Castle Conway II on slimming plan

[GeigenPrince]

I have a pair of Castle Conway II's (a three way bass reflex system 45 litres). Due to space restriction in my room I am wanting to refit the speaker system into slimmer cabinets, while maintaining the approximate volume within the enclosures - 45 litres - will this work effectively ?
The new cabinets would be 9.5 inches wide, 13 inches deep and about 28 inches tall.
Could I reduce the enclosure volume to say 40 litres and retune the port, which I think in the present cabinet is 37 Hz?
I am hoping that this particular bass unit (8 inch) would work well in a slightly smaller enclosure if tuned properly. I don't have the specs on the bass unit.
Also, could I tune the port in the smaller enclosure to say, 30 Hz to enhance the bass performance?( I listen to pipe organ music )

Your advice would be much appreciated !!

 

[alan-1-b]

Hi,

seriously, don't reduce the internal cabinet volume,
and especially not if you want to reproduce low frequencies well, such as down to 30Hz for organ,
{I like organ music myself - I'll recommend you a budget priced Bach CD with some good lows, and good performance of some interesting music, and some wonderful higher tones' voicings if you like Bach, ... ?
but I will have to find the CD later to get the correct title and catalog #}.

If you want to get balanced response down to 30Hz, ideally the cabinet volume should be increased a little, eg: to about 50 litres, but you can retune a 45 litre cab. to 30Hz with minimal bass problem for the 8" woofer of the type Castle used.
You will lose a small amount of power handling of the low bass before some distortion sets in,
thus set your listening level by ear if you play loudly.

Cabinet dimensions :-
Very important, especially for balanced bass response, is not have any one Internal dimension a harmonic multiple of any other,
that is no ratios of 1:2 , 1:3, etc ... nor similar ratios which have a common factor, such as 1:1.5 {think of it as 2:3} or 1:1.33 {= 3:4}.
If you can make one of the ratios 1:1.618 this would be good.

I don't have my table of preferred ratios here, but I'll try to find it and Post a three dimensions ratio close to your preferred dimensions if you like.

Yours is a little bit too close to 1:1.5:3 {= 2:3:6}, but,
are the dimensions you quote the External, or the Internal ?
I think you mean External, but as I don't know the thickness of the wood you plan to use, I can't calculate the Internal.

Also, do not place the tuned port at a quarter wave-length internal dimension of the height of the new cabinet.
What is the above and below internal distances from the centre of the port in Castle's design ?

For a tall slim cabinet such as you propose a shelf near the bottom and open horizontal slot in the baffle below is good for bass, but its not as simple to design and fine-tune as a circular pipe port.

What is your Critical dimension for the new cabinet - the width or the depth ?

 

[GeigenPrince]

Thank-you Alan-1-b for your prompt reply

My critical dimension is the width really - ideally around 9.5 to 10 inch max , assuming I would need 0.75 inch MDF to build the enclosure.

You asked about the port on the Conway - its centre is 2.75 inch above the inside floor and 17 inch below the ceiling of the cabinet.

I had hoped to use 0.75 inch MDF for the top, bottom and sides of the enclosure and maybe 0.5 inch for the sides. I would fit a circumferential brace about half way up the enclosure.

Best Regards

GeigenPrince

 

[alan-1-b]

Hi,

I have found a photo of the Conway II.
Are you sure it is only 45 litres internal ?
It looks a bit larger than that to me ...
It will be 19.75" internal height, from your position of port measurements,
but width and apparent depth seem to be large enough for greater than 45 L, thus do check and Post reply, so I can calculate some suitable dimensions.

1 cubic foot = 28.317 litres.

For your Critical dimension of Width:-
For an 8" driver, there will be about 6.5" of cone diameter, both in front and in back.
Cones do not work optimally when too close to solid surfaces {bad notch suck-outs at some frequencies, and non-linear compression effects} thus do allow at least 1" either side of the back cut-out in the chassis.
That is 6.5 + 2" = 8.5
8.5 + .75 + .75 = 10" external width
{yes, do at least use .75" thick walls, especially in a cab. of this size}.
I would use wider myself, with minimum 1.5" each side inside, but as you want a narrow cab., I think the above can be got to work.
What is the internal absorbent used, or that you will be using ?

Yes, I have seen very narrow cabs with bass driver stuck right up against each side wall inside, but these are compromised in sound, and even when quite good sound as result of other factors of their design, they would be even better if a little better spaced inside for the driver to breathe !
Fashion and demand to fit in small areas drive manufacturers to make less than ideal products !
Srereo imaging can be as good, and sometimes better, from a wider cabinet, thus ignore the current misunderstanding about that.

Circumferential brace is used by commercial manufacturers because it is cost effective to make and install, but better can be used,
such as T shaped, with the foot of the T against the inside of the baffle between the bass and mid drivers, and top of the T against the back panel.
Another T can be placed in the oppsite direction towards the other end of the cabinet,
Vertical H to support each side panel against the other could also be used.
Horizontal H is not as effective for side panels bracing if only one used in a tall cabinet.

 

[djarchow]

The other thing to consider when changing the baffle dimensions is how it will impact the XO. Assuming the XO was designed for that specific baffle and driver layout, changing the width will impact the diffraction signature and the baffle step. Which, depending on the magnitude of the change, might necessitate XO changes to match.

Regards,

Dennis

 

[alan-1-b]

Good point Dennis,

though only a few manufacurers were taking such into account back when the Conway II was made {about 15 years ago}, and from the look of the front of the loudspeaker in a photo I think Castle may not have given priority to diffraction as it looks to have a slightly inset baffle panel with the cabinet edges protruding a little, thus there will be various reflections at various angles and some frequency dependant peaks and dips in the forward response,
UNLESS what I have mistaken for a recess is merely an illusion formed by the black strip which is fitted to hold on the original foam grille.
{what is the case G.P. ?}

There will be some degree of less forward radiation of the lower midrange with a reduced width baffle, but such may not be a major problem, as the degree would depend on the cross-over frequency between the bass and mid drivers, {which I do not know, but will guess to be between the 600 - 800 Hz region}, and the slopes of the filters.

Of course with a new cabinet, G.P. can include curved-profile edges to the front panel to reduce reflections and cause a more even frequency-dispersion of sound around the listening area !

... and I'll bet you thought this was going to be a fairly simple project G.P. !!

 

[GeigenPrince]

Hi

Thankyou for your continued interest Alan and your input too Dennis !

The Conway II was reviewed by HiFi Choice in 1980, and they quoted the volume to be 52 litres. However, when I removed the drive units, I saw a false ceiling in the top of the enclosure which reduces the volume, as the internal height is 19.75 inches. !!

I can confirm that the baffle is inset by about 0.25 inch.
The xover freq. are 750 Hz and 4 kHz.

I was planning to use the same type of absorbent in the new enclosure - ie. 1 inch foam on sides /top/bottom, and 2.5 inch foam on the back panel.

So I should be using vertical H brace midway between front and back.

I am afraid crossover adjustments would be a step too far as I don't have the means to measure freq. response etc. I was trusting that the response of the new model would be smooth enough - given that this model was given a top rating in its review.

Yes -- I did expect this project wouldn't be too complicated !!

Oh yes-- the port --front or back ?

Thanks again

G.P.

 

[alan-1-b]

Little time for a month

Hi G.P.

unfortunately I will be busy working out of town for the next month, and will have little time spare and only very limited computer access time, thus it will take a while for me to address all the relevant matters for your project, but don't worry - I will get through them.
Are you in a hurry to finish your project ?

For the H brace - I meant for it to be placed to support the two side panels as those are larger surface area than the front and back panels for your cabinet preference, thus the sides wil resonate to somewhat greater degree than the back, and probably a little more than the front, {depending on the strengths of the edge joints].
The H cross-line should be a little off-centre so there will not be symmetry, and thus not doubling of a single resonance.
Whether the cross-piece is above or below centre should be decided after one has decided where the bass driver will be placed, so that there will not be a reflecting surface directly behind it.
Also, the the H should be a little off-centre vertical of the side panels, so no equal areas either side of it which would both resonate at the same fundamental frequency.

If you don't want to use two horizontal T braces between front and back panels, one will be OK to some degree, and its placement can be decided after the position of the bass driver on the front panel is decided.
Alternatlely, you could use a single T brace with its top turned through 90 degrees to vertically support the back panel, and its foot against the front panel.

BUT, before we design any further, given the age of your Conways,
if you haven't already, then carefully examine the surround of each bass driver to see if any cracks or splits are beginning.
Unfortunately such is common with some types of surround materials as they age.
It will be a large waste of time to design and build a cabinet if the drivers collapse soon after !

Check the surrounds of the midrange drivers also, as I cannot remember what Castle used in the Conway, and I cannot see in this photo because there is a circular cover over the mid drivers.

The tweeters Castle used were their own design then, and quite a good one, and should have survived OK, but look carefully to be sure.

Post your findings.
Even if deteriorating surrounds on any drivers, such may be able to be repaired, or alternate drivers used if repair is too expensive, or if unlikely to give long service life.

 

[GeigenPrince]

Hi !

I'm afraid I have to push ahead with this project as I have a limited window of opportunity at a local woodwork shop. I can use this facility free and will only be charged cost price for the wood !! Too good to miss methinks !

The bass drive units were reconed by Castle in 1996 since the foam surrounds had perished, they now have neoprene surrounds. The midrange drivers seem to have transparent edge to the cone, and seem to be in good order also tweeters seem OK.

Taking advice into consideration my proposed enclosure is thus- -

slightly larger internal volume = 49.5 litres with internal dimensions 22 cms wide, 33 cms deep, and 68 cms tall.

Port 2 inch diam. 5 inch long with a small flared baffle at the inside end. The port being on the rear wall mounted 16 cms above the floor of the enclosure.

The drivers to be vertically aligned close to each other such that the bass unit is just above half way up the front.

The midrange unit will be enclosed at the back by a 11 cm diam. tube - 18 cms long, closed at one end. This will be lined with carpet felt and filled with absorbent.

I am considering solidifying the side panels by bonding a 13 inch square ceramic floor tile to the inside surface of the side panels, just above half way up. This should stop any vibration of the sides.

The absorbent would be 1 inch foam on the sides panels and 2.5 inch foam top, back and bottom.

I also intend to mount these speakers on stands to raise them about 5 inches above the floor.

I'm looking forward to hearing the new Castle Conway's and will post my results.

Thanks again for your input.

G.P.

 

[alan-1-b]

needs careful design for optimum results

Indeed a limited window of opportunity, but using it to make a resonances problems' design and layout you may be
"looking forward to hearing the {next} new Castle Conways"
I'm afraid.

22cm and 33cm is ratio 1:1.5, and is almost worst for width to depth ratio, given how a speaker driver works !

Use ratio of width to depth of 1:1.4 minimum <---> 1:1.42 maximum,
eg:- 8.5" and 12" internal, and with 29.9" {30"} internal height.
{close to 50 litres}

Bass driver near to halfway of the vertical is also bad, work out a 1:1.618 ratio for bass driver position in vertical.

1'' foam may suffice on sides, though only absorbs 90 degree incident sound above about 2.5kHz.

2.5" on back is barely sufficient in a 12" internal depth, where should be a minimum of 3".
2.5" will be OK inside top if thicker on inside bottom - I'll Post more about this next time.

I'll calculate a, perhaps choice of, port vertical positions.

If you want it on back panel you will need at least 3", and better is 4" behind the cab. to wall.

You have the opportunity to make an optimum drivers' height baffle, thus I strongly recommend you listen first from a position about a metre in front of the speakers as they are now, with your eyes closed, and at a range of listening heights till you hear one where the sound becomes "coherent",
by which I mean, sounds as close to reality as possible.
Use simply recorded material - human voices and/or acoustic instruments, and from a variety of recordings.
This is not necessarily where the sound has the most high treble, as also can sound a bit thin there, but optimum is usually where the midrange information sounds most believable.

Electric Pop/Rock is not a good test for this as it contains too many reality confusing sound effects added during its mixing stage.

Next listen at the normal distance from the speakers that you usually do, or will be doing, and at various head/ear heights again till you hear the optimum, and this one should make more sense than the 1 metre distance position {which is for you to hear how sound adds, subtracts, and focuses}, then open your eyes and see the vertical positions of the drivers.
You may be on, or close to, vertical axis of the mid-driver, or you may be a little between mid and bass drivers, and if this is too low for seating then you can have an inverted baffle design - bass driver at top.

Position the drivers on the baffle to correspond to this determined coherent listening height, but with cabinet internal vertical ratio to suit bass driver and external stands height calculated then to give the optimum height when listening from your intended distance.

If this is not possible with a 1:1.618 bass driver vertical ratio, then I'll calculate a next best, or two ...
I'll do this in advance, because I have to go now, and I do not know when I will next have computer access - perhaps in approximately one week, I hope ...

 

[alan-1-b]

correction of a mistake in last Post

[
1'' foam may suffice on sides, though only absorbs 90 degree incident sound above about 2.5kHz.

[/B]

My mistake !
1`'' foam absorbs down to about 2.5kHz for 45 degrees' incident sound direction - {2.4kHz actually}- and only to 3.384kHz for 90 deg. , but as most of the high frequencies in the cabinet will be striking the side walls at angles the 45 deg. figure is more useful.

For the back panel behind a driver the incidence will be 90 degrees, thus for 2.5" absorption will be good down to 1.35kHz - if the foam is a suitable open cell audio type.

One quarter of the wavelength of the incident sound wave is the depth of absorption necessary for sufficient absorption.
How-ever, one need not fill the dimension to greater depth than one quarter of its size, thus for a 13" deep cabinet ideally use 3.25" thick absorptive,
3" will suffice, but 3.5" is better,
AND, unless an absorptive is very well designed, and optimally aligned with regard to its internal fibres' layout, it will not absorb well when thicker than about 3.5" <---> 4".

 

[alan-1-b]

Internal dimensions, continued , etc ...

As you are considering using a 13" ceramic tile on the internal side walls, suitable internal dimensions for a minimal coincident standing waves cabinet are, for W x D x H :-

9.25" x 13" x 23.25" for 45.75 litres
{23.5cm x 33cm x 59cm}

9.25" x 13" x 24.5" for 48.5 litres
{23.5cm x 33cm x 62.5cm}

I recommend you do not exceed 62.5cm internal heigth,
but if you have, then Post the internal W x D x H you have used and I will describe to you a, hopefully easy, H internal modification.

Place the bass driver at the .382 : .618 internal ratio of the Height,
that is 8.75" : 14.5" for the smaller cab.
and 9.25" : 15.25" for the larger cab.

If this is not suitable, than use a ratio of .375 : .625, but do not get further outside the .375 <--> .382 and .625 <--> .618 range.

Use the same ratios for position in Height of the Port if it is on the back panel.

Or, position the Port 2" above the internal floor {1.75" minimum}.

2" port will distort low frequencies.
Look on Wilmslow Audio web-site at their Flanged reflex ports.
Use at least their 61mm dia, or the 75mm if it is long enough.

More next time ,

 

[alan-1-b]

internal placements

The ceramic tiles should be placed off-centre of the vertical so that the untiled sections of each side-wall are not of equal area {and thus not of equal resonanct frequency}.
Place the tiles with their centres at the same ratio in the vertical as for the bass driver - .382 : .618 ,{or .375 : .625},
AND, most importantly, place them towards the lower end of each side-wall.
This is to keep the Centre Of Gravity of the loudspeaker as low as possible, as that will ensure better stability, and thereby a clearer sound.

The midrange enclosure you mentioned the size you propose of,
ensure it is no smaller than the internal volume originally used by Castle.

Post how you are proceeding, for interest, or what-ever ...

 

[Richmond]

It took 18 months to design the original 45 linternal litre Conway II cabinet, the best construction which was used in the Conway IIA/III which was using a carcass of Chipboard and MDF for the front and back baffles. (Chipboard had no end of consistancy problems - the most consistant at the time was from Ireland)

It would also be prudent to mention that not only was the midrange tube effectively used as a sealed box loudspeaker but also a form of strengthening/reducing internal refelctions. The bamf in the tube which was used on the later models was far superior to the orginal foam.

The top cavity (can be seen if you remove the treble) was again tuned using a combination of thickness of baffle to adjust the throat, and type of foam, constriction of foam as well as the cavity to reduce the peak in the sound pressure level near the cut off frequency (memory fails me to the exact frequency), also the cross bars where essential to deaden the vibrations of the side walls. Various combinations were used in the prototype, from 1/2 to a cut out panel with four holes.

The conway IIA Cabinet was slimar and was designed to take into consideration the improvements I made on the drive units. There was no change in part numbers 06.08 for the Bass unit, 07.01 for the midrange unit and 08.05 for the original 12 ohm treble unit - the crossover there was no varation bar to replace the fuses with the CASTEK or more correctly the PTC devices - RDE 090 (Bass) RDE 070 (Midrange) and RDE 050 (Treble) - I still have ample of these in stock.

These were fitted as standard on the Conway III and some later models of the Conway IIA - it would be prudent to mention that if you have a fused version and the fuse trips - cease using untill a replacement fuse is fitted (particularly the treble). The crossover board had the fuse fitted on the output of the crossover instead off the input (all models after the Conway II had the inputs fused) - If the fuse blows the result was a multiplication circuit which would cause the treble inductor to overheat and exceed the voltage rating on the Low Loss Capacitor.

The Conway II cabinet really needed a stand, this was addressed on the Conway IIA/III

As regards the drive units - the cabinet was designed first to get a flat as possible response first - then the drive units were designed. This was always my way and castle ways design out the problems with the cabinet and then units - not buy in the units and then design the cabinets to fit! as most of the manufacturers did.

We were very proud that everthing we did was made in house, where ever possible, using material sourced in the UK (alas in the last decade we had to source individual components from abroad - diecast chassis - "T" poles etc as we could not obtain them at the quality or cost need to survive.


- OK the original basis was the Conway I bass unit 06.02 but suffered from being too sensitive and was reduced down to the 86dB for 1 wat input.

Hence the reason for the 12 ohm treble unit -was fundamentally the same as the 8 ohm version but used finer wire. The early versions due to production inconsistances (resolved by using an aluminium weighting ring instead of an steel one) had a larger tolerance band than designed with a peak at 15-22.5 KHz, this was resolved for all castle trebles at the time by changing to Mackrofol N and a better controlled moulding tools, and consistance in volume of ferro fluid inserted into the gap prior to the coil being fitted. The ferro fluid did increase the power handling from a nominal 5 watts to 22 watts!

The midrange unit was brand new to the Conway II with an 11cm diecast chassis and indeed took the longest time to design and indeed is probably to most open to variation due to room temperature and humidity (optimum was 20 C) frequency response variations were observed in China.

The crossover incidentially was the same for the Conway II/IIA/III and Howard II (not the S2) The Howard also used the same treble unit and midrange unit but had 2 x 12 ohm bass units.

I hope that the above is of interest

Andy

 

[Richmond]

Hi

Forgot to mention.

The foam surrounds were upgraded to rubber surrounds in 1981 (Metflex) although open to temperature varations was a considerableimprovement.

Bar surround failures the other problem that did occasionally occur was pig tail wire fractures at the cone.

The later units had a Kapton voice coil but not until 2000

On the midrange unit if the surround in purple it is Triafol B or Yellow is Mackrofol N - watch for the surround lifting under the dust dome - can be resealed down with PVA adhesive.

The Conway II baffle was recessed as in line with the Richmond II and Kendal II - Icidentially the Conway II was Launched at the Harrogate Hi-Fi show (those were the days - the whole factory came out)

The Conway IIA had a flat front for refelction purposes the trim at the top and bottom was used to empahsis the wood finish above the foam front and was held on by a strip of 25mm velcro split in to 3 - hence some slight variations in the thickness of the velcro on the front.

Andy

 

[alan-1-b]

Hi Andy,

It is an honour to have you contributing to our discussion,
and I thank you very much for all the interesting information about the Castle loudspeakers' design and materials, etc ...

And here we were tampering with your careful cabinet design !

I particually liked that tweeter Castle used in the 1980s era as it seemed to render a more true-to-life treble than other tweeters in similarly priced loudspeakers from other manufacturers then,
and I suspect it still may compete well with some.
Why was it discontinued ?

I remember Castle designed, and probably built, a small loudspeaker for another Company in the 80s, "QED" if I am remembering correctly ?
It was apparently based on the "Clyde" model, but with some, undisclosed, differences.

I auditioned these, and particually liked their transient response - very realistic, and much better than the somewhat slow, or soggy, response of some other small and not so small other manufacturers' speakers then.
They were particually good with solo instruments and small group recordings, but unfortunately did not cope well with large orchestral works, even at moderate volume - OK, maybe I was expecting a bit too much from them in that application, but I was so impressed with their close to true-to-life portrayal of Timbre, etc ... that I had to try some orchesrtal recordings.

So, why didn't I buy a larger model of Castle design ?,
well, I didn't have sufficient money for such, and I was already into some DIY, thus I thought that with some SEAS drivers I could do as well, etc ...
I should have bought the Castle Clyde, or the QED version, and made a bass enclosure with a 10" SEAS driver, etc ... but I didn't have sufficient test equipment to redesign the passive cross-over to cross well enough to a bass driver, and thought I might mess up what I had liked about the QED/Castle sound.

I suppose those tweeters don't perform well low enough in frequency to be used with an 8" driver in a 2-way, but maybe with a modern 6.5" driver that extends without excess break-up sufficiently high into the upper midrange ...
Yes, I'm always thinking of applications for good sounding drivers, and many more than I've had the time or opportunity to implement.

I agree with your point about stands, as too many loudspeakers do not render realistic timbre when listened to on tweeter axis, but are often better on the midrange driver axis, or a little below,
for those designs that are capable of true-to-life timbre -{many aren't !}.

I reckon you had an interesting job !

I hope what you have contributed here will be of interest to who-ever may Search this Forum for Castle information,
and also interest GeigenPrince if he returns.

Thankyou again,