Conference hips::uk_audioo

    I know this note is old but if anyone is has any knowledy of this
    sub-woofer I would be interested.
    
    Two questions:-
     
    1) Is this wired in the typical Sat/Sw configuration. Ie. amp - SW -
       and out to Satellites.
    
    2) If all the signals pass through the Sw crossover, what is fed
       to the Satellites in terms of frequency range
    
    
    

    re .1....
    
    re Q 1) yes
    
    re Q 2) It depends on the SW - they have different models depending on
    what you are going to use as satellites. They produce a very good table
    of SW vs satellite vs price.
    
    Dave

Hello.

Following listening to some Live'94 demos, my existing system has many 
fine qualities, except really deep bass.
     
I have a relatively low powered Class A amp driving clear and open medium 
efficiency Linn Index II's.  The sound is great for music, but could do 
with a little help especially in my living room which is 22 by 14 foot 
with very high ceilings.
     
I have read some excellent reviews of REL Acoustics Strata, which is 
their smallest active subwoofer.  It can be driven by a line level or 
speaker level signals, giving a high pass filtered signal back into the 
main speakers.  Some details look interesting as it acheives a claimed 
flat response down to 20Hz using a 10 inch cone driven by a 60w MOSFET 
amp.  Unfortunately it's priced at the limit of my reach at �500.  I have 
a local hi-fi dealer who stock it and can allow me a demo with my own 
Linn's if I take them in. (e.g. Awards Home Entertainment Oct 1994
                                 What Hi-Fi (Listings page) Oct 1994
                                 Your Own Home Cinema (Best Buy) Autumn 
				1994)
     
Since I have fair woodworking skills, have used MDF in the past and am 
quite practical, I have considered purchasing a 12-18 inch sub driver for 
up to UK �160 and making either an infinite baffle or transmision line 
enclosure.  Initial calculations for enclosure size for the 12 indicate 
the thing might fit inside my bigger sofa!
     
I would be able to test-run the thing using a 70wpc amp from my car which 
includes a low pass filter.  If I like the sound then I intend to either 
purchase a second hand Pioneer A400 amp and use it to drive the sub with 
one channel and my rear channel with the other (unless I can bridge it 
all for the sub).  Another amp option is a UK �50-�70 integrated amp 
kit+psu available from Maplin which should be able to provide over 70 
watts into a single channel.
     
I'd like the idea of smooth response to 20hz for around UK �200.  Maybee 
with the new box being hidden inside the coffee table or chair. I can 
afford the REL the first week of December if the suggestions are to wait 
and get something properly made.
  
I am looking for deep, taught bass to underpin the Linn's and maybee to 
lowpass filter them to allow the A1 to get on with the music while the sub 
amp handles the under 100hz stuff.
   
Any suggestions as to obtainable response, a comparison to the commecial 
active subs more expensive hi-fi cans and any enclosure designs are 
welcome.  I don't understand how REL can get the response they have.
     
     
Best regards,
Rob.
     
     p.s. what response do you think I'm getting from my little Linn's?

Building a good subwoofer is a very difficult job, which is why most speakers
on the market either cost the earth, do the job badly or give up on the bass, and
most subwoofers are not worth buying!

The Linns are able to handle bass, they just roll-off too early (around 70 hz
I think). I suggest you do not use the filter as this will damage the signal path
to your main speakers - in fact the more expensive RELs do not have such a
filter.

The beauty of the REL active approach is the Subwoofer output can be setup to
intergrate with the natural roll-off of your current speakers thus filtering
the bass from your speakers is not required.  REL also recomend the subwoofer
is run from the amplifier output as it then takes the sound of your amplifier
and sounds faster.

I have heard a REL Strata compared to the other REL models and it does most of
the job of the others.  It sounds less controlled than the top of the range
�3000 studio but then it is alot cheaper.

Andy Leigh

(I own a pair of Linn Kabers and use them succesfully with a REL Stentor)

Thanks Andy,
I am very glad to hear from another Linn owner resorting to such an 
underhand tactic to steal some bass!

Having now read:
	An Introduction to Loudspeaker and Enclosure Design,
	The Loudspeaker Design Cookbook,
	Bullock on Boxes,

and a few reviews in Home Entertainment and now your comments.  I too 
understand some of the difficulty in designing a truly great non-boom box. 
I recently found out the local REL stockist allows a one week money back 
period, so I can try the Strata in my own room.

My only worry is that one review said it can only cope with Small/Medium 
sized room air volumes.  My room is 23x12 feet with a 11 ft high ceiling, 
probably giving a Medium/Large room volume compared to average.

I don't want high volume levels, but solid clear music and movies.  I am 
even hoping to have the Linns running at a lower volume level, as I think I 
was cranking them to compensate for the lack of Bass at low listening 
levels.

The Stentor is the UK �800 model isn't it?  If the Strata isn't enough I 
might look at that, but I think it would have to be very special.

Thanks again,
Rob.

Rob,

The Stentor has been re-resigned recently and is now �1500; the �800 model is the
Stadium.  I noticed that Hifi News has an article this month comparing the Strata
against the Stentor.

I went to a dem at this year's Hifi Show at the Ramada where all the REL models
were compared.  The room was much larger than yours and the Strata did quite a
good job, but the bigger the REL the better it controlled the room thus the
tighter the bass. The dem did not show any the Strata sounding lighter than the
others.

Trying one for a week sounds like a good idea as it takes ages to set is up
properly.

Andy Leigh

That's what I'm finding now.

Yup, I have the Strata now until Saturday, when I have to make a decision!

I find the Strata goes very deep.  I am having trouble determining which 
Lowpass filter setting to use.  The fine and coarse controls give a 
low-pass cut off at any 3Hz between 30 up to over 100Hz.

I have a Test CD with tones at 10Hz increments from 20Hz to 70Hz.  My Linns 
are good down to somwhere between 40-50Hz where the Strata extends this 
down to between 20-30Hz in my room in all the places I've tried it in.

On the plus side, 
I like the weight it gives to a kick drum, and I can now hear all the 
chords on a bass guitar, not just the higher ones.  Tori Amos' grand piano 
stuff sounds bigger than her other tracks.

Of course it's plain sailing for Movies (off Video, mostly off Laserdisc 
and very occasional NICAM TV material).  Gives a completely different feel 
to the train crash in The Fugitive since you can feel the derailed train 
sliding up behind Harrison Ford.   But I can get this effect with some 
cheeper home made boxes, I want subtlety for music.

On the negative,
I'm not sure if the Strata added blurs the higher notes of a bass chord by 
overpowering it's fundamentals.  I'm comparing this to the sound I have 
heard from various bass players in live bands.  This still occurs even 
if I lower the sub gain so that the 30Hz sound level is much lower than the 
40,50,60 and 70Hz.

Also it's strange with the CD test tones, there's a very significant output 
from the sub (alone) at even 500Hz, even with the filter set at it's lowest 
setting off course 1, fine 1 equating to about 32Hz in the manual.

I seem to be getting a peak at 40 and 50Hz.  Is this a production fault, or 
is is just the slope of the filter, or is it just where the 60watt MOSFET 
can't control the cone as well as my 35 watt Class A amp and smaller Linns. 
 It seems too wide a spread to be a resonance peak of my listening room.

Much as I hate to admit it, I think that UK �800 on the Stadium is a little 
out of my budget.

Cheers,
Rob.

Rob,

If you are finding blurring of notes I believe you have the gain set too high.
I first setup mine so the volume of a test tone was at a consistant level.  This
was achievable except one resonance in my room, however, I found this did not
work as it blurred the notes.  I brought the cut-off point and gain lower until
the main speakers output was not affected and there is no obvious output from
the REL.  The easiest way to detect if the REL is adding any input is to switch
it off and compare the difference.

I bought mine for AV use, but thought I would see how well it could intergrate
into the rest of the Hifi.  After a week of playing I had set it up well enough
to find the music diappointing when it was switched off, thus it stayed.
I have set it up three times since then each time I feel I have got it better.

To make a subwoofer "silent" at higher frequences it must be built massively to
prevent cabinate resonances.  This is one of the reasons why the RELs get more
heavy for better models.  I think the Strata is about 19kg whereas the Studio
is about 78kg!

This month's HiFi News has several graphs of the frequency response of the Strata
including how the treble is rolled-off.

Andy Leigh

From this string it is not clear to me whether the main speakers have been 
hipass filtered or not.  I assume not.


I find it useful to insert a first order hipass filter into the input path for 
the power amplifiers for the main speakers.  I personally encourage users of my 
line level crossover to go for 100Hz, pretty much regardless of the main 
speakers.

This has a few effects:

It unloads the main speakers of low frequency energy demands which often 
improves overall performance by reducing cabinet exitation and driver doppler 
distortion.

It unloads the main amplifers of current demands for low frequencies which 
usually improves sonic performance, especially of low wattage amps and tube 
amps.

It reduces overlap and thus interference of sounds from the main and subwoofer.


I havent been pleased much with lower frequencies (like 50-60Hz).   I thin 
the reason is multifold, but mainly it does less of the first two effects.

I do not usually encourage higher crossover points. That is mainly because of 
the problems of midrange sound coming from a subwoofer unless a brick wall 
lopass filter is used (and the difficulty building a good one).


Walt

Thanks Andy and Walt.

I bought Hi-Fi News yesterday, to bottom line is I'd like the Stentor, 
useful graphs in the review!

I've got the Strata's crossover down as low as it will go (30Hz) and I 
still get a combination peak, where the the low-pass filter only slopes at 
12dB per octave.  My Linn's appear to begin to slope off around 50Hz, so 
there's much overlap.

I am currently running the Strata with a Line level feed, but it does have 
another option of feeding it a Speaker level signal, which it will 
re-amplify for the bass and provide a simple 6dB per octave high-pass slope 
beginning at a)50Hz b)100Hz, both dependant on the actual speaker 
impedence.

I had avoided this because I assume my main amp degrades bass within.  
However I'm going to go and spend �25 on some extra cable and 4mm plugs to 
hook the bass units up via this passive high-pass provided by the Strata.

Thanks Walt, I have read about the doppler effect.  Perhaps the passive 
50Hz high-pass filter may provide less harm than these problems you 
mention.  For a future project, would you let me have any instructions you 
have handy for your active filter you mentioned?

I am also going to have to go out to a DIY place and buy a big concrete 
paving slab to place the Strata on.  This will be sited on 4 tall nails 
which I'll sink into the floorboards, to hopefully lessen any LF problems 
into the basement flat below me, as new renters moved in last night (since 
a 4 month empty period)! 

My Linn's are currently bi-wired.  I was planning on just routing the BASS 
tagged feed via the REL's filter.  Is the bi-wiring likely to be a problem. 
I have lost the factory fitted link to revert the Linn's crossover back to 
a 2 core pair.

Hey Walt, I still haven't given up on a pair of Isobarik subs, and will 
certainly continue the project if I can't live with this REL.  

Many thanks,
Rob.

I think you will find the final performance better when you feed the subwoofer 
crossover with the line level signals, But if it doesnt cost too much for the 
speaker cable, try both ways.

Since you already have the line level lopass (active crossover) for the sub, 
all you need to do in my opinion is roll off the main speakers a bit. I prefer 
to keep the hipass as simple as possible and have found a single pole (one 
series capacitor) works well.  I will grant you that when a 3rd order lopass 
and 1st order hipass are electrically summed you will get some amplitude 
deviation around the crossover point, but I found this preferable to the added 
complexity of a 3rd order or higher, or no filter at all.

Building a first order hipass filter for the inputs of a power amp is a snap.
You could take the approach that you want it tunable from say 50-100Hz, or you 
could pick 2-3 frequencies and select suitable fixed resistors for those.

I usually use a .1uf MIT RTX for the series capacitor. It is the finest 
sounding capacitor I have found.  To determine a suitable loading resistor (the 
load resistance determines the filter crossover point) you need to know the
frequency you want and the input impedance ofthe power amplifier.

The basic formula is:

 RL= 1/(2Pi*f*C)

RL is the combined input impedance of the amp and a load resistance.
f is the desired -6db point or crossover point
C is the capacitance

When you plug in a couple values for frequency and capacitance:

  1/(6.28*100*.1^-6)

	-and-

  1/(6.28*50*.1^-6)
 

You get

15.9K ohms and 32K.


These would be the required load impedances to get crossover points of 50 and 
100 Hz.  If the power amplifier has an input impedance (Rz) of 50K you can then 
solve for the load resistor (R1) to get the desired RL by:

R1 =  1/((1/RL)-(1/Rz))

for 50Hz Rl would be 88.9K  and for 100Hz it would be 23.5K.

So in this case a 100K pot would cover these and more, or a selection of 
switchable or insertable resistors would do the same job more cleanly and 
cheaply.



Assuming the power amp does not already have an input coupling cap and that 
you do not want to modify the amp itself I would implement such a device in 2 
small metal boxes that could each be plugged directly into the power amp inputs 
(using a male locking collar RCA plugs, like the Vampire, mounted to one side 
of the box and a pair of panel mount female RCA jacks on the other).

For demo purposes however you could forgo the boxes and just mount a film cap 
between male and female RCA center pins, the R1 resistor or pot between the 
male RCA center pin and its shield, a wire between the male and female RCA 
shields, ans some electrical tape to keep things from shorting out.  Plug that 
into the power amp and the interconnect from the preamp into the other end.

If the amp uses an input coupling cap and you are not adverse to making mods, 
then all you have to do is add or replace suitable parts. For example if the 
amp uses a 10uf input cap, you can forget loading it - replace it with a .1 and 
go from there.  This might even improve the sound further if the original cap 
was not the greatest to begin with.
 
Walt

Thanks!

I tried the Passive high-pass filter in the REL last night, 
it really seemed to tame the peaking resonance which has troubled
me.  Problem is, as predicted it made a real hash of sound through my 
main stereo pair.  I'm anxoious to build Walt's suggestion for a 1st order 
active high-pass.  I have to do it quickly since tomorrow is the last day 
I have to return the REL under the trial agreement.  I've got the day off 
for a lot more listening.

I've obtained a catalogue from a local Maplins electrical supplier.  
They don't seem to list such an animal as a MIT RTX capacitor.

I have found the following types are carried with a .1uf rating, 
I will try to mail order a MIT RTX for a long-term solution, but in the 
meantime could anybody possibly recommend some of these available which may 
work adequately for a demo

Capacitors .1uf:
Voltage	Tolerance	Type		Price			My pageRef
(DC)
16	-20+80%		Minidisc cementcoated ceramic		493
25	+-5%		Polystyrene film  and tin/lead foil Philips 494
35	+-20%		LC Tantalum resin-dipped.		497
50	+-10%		Monocap plate ceramic			493
50	-20+20%		Disc ceramic				493

63	+-10%		Mini Metalised Polyseter Film WIMA MKS 2  495
63	+-20%		SubminEster				495
63	+-20%		Minielect				497
100	+-10%		Monolothic Ceramic Resin Dipped plate ceramic 492
100	+-10%		Mylar Film				496

250	+-5%		Poly Layer Siemens			494
250	+-20%		Polyester 20 Low-inductive wound cell 
			of PETP film				495
250AC	+-20%		Interferance Suppression metalised film 
			PETP Cap				496
1000	+-10%		HV Cap Industrial Capacitors Metalised 
			polyperopylene				496

I cannot find my amp's input impedence in it's specs, they merely state 
200mV - 80dB.  I am racking my brains to remember how to measure and 
calculate impedence from my O'level.

In a contingency, I have plotted  Walt's supplied R1 formula against 
Amp input impedences of 30-100 Ohms, and noticed a 27K ohm R1 gives me a 
-6dB point between 110-75Hz, all of which should be within the range of the 
active lowpass within the REL.

May I again list some Resistor component types available to me in this 27K 
ohm rating:
	Carbon Film 1.8W
	Metail Film 0.25W
	Metal Film 0.6W
	Metal Film 2W
	Wirewound sealed in cement 3W

Many, many thanks for any timely advice, since I'm out tomorrow.

Cheers,
Rob.

I assume in the following...

>In a contingency, I have plotted  Walt's supplied R1 formula against 
>Amp input impedences of 30-100 Ohms, and noticed a 27K ohm R1 gives me a 
>-6dB point between 110-75Hz, all of which should be within the range of the 
>active lowpass within the REL.

	...that you meant to say 30k-100k Ohms?


You have 2 caps that appear to have potential:

25      +-5%            Polystyrene film  and tin/lead foil Philips 494
1000    +-10%           HV Cap Industrial Capacitors Metalised 
                        polypropylene                          496

I have some experience with the first one since that is what they use in the 
Philips CD players (different values) low pass filter.  It is a good notch down 
in terms of sound quality from the MIT polystyrene, but it is probably better 
than most of the others on your list.  

The second one is interesting because it has a decent dielectric and it is 
generally believed that higher voltage constructed caps sound better than 
identical ones made for lower voltages.

MIT is US made and may or may not be available to you locally.  I think Michael 
Percy and Sonic Frontiers both export and could serve you. Other brands that I 
find very close are Ultra-Cap from IAR and Sidereal Cap. I think I recall 
someone mentioning the Sidereal is distributed in the UK.



For resistors, you can go with:

        Metal Film 0.6W

The wattage doesnt need to be that high (.25 is plenty), but there is evidence 
that higher wattage resistors of a given technology are quieter and sound a bit 
better.

Here I recommend you go with Holco H2 or H4 if this works out. Over here they 
are not at all expensive, and I think they are British made so should be 
available to you.  Again Michael Percy and Sonic Frontiers carry them on this 
side of the pond.


Measuring input impedance may be as simple as connecting an ohm meter to the 
powered off input and ground of your amp.  That isnt always true though because 
if the amp uses a coupling capacitor or low input impedance gain devices the AC 
impedance will me masked to your DC measurement.  It is best to either ask the 
manufacturer or check any magazine reviews of the amp that provide a spec 
summary. It is possible to measure the impedance fairly simply if you have a 
signal generator and AV VTVM and a resistor of around 100K Ohms (known value) 
using the voltage divider rules.

Best of luck.
Walt

Thanks again Walt.  May your cone never sag or your doppler appear in 
public!

The 27K ohm Polystyrene film Philips I specified was an error in the 
catalog table, so I had to go for the 1000volt Industrial Metalised 
polypropylene caps at UK �1.20 each.

I am waiting for a component importer to call me tomorrow, about the MIT 
polystyrene and his ad states he stocks Holocos too.

I'll make them up into a small case per RCA phono tonight.
Thanks.
Rob.

I sucessfully built the 1st order filters for a test demo.  They really 
worked well to tame the crossover overlap around 40Hz, degrading the stereo 
pair MUCH less than the passive crossovers built into the REL.

I called Musical Fidelity to find the input impedence, which is 47k ohms 
meaning the 27K ohm resistor giving a 93Hz high-pass frequency.

I gave a REL Strata a good chance on Friday and Saturday, but the bottom 
line is I prefer the music out of the Linn's alone.  The Strata gives a 
firm bottom octave and is impressive, but I just feel the thing makes the 
music plod a bit.  

Perhaps it's due a lot of mass loading to the single long-throw 10inch 
woofer driver to get it's resonant frequency down to allow the 40 litre box 
to reproduce around 20Hz.  As I would expect the on-board 60W "DC coupled" 
MOSFET to keep the pace of my A1. 

I felt a little sad taking the Strata back on Saturday with no quible 
whatsoever from Reading's B+B Hi-Fi store, or it's very helpful Manager 
Dave.

I talked to AP Electronics, a division of AUDIOKITS Precision Components 
about the Holcos and MIT capacitors.  They don't know of anyone in 
the UK carrying MITs but swore blind their Soflex equivalent beat the MITs 
in a blind listening test.  

A pair of Soflex and Holco H2's have just arrived for UK �5 including �1 
postage.  I will build another active filter to compare against the 1000v 
cap and standard resistors in my demo version.  Thanks again, Walt.

So, back to the drawing board at designing a pair of isobaric boxes. I am 
inclined to lowpass the subbass very low, but may still yet need the 
service of Walt's nice and simple 1st order highpass above.

One annoying thing, is that my Amp doesn't have a separate pre-amp output 
so I'll need a filter for each phono input line.

Walt, do I take it that a 1st order low-pass is more complex than your 
high-pass?  My books doen't seem to cover Active filters for more than a 
passing mention.

My enclosures will probably be 18mm MDF with the rear and sides being 
surrounded by 9mm MDF with 9mm of sand in between.  

I'm looking at isobaric 10 inch drivers, in a ported enclosure if I can 
find an electrician to help measure the Theile-Small parameters of the 
supplied drivers.

Failing that, I may go the the manufacturers numbers for calculating the 
port. Or just go for a closed box design with a .707 system Q.

Cheers,
Rob.

>Walt, do I take it that a 1st order low-pass is more complex than your 
>high-pass?  My books doen't seem to cover Active filters for more than a 
>passing mention.

I recommend a 3rd order lopass.  I know that sounds like some sort of heresy 
but I have reasons.  First among them is that a first order lopass will not 
attenuate higher frequencies (e.g. above 200Hz) enough to keep from muddying up 
the midrange.  

I have found that it is not difficult to tune a 100Hz 3rd order Butterworth 
lopass with a single order hipass. I start at 100Hz with the hipass and have 
found it is seldom necessary to do any adjusting unless the main speakers have 
a significant bump below 100Hz or a suckout above. 

By the way, the hipass cutoff frequency need not be precisely 100Hz, or 80 or 
60 etc.  If you map the single order filters attenuation curve you will see 
that a shift of 10Hz or less produces an insignificant change in attenuation at 
any given frequency.


I use a cookbook single stage VCVS Butterworth. I will be glad to mail you a 
schematic of my crossover - send me mail offline for an address or FAX number 
to use.   I am convinced that part of what makes my low pass filter a big hit 
is the choice of opamps.  I know that in theory anything better than an NE5534 
should be unnecessary, but believe me the things you found unsatisfying about 
the REL (plodding base line) could be no more than an underestimate by the 
product designer of the influence an opamp in a filter will have on sub-100Hz 
sound quality.   The absolute best bottom end (tight, detailed, dynamic, 
rythmic) I have been able to produce was the result of replacing NE5534 opamps 
with Analog Systems MA362.  In a system where the sub is lean to begin with the 
Analog Devices AD711 will add warmth and impact with only a little loss of 
detail.  By comparison the NE5534 falls flat.  The TL071, LF356, and a host of 
others sound lean and dry. Of similar importance is a really good DC supply, 
third selection of the right polystyrene caps (not the correct value, but the 
brand/design).  Yea, I know...I had a lot of trouble with this too because I 
cannot understand why it should be so.  I guess I dont hold those beliefs so 
close that it masked me from hearing the improvements that occurred - 
fortunately when I was doing the experiments it was motivated and funded by my 
friend Kwame - otherwise I would have not even tried it.


Walt

Thanks for the thread so far, I have a little news and an audition.


I obtained catalogues from a great many dealers in my search for some 
drivers for my planned subs.  One that I came across was from Wilslow 
Audio, which also detailed their 5 ready designed enclosures.  

I'll get technical here and talk music at the end of this note.

1)What you get
---------------

 Two are push-pull ported compound isobarics, two are single driver ported 
 and the third is a small push-pull isobaric bandpass.  All kits include   
 passive crossovers which appear to have a passive low-pass and high-pass  
 around 150Hz.  I cannot vouch for the quality of the components used in   
 the crossovers.  Cabinet material is 18mm MDF.  Veneer is extra.

It looks like Wilmslow Audio don't aim these at active users, since they 
are listed by their relative efficiency.  

I couldn't resist a little reverse engineering... (I can post the 
Thiele-Small parameters and predicted response plots if anyone wants)
-------------------------------------------------------------------------
 Their �180 CPP kit (as appears to have begun this thread in 114, aug'89)  
 appears to consist of a 30 litre enclosure housing a pair of (�46 each)   
 10in Peerless 250mm drivers in a push-pull isobaric configuration.

.The �279.00 Super PP kit as above but around a 50 litre enclosure for the 
 pair of (�72 each) 12in Peerless 315SWR drivers.

.The �226 CDVC uses a single (�102.50) 10in Volt DVC 220 twin-coil driver  
 in a ported enclosure.  
 
.The �281 SuperDVC is as above but uses a larger (�121.25) 12in Volt DVC250 
 driver.

These appear to have a longer throw and pose the question; 

Q: ignoring enclosure practicality, is a compound of two 46 drivers going  
   to give more control and less distortion than a more expensive single   
   driver?

.The final design appears to have around 6inch drivers and appears to be   
 for car use, with an enclosure the size of a brief case.

-------------------------------------------------------------------------
After a very friendly invite, I took my Linn Index II's and Musical 
Fidelity A1 amp along to audition them all passively driven by the one amp.

2)Kind Room
------------
First thing I noticed was the the demo room was half the length, width and 
height of my listening room.  With the Linn's backed up to the longer solid 
brick wall.  The A1 was connected up and we went for a great little pub 
lunch for an hour while it warmed up a little.

On listening to just my system, feed by a competent Denon CD, 
playing The Brand New Heavies - Brother Sister.  I was very impressed by 
it's warmth and balance in the small room. Much like I had in my previous 
listening room.

3)Relative efficiency
----------------------
I will only detail my listening of the CPP, since all 4 others seemed to 
lack the middle efficiency of my Linns, making them inaudiable.  I would 
have loved to audition them with active crossovers.  Even a non-critical 
friend confirmed that the CPP sounded the best of the bunch in these 
conditions (she also now knows what a warning about 3hours drive, 2 hours 
listening and 3 hours drive means)!

4)Musical detail thru the CPP Subwoofer
----------------------------------------
On plugging in the CPP, I found a little more detail on the mid-bass, but 
perhaps less extension deep down.  The sound was not light-weight, but 
probably didn't have the roll-off hump I assume is designed into the 
Index's to make the smallish box and driver combination sound full.

Surprisingly, it's passive high-pass didn't trash the stereo pair as the 
REL Strata's did.  Overall a different sound, I personally preferred the 
sound in this room without the CPP.  However an advantage to me back home 
will be that the CPP can be sited against a wall, next to the Linns which 
are backing under a less solid large window in a large room and are unhappy 
for it.

Nothing I heard here had the absolute extention of the similar sized active 
REL Strata with it's single 10 inch Volt Driver in a 40 litre ported 
enclosure, with it's active Linkwitz-Riley lowpass extension.

5)Delivering the Goods
-----------------------
I expect the kit to arrive next week, and hope that it will have more 
control than the REL, if not the extension.  I also hope it's passive 
crossover is equally good at home, but expect to build Walt's active 
low-pass and borrow an amp from my car for auditioning.


Since the kit's cost over the drivers alone was only �90, my reasoning was 
to try a single kit, then if still inclined to design my own, buy another 
CPP kit.

I can then design my own, perhaps twin driver bandpass enclosure with one 
driver reversed (either separately enclosed or compound).  I can then 
reference how this sounds against the first CPP design.

If I prefer the bandpass, then remake the first, if I don't then rebuild 
the second as a CPP and hear the improvements of identical sub-woofers 
removing the single point source problems in my room.

6)Tuning by hand and mind
--------------------------
I have found a few shareware PC programs which should work with my PC's    
Soundblaster sound card, which I hope to use to measure the results.    
They give real-time frequency response plots from it's line and mic     
inputs, one also shows the peak frequency.

Q: Does anyone know any method of getting such a card to produce a    
sine-wave of each frequency from set points between 20Hz to 100Hz?  I'd    
love to be able to use it to tune the enclosure posts exactly with the 
drivers, as recommended in every speaker building book.


Any comment appreciated.


Cheers,
Rob.

Q: ignoring enclosure practicality, is a compound of two 46 drivers going  
   to give more control and less distortion than a more expensive single   
   driver?


No clear answer there.  It depends on whether the "more expensive" single 
driver produces distortion at similar frequencies and SPL that are much 
smaller than the less expensive one.  The improvement in linearity for 
compounding should be in the 10-20dbv range (x3-x10 depending on the exact 
configuration) for a given driver. Is the more expensive one lower by x10 or 
more? 


Walt

    Just out of interest, don't Linn Products hold the patent on the Isobarik
    speaker enclosure design ?                        

I've now received, built and lived with the Wilmslow CPP sub for a while.

Presentation
------------
The kit was excelently packed, cnc cut 18mm mdf with all I needed including 
pre-build crossovers, cable and gaskets.  All I needed to find were some 
large g-clamps and some PVA wood glue.  After two sessions it was done and 
dry, looking quite ugly in it's plain sandy MDF finish with grey thick pvc 
drainpipe port tube.  Finish is my last priority, especially as I may 
re-work the cabinet.

Passive
-------
I first wired it as Wilmslow sell it, passively, using the supplied 
crossovers.  They appear to be in-house made copper wound inductors about 1 
inch in height, with capacotors who'se name escapes me.  The crossover 
frequency is around 150Hz.  The left sub-bass channel goes to the forward 
facing internal driver, with the right going (out-of-phase) to the backward 
facing driver mounted on the outside of the cabinet.

The efficiency matching is roughly similar, but only done by ear (when I 
selected this model).  I noticed my Linn's appear to want to be 
out-of-phase to the outward facing woofer, I also noticed this at the 
Wilmslow demo.

The difference was all I had hoped for.  Suddenly I had all the 
fundamentals if all but the lowest guitar notes, kick drums do and the 
dinosaur is Jurassic Park shifts some wind as it's front feet land after 
stretching to an upper branch.

Positioning seems less critical than the downward facing driver'd REL 
Strata.  I can now place the subwoofer on the inner side of my right Linn 
Index.  My windows don't rattle as the Strata made them do when it was in 
this otherwise ideal position, I guess again because the driver and port 
face forwards, not down.

On the downside, I have noticed a slight loss of detail on the upper bass 
note registers and vocals.  The damage is not much and certainly not as 
objectionable as the REL Strata's passive crossover I tried.

I noted that as predicted by the Thiele-Small models, the Wilmslow CPP 
didn't manage a 30Hz test tone.  I'd say the STRATA manages 25Hz the CPP 
only 35Hz.  On some music, slightly less power and extension was being 
managed.  On everything I tried, there was very little overhang or plodding 
of bass notes, everything worked very well.

On the whole this is superb for the cost.  I think my A1 amp drives this 
well, since it can handle tricky loads much better than inefficient 
speakers.  However I suggest anyone who was speakers they liked which now 
sound lost for bass in a larger room to try this out.

Active
------
Here's where Walt should start taking his share of the blame.  After I had 
built his Active High-pass crossover, I just had to try running the CPP 
actively, without the passive crossover to hinder things.

After two weeks, I pulled the bigger amp out of my car.  It is about 3 
years old, known in the UK as a �250 Kenwood KAC-821.  To put it into 
perspective, here the specs say it should deliver:

    Power Output
	Normal (20Hz - 20KHz, 4 ohms, less than 0.5% THD)
					75W + 75W
	Bridged into 4 ohms		150W
    Max Power output (1 KHz, 4 ohms)		
	Bridged				280W
    Damping factor (100 Hz)			More than 200

To compare this to the REL strata, which has only a single driver 
controlled by with what is ambiguously listed as:
    Amplifier type	DC coupled complementary MOS-FET 60 Watts RMS.

So in my Active test, each CPP woofer has available at least the same, if 
not twice the power of the UK �500 REL Strata.  Costing 20 for Walt's 
high-pass active crossover parts, 180 for the CPP and a theoretical 250 for 
this amp which includes an Active low-pass bass crossover.  Interesting as 
it works out about the same cost as the Strata! 

With the amp's variable low-pass at about 80Hz and Walt's  high-pass 
operating at about 90Hz things sounded just right.

The difference.  Wow!  Lower? No.  More controlled and clear, YES!
The improvement is unimaginable, kick drums have power and punch, yet other 
instruments are uneffected by them.  Vocals sound very clear and the image 
is more central (probably due to the passive 150Hz cut-off allowing some 
directional sounds to come from the CPP itself).

I have the Kenwood gain set a whisper off it's lowest setting, giving a 
subtle sub-bass underpin, just as I had the REL.

A friend who had heard them both surprisingly described the active CPP as 
"having more bass".   I myself consider it to have slightly less extention, 
but much more control and detail, more detail in both bass guitars, drums 
and vocals and cymbals.

With this in mind, I intend to build a bandpass cabinet (25-80Hz) to house 
a single CPP driver and see it if actually does manage to get a lower 
response (system Fs Hz), if so, build a second and position each next to 
the appropriate Linn speaker and easily re-match the levels using the 
Kenwood.

I am hoping for suggestions on the bandpass cabinet frequencies.  

From a modeling program (BoxPlot for Windows 2.2c) I appear to be able to 
model 25Hz-80Hz with a reasonable sized front and rear enclosure volume and 
an estimated 102dB peak output trade-off vs an estimated 110dB available 
from the current CPP configuration.  

Since the final pair will then manage 102+3dB (as a pair), I wonder if the 
around half peak volume ability will de detrimental?

Any comments welcome.

Rob.