Sunday, 28 February 2021

Linn Tukatan

 Linn's little Katan standmount from the 90s and early 2000s has an excellent reputation, particularly in active form, and quite right too, its not the last word in refinement but the music it produces is very enjoyable indeed.  In Linn part number terms, it is fitted with the 038/2 tweeter and 039/1 mid-bass driver.



Its predecessor is the Tukan, is similarly small, but lives in a more conventionally shaped cabinet.  It has the older drivers - the tweeter being the 015/x family and the mid-bass the 016/2.  It has a reputation as being a very "fast" speaker, but there's a certain hardness and lack of friendliness to the sound. Its also considerably bass shy compared to the Katan.

Curiously, when taking either of these speakers active (aktiv in Linnspeak) they both use the same analogue active filter cards, despite their very different drivers.  So the crossover points, slopes and relative loudness of the drivers are shared.

Also, the driver cut-outs in the cabinets are identical allowing the drivers from the Katan to be fitted into the Tukan cabinet.  So I gave it a go.  Even with the passive crossover, the Tukatans are a huge improvement on Tukan, being more flowing, warmer, far smoother in the treble and they now have that Katan musicality and friendliness.

Here are the Tukatans:


Inevitably, of course, a set of Exakt filters had to be created and were built pretty quickly.  Below is the FR resulting for Tukatan Exakt (ignore below 120Hz as the room is having too much influence there).  A great fun pair of speakers!

Monday, 15 February 2021

And Now For Something Completely Different. And Pointless. Linn Formation is Born.

Sometimes, regardless of all the DIY that needs doing, the ongoing 8 projects that need attention, the exercise that should be taken and the need for a useful outcome all goes to the wall and something flippant takes priority.

Perhaps, when on the UK's third lockdown, rational thought takes a little bit of time off? Perhaps there was just too much idle time spent on ebay? Perhaps that old Linn LK box that's been in the back of the workshop stack finally found some "use" and was calling out? Perhaps its a very unusual audio version of a midlife crisis?

Whatever, here's what occupied about 25 to 30 hours of my time over the past few weeks - a "could it be done" rather than "there's a good reason to do this" kind of project took shape.

Linn used to produce pretty much all of their range in what is known as the "LK" box.  From the Karik CD player to LK100 amplifier, from the Classik all-in-one CD/amp/tuner to the Numerik DAC, the "LK" box was the cornerstone of the Linn industrial design approach. Simple. Solid looking. Understated. At about 2/3 the width of the traditional 420mm (ish) hifi design, it was different, fitted discretely into homes and is still thought by many as Linn's peak time in terms of audio performance per £, $ or Euro.  The LK design came with the first electronics - the LK1 pre-amp and LK2 Power amp combination, introduced in 1985 you can read about the luke warm reception given by Stereophile here.  These items were different widths, but, apart from a few exceptions such as the Lingo LP12 power supply, most of the following products followed the form factor of the LK 2 amp. By 1996 Linn had started to change direction with products such as the AV5103 pre-amplifer and in 1998 with the CD12, and by 2003 to 2006, all the LK range had been replaced by the newer look now known as Majik and Akurate/0.

LK2 and LK1 - the first examples of the LK enclosure design

It looks like the Majik-i was possibly the last of the style to be replaced in 2003, although I haven't done exhaustive research. That's around an 18 year lifespan. During that time Linn provided a multi-room system under the Knekt banner, many components of which were also housed in the LK unit. That system was proprietary and is now considered to be at least 10 years beyond when it was obsolete.  But there was a lot of it sold and if you're patient enough, non-working units come up for sale for very small amounts of money, given the quality of the box they're housed in.  So I've picked up a few over the years when they've popped up, imagining they might come in useful for a project or 2 at some point.


A couple of examples of LK Intersekt

That's the first rambling pre-amble.  The next is kind of related, but not directly.  One of the joys of the LK design is that its pretty much timeless, understated and the exact opposite of the bling we see all too often in so called "high end" audio.  It was also an antidote to the omni-present Japanese mass market products of the time. A time when the more knobs, switches, lights, dials and meters was the battlefield on the shelves of Currys, Comet and Lasky's. When, as a teenager, my brother went off to university and bought himself his first hifi, products from the likes of Linn were beyond aspirational. So he had the ubiquitous Dual CS505, a pair of Wharfedale Glendale XP2 speakers and, relevant to this tale, a Pioneer SA-510 amplifier.  So, in this project, 2 elements of nostalgia have come together in a loose association - in direct contrast to my normal desire for understated and inoffensive black products, here we have one of the most understated products of the hifi World combined with something to give it some of the flash of its lower priced contempories.

Back to that ebay question.  What happened to all those Pioneer amps which possibly sold in 10s of thousands in the late 1970s and early 1980s?  Well, there are still quite a few about, but for what they are and how they perform today (I tested one about 5 or 6 years ago against a NAD 3020), they're vastly over priced. But why the Pioneer amp anyway?  Well, in a market sector full of features, the Pioneer Blue Line power meters on the SA-510 and many other products of the time were always one of the best looking meters on the market and they would fit well into the nostalgia angle for this project (if there ever was one - if there wasn't I can make one up in hindsight!).

Pioneer SA-510

And then it occured that the Pioneer cassette decks also carried a very similar style of meter, and there could be more choice there, particularly as they are mechnical devices prone to failure, but I didn't need the tape mechanism, just the electronics.  But the LK box is a good deal smaller than a full width cassette deck, so there'd have to be some speculation (read, crossed fingers), that getting a set of Pioneer meters into an LK box would be physically possible. The other problem was that there are some Pioneer behemoth decks out there that still attract high prices - hundreds of pounds.  What I needed was something that was enough above entry level to have the blue meters, high volume of sales and with a broken mech.

After a few weeks of ebay monitoring, a suitably non-working CT-300 was found and picked up for a very small amount of cash.  In good physical order, with working electronics but a dead mech.

Pioneer CT-300

Its easy to get into these machines - just 4 screws removes the top and side panels.  It highly likely that new belts and a bit of a clean would've had this machine back up and running, but given the depth of the wear grooves on the main head, its likely to have been less than useful - maybe replacement heads are available. Anyway, that wasn't the objective.  A quick measure up of the mainboard proved that it could fit into the LK box with enough room for the transformer too.

The idea was to transfer enough electronics into the LK box to enable a connection from the "tape out" output of a pre-amplifier as a feed, then to have the LK box display the meters following the peaks in the music. In this way the meters live separately from the music signal path, hence avoiding why there aren't any of this kind of stuff in the Linn products - the potential for a negative impact on sound quality. So the end point LK box needed to fire up the meters on power up, and start to monitor the incoming music signal.  The easiest way to achieve this is to set up a tape deck in pause/record as if it was about to start recording the music. But with no circuit diagram and no idea what all the electronics on the main board are doing, how to achieve this?

It has to be said that the internals of the Pioneer are nowhere need as sharply etched nor finely executed as the exterior. In fact, they're fairy messy.  At least with all the cables from the board to the mechanism it did make the job of disconnecting the 2 a lot easier than if the mech had been mounted on the board.

Under the hood of the donor deck

The first thing to work out was which cables told the electronics that there is a tape in the mech and that it is not record protected.  This was pretty straightforward as there are microswitches so, using a multi-meter, it is possible to check if these are closed circuits or open circuits when there is a suitable tape on board. To make life unusually easy, both switches are open when the tape is present and recordable, so its easy to snip off their cables at the board end and that makes them appear permanently open circuit - the electronics now think there is a recordable tape at all times.

Removing the rest of the cables to the mech was a long and laborious but straightforward process.  The deck was put into pause/record with an incoming signal making the meters do their thing. I then switched off, disconnected the mains and clipped off one cable from the mech to the board. Switched back on, set up pause/record and checked the meters were still responding. Then repeated these steps for each cable, one at a time. In this way I'd know which cable I'd clipped last, should the meters stop responding.  As it happens, every cable to the mech was cut at the board end with no ill effects on what I wanted the board to keep doing. Next were the output cables, again, no issues. Then off came the front panel, base panels and out came the mech.  The level of engineering in these products is remarkable given they sold for around £150 back in the early 80s.

At this point, after stripping the Intersekt of its internal boards (about a 15 minute process of about 20 screws), I did some more careful measuring of the board and how it would fit into the LK space - orientation and position to allow room for rear connections, the record level control and the transformer. Satisfied that there was a plan in mind each of the cables from the transformer to the board was numbered with 2 labels about 20mm apart and then each of those cables was cut through, between the 2 labels.  


Bench testing with the board still in the chassis, but with the mechanism removed

With the board out of the deck (meter display still attached), the transformer came out next and the chassis of the deck is ready to go to the recycling centre to be part of the next Nissan Leaf or similar. I then installed temporary insulation around the mains side of the transformer and re-connected all the outputs to the board on the workbench.  This allowed another check to make sure everything was working and there were no dependencies upon connections to the chassis of the deck.  All working well. The slim play / record / pause etc. buttons on the front panel are a simple top pivot design with the bottom end of the button connecting with a micro switch mounted on the board.  So the next thing was to check with the multimeter on if the pause and record switches were electrically open or closed when in their engaged position. These were closed when activiated, so a small loop of wire was soldered into place between their pins to make them permanently "on".  All connected back to the mains confirmed that powering up the circuitry put the system automatically into pause/record and the incoming signal was fed to the meters. The record level control also remained fully operational.

Sorting out the location of the transformer, creating a mounting for the Pioneer 115V / 240V switch (the plastic round unit just above the earth wire) and temporary connection of LK mains switch to the transformer (later this connection was made permanent by soldering the wires together and double wrapping the joint in heatshrink insulation). Of course, all the appropriate fuses and earth connections are retained for safety purposes.

Now its time to trial layout everything inside the LK box, choose the length of plastic "stand offs" to support the board off the base of the box, mark up for mounting screws, check clearances for electrical safety, allow adequate room for the meters to be mounted on the back of the front panel, make sure there are no obvious places where components might not get a little air around them etc. Then the chassis was drilled to accept a mix of M3 and M4 bolts, depending on their role - M4 for the transformer, M3 for the stand offs. Each hole had to be filed to remove burrs etc.  The chassis is all aluminium sheet, so drilling and filing is easy work.  I did make a small error at this point, choosing 30mm high stand offs on which to mount the board, giving lots of access room underneath for the mains cables etc., but it meant 1 capacitor was about 3mm too high for the sleeve to slide back on, so the standoffs were all removed and reduce to 24 mm in height - allowing enough "headroom" but not creating any issues for the cabling.

Now the meter display cables were all labelled up twice, cut through and extended to allow them to reach the new location of the meters in relation to the new orientation of the board.  Compared to the original Pioneer orientation, the board is rotated through 90 degrees to give enough room for the transformer - you can see how the front panel headphone / microphone sockets and tape type selectors are up against the bank side panel of the LK box.

Above and below - extending cables for power or to the meters.  All the internal cables are very thin solid core copper - like telephone cables. Relatively easy to work with as they stay where you bend them to, but easy enough to snap off. Lots of photos were taken during the build from various angles.  Only once did I need to refer back to them to work out where a cable had sheared off level with the board, but it could've happened several times. Here a cable is extended by twisting the 2 lengths together, soldering the joint, straightening and then fitting with heatshrink insulation

Trial installing the board in its new orientation and checking it still works before re-routing the input cables (the pair of light grey cables travelling from left to right in the picture) to phono sockets on the back panel. On the left of the picture you can see the headphone / microphone sockets facing into the blank side panel of the LK box. Across the back of the box you can see some of the many holes left behind by the removal of the large number of connections required for an Intersekt

Now the captive lead for the audio input needs to be re-routed and connected up to phono input sockets on the back panel of the enclosure.  Note, the output signal cables have been removed entirely - there is no intention of this box being in the signal path, so no need for outputs. A pair of gold plated phono sockets were unsoldered from the Intersekt board and re-used here.  The Pioneer's "record" indicator LED needed longer wires so it could reach the small hole in the LK's front panel to take up its new role as the "power on" LED.

Extending the cable to the "record" LED so that it can be installed next to the power switch on the LK front panel

Re-using a pair of phono sockets from the Intersekt board, replacing the Pioneer's orignal flying input lead

Attention then turns to the exact location of the meters on the front panel.  They could have fitted pretty much anywhere along the front, but I chose the left of centre location as it kind of reflects the way the control / display panels on Classiks and Kairns are located.

The front panel looks like a chunky casting, but that's just a deception of the simple but very effective industrial design. Its is an alloy casting, but it only about 1.5mm in thickness.  So this is easy to cut due to the thickness and the relative softness of the material.  An 8mm drill and a jigsaw with a very fine blade gets the main work done, followed by some gentle filing and sanding.  In the location I wanted to use there are 2 cast bosses on the back of the panel, presumably drilled and tapped on some models to provide support for a component or board.  I used a powerfile to gently file these away, being very careful not to go too far and create unwanted holes in the panel.  Marking, filing and cutting pictures below.

Marking Up

Unwanted bosses (not the first time that phrase has been used, I'm sure!)

Bosses no longer getting in the way

Cut complete and edges filed

Prepped ready for painting

Now for the most laborious part of the project - painting the front panel.  There are about 8 coats of primer (2 of these would not have been required, if it wasn't for dust getting in the paint) and 6 coats of satin black. First primer coats flatted with 400 grade wet n dry (used wet), later coats with 800 grade then each coat of black flatted with 1000 grade until the last coat. Then the panel was baked in a 50 degree oven for 40 minutes. Rattle cans of Acrylic paint are sufficiently effective.

Finished satin black panel - a shame to loose the branding, but unavoidable

Final stages of assembly are pretty straightforward. The phono input sockets mount into existing holes on the back panel, as does the level control - something that should only need to be set once to match the level of the input signal from a pre-amp. I filled the rectangular holes along the top part of back panel with a small strip of black aluminium left over from another (half completed) project and some araldite adhesive. I really should have cut out the back panel and created a new one, but it was getting to the point where the time and effort had gone far enough.  So I chose to fill all the remain holes with 9mm rubber blanking plugs. Which I thought would be quick and easy, but in the workshop at -4 degC those things are tricky to work with!

So that's all done. Pointless, but proved that what I set out to do is actually possible. And maybe its one of those "because its there" challenges. Now my workshop system can have a Linn VFD VU meter display box without impacting on the sound quality.  There are simpler, easier and quicker ways to achieve this with a kit off ebay, but it wouldn't be quite the same, and it wouldn't be those iconic Pioneer meters.

I call this new box Linn Formation - Pioneer Blue Line Edition.  Here's the finished article.

The completed layout. Transformer with insulation on all the previously exposed pins, phono input sockets and input level controls on the back panel to the left of this photo, wiring now at least a little bit neater than Pioneer's

First test after the build

Back panel with over 30 blanking plugs. Silver coaxial left / right input level control and pair of input sockets. Top rectangular holes blocked by an aluminium strip bonded to the inside of the back panel

Saturday, 6 February 2021

Inside Linn's Akurate Kontrol/0/D Pre-Processor (Exotik DA)

 In the early years of the 21st Century Linn was still fully on board with multi-channel audio / surround sound with the Akurate Kontrol product.  They're back into surround again now, with the optional module of the Exakt DSMs - albeit with a lack of choice of centre speakers and no subs at all in the range - perhaps they'll arrive soon?

Introduced in 2006, the last dedicated (audio only, no video switching) pre-processor from Linn was initally branded Exotik and with the digital processing option, the Exotik +DA. About 3 years later, when Linn rationalised their products into three distinct tiers - Majik, Akurate and Klimax - the Exotik was re-branded as Akurate Kontrol. To distinguish this from the later stereo only Akurate Kontrol/1, the product is generally referred to as Akurate Kontrol/0. Part way through its life the power supply was changed to Dynamik which was also offered as an upgrade to existing products, hence the full nomenclature of the version in these pictures is Akurate Kontrol/0/D.

It was an excellent sounding processor and acquitted itsself well enough for stereo music to justify its place in the Akurate tier.  Covering Dolby ProLogic, Dolby Digital and DTS, it was equipped to offer audiophile grade cinema sound.  It was never directly replaced when Linn decided building its own fully licenced products was no longer cost effective at their sales volumes.  The latest foray into surround relies upon the use of a module built by a partner supplier.

The excuse for opening up this particular version was for the conversion to the workshop system colourscheme...


General View

The digital board (the " +DA" on the Exotik version)

Main board - notice top left that its origins are from the much earlier Kinos pre-processor product

This small board allows the user to choose between MM phono, MC phono or line level on one of the analogue inputs

Rear of the front panel / display

Across the main board

Mainboard dustscape

Refinished for the workshop system

Friday, 5 February 2021

Product Review: Songcorder Software by Meraki Acoustic

The Linn DS (digital streamer) family is, at the time of writing, just over 13 years old.  The first internationally recognised true HiFi company to take streaming seriously, Linn launched their top of the range Klimax DS as a market leader.  There have been several iterations since, all versions able to be upgraded to the current specification.

But the family has grown since then.  There are pure DS players (network connection, streaming capability, onboard stereo DAC), DSMs (all the above plus pre-amplifier capability, including analogue inputs) and other DSMs that do what DSMs do but have a built-in stereo power amplifier. Spanning the complete range too - from the designed-to-be-hidden Sneaky, through Majik, Akurate and Klimax ranges.  Most recently the  modular Selekt DSM has been launched which can be all sorts of combinations of streamer, pre-amp, all-in one stereo, all-in one surround sound and more.  Surround modules can be added to DSMs too. Then we throw in the choice of outputs - traditional stereo analogue, Linn proprietary Exakt digital link to Exaktboxes or Exakt integrated speakers and a combination of analogue & Exakt on some units.

That's probably pretty confusing to the uninitiated, and understandably so. But there's even more to it than that. Software. Without software these players would be expensive (or very expensive) decorative (or in some cases not so decorative) items.  The software that makes them operate is vital.  But one of the beauties of the Linn DS / DSM is that the software continues to evolve. And every DS / DSM since that first one back in 2007 is capable of being upgraded to today's software versions. And Linn releases the software updates without charge. That's right, a 13 year old product is still being upgraded by Linn on a frequent basis - and it doesn't cost the original, nor subsequent, owners a penny.  Buyers need to keep that in mind when they think Linn gear is too expensive. Some of these software releases have unleashed even greater sound quality and musical enjoyment from the streamers, not just new functionality.

So even though Tidal and Qobuz streaming services weren't available back in 2007, they are available on the DS/DSM today through those software updates. As is the ability to tune the output of the music to counteract the worst impact of the room on your music.

Why do I mention all of this? Well where there is software, there needs to be ideas. Creative ideas. Ways to make hardware perform tasks that weren't originally envisaged. New ways to exploit, new services to offer, new functionality for the user. And that's where Meraki Acoustic comes into the Linn DS/DSM picture.

Many of the DSM devices have a built-in analogue turntable input - MM or MC - and this was the inspiration behind the Songcorder app from Meraki Acoustic. The DSM turns that incoming analogue turntable sound into a digital signal. What if there was a way to capture that digital signal, feed it to a computer and capture it as an audio formatted file? Could that be a way to digitise all that precious and delicate vinyl? Could a DSM customer capture their favourite music in format that doesn't wear out? Would they finally be able to get that vinyl only release from 1976 into a digital file to listen to on the move, in the car, in rooms without a turntable?

The answer is yes - with Songcorder.  Meraki kindly sent me access to a copy to download to investigate further.  As most will know, although I have a turntable, its not the main source in my system.  Even so, I thought it would be interesting to give Songcorder a try.

The download was easy as was install on my up to date version of Windows 10 running on a Dell Latitude E6440 laptop which has a 3 year old Core i5, 8GB of memory and an SSD.  I tested it along with my Ariston RD80sl / Linn Basik Plus / Goldring 920 turntable and my Linn Akurate Exakt DSM through its built-in MM phono stage. The instructions recommend a wired connection for the computer running Songcorder and for the DSM - wifi being "potentially" troublesome.  In practice I found wifi to be almost always troublesome so stuck with an ethernet cable throughout, which was totally stable. Your DS or DSM must have Songcast enabled for Songcorder to do its thing.  It must also be listed in the compatible devices list here:

If you're as old as I am you'll remember pressing pause and record on your tape deck, gently lowering the stylus onto the record and then pressing the pause button before the track started - the way we used to be able to play our own music in the car or the kitchen or later on a Walkman or similar.  Well Songcorder is similar in concept - whatever you're playing through the DS or DSM will be captured and recorded as a digital file ready for use with your streamer or loading to your DAP.

Once installed and fired up, this is the simple interface: 

Select your wired connection in the Network Interface drop down, select the Linn DS/DSM you want to record from (the software found all 3 Linn devices on my network) and select which directory you want to drop your recording into, and that's it. Nothing fancy but nothing complicated nor confusing - refreshing.  I really like the simplicity.  Start your turntable playing and click on that big red button. Done. Of course, you need to keep on top of what's going on.  You could record a whole album side and leave it at that. Or you could then use some software tools to break down the side into individual tracks. Or you can do what I did and capture individual files for each track on the LP.  When recording you get the simple screen below, just click Stop when you're done.

For each vinyl recording Songcorder creates a file at 24 bit 192kHz resolution using my Akurate Exakt DSM - I suspect that this is the resolution used by Linn when converting analogue to digital. Recordings from earlier Majik DSMs would be resolved at 24 bit 96kHz due to its lower resolution ADC stage. Songcorder generates what looks like a unique filename in FLAC - a widely supported format that can be converted to other forms by many of the popular digital music file handing programs available. You'll need a music file editor (I use TagScanner but there are lots of choices) to rename the tracks and input your own meta data such as Artist, Album, Track, Year, Genre etc.

Only one element of Songcorder I didn't enjoy is that it goes back to its default folder every time, rather than remembering where you prefer to capture your recordings. Its a very minor point and doesn't take away from the fact that this is a superbly simple tool that does exactly what it sets out to do.  The good news is that this gave me the opportunity to test the bug reporting process and happy to say that work is underway to address this as I type.

Sound quality is excellent - when playing back a vinyl recording, as far as I could tell, there was no degradation from the original analogue conversion I was listening to as recording took place. 

Whilst vinyl is the inspiration for Songcorder - to preserve those rare recordings or to create back up copies, for example, I did find that Songcorder will capture any stereo audio input on the DSM that's connected by toslink, SP/DIF or analogue inputs. Streaming works, but HDMI was less successful - theoretically it should work with stereo unencrypted content, but I didn't find any. Streaming works equally as well as the analogue input and captures the file in the resolution of the original source.  Of course an anlogue stereo output from such devices would work. Songcorder picks up the signal before any of Linn's SPACE Optimization or Exakt processing is applied.

Of course, one should be careful about copyright issues and only capture music already owned, or material that is free from copyright. 

An impressive piece of very effective software - I was so pleased by its performance I have purchased a copy, an equally smooth and simple process. At the time of writing the software cost a mere 20 Euros and is highly recommended.

Monday, 25 January 2021

System Synergy and Expectation Bias

A couple of phrases you might hear or read from time to time when researching or in dicussion about music playing hardware.  What are they about, and why would I be writing about them right now?

Sometimes how to phrase an audio experience is very elusive.  Some readers might say that continues to be the case, even for published posts!  This particular case has been more elusive than usual - its been a year since I made my notes on this project. How to frame those notes have been through about 6 iterations and I'm not sure I'm there yet. But a lot of time and effort went into this, and its time to just put it out there, good or bad, as the products involved might be old hat before I'm content with the wording.  I wanted to not just list the reviews, but to try and put them into some kind of relationship - finally settling on the synergy and expectation bias frame to hang it all from. So here it is, some kind of a review from circa Christmas 2019:

A 6 week or so period of frenzied (at least by my standards) box swapping, mostly for research purposes, but also to inform some purchasing options, mainly not for me, but for a good friend of mine, who's system features here. And it's all given the opportunity for another few blog posts too.  Elsewhere on the blog you'll soon find some impressions of some of the products involved so I won't go into huge amounts of detail here, but will cover a "flavour" of the systems involved. Long form, no photos and no apologies for that.

Coming to Terms

What do these words mean, in an audio system context?  Well, let me have a go at describing how I would use them, so these aren't definitions as such, but more setting the context of this post.

System Synergy - when putting together a whole series of components into a system - CD player, turntable, amp, speakers, cables, the room - do they work together in the way you would like them to?  Or do they lead to some kind of imbalance in a system?  Do you hear a coherent musical message, or is there something that grates like a bloated bass when you wanted tight tunefulness or a laziness to the sound where you wanted pace and energy?  Well, when it all comes together to meet your hopes, that, to you, would be a system that displays good synergy - the components complement each other to give an overall positive result.  If it doesn't work, the synergy's not there, then there needs to be a further search to find what's causing the imbalance, to work out what needs changing or improving.  Note that there is no mention of cost in this description - it has nothing to do with cost, it has to do with balancing characteristics of components to get a good result.

Expectation Bias - we'd probably all like to think we're impartial and can evaluate what we're hearing without regard to what we think we're going to hear or prefer.  Are we going to prefer one brand over another because its been a favourite for years? Are we going to like the sound of the component that has the nicest finish or a preferred species of wood on the front? Will we expect the more expensive bit of cable to sound better than the cheaper one?  Of course we all have these biases - its part of what makes us human.  I think its possible to try and minimise them by being concious of the issue, but probably not to eliminate it.  Blind testing (where you don't know which component is playing) is often advocated, but it just introduces different problems (such as an unnatural focus on the listening process), so I don't consider it to be the answer.  Just accept that a level of bias is inevitable and that it influences your opinion, but try to minimise this as much as possible.  That's one of the reasons why this blog is just opinion - there's nothing scientific about most of it, and that's just the way I prefer to keep it - the disclaimers make this clear.

So How?

If we're susceptible to expectation bias, how do we know if a system really has good synergy or if we're just expecting it to have good synergy?  Well I don't think there's a way to be totally objective (objectivity is restricted by the measures used anyway, as there's no way of knowing if a set of measures is comprehensive or relevant enough, that's why science keeps evolving) so just embrace the subjective and try to deal with it.  Its also worth noting that listening to music is meant to be an emotional experience and therefore, to me, it seems that choosing a system on some level of emotional evaluation is entirely reasonable.


Mustn't leave the room out of the equation either.  Lots of impact on the way a system sounds comes from the room and the interaction of the speakers with the room, so I've included that in these musings to try and cover a little of the less obvious.  Where possible (ie with the Linn front end) SPACE optimisation was set up for each system.  What is often referred to as "Tune Dem" or "Tune Method" was used to position speakers and adjust SPACE.


I'll go straight into listing the systems that I listened to over those few weeks a year back and that have fed this post, along with my expectation bias - as without that the rest of the post will be a bit meaningless.  The premise of the post being that my expectation bias wasn't met and I'll describe where and in what way.  Perhaps this is part of trying to be aware of my own bias and trying to rationalise how I have, at least partially, probably, tried to put it to one side and listen to what's actually going on. Still a big "if" though.


I won't go fully into every detail of the systems here, such as precise cables etc., but the main components that differentiate the systems.  I suppose its fair to say that my expectations are in comparison to my everyday system.  They're based on a mix of hearing 2 of these systems frequently and hearing the other components in other systems. It is fair to say that every single one of the components in these systems is highly respected, so there are no "duds" at all, and that's where the synergy element becomes so intriguing.

System 1: 

Melco N1ZH server, Linn Akurate DSM / Exaktbox Katalyst streamer, 3x Lejonklou Tundra Stereo 2.5 power amps, PMC Twenty.26 converted for active use, medium size pretty much rectangular room. This Exakt system uses filters for the PMCs available from, not the standard versions.
Expectation Bias 1: this is my everyday listen - its taken a long time to get to this system - it has a great deal of  detail but also a very musical flow to it.  Its not the last word in dynamics and slam but I can listen to it for hours (and do so, most evenings), following the tunes, enjoying the emotional impact of the music.  Its not perfect, no system ever is, but its what works very well for me.  What's my expectation?  Well, its more a reference point for comparison than an expectation really, but I would say that I expect it to be very difficult to beat in this evaluation with other systems having some benefits but not the overall balance of this system.

System 2: 

Melco N1ZH server, Linn Akurate DSM / Exaktbox streamer, 2x Lejonklou Tundra Stereo 2.5 power amps, Kudos Titan 505 on their dedicated stands and configured for active use, medium size pretty much rectangular room
Expectation Bias (comparative to System 1): more detail, a more dynamic presentation, a touch of brightness. Perhaps a little looseness in the bass.

System 3:  

Melco N1ZH server, Linn Akurate DSM / Exaktbox streamer, 2x Naim 250DR power amps, Kudos Titan 505 on their dedicated stands and configured for active use, medium size pretty much rectangular room
Expectation Bias (comparative to System 1): more detail, a more dynamic presentation, a touch of brightness, possibly assertiveness approaching aggression, more impact in the lower frequencies.

System 4:  

Melco N1ZH server, Linn Akurate DSM / Exaktbox streamer, 3x Naim 250DR power amps, PMC Twenty.26 converted for active use, medium size pretty much rectangular room
Expectation Bias (comparative to System 1): a more dynamic presentation, a touch of brightness, probably assertiveness approaching aggression, a loss of musical flow.

System 5:  

Naim NS01 server, Naim NDS streamer with 555DR PSU, Naim 252 pre with PSU, SNAXO crossover, 2x Naim 250DR power amps, Kudos Super 20 configured for active use, a medium sized room but L shaped, using the larger part of the L, which is smaller than the other listening room.
Expectation Bias (comparative to System 1): a system I'm already very familar with - a more dynamic presentation, too bright for my taste, assertiveness approaching aggression, a loss of musical flow somewhat compensated for by the enthusiasm.

System 6: 

Naim NS01 server, Naim NDS streamer with 555DR PSU, Naim 252 pre with PSU, SNAXO crossover, 2x Naim 250DR power amps, Kudos Titan 505 on their dedicated stands configured for active use, a medium sized room but L shaped, using the larger part of the L, which is smaller than the other listening room.
Expectation Bias (comparative to System 1): a more dynamic presentation, slight brightness, assertiveness approaching aggression, perhaps retaining the music flow, extra weight (because of the smaller room mainly), more information delivered.

So there we are, cards on the table. Something else worth mentioning - those who say "all competently designed solid state amps that measure within a gnats whisker of measurement-of-the-week all sound the same".  I have an expectation bias about that too - its just not true - I've heard far too many amplifier changes in systems to know that this is not true.  Unless I've only ever heard incompetently designed amps with bad measurements, which is also not true.


So here we go, here are the descriptions of systems 2 to 5 and a conclusion, that's not what I thought it was going to be:

System 2: the biggest suprise for me really.  The Titans give a far more detailed presentation than the PMCs, its suprising how much more they reveal in the recording. The extra brightness is certainly there, as per expectation, but is this just brightness or is it more detail too?  I'm going with the latter as there's more expression too in the vocals.  In fact the greatest strength of the Titans in this and the other systems is their vocal expression - its quite outstanding by any standards of the many systems I've heard.  Another element that was as expected is that, compared to the Titans driven by the Naim amps, there is a softness in the bass with the Tundras.  And here I think we see the first of the synergy issues - perhaps the Tundras and the 505s don't sit all that comfortably together as there was another element here that I wasn't expecting as the Tundras have never sounded like this in any other system I've heard - a sense of relentlessness.  Not shouty aggression, just a bit too much going on and losing some of that lovely flow that the Tundras present in partnership with speakers such as the PMCs and Graham Audio LS5/9. That, juxtapositioned with the lack of pace in the bass lines, meant that this wasn't the best of the combinations compared here.  There was a fair bit of experimentation with toe-in with the Titans in this room as there's definitely not the width or precision of soundstage compared to the PMCs. Don't get me wrong, we're talking about a decent sounding system, but in absolute terms, it lacked the right synergy.

Expectation bias 0: Systems 1

System 3: its long been a thing, a Linn front end with Naim amplification into a pair of speakers of choice.  Traditionally, of course, that was an LP12 and a passive Naim pre-power combination. Here we bring the tradition up to date with Linn streamer/pre and Naim power. Yep, more impact in the lower frequencies, no doubt about that - nearly reaching as low frequencies as the PMCs too - impressive. Mind you, the 505 standmounts on their dedicated and structurally massive stands are actually much more intrusive in the room than the much more constrained PMCs (which I always use the the grilles on). Definitely more detail from that (very) expensive tweeter - is it too bright? Its touch and go and track dependent, sometimes just tipping over the edge. But here's the thing - music, because listening to the music on this combination was very engaging - but its different to the regular system 1. There's an immediacy without shoutiness and that detail mentioned earlier but there's not the same musical flow that a Tundra delivers (in partnership with speakers they work happily with!). Whilst the Titans have very impressive bass power and immediacy, they still didn't quite fill the room like the PMCs do. Vocals are excellent again and there's plenty of engagement into the meaning of the music. This is good, but is very much along the lines of the expected.

Expectation bias 1: Systems 1 

System 4: before I go any further, just consider this: moving from Tundra to Naim power amps is a 50% price hike in amplification prices. And they do have a very different mix of abilities. I can also say that the Naim NAP250DR is a heavy beast.  Not too easy to lift onto a shelf and keep the shelf edges in pristine condition.  Those massive toroidal transformers are very heavy indeed. And noisy, good grief they're noisy. Not through the speakers, no issues there at all. But my goodness do they hum. I know my incoming mains typically carries about 2.5% distortion (the PS Audio regenerator displays this information) - I'm not sure if that's high, typical or low, but its certainly enough to get the NAPs a buzzin'. Just to be clear - neither of the power amps in this sequence of tests were plugged into the PS Audio, only the low current source electronics are fed from the regenerator. So there is no doubt that the 250 boxes have a tighter grip on the PMCs. There is more slam, there is more immediacy. There is a little more enthusiasm in the upper frequencies but it never becomes wearing in this system. All of this as expected.  What I didn't anticipate at all was the way everything hung together as a piece of music in very much the same manner of the Tundras into these speakers. Even that very taughtly controlled bass didn't disrupt the ability to follow the tune. Very good indeed. Except for gaps between tracks when the boxes from Salisbury irritated me significantly. Expection or surprise - calling this one a score draw.

Expectation bias 2: Systems 2

System 5: nearly all change.  Change of front end, speakers and room. As I'm very familiar with this system as it as grown and developed over the last 5 years with the Super 20s, then my "expectation bias" notes above are really all you need to know - they're more about experience than expectation. So let's call this one a no score draw!

Expectation bias 2: Systems 2

System 6: Would we expect (or have a bias!) that the 70% more expensive 505 with their stands would have a family feel when contrasted with the Super 20 floorstanders? Would there be a family resemblance from the North East parentage? Yes. And No. The Titans have that same family immediacy but the way they feel more "grown up" was far greater than my expectations.  Especially in the treble - where the Super 20 appears to major on detail, I find it gets too much after a while. What seems to be the case with the 505 is that the detail is real and it knows that it doesn't have to try too hard. I was a little worried about the size of the bass that comes with the upgrade to Titan (yes, the standmount has far more bass output than the floorstanders, in part due to their isobaric loading) in this smaller room, but it worked very well indeed - and there's an important reminder that system synergy includes the room.  Here, of course, there is no SPACE optimisation to make the best of the room so its all down to component choice and speaker positioning. What was just a touch underwhelming from these speakers in the bigger room is just right here. I don't get the same sense of ease of musical delivery compared to the Linn front end with Naim amps and these speakers, but I can see how their immediacy could appeal. No stridency, no experience of fatigue.

Expectation bias 2: Systems 3


So you can now decide if any of this tells you anything about expectation bias or system synergy (which needs to include the room).  I certainly was surprised that the weight of the 505 could work in the smaller room, that the Tundras and Titans just didn't get on together at all and that I'd enjoy the Naim power amps in my regular system anywhere near as much as I did.

I wish the Naims amps didn't have 2 faults: that hum I can't just dismiss nor excuse, and their price. So I shall continue to be extremely happy with the Tundras as they work very well indeed with the PMCs.  As for my friend with the largely Naim system - he bought the 505s and has them working even better now the speaker cables are Super Lumina.

Wednesday, 15 July 2020

LINN Exaktbox-i: Photo Comparison Between Majik (Pre-Katalyst) and Akurate Katalyst Versions

Linn's carefully hidden Exakt separates system consists of the following key components:

An "Exakt ready" streamer / pre-amp (such as Klimax DSM / Akurate System Hub / Majik DSM amongst others)
A proprietary Exaktlink cable which uses standard Ethernet cables
An Exaktbox which is essentially a digital active crossover and digital signal processor, amongst other things (such as Akurate Exaktbox10 / Klimax Exaktbox)
Power amps, which can be from Linn or any other manufacturer.

However, there is an option to combine the last 2 elements - the Exaktbox and the power amps.  This is called an Exaktbox-i and combines 8 channels of DAC / digital active crossover and 8 channels of power amplifiers in Linn's normal just-a-bit-narrower-than-standard box.  There's really only one of these in the range, but it has evolved in the 4 or 5 years it has been available:

1 - Intitially it was a Majik Exaktbox-i containing Majik level DACs and 100W power amps, all in the Majik level casing
2 - Then it was re-badged the Akurate Exaktbox-i but still contained Majik level DACs and 100W power amps, but now in the more substantial Akurate casing.  This version was very short lived, soon replaced by:
3 - Akurate Exaktbox-i with Akurate level Katalyst architecture DACs but sill with Majik level 100W power amps in Akurate level casing

As is often Linn practice, those who bought the earlier products don't get left behind when there's an upgrade.  So it is possible to have a Majik Exaktbox-i upgraded to Akurate Katalyst Exaktbox-i internals but retaining the Majik casing.  The first Akurate Exaktbox-i can also be upgraded but I expect most of those were done pretty much automatically as there was a deal where the cost to upgrade was only the difference in price between Katalyst and non-Katalyst versions.  It would be very odd to find one of these that hadn't been upgraded, given the small price to do so.

I've been using an original "pre-loved" Majik Exaktbox-i for a while for Exakt filter design activities.  For some reason, now seemed to be the right time to take that up to Katalyst specification.  So that has allowed me to make comparisons between the circuitry of the non-Katalyst and Katalyst versions.

A similar comparison of the Akurate Exaktbox10 (April 2018) is here.

Much has been written about the musical capability of the upgrade from Linn's previous generation DAC to the Katalyst DAC, and I can't think of anyone who has anything but praise.  An example is here regarding the upgrade in Linn's onboard active Akubarik speakers.  So I'll not write about that here as it is well trodden path. With the upgrade being applied to the Exaktbox-i back at the Linn factory, there's no ability to directly compare the 2 products back to back.

To the photos.  Note that every single audio board (DACs are combined with the power amps) and the Exakt processing board are changed in the upgrade - the photos with the green circuit boards are the Majik non-Kat version, the black circuit boards are the upgraded Kat version.  The PSU and the casework is untouched, save for the addition of a Katalyst sticker on the underside.

Externally the box is unchanged


The most obvious difference from an overall perspective is the change from green to black and gold boards

Each side of the enclosure has 4 channels of power amps (the chips bonded to the heatsink are the obvious components) and 2x stereo DACs.  The DACs above are the cross-shaped cluster of an oblong chip surrounded by 8 smaller chips

Here you can see the difference in the DAC architecture - now we have the 2x stereo DACs which are the 2x flat square chips - the board layout is very different

The "flyover" board down one side of the enclosure is the Exakt processing board.  The above original layout has 3x FPGA processors - the large flat square chips. The one on the left is the "master" processor with the other 2 processing 4 channels each
On the new board there is one more powerful FPGA chip in the centre doing the work formerly done by 3x separate chips

Board labelling for channel 5 to 8 DAC / power amp board has been relocated

Relocated labelling for the Exakt board

The increased complexity and density of the components surrounding the DACs is clear in this comparison.  The Katalyst DAC has multiple different regulated power supplies which probably explains this increase in components


Xlinx FPGA that does all the Exakt processing

DACs and power amps for channels 1 to 4 live in the shadow of the Exakt board