Sonab

Back to Retro Page


Sonab Project - Rejigging the OA6 Mk2a

I have come late to the appreciation of Sonab speakers, but now find them quite likeable and intriguing. At first glance they appear to break many of the rules of design that we accept as right and proper. They commonly use upward firing bass-midrange drivers and tweeters facing anywhere but directly forwards, usually at 45 degrees off-axis.

I've come by four different models in recent years, and some of them have needed a bit of attention to get them working and presentable cosmetically. The latest is the most unusual of the four. It's the OA6 Mk2a, which was conceived as a flow-on from an earlier model OA6 active speaker - driven by internal valve amps, no less!

Sonab's models "OA" are code for Ortho-acoustic then the number of drivers, although this varies a bit. Examples: OA5 has four tweeters and a bass-mid, total five drivers. OA6 indicates four tweeters and two bass-mid, while taking a slightly different tack, the OA12 has one bass-mid and two tweeters, the OA14 one bass-mid and four tweeters.

Back to the OA6 Mk2a, the subject of this article. The top baffle has a similar complement to the OA5 Mk2, namely an upward firing Philips 8 inch bass-mid (9710) and four small Peerless tweeters, aimed at all points of the compass. Then there's a second Philips pointing downwards, which is where the story really gets interesting, since this one is powered by a solid state amplifier that they refer to as an "energiser". This, ideally, delivers hefty bass, while the top array does a fine job on midrange and treble.

It's a partly active design, or with built-in subwoofer if you like. The trouble with this design, according to all I've read about it (and there's not enough info around, as far as I can see) is that it's a one-off special which uses an uncommon type of silicon transistor in a special amp circuit that not every technician is going to be able to fix when it goes bad. Forums tell of amp modules going up in smoke, while on the Carlsson Planet site you can read (if you translate it from the Swedish - rather imperfectly via Google) that the design is also known to stress out if pushed hard and take out at least some of the drivers! The contrary anecdote to this was given by none other than Len Wallis of Len Wallis Audio, who once owned a pair of these himself. He told me about running them very hard one evening in party conditions, which they coped with for hours, until the the bass section stopped - but by the following morning they started up again! Maybe just a temporary overheating issue there, with no lasting damage.

Anyway, having read all the horror stories and not wanting any smoking ruins, I decided to go around the problem and re-jig the OA6 Mk2a as passives. Before going any further, I have to say that everything I've done is reversible, so if in the future I or any other owner wants to go back down the active path, they can be restored to original specification provided all those special parts and technical know-how are available. I have kept the amp modules, so they can be screwed back into place very easily. After that, who knows!

Now for the detailed case study. The pair I acquired were something of an unknown quantity, so step one was to assess what was working without risking damage by using the amp modules. The cabinets were gaping at each corner, so would need filling and retouching with a suitable colour to match the veneers. Firstly, I applied a signal to each speaker to see if the top passive section was working. One did produce some sound, the other didn't.

Next I removed the top Philips 9710 from each, finding that both had been water (or beer!) damaged, so were in need of a complete overhaul or replacement. Then I looked at the bottom pair of 9710s, and found them to be in good condition and working, although one needed some work on one of the terminals before being reinstalled up on the top baffle.

Both amplifiers were removed and put into a carton for storage. At this point I tested both speakers, now resembling OA5 speakers, with an amplifier, and the results were pretty good. There was a lack of bass, but a good lot of midrange and treble, even though the Peerless tweeters had deteriorated quite a bit, at least externally.

But the next issue was what to do about the bass drivers down below. Should I buy a pair of secondhand 9710, costing lots, and needing amplification to produce good bass? I decided not to, for now anyway, but to press ahead and explore the passive mode further.

I wanted a cheap test-bed speaker, and WES had two that come close to what's needed. One is a lot like the 9710, being a full-range (with wizzer-cone) 8 inch, model DC8-30, which retails for about $26. Right below it on page 3-100 of the WES catalogue is the SP8-30, which is an 8 inch woofer with a healthy 90dB sensitivity. It's more bass oriented than the DC8-30, so it was selected for the job at the same very reasonable price.

These woofers are a bit smaller than the 9710, being 205mm diameter as against its 215mm. So fitting into the same cutout was ok, but the bolt-holes don't align properly, which is where the next bit of fudging happens. Bear in mind that this is still an exercise in testing a theory, so a bit of fudging is ok as long as I don't disfigure anything! It just happens that you can still use three of the existing threaded holes as follows: one as intended, and two outside the rim. What? How come?

The lower woofer housing (and the woofer itself) is covered with a stiff mesh. This becomes the plate that holds the woofer in place. One bolt goes in as normal, then two go through the mesh into two threaded holes but just outside the rim of the woofer. The fourth hole is not useable, so a little bit of metal is formed up by yours truly to act as a clamp (see picture of bottom of speaker) and this uses one of the bolt holes nearby which is for the lower metal frame - the frame that gives ground clearance for the woofer and originally also space for the bass "energiser".

Now we have woofers installed, which have their own leads (blue and grey), plus red and black wires going up to the top array. The time has arrived for a live firing! I used a vintage Luxman R-3055 receiver for this, with speakers A and B switching. A was connected to the top set, and B to the woofers. Fingers crossed!

The results were better than I expected, a blessed relief. Full-range, spacious sound. I now have a pair of speakers which produce ample bass without any EQ being applied via the bass tone control. If the Loudness button is engaged, the bass fills out noticeably, perhaps even a bit more than I want, but not grossly.

The wiring needed tidying up, so I've made brackets out of plastic angle and installed two sets of three-way binding posts for each speaker, so biwiring continues to be possible. Alternatively, the use of shorting bars can take us back to running off one set of leads, although with a lowering of the overall impedance. The sensitivity seems fine, not needing more than 1 watt per channel from the Luxman to play at normal listening levels - in fact as with most of my speakers you have to crank it up quite a bit to even register one watt! The LED indicators on the R-3055 start at 0.1w, then leap to a giddy 0.3w before going the large hog at 1w. I can't remember the last time I saw the 3w indicator light up. This is salutary for all those hifi tyros who insist that lots of power is essential. Not always, and for most people probably not that often.

Conclusion: That's where I'm up to, and there may be further developments. These could go along lines of (i) better woofers, (ii) some alternative active amp solution, or (iii) getting the original "energisers" up and running again. That would, however, need my confidence to be restored by more info on the circuits and spare parts situation. I'm not keen to go active again with those old amps when a passive solution seems quite ok. There's a fourth option too, namely bi-amping, which can be done anytime now, no further mods required!

Note: If anyone has more information on these speakers apart from what I've already picked up via Carlsson Planet, please email me, gforgie at homeentertainment dot com dot au. Schematics, parts lists, commentaries, all welcome.