LOUDSPEAKERS 2.

Content :-
Testing the response of speakers with pinknoise,
Test methods. Sub-woofer recipe with explanations.
Design plans for 3 way floor stander speaker with full explantations with 10 good reasons for using this design.
Chart for crossover filter values, notes.
Microphone recipe for making a home brew microphone for testing with pink noise and a PC with sound card and program.
                              ---------------------------------------------------------------------------------------
Testing the response of  loudspeakers.
I have the right test equipment to analyse what is wrong with the response of any speaker and to make sure
any design modifications I do will *always* ensure they sound better when I am done. After making many initial
calculations on crossover component values I then build the speakers and measure the acoustic levels. I then adjust
L, C and R values until the response becomes as flat as I can obtain without using too many crossover components.
I always run some music to see if I have done any good, and if it does not sound well I may
shift the Xover frequencies marginally to get a slightly better sound.

I use Zobel RC networks for equalizing all the impedances of all drivers and usually use what are over damped LC or CL filters that have a gentle rate of attenuation and low phase shift up to the -3dB point and then the rate of attenuation increases to an ultimate 12dB/octave attenuation rate at an octave away from the -3dB point.
The SEAS drivers have such a flat response contour that there is never much need for extensive LC resonant
compensation series or shunt filters.

Before I got a computer in 2000, I made a purely analog test rig for testing the response of  speakers accurately enough to reveal all the peaks and troughs in the response along the audio band between 20Hz and 20kHz in a room, without reliance on an anechoic chamber.
Anechoic testing is very accurate but what listening room is ever as sonically dead as an anechoic chamber?
The method I used involves using a "pink noise" signal fed to the speakers under test and filtering the sound energy at 33 different frequency bands along the audio spectrum between 20 Hz and 20 kHz.
With a calibrated microphone and amp I measure and record the level of energy in each band and plot a graph of the response.
I take response readings at 4 different microphone positions and average the levels expressed in voltage dB to obtain what is a fairly good indicator of the actual acoustic output from a speaker.
The 4 mic positions are usually at 3.5M +/- 0.5M from a speaker and nearly on axis and between 0.7M and 1.2M above floor level.
The response will vary slightly with each of the 4 mic positions because even with randomly varying
test signal frequency and phase shift and level there is still some effect from room resonances and reflections.
Resonances in all domestic rooms will always prevent making accurate graphs of responses using steady musical tones.
The results I get in my listening room which has very nice acoustics and a good size suits 95% of people since their rooms are similar, and with similar resonances and carpet/furniture, and they find they have little use for a tone control or graphic equalizer, and there are no inexplicable absences of some musical frequencies or irritating  peaks or resonances at other frequencies.
The bass will be a tight foundation for the music as the speakers allow, and treble will not be too low or too bright.
Unlike so many speakers made to produce too much treble on-axis, mine always sound just right on axis with good imaging. 

In layman's terms, pink noise is a test signal that when played through speakers moderately loudly
sounds just like standing beside a nice big waterfall. This sound is generated in my test gear by passing a tiny current
through a transistor in the opposite direction to normal flow and amplifying the noise to line level for a power amplifier.
Sound energy in this signal contains all frequencies changing randomly in level and phase and frequency between 20Hz and 20kHz and has the same amount of power for each of the 33 selected bands of frequencies.
During the testing regime the effects of variations of capacitor or inductor values used in the crossover filters can be closely tracked to ensure that the final design will give the flattest response possible.
All my speakers are not just designed on a computer, or reliant on simplistic calculations but are thoroughly tested and
trimmed for response flatness in the real world. I have tested many mass produced speakers and very few ever measure as well as my own. The only time I have needed the computer aided design process was for matching the woofers
to a ported box. There are plenty of box matching programs now available. Win ISD is a good one to use.

My test room is my own lounge room with average sizes 7.5 metres x 8 metres with high sloped ceilings and carpeted concrete floor is what most visitors say is an ideal room for hi-fi, and when I get speakers to sound right in this room the speakers
usually perform very well elsewhere because the response has been tailored to suit the common listening position in a similarly furnished room.

If you have read closely, you will know I am not completely in favour of using a sub-woofer.
However, I have two clients who both started with Vienna Acoustic 'Mozart' speakers and both independently
concluded their sound could be improved by adding the missing octave below 60Hz.
Both had tube amplifiers I have built for them which they prefered to the solid state amps thay had previously used.

Subwoofers recipe...
The driver was a single 300mm Peerless XLS subwoofer driver. It is rated for 90.6dB/watt/M at 117Hz
but at 30Hz produces 83dB/W/M only, so you need a good amp for between 25Hz and 60 Hz.

As a guide, for most sub bass frequencies in ordinary music an amp with at least half the power of the main amps
would be ample for most people.
I have measured the average sub levels in classical and jazz and it is much lower than the main channels But where youth are desirous of hughmungous bass sensations the sub needs to have at least twice the power of the main amps. The sensitivity for sub bass is much lower than above the crossover point because a large amount of energy is produced by the port on the sub and not by the front of the cone. As cone dia reduces the need for power increases as frequency reduces.

The sub enclosure I have made with 33mm thick MDF has an internal volume of about 86.5 litres not including the port which can be a 100mm dia pipe of 380mm long, or a rectangular port 88mm x 88mm x 380mm formed
in a corner of the box with scraps, or a "shelf slot" port of about 320mm wide x 24mm thick x 380mm long.
The port shape can be any shape as long as the cross section area and length are the same.

The internal box dimensions were 530mm x 510mm x 320mm.
In the samples I made I used a "shelf" port, so that the port  "hole" was slot 24high x 320mm wide
and 380mm mm long and using a spare piece of 33MDF to form the "shelf" across the 320mm direction.
I had a joinery shop cut up a sheet of 33mm thick MDF for me and at my workshop I glued the peices together by standing them up on each other on a generous bead of glue while keeping everything perfectly square and taped together as I went.
When the 4 sides and and bottom were glued, I waited a day for glue strength then glued the
port shelf and top which was then ready for cutting out the speaker hole with a jig-saw.
Next day I carefully drilled lots of 8mm dia holes around the joined sheets about 70mm deep at 120mm centres
to allow 80mm long dowels to be slid into the holes on plenty of glue to hold the glued panels together better.
No screws were used.
The ends of the dowels were sawn off and planed smooth, holes filled, all external edges were given a
pencil round and all well sanded down. I applied 3 coats of a water based grey metallic acrylic paint.
The driver and its terminals were installed.

One channel of a solid state amp was employed to power the sub.
I built a special active filter to accept the stereo signals from the preamp and fiter out the bass signals
as a mono signal with 3 switchable cut offs could be chosen at 30Hz, 45Hz, or 74Hz.
The initial attenuation rate past each pole is gradual because there are 3 cascaded 6dB/octave
filters each driven with a simple emitter follower solid state signal amp, but an aditional fixed filter with a pole at about 200Hz gives an ultimate attenuation rate of 24dB/octave.
There is not much sound below 50Hz in most music. The sub should be set up in what is the best position
to give a flat response at the listening seat between 25Hz and say 100Hz, something which is very
difficult to achieve without properly measuring the response. Most audiophiles will just guess the
box into a position that can be approved by their suffering wife, who probably is at a complete loss to understand the need for a sub. But at least she might get a nice place for a vase of flowers or a lamp, or heaven forbid, a picture of her mum.
Phase of the sub output can be reversed by swapping speaker cables, and thus get a better
transition of response between low bass and sub bass.
While listening to a variety of music, the levels for the sub and the adjustment for cut off frequency can be
made for the best sound without sub "bloat", or too much sub bass, which sounds dreadful.
When the main amp is switched off so only the sub is left to work with bass signals, there should only be
a bit of what sounds like very unmusical rumble, with voices only just discernible when the 74Hz cut off is selected.
One learns easily most people can survive without a sub-woofer for music. But a sub is good for movies on the HT system because movie makers have deliberately engineered low bass signals into the sound track.  For myself, I prefer a better story with less explosions.

Here is the design sheet for a single unit 3 way full range speaker.

Design sheet, 3 way speaker.

The above speaker design features the following :-
1, Sloping sides to make the speaker tall enough for good treble midrange height,
2, Has enough volume for the bass and midrange rear chambers,
3, Has some freedom from standing waves between sides inside the enclosures,
4, The speaker is unlikely to be toppled over to kill a child,
5, The project extends the woodworking capabilities of would be joiners wanting to improve by being forced to work with
angles other than a boring 90 degrees.
6, Have slightly better appearance than rectangular boxes,
7, Have some time alignment with the front baffle lent back,
8, Allows someone without a fetish for piano black or exotic and endangered rare species of timber timber finishes
to use cheap but effective material which can be painted any colour to suit she who must be obeyed.
9, The "pointy top" and well rounded small amount of baffle around the tweeter will aid dispersion, reduce beaming
and improve imaging.
10, The angle chosen for the "lean" of every side off the vetical is the same so setting out the cutting of the
sheet of MDF or other material is fairly waste free.

The panels can be cut from a single 1,200mm x 2,400mm standard sized sheet of 33mm thick MDF.
Each edge will need to be slightly bevel cut to make it all fit properly together and even though the above looks quite simple, the woodwork is quite tricky because of the angles of cut and bevels involved. Some DIYers will probably curse and swear when they return to the timber store to buy more after cutting something wrong.
As I said above,  I shan't imitate or promote mediocre speakers of the world while I am alive and nor will I promote
your tendency to be lazy.

Really keen speakerologists will use one layer of 16mm+ plywood board to build a box which is less on each overall dimension. Then they can fix in in the bulkheads and braces easily. Onto this first box there is a well flattened cotton cloth
glued on with no overlappings with generously applied latex based glue.
A second finishing layer of 16mm MDF which has a different density to the plywood is then latex glued over the cloth covered first layer. This will produce a box which is much less resonant than plain 33 thick MDF material, however with 33mm+ wall thicknesses the resonances are reduced a lot, but with more flexible laminated panels with the same weight overall, the woodwork resonances will be minimal.
When all the latexing is completed and cured after some hours any gaps in the joins can be filled with epoxy
glue and the external panels dowelled and sanded smooth for a paint finish.
There is little advantage on gluing lead panels to the inside of the box unless it is done using 2mm thick and cover 50% of the areas of internal front, rear and side panels. Neutral cure silicone should be used to hold it in place.
Extraordinarilly really keen diyers will seek out the type of material used by the makers of Wilson Watt-Puppy speakers
or cast their enclosures in reinforced concrete or build them out of sawn up concrete footpath slabs
which are siliconed together. There 1,001 and one things that are all going to be better than
plain old thin MDF board or plywood.
The basic idea is to make boxes that are acoustically dead. One client of mine built a very good pair of enclosures
using 35mm thick Australian blue gum planks normally used for roof trusses.
This material has a density slightly more than water because I put a small offcut in a bucket and it sank,
so it must have been over 1,000Kg/cu.M.
Most MDF and pine based plywoods are between 500 and 700Kg / cubic metre.
But whatever real timber is used it must be kiln dried, not just air dried lest it move too much after the speaker is built.
My own speakers have solid real timber planks for the front and back which is fixed on a foam gasket to allow
expansion and contraction and thus prevent cracks appearing where timber is fastened to to opposite direction grain.
Speaker enclosures can certainly challenge the woodworking skills of anyone, because they must be made to avoid resonance,
lest they end up colouring the music. Many diyers do not appreciate trying to avoid the philosophy behind a violin.

I used Win ISD to confirm that the chosen Peerless drivers are not a bad set to use.
The top chamber is about 7 litres in volume, about twice the minimum size recommended for the 5" midrange.
This sealed chamber should be filled with high density polyester wool which is normally used for acoustic damping panels
fitted into timber walls betwen frame studs, but the wool can be light density insulation bat material packed tight enough
to have twice the loose volume crammed into the enclosure. 

The bottom chamber is about 47 Litres allowing for ports and braces etc, and about right to get a good bass response
from many modern 8" dia speakers such as the Peerless 850136.
The wool filling should be light density and fill should not be crammed in but about 1/2 fill the
ported chamber without obstructing the port oprning. It should be secured with a few staples and threaded up around the braces of the enclosure.

Another driver which has a cast basket and looks better is the 850490 but the response isn't quite as good.
The Win ISD predicts that what I call the reference response of 0dB is flat between 40 and 50Hz, then rises +3dB to 300Hz.
The response below 40Hz shows -3dB at 31Hz.
The midrange I chose is the 5" 850489 with a phase plug. Its response in its midrange band is at about the same level
as the bass speaker level at 40 to 50Hz.
The crossover F should be at 300Hz  and 3kHz and the exact schematic for crossovers can be designed by trial and error to get the best seemless transitions between drivers; sometimes an overlapping of the crossover points gives the best bass.
The crossover point of around 300Hz will remove the rise in response from 50Hz to 300Hz of the bass speaker.
The HF dome tweeter is the 1" dome HDS unit, Peerless 810653 and this should be as good as anything from Seas.

I have not provided everyone with a crossover schematic. The last time I built a very similar speaker was with slightly different Peerless drive units using a 10" CSX Peerless bass speaker 850146 which tended to have too much bass.
Serious DIYers or speaker makers attempting to build such fabulous speakers as decribed here should have a solid knowledge of passive LCR filter behaviour and own a good test method for getting the response to sound well and measure well, and have a sixth sense  required to understand the vaguaries of coping with the unpredictable acoustic results when applying logic and commonsense to speaker production. Things don't always work out as imagined.
But for those who are going to attempt their own crossovers here is a table with high and low pass filters.

LCR filter chart.

This chart is useful although there is not a combined midrange BAND PASS filter shown which may have
LF pole and HF pole ( a pole is a -3db cut off point )

For a midrange one may series a high pass with a low pass filter to get a bandpass character.

For the crossovers to give the wanted attenuation rates the signal source must be low, and preferably
1/20 of the value of R shown, so that where one may have 6 ohm speaker impedance the amplifier output resistance should be 0.3 ohms or less. 0.6 ohms could be used but lower rates of attenuation begin to be seen.
The low Ro needed is quite feasible using vacuum tube amplifiers.

Speakers have varying impedance with frequency and for the electronic response to be as desired the R value following
the L, C, or combinations of L&C must relate to the L and C values as the formulas indicate.
Using say twice the R value shown will nasty peaks in the response, using half the R value will give an initial slow roll off
and may result in a response dip near the crossover region.
The various filters shown give their phase response at the -3dB response point.
It is extremely difficult to have a perfectly smooth response with 12dB/octave filters and it is an art to get a crossover
optimised for a smooth response. Any change made to the crossover frequency will require different
LCR values, so if just one L value is changed from 1mH to 1.5mH, then the C must also change to suit the given R value.
The impedance of each driver in the box must be measured and plotted before any construction of the crossover is possible. The inductive nature of drivers in their upper bandwidth means their impedance will rise with frequency.
The impedance at the middle of the bandwidth of a driver is usually mainly resistive in nature.
To counter the effects of the rising inductive nature of a speaker a series R&C Zobel network is connected
across the driver to that as F rises the impedance remains flat from the midband value.
The R value is then about the same as the midband impedance value.  The reactance of the C will be equal to the R value
at the F where the driver  impedance rises by 3dB. This is a guide only, and values of R&C must be trimmed
for a flat impedance that will approximate a resistive load which is then the R value shown in the crossover chart.
The L&C values for the crossover must be designed around this R value. 

Although I have my own home built pink noise generator and switchable bandpass filter the easier way to know what you are doing is to use a PC based program to display the response from a calibrated microphone sending a signal to a
PC sound card with pink noise from the program used for such testing.
The response will be a graph line on the screen and all a lot faster than using pencil and paper to make a series of dots on a page of a notebook to describe the response.
I use pre-tested and calibrated electret mic inserts from Hi-Q International in Melbourne in my home built microphones.
They cost less than $2 each. But other companies may sell electret mics with guranteed +/- 1dB response accuracy between 20Hz and 20kHz.
Such inserts are about 10mm in dia and two wires can be soldered to the rear and then the mic is fitted to a 300mm long length of 12mm copper water pipe. The mic is padded in cotton wool where it is flush with the pipe end to prevent
hard contact with the pipe. An RCA socket is fitted to the far end of the pipe, and the cost of such a good mic is $10
and some time. I use an old solid stage mic stand  and clamp to alter mic height and position. I have 3 such mics, and occasionally a change mics to check on each other for drift, but after several years there has not been much change and I continue to be able to reform older speakers and have them sing and build the occasional new speaker that is second to none.

Back to Index Page.