OB based on idea of CS2

[Okvarbud]

Hello to all at this great forum,
this is my first post here, and my first project.

I've been in audio for way more than 20 years, and for the last three years happiest listener ever having a system with Emerald Physics CS2 speakers. I've made some adjustments to both speakers and Behringer DCX, configured my own EQ scheme and I think I also nailed a good amplification for such system - custom made MOSFET for bass/mid, and 2A3 SET for mid/treble. I've been also checking other open baffle speakers concepts, had some of those in my room for a while and listened to some other elsewhere. However, I feel CS2-like concept is best for me for many reasons so I decided I'd like to improve that concept if possible through some DIY. My desire for improvement is more result of situation in which I can afford the expereiment than being dissatisfied with speakers as they are now. I consider CS2 a milestone among commercial speakers concepts and want to express kudos to designer Clayton Shaw for bringing them to life.

I'll make a new baffle from two sheets, keep the original CS2 baffle width, and use Dayton 12" waveguide and Selenium tweeter like in CS2. Having heard some other OBs and many other speakers I think I can't really do better than that without spending way much more, but instead of Eminence Alphas I ordered Acoustic Elegance LO15s so I expect biggest improvement in overall sound in their range (up to 1kHz).

I mostly figured out in detail how I would like to make these new ones, but still have some doubts. I’d appreciate your advices.

Main issue is that I'm not sure which shape is generally considered better for this concept, A, B, C or something else? Either way I'll round the edges of the baffle to get circular edge hoping to minimize diffraction. Complete baffle will be around 7 cm (3”) thick.



Which baffle material is recommended, plywood or MDF?

What material and how thick you would recommend for applying between sheets (glued to one sheet)? I see many types from EAR at PercyAudio but have no basis for decision which one to use.

Thank you.

[Keriwena]

Wellllll......

There's a huge thread here,
What are the characteristics of a better material for enclosure?
with so many diverse opinions, theories and proposed solutions I really can't offer you a simple summary.

I guess, given the thickness you allow, I'd laminate together as many different materials as possible. Seems to me, some sort of averaging is the best approach, as in the musical instrument world where resonances are desired, solid materials rule. So do the opposite.

[twest820]

Also look at Gainphile's horn builds. Doing a decent job of addressing the front to backwave mismatch created by the horn is not exactly trivial.

[gainphile]

If you could find a way to magnet-mount those big 15s like linkwitz phoenix then the importance of baffle material is nil!

I really like the 2nd design!

[twest820]

Provided the magnet mounts and driver baskets are decoupled from the baffle and there's good field cancellation so the net load on the baffle from the front and back waves is also nil. Both are somewhat tricky to achieve. I agree magnet mounting's less demanding of the baffle than rim mounting and there's no shortage of magnet mounts to look at---a good reference design is the Isiris.

[SAC]

Have you seen the CS2 development ideas on this page? May be of use to you I suspect.
Modding the Emerald Physics speakers and more

[Okvarbud]

Thanks to everyone for responses and suggestions. I made some further searching, so this is where I stand now;

Magnet mounting certainly has its meaning, but will be very difficult to make well for a several reasons. Drivers are not made to be held like that, when you use more than one it is almost impossible to align them well, plus you need to align them to a baffle, have their surface flush with the baffle, but not have a firm contact with the baffle. All this together may bring some other compromises. Perhaps it is easier to do it well when project is based on one bass driver, but with more than one it is quite difficult. Not that I’m against challenge, but this might turn into nightmare.

Then again, how much of a factor this is when a baffle is well made? So for a choice of materials, I think I'll go with two layers for one baffle, like I planned before, but with mixed materials, front in plywood, back in MDF, EAR SD40AL in between fixed to front layer. Baffle will stand on two spikes and have a metal leg behind which will attach to a baffle somewhere above the woofers. I’m not sure about that now, perhaps I’ll go for some sort of a spine instead of a leg.

As for drivers choice, after seeing graphs of Jacek’s SEOS 15” with B&C DE250 it seems like a no brainer any more. A choice of SEOS waveguide also makes decision of a baffle shape easy. So, with AE LO15 bass/mids my speakers will look something like this;

[Okvarbud]

Originally Posted by SAC

Have you seen the CS2 development ideas on this page? May be of use to you I suspect.
Modding the Emerald Physics speakers and more

Thanks, I've seen that page before and tried some of those suggestions. However I don't agree with some "night & day" improvements descriptions.

Only real night&day difference is made with EQ scheme. This is a foundation of design and needs to be done optimal, which is of course specific to every system and room. It takes long time to be done well. Factory settings are only starters.

I have modified DCX, full mod by Selectronic in France. It is better than stock, but stock will also work well if EQ is well made.

I also applied Dynamat Extreme to Alpha's baskets and that was also not necessary in my opinion.

What worked significantly well for CS2 after EQ was done right is SET amplifier for compression drivers and opening of the back of compression driver thus making speaker a dipole in full range. I needed to attenuate treble for a half or 1 dB comparing to previous EQ setting, but improvement in naturalness was really a surprise. That is why I now want that my new speaker is also a full dipole. If that is not a requirement, I would decide on BMS 4550 instead of B&C DE250, but BMS cannot be opened in the back due to its unique design.

Removing mesh from Selenium tweeter opening and smoothing a connection of tweeter tube and WG throat with Patafix tack also made complete sound a bit more natural.

[Okvarbud]

Since I now have all major parts, I'm resurrecting this thread.

I'm at perhaps last major decision and need help.

As at above picture, I'll have B&C 250DE compression driver in Autotech SEOS 15" horn placed above a pair of Acoustic Elegance LO15 woofers. I plan to actively cut with BUT48 at around 1kHz. Setting at my current speakers is BUT48 1.24k low-pass for woofer, and at 1k with same slope high-pass for tweeter. This will be my starting point for new speakers too.

Is there a calculation for exact unique optimum distance between tweeter and top woofer, and also between two woofers? If so, which distance this is?

[Rudolf]

Originally Posted by Okvarbud

Is there a calculation for exact unique optimum distance between tweeter and top woofer, and also between two woofers? If so, which distance this is?

Download this program and see for yourself.