power, into 8 Ohm if we wire the drivers series/parallel. Alternatively 4pcs LM3886 could be employed Bridge/parallel per driver with around +/-25..30V supplies to keep the dissipation per chip lowish. Probably better to bridge one of the Big Tripath Modules for nearly 2KW power.

If we drive 8 Drivers series parallel with 2.83V in total we get 68db/2.83V @ 25Hz, 82db/2.83V @ 50Hz and 99db/2.83V @ 100Hz. With 1K2W (or 31dbW) we can have theoretically 99db/1m @ 25Hz. Placing the sub into a corner will boost the LF level by another 4 - 5db, so I think a pair of subwoofers would make a good figure for larger rooms.

The material cost would likely be < $ 1,500 for a pair of these and probably < $ 750 for a single unit, with an output capability in line with some very "high end" Subwoofers in an increadibly compact format.


ChocoHolic 1st January 2004 10:16 AM

...yes the costs were somehow inconvinient...
But 5 years ago I did not find any dedicated URPS-drivers. At this time I also looked at Apex, but
did not get what I was looking for...
About 3 years ago suddenly subwoofers with interresting drivers were coming from every
end... I think I was simply looking some years to early.
Now I feel several suitable drivers are available and also switchmode amps are more common,
this is why I encouraged this thread.

Kuei's project looks really interesting. And this drivers will for sure enable him to get even better
results than my old design. Much more volume displacement per box volume.
The Apex drivers look impressive (... will have a look to their Web-Site..)
And this at reasonable price. Yeah, cool!
But do you really want to place it in a corner? My personal trial always ended up
in a position with some distance to all wall and corners. Otherwise I got booming
resonaces of the room... OK, more pressure, but much less pleasure to my ears.

I could think now to set up a new subwoofer, too.
...with some more updated speakers and a switch mode amp...
and then I could use my nice small ESOTECs for new satelite speakers....
Unfortunately such a project would take me 5 years... Cannnot spend time every day
for this hobby :-(( !!

JC, well even a DIYer does not need to do every component on his own.
You can buy such systems, too. Unfortunately the real good systems seem to
be quite expensive...

mikee12345 1st January 2004 02:21 PM

I love big bass,and Large boxes
I notice

1) small woofer
2) attempting high spl
3) attempting low frequency
4) 28db EQ

These things are not associated with low distortion.

you dont mention xmax at all..

It does look nice.


WOW! OK, I really tried this concept in an excessive way.
Definately, you did!

To me changing the amplifier output impedence is just fun,i think you can reduce the damping factor slightly,but remember it will change the speaker Qts factor...mucking up your alignment

Personally- I would prefer a trusty 15" jbl to do the job,or basshorn.

Have you measured SPL and distortion?
i would like to know it


ChocoHolic 1st January 2004 03:09 PM

Hi Mikee!
In fact I cannot measure the distorsion and I agree on your concerns
regarding this point. I think the large voice coils of the Dynaudio speakers
are the best way to avoid that the bending of the speakers membranes.
This was the reason why I did not like most of the high excursion speakers
five years ago. They were typically running 2" voice coils for a 12" speaker....
Well, my ears say that the resulting distorsions of the speakers are OK.
But I had to struggle with the distorsions of the amplifier!!!
As the box defeat about 28 dB at 30Hz (I coupled the system below 40Hz)
which is the desired signal to hear, but do not defeat at the frequencies of most amplifiers distorsions.
This system requires a real low distorsion amp.
I had not expected this and had to redesign the amp... and now it is fine.

X-max of the small ESOTECs is 15mm (peak-peak). Linear about half of it.
So the 600ccm are peak-peak. Sufficient for normal levels in typical living rooms, but not for disco.
Also at the limit for real home theatre.
But I estiamte the limit of my neighbours around 200ccm anyway..... ;-)

In fact I also love the big JBL drivers, but without electronic correction they also will not make a flat frequency response down to 15Hz....
But the real worse thing is that I would have to place it in the hyper space and beam its sound to my small appartment....
Unfortunately- up to know I did not manage to design such a system ;-)

ChocoHolic 1st January 2004 03:54 PM

2 Attachment(s)
Just one more point.
You influence the damping of the system very much by the impedance of the amplifier, not only marginal.
If we look to the simplified equivalent schematic of a speaker (attached),
we can see an LC-R-circuitry.
L and C reflect the resonance system of the speaker( dominated by moving mass and the spring). Rem reflects the mechanical dampig, which is typically weak compared to the damping of the eletromagneitc loop. R is reflecting the DC-resistance of the speaker.
The electromagnetic loop is electrically closed through the signal generator/amp.
The internal resistance is positioned in series to the Rdc of the speaker.
By changing the internal resistance of the amp to negative values you can definitely influence the damping in a dramatic way!
Of course this is changing the complete set up of the box. Yes, but on purpose.
Optimised adjustement will also depend on the room and the position of the box. Different situations require different damping of the system.
In fact proper placing a subwoofer is much more difficult than normally promoted....

mikee12345 2nd January 2004 01:37 AM

2 Attachment(s)
I know that YAMAHA AST cancels out the speakers Rdc with this method.

quote from a loudspeaker book(forgot the name)
' this results in higher than usual port output'

for 16hz=

:) :cool:

All times are GMT. The time now is 07:42 AM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 18.75%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio