[1.2] Operation and Features
[1.3] SLA-1 Innards
[1.4] Pics of the Insides
[2] Long Term Plans
I've been thinking about the amp end of things for a while. I looked at the AKSA kit amplifier and had my ear to the ground hearing about the Samson Servo 170. The 55W/Ch AKSA kit does not include chassis or a transformer costing about $385 Canadian verus the SLA-1 at $300 even. The immense success of the DI/O in my playback system persuaded me to become a repeat customer.
The guys/gals designing at ART are turning out some pretty decent stuff (DI/O for one) these days with great price/performance ratio. Hinting stronly to my wife, I recieved a belated BD gift in the form of a SLA-1 over the weekend. I was tuned into the Audiogon SLA-1 infomercial thread prior to it being deleted and am aware of some of the underhandeness going on with regard to false pseudonyms posting about their own companies. Regardless, I sensed something very real underlying all of this. I have the original post that mentions the SLA-1 in the same breath as Mark Levinson and Pass amplifiers.
The specifications on ART's home page show the ART SLA-1 to have high power and low distortion. The one indpendent review found was one of the SLA-1 and partnering SLM-1 speaker. Being down under, the price premium for shipping this product there is quite steep. There is a PDF available. The manual is currently not available.
The following links will direct you to initial discussion threads related to the SLA-1 discovery, aftermarket upgrade services, and competing amplifiers choices offered by Samson. Various links are distributed in this document in each subsection. A reservoir of "design info" stored at Yahoo groups can be accessed by following the links in the 2.1 Mod subsection.
Any component evaluated needs to be taken within the context of the system it is evaluated in. This section covers short term and longer term impressions of the SLA-1. The latter section will be updated over time and as the SLA-1 is evaluated in different systems.
My long term reference has been the integrated amplifier that I have used for the last two decades. It is not perfect. Its imperfections to me are known and I can to some degree filter them out when comparing components ahead and behind it...to some degree. To a large degree, the changes that I have wrought on the 3020 are very similar to how the SLA-1 comes in stock form: No tone controls, simple gain path from line level to to amplifier input circuitry. The NAD has the benefit of the following: Seperately regulated predriver for the amplifier output stage, Boutique caps (Wondercap), Updated internal wiring with Kimber 8TC, DC coupled preamp output, ALPS potentiometer and most impotantly a host of subtle mods/cap types to tune it to my system over time...most of which I have forgotten. Out of the box, the SLA-1 has some of the NAD "sound/imperfections" removed and this has encourage me to listen to some CDs which I have not listened to in a while. The balances are close but I'll have to wait until I compare it to my friends 50W reference Tube Soundtech amplifier based system to sort out the signature of the SLA-1. From what I have heard to date, the designers at ART have done their homework. I could live with it for quite a while.
I presently run a highly modified NAD 3020 that is the equal of more costly amps in terms of the first Watt (probably more) of amplification. It has served very well in the role of a microscope to sort out deficiencies in upstream cables and sources. Out of the SLA-1 box, last night, I still find the 3020 to win in terms of information retrieval and overall balance. The SLA-1 sounds surprisingly close, though. From turn-on, the 3020 normally takes about an hour or two to warm up to its sonic splendor.
Within 1/2 hour, I opened up the top, disabled the fan and gave it more of a listen. Updating the power cord to an Belden 83802 based unit gave a bit more resolution. The next morning it sounded somewhat better after having been left on all night. I'm not sure if it sounds better with or without the ground "lifted". It sounds promising. I do not know if it beats 95% of the amps out of the box (as claimed by Don) but it sure has potential at this early stage. Perhaps my little 3020 beats 96% of the amps
The amp is stable driving 2/3 Chris Venhaus CAT5 loudspeaker design.As these cables are fairly capacitive, the fear of the output stage blowing up and frying my speakers was in the back of my mind. I tested this first on my "cheaper" Radio Shack LX5 speakers before moving onto my Axiom M22ti.
The urge to unplug it has not hit...which is a good initial sign. I am beginning to listen to CDs from start to finish (always a good sign) at this early stage.
Using a wider variety of music over the last few days shows me that it is getting better as I leave it plugged in. I did not expect my jaw to drop on the first day but good systems have a way of proving themselves worthy over time.
On the construction side, I noticed internal construction details to ease/lower cost of construction related to the snap on connections between the power tranformer outputs and the full wave rectifier and the connections between the Attenuator controls and the main amplifying stage (read mod potential).
I will adding further comments to this section from within mostly the context of my system with specific references to how it fairs with different types of music. Stay tuned. I'll might organize subsections to reflect this chronological updating of the evaluation. This might be week to week, month to month and include discussion from the various other parties who are also evaluating the unit. I will also try this amp in another system and report back but the basic rule of thumb for this evaluation is to use the unit "as is" with few to no mods. Section 2.2 will address the long term finessing/modding of the unit and resultant gains from doing so.
"This speed and transparency really surprised me and continued to surprise me over the next few days- everything I played, from gigantic Mahler to intimate Patricia Barber, seemed to have acquired a snap and drive that floated it out of the grunge and actually opened for me a new sense of musical priority"
"...when I put my other amps into the picture the focus softened, things slowed a bit, voices and instruments thickened and somehow seemed “hotter” in an unnatural way: less revealing of time and space... For now, the kid stays in the picture."
"...it can sound a bit lean and at times even threadbare"
"...I find the SLA-1 a worthy, special little amp that
can enliven the listening experience with its speed and
transparency. But it's tonally thin, upwardly balanced, and
can sound harsh unless carefully matched both up and
downstream."
"To these ears, the ART is a very clean sounding and powerful little devil. Whether it is as good as D911 claims remains to be seen, but it appears to be holding its own in my system"
"I want to give the ART more time to break in and
experiment with some different power cords and tweaks before
commenting further on the sound, but I think it is safe to
say that the amp is pretty damn good, perhaps even
remarkable at its price point."
.
June
13, Update: Walter Salas
"The amp is dynamic and clean, with precise placement of instruments in the soundtage, but can sound hard or lean at times and does not quite have the extension/dimension or musicality of reference quality amps costing thousands of dollars more"
"It is a credible performer even in audiophile systems... but I also know that I would choose the Plinius (SA-102) 10 times out of 10."
With the above experiments from the morning in mind, I spent some more listening to the ART SLA-1 in the company of my goldened ear friend Danny with his LT1362 modded DI/O. This unit has the default output level set to almost 7V RMS (compared to my DI/O set to 2V RMS) plus some recent critical fine tuning changes by Danny. I'll refer to this unit as "Danny's DI/O". To cut to the chase, using the supplied rubber feet with the SLA-1 is key to realizing a fullness in the bass and freeing up the amp's dynamics ( The styrofoam packaging blocks make for poor substitutes for the supplied feet but ease the return/exchange of the amp should I have decided to return it...NOT!!!!). Using a SACD/CD recording by Marc Vallee playing an acoustic version of U2's “I Still Haven’t Found What I’m Looking For”. The 7V RMS of Danny's DI/O was ably accomodated...it also bought the sonic level of my overall system closer to level/nature/preferences of my friend's tube based system.
Low level reverb information, bass rhythm details, drive and transparency were now much improved. The limitations of my DAC were more clearly obvious while the advantages of Danny's DI/O (only hinted at earlier seesions with the NAD) was much more tangible, obvious and evident. The SLA-1 is a transparent and fast sounding device that will expose a good signal. Notes come and go in a way that allows the music to breathe. With cork and blutak in hand, Danny made a few more adjustments over the course of the evening. This resulted in a lack of sibilance,quieter background and larger soundstage. Songs seem to play twice as long as their marked times...as if in suspended animation. Notes just seem to float about the room.
With my speakers, Danny was not able to clip the amplifier playing an extremely dynamic recording of a large church choir from Chesky's demonstration disk while playing it at comparable levels to his setup. He estimate's that it was putting out over 50W. The green "signal" LED stayed on continuously. His kilobuck amplifier (with his ATC monitor speakers) will audibly clip. Although I do not normally play at this level, it was interesting to see the SLA-1 hold up to playing levels derserving of the evil eye from my neighbour.
So for me in my system, with the appropriate upstream sources, the SLA-1 can hit a very decent sonic level almost right out of the box (ie. Given a few days of burn-in and a borrowed previously broken-in power cord). Do not evaluate it without the supplied rubber feet attached and use a good source. I ended up turning off the amp around midnight (5 hours later) and thinking about how I needed to upgrade my frontend/DAC to the standard of Danny's unit. I have been well aware of the compromises in my NAD 3020...it is a as someone said a "rose coloured frosted piece of glass" that I have grown to know and like/love for all its warts. The SLA-1 is a revelation. I've had my NAD 3020 for over two decades. Anyone out there want a highly modified 3020...? For a slightly different perspective, Tim Moorman posted his thoughts on the SLA-1, below:
"...certain mid and upper mid range frequencies highlighted or projected slightly forward of the mix..."
"Tonal quality is poor. Lacks the rounded organic tones of good tube or excellent SS gear. Some congestion. Not close to my SS Plinius 8150 integrated, which, admittedly, is far more costly, and pales in comparison to single ended tube stuff. "
"The clarity and speed are good, but at the expense of reduced mid bass and bass involvement...Probably nothing close to it in this price range. I like the attenuation for each channel, compact size, and bang for buck. Good starter amp."
As a experiment, I put the OEM Unicab cable in and gave it a listen. Although the tweak above improve the overall system playback even with the Unicab cable, putting back in the Belden power cable wrought more easily heard changes.
...voila...perfect CD playback forever. You can now use the "next" and "previous" buttons to access any track with an guranteed improvement in sound.
Well it did not quite have the magic that it did in my system. The double run of Kimber 8TC and ATC speakers seem to be less of a load on the amplifier than the CAT5 cabels judging from the much cooler heatsinks. The system responded to power cable changes (Jon Risch to Belden SJT and back) and interconnect changes (Van den Hul Thunderline to sliver interconnect) to bring back the air and some of the dynamics but overall things sounded lean and the soundstage was small. Music seemed to be in a hurry and details seemed to be glossed over. The bass seemed anemic compared to the tube amp. These comments reflect a constant evolution and improvement in the standard of replay that Danny and I have colletively been able to make over the last 3 months. The level is a very high level of playback. Resetting our high expectations, I can say that we heard potential, but the ease and naturalness was missing. Patricia Barber was just a recording. The standard comparison tracks indicate to me/us that it seems to be a frontend filtering of information problem probably assoicated with the opamp, passives, or wire. The impression of speed of the amp and analog sound could still be detected but timing and space attributes got lost in the mix.
How do I resolve the conflicting signatures for earlier listening sessions...I'm not sure. The double run of spaded Kimber 8TC was arkwardly attached using some adapters. I should have bought the CAT5 cables over. Maybe the speakers are truly revealing of the upstream components and my comments about the early gain paths in the SLA-1 have some merit. Anyways, I had returned the NAD 3020 back into my system in the morning and gave it a quick listen with the "skip track access" and borrowed Van den Hul Thunderline speakers...very sweet and a level of detailed and warmth was added into the original setup. It sounded quite listenable. Bottom line is that synergy is a key apect to system setup. Trust your ears and DIY your way to synergy.
I had a mild headache resulting from being over some details on a project for most of the day...so my listening acuity and comments that follow reflect mostly what Danny heard. Prior to plugging in the SLA-1, the Soundtech Tube amp with the new JR interconnect was audioned. As Danny put it: "I'm able to listen and hear more information at lower volume levels with greater enjoyment". His conclusion is that the cable and shield have lowered the noise floor and increased the signal to noise ratio. To me the Norah Jones album sounded non sibilant in his system for the first time. In its place was a better integrated mix with certain textures to plucked string that I had not hear before but without any sibilance and all very relaxing. On the Mary Chapin Carpenter album (Come On Come On), I got the best impression of the room size and acoustic that I have heard. You could sense the witdth and depth of the recording venue and almost even sense the height of the room...quite special. Some of my listening preferences are a result of getting used to what is not right...it is that sound associated with "silver" where certain parts of the spectrum are emphasized giving the sense of more information and a larger soundstage. The quieter background of the JR cable seem to have dminished this euphonic colouration. Putting in my next favourite cable, the Van der Hul Thunderline, confirmed this. With the JR cable back in, the SLA-1 was auditioned and comments are made relative to the Soundtech amplifier.
As I have already let the cat out of the bag about the improvements(dynamics, soundstage, ease and naturalness) in performance over the previous audition, there still remained some room for improvement. Cold out of box, it initially sounded lean in the midrange but with a upward tip to the balance (this was alluded to by other reviewers) and a lightness to the bass. It still sounded very analog but very low level detail was missing.
Playing an extremely dynamic recording of a large church choir from Chesky's demonstration disk through his ATC speakers gave a wide and deep soundstage using this wide dynamic range recording. The volume controls were at the 12 o'clock position making the organ hiss very audible....the amp hardened at times on peaks at this volume levels but did not clip audibly and gave a good account of the recording.
With more warmup, some of the warmth and naturalness returned. It seems as if the stock SLA-1 errors on the side of omission did not compliment the JR cable's neurality. Putting in the Thunderline cable confirmed this as seemingly missing detail and balanced seem to have been restored (to me). To Danny, it was brighter and a more outlined presentation (what some call mid-fi) but with less inner detail due to a raised noise floor (Note to myself: Try to figure out this aspect of Danny's called noise floor and its sonic signature). If I assume the signature with the JR cable is the right one, then I I think I can hear that signature that I associate with poorer quality resistors, capcitors, wire and opamps. Eliminating this signature and improving the subtle sonic quality that makes a system listenable for hours at a time is mark of success. In its default form today, I can easily filter these slight flaws and enjoy the music. It is only in comparison to knowing what is possible that will lead to a slow evolution of the SLA-1...hopefully to the level of the Soundtech Poweramp One Tube amplifier
Just finished listening to a session of Miles Davis ("ALL
BLUES" from Kind of Blue album)...basically just piano,
trumpet, bass and drums....I've never heard it sound so
together:
The amazing thing is that the SLA-1 was not continuously
powered on (repercussions of the Eastern Seaboard Blackout),
like it normally is...I was not expecting much from it in the
first 1/2 hour. The lower powered electronics like the
transport and DI/O were (sorry Kyoto). I pulled out a DIGITAL
TEST CD by Pierre Verany release around 1988. In it is a
recording of Fireworks. The accompanying instruction manual
describes the cut as follows:
I've listen to ths cut (eventually) on every system that I have
ever owned when I thought I had it fully tweaked out over the
last ten years. I used to live on a third florr balcony
apartment where you could stand out on the balcony with the
wind blowing, cars passing by, and the voices of people taking
in the street....this is the first time that this recording
sounded like I was standing on the balcony listening to
fireworks iwth people ooing and ahhing at the explosions. In
the past, my imagination strain to try to transport me there.
It was uncanny this time. I think it is mostly the finishing
touches of the JR cable.
Note: See the link above that describes a tweak...it turned my system on its ear and most of the criticism leveled at the SLA-1 has been lifted. The following summary pertains to it prior to Day 8's changes and listening).
It is still early. Some of the early adopters/reviewers (3 so far) have the relevant experience and ears to make an assessement of the SLA-1 within the context of their systems and tastes.
The positive comments are:
The negative comments are:
The summary above is my attempt to group "common" observations into a succinct summary from each of the three reviews. There is a slight filtering upon my part which I will address below. .
Some of the statements above come from user's with very extended LF speakers and tastes. For me, #1 is not so much an issue providing everything above the last octave is tight, controlled and fast. The smallish filtering in the power supplies is sometimes eclipsed by competing 100W amps by as much as 7x (ie. 35,000uF versus 4,700uF). This might lead to some of the comments regarding congestion and deep bass at loud volumes. LS3/5a users need not worry or comment. Issues #2 and #3 were not so much an issue with Danny's dac but were noticeable using my DI/O with the attenuated gain. For me, on femalie voice, it amounts to a slight edge and un-natural sharpness to the pitch and lack of chest/warmth. Issue #4 was made by a tube lover and I can understand his perspective. The high's comment is something I need to ponder as I did not find it particularly hard, bright or spitty...again a case of erring on sin's of omission? I'll pay more attention to these points with more extended listening when my frontend is updated. The overall impression the SLA-1 and Danny's DI/O was that I just listened to one CD after another and took off my analytic hat.
As mentioned above, the strength of the amplifier is mostly that of the sins of omission. If the sonic signature of amplifier in due to the first input stage, then op-amp substitution may very well help SLA-1 on the road to getting a fuller and more round orgainic tone. At the same time it is desirable not to upset positive speed, transparency, and dynamic nature of the stock unit. It is entirely possible that a more a "better" opamp in the voltage gain stage might lead lead to less synergy with the bipolar output stage. The AN6554 and the present passives could have easily done a lot worse.
If the issue are in the second stage, hopefully it does not have to deal with supply regulation and that beefing up the supplies can help. In this case, passive swaps of critical caps and gain setting resistors may address some of the weaknesses. The modification section will capture this as a work in progress. ClaudeG...where are you when I need you.
P.S. An excerpt of this SLA-1 review has been posted in the review section of audio asylum...you can post your thoughts on this review there.
1.1.4 Updated Conclusions, May 2004
Whoever said that hindsight is 20/20 was right. Given the above status , it is not surprising that changes in the system were to be necessary to get to the next level. On re-reading, I'm surprised that I enjoyed the sound...it was bright and detail but I would say the tone was missing. Today the sound is bright and detailed but without being harsh or strident. The major changes in order of impact are as follows:
Each and everyone of these elements giving me about 10% more than what I had surmise to have had in the era of my "Preliminary Conclusions".
The SLA-1 amplifier is made for the "pro" market. This means balance XLR inputs and unbalanced 1/4" phono plug inputs (a la the DI/O)are used. I used a RCA femaile to 1/4" male phono plug adapter to enable the use of standard RCA terminated cables. The output is 100W into 8 ohms with a 1 dB bandwidth between 10 Hz and 40 Khz. The PDF for the SLA-1 User's Manual is available.
The SLA-1 is based upon two large printed circuit boards to the left and right of the centrally mounted toroidal transformer, a front panel PCB and a small board behind the transformer used for the IEC connector/fan regulator/ground lift switch.
The power supply and most of the weight is due to a centrally mounted toroidal power supply with a center tapped 76V output (i.e. 38-0-38) and a secondary 13V RMS output consisting of two green wires. This is used to power fan section that gets a 1/2 way rectified DC 7808 8V regulated DC. I can detect a faint hum coming from the transformer. The full wave rectifier bridge (unmarked) is heatsinked to the left channel...updating this to soft recovery types may be more difficult unless one uses higher rated parts and no heat sinking. The main power supply uses 4700 uF 63V (Metacon) units for the positive and negative supply (mounted on left channel output board). These provide +/-50V rails. For 100W output into 8 ohms, the outputs need to generate a 40V peak. The input, volume section and pre-driver section are powered off +/-15V derived from zener diodes off the +/-50V main DC supply.
The remaining 20 electrolytics (for each channel)are made by Tocon (whom I have not heard of) consist of about 4 x 220uF units, 1 x 100uF unit and a ton of 10uF 50V units. There are some green film units (about four of them) are used in various parts of the amplifier with values of 0.1uF and 0.001uF.
The 1/4 watt resistors are 5% units. Larger 1 watt units are found for the 10 ohm stabilization network, 2 x 1 kohm unit, 2 x 470 ohm units, a 100 ohm, and 2 x 47 ohm units, unit.
The quad op-amp Matsushita AN6554 uses two of the units for left and right channel in differential to single ended gain configuration providing 3.52 dB gain (1.5x) for single ended inputs. The 14 pin package uses pins 8,9 and 10 for the left channel and pins 5,6 and 7 for the right channel. A 22K and 33k feedback resistor set the gain for this stage. A second pair of these is needed for handling the balanced inputs. Two input DC blocking caps and and output cap complete the circuit. The phono plugs TRS (tip ring sleeve) provide IN+, IN- and ground connections for differential inputs. Single ended Phono plugs will short IN- to the sleeve/ground and provide single input via the TIP portion of the plug. Both the XLR and Phone plug inputs provide for balanced inputs. It should be possible to flip the phase of the amplifier by feeding a input to the "RING/IN-" input and grounding the "TIP/IN+" to ground/sleeve.
The outputs from the input stage go to a 9 pin header and connect to the front panel PCB. A third op-amp (of the four in the AN6554) provides a Unity Gain buffer from the volume control before going to amplifier section. The fourth op-amp is provide for phase inversion in bridge mode.
The output stage is a complimentary Darlington configuration with about 10% local feedback. The output side of the Darlington are two pairs of 60 Mhz output transistors: 2SC2837 (NPN) and 2SA1186(PNP) complimentary bipolar transistors are used for each channel and arranged in common emitter configuration . These biasing current was found to be 47mA in the left channel and 32mA in the right channel. The input side to the darlington consists of a complimentary pair made up of 2SC3421/2SA1358 devices (10W, Vceo=120V, 1A, ft=120 MHz).
This is driven by 1/2 of the 4558 opamp. Feedback components set the gain to about 25 dB (18.3x). Signals coming in are both high and low pass filtered. The output stage contains a classic zobel network to stabilize the amplifier. There is no DC servo control so large caps are needed to provide for high pass filter that limits the amplification of subsonic signals.
A generic 812H relay provides a delayed connection to the speakers and instantaneous disconnect to them during power on and power off , respectively. This means that there is not turn on transient or turn off transient. The same relay disconnects the output in the event of overheating and shorted outputs.A second sourced UPC1237HA chip is used to sense when to open the relay breaker and protect the chip. One is used per channel. Two small signal transistors (Q19 and Q20) are 50V/150mA 2SC945 NPN devices (ft=300Mhz) and 2SA1015 PNP devices(ft=80 Mhz) make up a current sense circuit to feed into pin 1 of the protection IC. These general purpose transistors are also used near the bottom of the left channel board and in the front panel board. My guess is that they provide for compare functions for the green signal LED, red clip LED and the overload LED.
For an inexpensive amplifier, the ART is feature laden. It is possible to simplify the circuit further by disabling some of these features. Parts swap are the order of the day for electorlytic caps and the signal gain resistors. Care needs to be excercised if replacing the opamp in the amp stage as a higher gain/wider bandwidth opamp could destabilize it. I am somewhat dissapointed in the lack of a DC servo to null out output offset. A phase inversion feature could be implmented easily enough but the detailed implementation needs to be worked out.
Click on images below for enlarged view:
For a comparison to someting else, here are some pictures of the State of the Art Class A Plinius SA-102. It has 4 pairs of output devices versus SLA-1's two pair; Two toroids (versus one for ART); soft start circuitry to avoid high inrush currents; larger heat sinks; 25 more watts into 8 ohms, 70 pounds more to lift, and a whole lot more money.
As little "orphan Annie" said, "I think I'm going to like it here". My gut feeling is that I like the road this little unassuming unit is leading me toward. Already, some of the signatures I hear can either be trace to the upstream sources while some are due to the unit itself. Hopefully, t ime, experiments, mods and more listening will sort these aspects out.
I've already hinted at the ground lift position as having an effect on the sound. RCA jacks are probably soon to follow. Over at AudioCircles, Occam has created a thread to capture and express his thoughts on potential mods...please feel free to jump in and contribute. Most of the major stuff(passives, actives) will not happen until schematics are available, so sit tight.
For members of the Art DI/O DAC Yahoo group Diomods, I have created a SLA-1 information folder and stockpiled a number of specs (UPC1237 protection circuit chip and 4558 dual opamp) and the hard to find SLA-1 User's Guide. Your's free for joining the group. Feel free to add to the file collection in the folder. Its only been a few days but here is a first stab at mods to the SLA-1:
June 7 Update: In terms of the major stuff, I'm going to let the unit ride for a while based upon my last listening session in Section 1.1.2. When schematics become available, the following experiments will be performed as more long term extended listening tests pinpoint the synergy in the system. As it sits (near stock), the SLA-1 designers have been able to nail a sound that is just plain "sweet".
For the amplifier stage upgrading the JDC 4558 may be worthwhile. Only pins 1,2 and 3 of this dual opamp are used. Good choices for opamps would be ones with similar open loop gain and slightly better bandwidth behaviour for stability.
This will be based upon trying to create synergy in the system by altering cables, cable polarity, source component configuration, and simple bypass experiments. Most important will be using the amplifier in stock fashion in different systems with different source components, speakers, interconnects and speaker wires. May take some time but I have a few co-conspirators already. .. ........
