High-end 3-way active Avalon look-alike

Hi All,

I promised some time ago to post a topic on the construction of my new loudspeakers.

Firstly started off as a 4-way system, using an Eton 12" woofer, Accuton C2-220 low-mid, C2-79 mid and C2-12 tweeter. Design was made, components bought and due to active set-up the required amplifier channels purchased (i.e. 2x Classé CAV-150)

The 4-way concept daunted me a bit, but what the heck..

.. And then came the C2-90 midrange unit. What a beauty!! Functionally merging the C2-220 and the C2-79 into one unit. So bought a pair of these too. I'm now the proud owner of a collection of Accuton units that will allow me to open a shop.

So it's going to be a 3-way after all with the Eton woofer, C2-90 mid and C2-12 tweeter.

Filter is the SAW-30.3 of Thel which is an amazing piece of circuitry. Combined with a dual-mono no-compromise power supply this unit operates flawless.

At the moment I'm busy building the cabinets. I keep a minute log of the process on a Dutch forum: Topic on Dutch forum

Sorry for the language but there are loads of pictures that make up for that.

Some pics: (click on them for better viewing)

Design, front view


Design, side view


Internal bracing, current status


The baffles:


Hardware!! (maybe I'm goint to use both nevertheless as mono blocks... )


Will keep you posted. Most recent updates will be on the Dutch forum, so if interested, keep checking that one... The thread there is very visual, so Dutch is not required.

Sorry for a long silence guys!

Most of the reporting about this project goes on on the Dutch HT-forum, so that topic is most up to date. (see link in previous post.)

I haven't been slacking around though, despite a hot summer here in Holland.

Short overview:

Glueing one of the cabinets


The top- and bottom sides of the baffle were not yet angled like the rest of the cabinet


After some creative routing and sanding they were


It turned out that because of using G-clamps during glueing there was quit some tension in the cabinets and the top and bottom panels took the pounding. In other words: they were bent. So I set up a reference plane above these sides and routed the cabinets square and flat:

Reference plane by two parallel Aluminium L-shaped profiles:


Routing


Result


After the trimming of the edges a 6mm MDF layer was glued on.

Both cabinets built:


Next step is cutting off the corners. I was a bit apprehensive about that. Screwing things up at this stage would be very frustrating....

So for the first cut I put aluminium strips on the cabinet to guide the hand-saw, but that was a nightmare. The saw was biting in the aluminium all the time. So in the end I just did it without strips. With a bit of concentration and making sure that the guiding lines are drawn up correctly it was actually a piece of cake:



The cut edges are covered with 3mm MDF to get a smooth surface:



And that is what one cabinet looks like when all corners are cut off:



Starts to look like something decent.

At the moment I'm completing the step of cutting the corners off the second cabinet. After that I need to vineer them. Still looking for a nice vineer. I think it will be bird's eye Ahorn.

So that's it for the moment. Keep checking the Dutch HT-forum for the latest pictures. Also the pictures shown here are just a subset of the pictures posted on the Dutch HT-forum. So for more pictures you can also consult my topic on the Dutch HT-forum.


Cheers, Fokke.

Thanks for the feed-back guys!

Things have been a bit busy lately. Also, the project now comes to a phase of a lot of tiddly stuff as the cabines have both been built:



I chose the vineer, it's bird's eye maple allright. I bought two sheets of Lignaflex (vineer on paper backing) and two strips of real vineer. The thought is that Lignaflex is great for big surfaces (really easy to apply and no fussing with gapfree merging of more than one strip) BUT... on the facets the paper backing really shows due to 45 degrees angles. This is really not nice to see.

So the tactic is to vineer all "square" surfaces first (rear, sides, top and front) with Lignaflex. When flushing the vineer on the facets, the ugly paperbacking edges will appear. Then the facets will be vineered with real vineer (without paper backing). Flushing that with the top, sides and front will then result in edges that are only wood, and no paper backing will show.

Glueing the Lignaflex will be easy, I already have experience with that. The real vineer on the facets is trickyer. I made a test piece on which I ironed a piece of vineer. The process goes as follows:

1. spray vineer with a bit of water to make it damp, and flatten the vineer between two boards of MDF with some weight on it (approx. 20 kg)

2. Cut vineer to shape

3. apply normal white (PVAC) glue to both vineer and workpiece

4. Wait till glue has dried transparent.

5. Place vineer on workpiece. Because the glue has dried, the vineer can be moved around freely.

6. Heat up iron to silk temperature and press iron on vineer for approx. 10 seconds. after that , remove iron and apply pressure with a flat block of MDF of about the same size as the iron foot untill the glue has cooled down (20 seconds). Treat the whole surface like that to make sure that the vineer is glued entirely to the workpiece.

7. and voilà:



Perfectly FLAT!

I was surprised by the ease of this method. Because there is no wet glue that threatens to dry or contact glue that sticks as soon as there is contact (always the wrong spot if you make a false move).

Because of the short contact pressure (30 seconds total) it is ideal for surfaces that are uneasy to put pressure onto. Also you can work at a relaxed pace, there is no reason to hurry.

As for the bass cabinet: it will be a closed box of approx. 70 litres. I ran simulations with the box volume ranging between 60 to 80 litres, and the impact on frequency response is minimal. In other words: a very tolerant driver.

I chose a closed box because I can always add a vent if I want to, and because I prefer quality of bass over quantity. I suspect that the 12" driver will deliver plentiful anyway.

This week-end I will be cutting out the grill panels, and I will dare to start vineering the backs of the cabinets!!

There is one concern that I might have thougth a bit lightly about: the real wood vineer is slightly darker than the lignaflex. I will have to see how it matches (maybe sand the real wood back a bit to make it lighter). Otherwise, I will have to get another batch of real wood vineer.

More later!

Cheers, Fokke.

Vineering is almost finished now, and the cloth grilles are made. Some pics and explanantions:

First I did some Lignaflex vineering of back side, sides and top.

Glueing. I use a simple foam roller to distribure the glue evenly and thinly.


Ironing. It is useful to "calendar" what part has been done and what not. You don't want to miss any spot.


After trimming




Grilles:


I decided to use magnets to fix the grilles to the speaker cabinet in stead of the normal nippies. For this, I purchased special superstrong magnets from Supermagnete, a German company.

One magnet goes into the baffle, just under the vineer skin, and one inthe grille.


The sthings stick like a *** ***! When I hold the grille near the baffle, it just "jumps" into place!


Nice detail of a vineered baffle:


After vineering baffles it's time for the facets. After extensive consultation of vineer specialists I decided to use the vineer I already purchased. Oiling it and in due time the colouring will come together.

With flash the difference looks shocking, much more than natural light...



So that's where I am at the moment. The second cabinet is being finished (three more facets to vineer). Then it's time for routing the speaker holes... I already did a routing experiment in my test piece. No worries for spliters!



Cheers, Fokke.

tungsten said:

thank you very much for the supermagnet tip - something I will definitely use in my Eton/Skaaning/Seas speaker.
Which magnet size are you using? How much does your grille frames weigh?
I noticed that your grille frame does not have rebated inner edges - are you going to use felt?

Click to expand...

I think the cloth will finally touch the cabinet first, and maybe I'll cover with a thin layer of adhesive felt.

The magnet size I used is 8mm in diameter, and 3mm thick. (S-08-03-N)

If you want to be absolutely sure or think that there will be more distance between the magnets than just a layer of vineer, use the S-10-05-N, these will be more than sufficient.

Sjef said:

With all respect but I have find the Thel crossover really a piece of crap. killed all the air and low level detail in my system. (I did use it with the Thel accu supply)

Click to expand...

I don't have that experience at all. However, the applied amps are very important, as are cables. First of all, all amps need to be identical. I used to have a setup with Denon POA-6600's for bass, Audio Analyse A9 for mids and Audio Analyse B90 for highs. Sounded clean, but as you mentioned killed all homogenity and refinement in the system. It was like listening to a music box.

The turnaround came when I tried a pair of Hypex Active Home III units (ridiculously cheap) incorporating three lousy power op-amp poweramp units. They just blew my stack of amps and fancy THEL filter away!! The reason was plainly the equality of poweramps.

So after purchase of a Classé CAV-75 6-channel HT poweramp the problem was solved. Then I found out that also the interlinks between filter unit and poweramps are crucial. As I needed 3 pairs of identical interlinks, I went for affordable vdHul D102MKIII Hybrid interconnects, as I had always been pleased by the all-metal predecessor. However, also these interlinks killed off the whole thing compared to crappy Oehlbach cables of different sorts. Then I came across the "Siltech effect" auditioning a pair of FTM3-G3 interlinks. Downright stunning, I was sold!. Later on I tried a pair of Snow Lake G5's between Wadia CD-player (no pre-amp required and not present at the time) and filter and things really got scarily realistic.

So I hunted down a couple of FTM3-G3's. Furthermore, I constructed some DIY interlinks because of the frustrating price tag of a pair of Signature Snow Lakes (the known "cork screw interlinks) first based on teflon tubed silver wire (useless) and later on using Siltech M7-G5 internal wiring. The latter have sound signature of the snow lake, but slightly lack the extreme detail of the snow lake. Also I made a cable based on M10-G5 (thicker internal wiring) that does every thing wrong in high-mids and highs, but has the exact bass performance of the snow-lake.

So I use that bass-cable to connect the low-freq output of the THEL to the poweramp, and FTM3-G3 to connect mids and highs. The M7-G5 based cable connects ML28 to THEL filter, and the Wadia is connected to ML28 with a balanced vdHul MC-Gold interlink, that has almost the performance of a Snow Lake for a fraction of the price.

I added the ML-28 after a comparison with <-> without where the ML-28 provided a darker background, more loose and deeper bass and a more spatial and bigger sound stage compared to connceting the Wadia directly to the THEL filter input.

The morale is that it takes some effort to make a THEL-filter based system sound right, the surrounding components are of utmost importance. However, if your system didn't sound right, it is IMHO a bit short sighted to blame it on the THEL filter.

Once you got your act together, the THEL filter will reward you with a stunning sound stage, clarity, low-level detail and ease of listening that I rarely find available in high-end systems.

Time to update this topic, as I start to clutter other topics about veneering with recent pics of my project.

Anyway, veneering is done, and the last bits are done too. They are playing at the moment!!

First some pics.

First of all, the driver rebates and holes were routed.


This job needed utmost concentration since the speaker was already veneered, hence no margin for error. Before every step I made a test in a scrap piece of MDF. When done, that scrap piece looked like this:



Also all the interior details came back out. I almost forgot about those...




Once that was done, the finish was in sight, and it got really hard not to loose my patience...

Finishing was done by applying Danish Oil. This is amazing stuff. Also, the final result is not because of one layer, but a total of 7 layers were applied.

After one layer it already looked like this:



Application of Danish oil is easy. You just rub it on liberally with a piece of cloth, and after 2-3 minutes remove the excess oil with a dry cloth. Due to the short timespan between rubbing on and removing, the cabinet cannot be treated in on go, but needs to be split up in different phases.

Not every layer has been applied in the same way.
Layers 1-3 were applied with cloth, and excess oil removed with cloth.
Layer 4 was applied using P600 waterproof sanding paper and excess oil removed with paper kitchen towels.
Layer 5 was rubbed on with cloth and excess oil removed with paper towels.
Layer 6 was applied with P1200 sanding paper and excess oil removed with paper towels.
Layer 7 was rubbed on with cloth and excess oil removed with paper towels.

Then it looks like this (after mounting drivers):



By the way, mounting the drivers can easily be done using the ironing board (the ironing equipment has been frequently abused during this project)


Also the grilles were completed. In the hole for the tweeter some Needle felt was fitted:



Then the cloth was fitted with hotmelt. To do that, you first need to apply a stripe of hotmelt along the edges of the frames, on the rear side. Then after cooling down hold the cloth on top of the stripe of hot melt and melt the hotmelt again with an iron. due to pushing down the iron, the hotmelt wil soak into the cloth and glue the cloth to the frame. Some hotmelt ends up on the sole of the iron, which can easily be whiped off. Very frequent whiping is recommended to keep the iron and the cloth clean.

So after some messing about it looks like this:



Also the rear of the speakers is not to be sneezed at:


This has been an amazing journey in woodworking for me. I learned so much, and have gone through quite some barriers.

Now the next phase is waiting, namely breaking in and setting up my filter. Sound is already promising, but nowhere near finished. It will take at least 6-8 weeks of intensive burn-in before the final set-up can be done.

Also my filter's low->mid crossover frequency is now in a fairly low range of 66-257 Hz. This is slightly too low for the mid. Ideally would be 350-400 Hz. For that, I'll have to purchase some new capacitors for my filter.

I already noticed, that small changes in volume settings between lows, mids and highs have a considerable impact on sound experience. Small adjustments need to be made at a time.


At the moment, the sound can be characterised as relaxed, dynamic, clean, laid back. Also micro detail is missing due to the fact that the units are not broken in yet. Spatial imaging is also promising, but held back at the moment.

Will keep you posted on future developments!

Cheers, Fokke.