Friday, October 29, 2010

Marantz 20B Sweet Rockin' Bad Boy

My Marantz 20B definitely has more personality than the average tuner; it has a Bad Boy attitude.  But it does sound sweet, particularly on strong commercial Rock n' Roll stations which might seem unlistenable on a geeky tuner like McIntosh MR 78.

 It doesn't work for the local classical station, my favorite station, it suffers even more from subcarrier noise, or something like that, than most tuners, so it's somewhat too noisy for seated listening, in addition to projecting the usual birdie I get from this station on most tuners.   With the Marantz, the birdie is localized outside the speakers, out-of-phase.  Low compression classical music also exposes noise much better than highly compressed Rock.  The modified Sony XDR-F1HD is the only tuner I have which gets this station perfectly, though modern digital tuners like Yamaha TX-1000 suppress the noise and birdie so well you have to get up to the speaker to hear it.  Analog tuners always get this station with some noise, but the 20B may be the noisiest I've tried so far, not that it's super noisy, but just noisy enough not to be fully satisfying.  I could enjoy this station on a Kenwood KT-8300 despite its noise, but the 20B just goes a bit too noisy.  It will be interesting to compare it with tube tuners.  I would not be surprised if some tube tuners actually sound quieter.

It works much better for a slightly weaker college Jazz station (which ironically has an actual HD subcarrier which creates no birdie), but still does nothing special like it does for Rock.

It totally enhances rock stations with a nice bass which seems to go underground (is it boosted?) and liquid mids and highs.  In the mids and highs it sounds a lot like tube tuners (and also has a rather laid back sensitivity curve like they do, hence the noise level on weaker stations).  There's also a kind of loosey-gooseyness that seems to open up the sound on otherwise highly-processed commercial stations.  The over-processed sound seems less processed and predictable.  I sense there is a kind of over-arching transparency that comes from the very wide bandwidth, but other than that this tuner seems more "musical instrument" than "precision instrument", and may also achieve some of its enhancement of rock stations through selective coloration and obscurity rather than the highest transparency.  Whatever, it works on commercial rock stations.

Listening through the noise to the classical station, I think I can also hear a bit of distortion which is somehow not audible on rock stations.  This 40 year old 20B looks untouched and after that many years, it's amazing it works at all, and like any tuner of this age,  it certainly needs to be refurbished and aligned.  Perhaps that will make it cleaner and quieter?  No, undoubtedly alignment will make it cleaner and quieter, but how much?

Anyway, right now it's a tricked out low rider car with fancy hydraulics and squeaky springs, just what you need for Rockin'.  Let me count the ways it shows attitude:

(1) Electrical leakage.  First plugged in, I felt a tingling from fairly strong leakage from the faceplate, though probably not bad enough to cause injury.  So I reversed the plug, a procedure all audiophiles from the 1960's are familiar with.  At first, this seemed to kill the leakage completely.  Then I noticed it's still detectable if you rub your finger along the edge of the front panel.  If you hold your finger down, you feel nothing, so this is very small leakage, not dangerous, but noticeable.  This unit from 1970 has no grounding plug, if it were sold today it HAVE to be grounded, and then you wouldn't notice any leakage (but you would instead notice an annoying hum from the ground loop).  So, the Bad Boy gives you freedom from hum but leakage instead.

(2) Interstation WHAM!  Unfortunately, and in contradiction to what the eBay seller said, the muting on my unit does not work.  Between stations, there are noises strong enough to rattle and blow your woofers if you are not careful.  It's like the usual output voltage is 1V but the interstation noises are 7V.  (This is a guestimate based on the LED's on my Lavry, not an actual measurement yet.)  But if your system has tons of headroom, like mine, it's fun to twirl the dial and shake the whole room.  Who needs actual music?  There's no volume control on the front panel either.

(3) Ringy chassis.  The chassis is nicely made out of very solid metal.  It rings nicely when you hit it.  So don't hit it.  Or hit it with rhythm.

(4) No stinkin' meters, man.  Yeah it's got a scope.  Not really a big enough scope IMO, the 10B and Marantz 150 had larger scopes, but it's better than no scope.   The combination of scope and very wide IF bandwidth is a license to off-tune.  When you have meters, it's like you have a little geek telling you "tune here or else."  With a scope, even if not adjusted, you see what's happening in the IF, and you see where the ultimate detection limits are, but not so clearly where the exact center is.*  So you just tune it "in there" and it works.  By the way, since this is a very wideband tuner, perhaps the widest band tuner I own, the detection limits are waaaay out there.  And somehow they arranged it so the stereo decoder works all the way to the limits.  In fact, it may be broken in my unit, my unit never seems to go to mono (judging from the sound, actually my Stereo light never comes on, may be just burned out or something worse).

*That's the way the most useful scope display, "multipath" works.  Now unlike the earlier Marantz 10B, the 20B also has a "tuning" mode to the scope display.  That changes the line on the scope from horizontal to vertical and you tune so that the vertical scope line appears inside the bullseye.  That's an extremely boring display so I tend not to use it.  Also, it turns out it isn't accurate anyway unless you calibrate it first.  To calibrate (I discovered this without reading the manual) you run the dial pointer all the way past 108.  That seemingly forces the scope line to its intended center.  Then you adjust the vertical and horizontal controls as required.  That whole procedure seems Bad Boy to me, especially turning the dial past 108.

There's also a Left Channel vs Right Channel vector mode to the scope display.  If you're wondering what your stereo separation actually looks like, there it is.  Now real bad boys keep their Marantz 4400 receiver scopes set to the vector mode, because it is also "light show mode".  I've seen blogs where no bad boy present actually understood the other scope modes.

Marantz has decided not to give you an intensity control for the scope.  They don't trust you quite that much.  Anyway, the intensity was just fine, I thought of turning it down a bit to preserve scope life, but couldn't.  A Bad Boy certainly wouldn't bother turning it down, only up.

Most analog tuners that I've come across have had malfunctioning center tune meter anyway.  They never seem to be tune best at the center of the meter because the meter itself is off.  So allowing (even forcing) you to calibrate your own scope meter allows you to make sure it's always right.  Or always wrong.  Who knows if you can really trust the "past 108" tuning calibration voltage?   Anyway, given this kind of functionality, you could also adjust the center tune to whatever you want, whatever works for you,  Bad Boy, even if it's a particular amount of off-tuning.

(5)  Don't mess with adapters, just cut the RF cable.  Now perhaps this can be excused since this is such an old tuner, but I've always been annoyed at tuners and receivers that don't give you a proper F connector for your 75 ohm coax.  Marantz kept up this for quite awhile, the Marantz 2270 receiver I bought in 1974 also had no F connector (and worse, that one had lousy spring terminals) .  I'm not quite sure when Marantz started putting F connectors on their equipment, I think the Model 2130 from 1976 had one, but it seems like Marantz was bucking the trend here for quite awhile.  Why?

Well possibly because it's actually cheaper and easier to do things this way, unless you're worried about re-using your cable later.  Bad boys don't sweat such details.  Run the cable in from your antenna, and then Marantz gives you license to cut the cable at the exact length required to reach your tuner.  Then get out your big old slotted screwdriver and screw it down.  This one has nice big screws.

Back in the 1970's I used to worry that kind of connection was not good for the 75 ohm characteristic impedance.  But it's probably doesn't make much difference for a half inch of bare wire.

Once you've done this, of course, you are not going to go back to that other geeky tuner with an F connector, like a McIntosh MR 78.  But you weren't planning to do that anyway, were you?

(6) Right on top, left on bottom.  Most audio equipment has adopted the convention of putting left channel on top, right channel on bottom.  Marantz puts right channel on top, left channel on bottom.  For decades my conspiratorial mind has been pondering if there is some kind of political significance to this.  But even if it wasn't an attempt at making a political statement, clearly Marantz was bucking another trend.  Also, annoyingly, the channels are labeled A and B on a big label.  You have to read a smaller label to see which is intended to be right and left.

Now there are lots of bad boys in audio.  Here's a short list:  Dick Sequerra, James Bongiorno,  John Curl,  Dan D'Agostino, Nelson Pass.  These are the kind of guys who give you want you want, even if it's over the top, ignoring the nannies who say you don't need it.

And now I can see I'm a Bad Boy too, and I like this Bad Boy tuner for what it does best.

[Disclaimer for geeks: this is a "first impression" review from my first night of listening.  No measurements or A/B tests were performed.]

Thursday, October 28, 2010

Marantz 20B Mystery Solved

Now that I've actually printed out the 20B schematic, I can see there
are only 4 filters, not 8 as I had previously thought.

The 20B schematic from MarantzPhillips included both Model 20 and Model 20B
schematics for the IF. Each shows 4 filters. I thought they were both part of
the 20B. The 20B version has the tap for AGC return to the front end, and a FET
replacing the first bipolar, but is otherwise the same.

So the mystery I proposed is solved. The Marantz 20 and 20B were clearly
designed as an audiophile tuner, with 4 filters (not uncommon in such) for a
single wide bandwidth. I believe that the 20/20B may have a wider bandwidth
than the 10B. Actually there was discussion previously here that Marantz 10B
does not have sufficiently wide band for lowest possible stereo distortion. 
Also, IIRC, the 10B has 6 filters, I need to go back to check my 10B schematic.

I don't know how this compares with other tuners, but each filter module in the
20/20B IF seems to have 8 reactive elements. So that would make 32 reactive
elements (poles?) in all.

WRT the 20/20B issue, I now admit that is an interesting question. It is quite
possible that either of these things in the 20B could have a slightly
detrimental effect:

1) RF amplification (could add noise and distortion)
2) AGC circuit (operates like slow feedback)

I'm not sure how fast the AGC circuit operates. I think it would be interesting
to replace AGC circuit with a switch that would either use normal AGC operation,
or set some fixed levels of amplification. And, having done that, remove the RF
attenuator too.

I think in my case, though, I'm going to need the RF amplification. My 20B
arrived this morning so I look forward to unpacking it tonight.

Wednesday, October 27, 2010

Marantz 20B from below

I found the above picture at ImageShack linked from AudioKarma showing all the stuff inside a Marantz 20B from the bottom.  This is someone else's unit; mine has been shipped but has not arrived yet.

This is just as impressive as the 13 page schematic.  There's lots of stuff visible here, and lots more inside boxes to the left and on top.

But while I'm high fiving myself over finally getting this tuner, a question arises.  Why isn't the 20B better than it is?  With all that stuff, you would think it could be one of the best of the best.  (Some people think the 20B is one of the best, but that's unusual, the 10B is more widely known and praised.)

OK it might be a good sounding tuner, but why isn't the selectivity one of the best since it has an outlandish 8 filters, and they are always active?

I wonder if there isn't some kind of design error here, as apparently there was with Marantz 120, though for different reasons probably.

I believe that the Marantz Model 20B was one of the last Marantz components designed by the legendary Marantz designers Sid Smith and Dick Sequerra.  It was intended to replace the 10B, and used a similar IF design with 8 LC filters always active (no wide/narrow switch).  It was manufactured in the USA at Woodside, NY and later at Sun Valley, CA.  Actually, the design was first used in the Model 18 receiver, then became the 20 and 20B tuners and 19 receiver.  The models 19 and 20B had a revised version which included a mosfet RF amp to improve sensitivity.

The 120 was the first high end Marantz model produced by Standard Radio of Japan, which had been purchased by Superscope to be the factory for the less expensive Marantz products.   It also uses an outlandish 8 filters, but this time they are ceramics, which are less expensive but can be very good.  But it appears there is a design error, the IF strip is not loaded properly, and it appears the 120 cannot be made good without completely replacing the IF.  (This is documented at FMTunerInfo.)  After the 120, Marantz temporarily switched back to an LC filter design for the models 125 and 150, then finally got the hang of ceramic filters with the Model 2130.

Though their pedigrees are entirely different, the 20B and 120 appear to both have 8 filters yet be lacking in selectivity.  It is known that the 120 has a design error.  So that makes me wonder about the 20B.  Though I find it hard to believe that Dick Sequerra would have made such a design error, if he was the 20B designer.

Another possibility goes like this.  It was hard to achieve the highly touted selectivity of the 10B using transistor circuits because the impedances of the transistors in common collector mode is too low.  Or perhaps the gain is too small, or they are not as linear.  But this design (actually a re-implementation of the 10B with transistors) worked reasonably well and sounded good.  Given the difficult transitional situation of Marantz (acquired by Superscope in 1964, moved to Sun Valley and outsourced most manufacturing to Japan in the late 1960's), they chose not to invest resources in an entirely new design to make the best of transistors.  Perhaps they were even able to use surplus 10B parts, such as the filter modules.  So they went with what they had, even though it would not have been cost-effective for an original design.

Surely Dick Sequerra would not have wanted to do things this way, but he was nearly (or already) out the door anyway, possibly even working on contract.  After Superscope acquired Marantz, the old staff stayed on for a few years but eventually got ticked off by the Tushinsky's and left.

For what it does, the 20B does not look like a cost-effective design, but perhaps it could be made much better with some new thought.

Smith and Sequerra did get another chance to make the best possible transistor tuner later with the Sequerra Model One.  That tuner is even more outlandish and expensively built than the 20B, and includes balanced circuitry in the RF and IF stages,  a high end design feature I've never seen in any other tuner.  Still, reports I've heard suggest that the Sequerra Model One isn't the best of the best either, in sonics and selectivity, though the cool panoramic analyzer is worth the price of admission (and may explain why the FCC bought most of them).

One thing is unarguable, and Sequerra has even admitted it himself.  He should have figured out how to make better equipment at lower prices.

Tuesday, October 26, 2010

Examining Marantz 20B schematic

This tuner is very impressive from the 13 pages of the official schematic, which I have now obtained from MarantzPhillips.  Unfortunately it can't be saved or printed properly, at least I can view it onscreen for now.

The IF alone has the 8 LC filters, and of course they're all full time there is only one bandwidth.  Each filter looks to have about 8 primary reactive elements in addition to fairly complex power supply decoupling which includes chokes and capacitors everywhere.  Each of the 8 filters does appear from the schematic to be some kind of "filter module."

Funny, never has a tuner with so many filters been made so un-selective.  Legend has it the 20B is less selective than 10B, may have something to do with the gain or impedance of the active devices.  OTOH, some say the 10B bandwidth wasn't sufficiently wide for low stereo distortion, so maybe this is better for sound if not for receiving weak stations.

OK possibly it is less complex than more modern tuner because it's just brute force discrete circuitry instead of IC's.  But there is a lot of circuitry. There is also a 4 transistor circuit board "limiter".  I do not believe there are any IC's, so it's all discrete.

In the front end, there is indeed a dual gate mosfet in the first tuned circuit.  There's a jfet that appears to be part of the mixer.  3 caps one inductor in first stage, 6 caps two inductors (transformer) in second stage .  Mixer is not balanced.  BTW I think Sequerra Model One has balanced IF as well as mixer.

The MPX oscillator is quite complex, 10 transistors, one FET, one optocoupler.

I think detector is ratio detector.  Detector like most other blocks is in a shielded can.

The multiplex matrix is non trivial, 4 fets, 2 transistors, two optocouplers, and 4 diodes.  The diodes are shunted with caps and have cap+resistor loads.

About the only thing that looks a little unsophisticated is the power supply, though it is quite large.  Within the power supply itself, there are 2 unregulated voltages for scope and 3 regulated voltages for tuner.  The regulators are all shunt regulators, with 2 and 5 watt series resistors.  I notice that on the front end board itself there is an additional elaborate capacitance multiplier regulator, and there might be other sub regulators on other boards.  Most serious transistor tuners have had multiple series regulators, sometimes with active feedback.  Series regulators are far more efficient and can regulate better generally.  Shunt regulators may have lower noise.

The output stage is a flat two transistor cascade (all the countouring having already been done).  Seeing as it has a 110K input impedance it could be bypassed though capacitive coupling is required because it rides on the matrix +4.8v and -4.8v for other channel.  There is no input or interstage cap on this board, only the output 1uF cap.

A DVD-Audio winner

Finally got back to listening to discs, a nice previously unwrapped HDAD+ version of De Falla's The Three Cornered Hat.  This is a classic recording remastered from the original master tapes to produce a DVD-Audio with 24 bit resolution.  If you have a DVD-Audio player (like all my Denon's) you can play the DVD-Audio version with 192kHz sampling rate.  If you only have a DVD-Video player, you can still play the version with 96kHz sampling rate, still with 24 bits.  A CD (16 bit) is also included, I plan to copy that to the hard drive used by my Sonos system.  I think that all good music should be distributed like this, with both CD and "advanced resolution."  (Note while the disc includes some pictures, it is not a video disc, there is no reason to listen with monitor on.)

This disc is incredibly liquid, dynamic and theatrical, a treat to listen to.  The incredible dynamics do not seem to be weakened much by just a tiny bit of noise, probably coming more from the microphone amplifiers than the 35mm recording.  Other than the noise, it sounds so good, one wonders if maybe it sounds extra good because of the noise.  Often modern noise-free recordings just don't quite have the depth of this recording.

Classic Records makes these HDAD+ packages and also regular HDAD which do not include the CD.  Before coming up with the HDAD package, Classic Records made DADs which would play on all DVD video players at 96Khz, and I have quite a few of those.  (PULSE is one of their best, btw.)

My feeling is that 96Khz is all that's needed, 192Khz may even be generally inferior for technical reasons (see essay by Lavry).

In my system, I cannot play 192Khz as well as I can play 96khz, because my Denon 5900 cannot output 192Khz digital directly.  So if I need to play a 192Khz disc, I take the analog output of the Denon and convert it back to digital with a MSB PAD-1 analog-to-digital converter which operates at 96khz.  Either way, digital gets fed to my digital equalization preamp and digital crossover.  But no doubt some noise is added when I have to convert analog to digital, as I do with 192Khz discs, DVD-Audio discs in general (which usually suppress the digital output at the highest resolution, even if it's only 96khz), HDCD's, and SACD's.

So in the case of this disc, I could choose either the direct digital output at 96Khz (and 24 bits resolution), or the analog output at 192Khz converted back to 96/24 digital by the MSB PAD-1.  One would think bypassing the conversion would be the best, though sometimes I have liked the converted higher nominal resolution versions, which can reduce harshness a bit (useful for Red by King Crimson).  Last night I only bothered to listen to the direct digital at 96Khz.

The MSB is also connected to my preamp using AES/EBU over balanced wire instead of spdif.  But since my preamp has only one AES/EBU input, I need to make a connection change to switch from the Lavry (currently used for FM radio) and the MSB (currently used for high resolution discs).  I've decided the simplest and most effective way to switch XLR is to have short XLR cable that can connect to one of other cables that actually connect to the equipment in question.

Sunday, October 24, 2010

Finally I get what I want, a Marantz 20B

Not that I am expecting this to be my best performing or sounding tuner...  I do not expect it displace my modified Sony XDR-F1HD in that position as my #1 tuner in both weak station pulling power and sonics.  So perhaps, I shouldn't have bought this.  I am not a nostalga-phile.  I try to pursue the best possible sound using the latest technology (such as I can afford), not relax into some 50's or 70's sensibilities.

But I just had to because I've been studying them for nearly 40 years since I first saw one at a new high end stereo business (one of three) in my school age hometown, and I hope it's pretty good sounding tuner, fun to play, and a useful tool because of the oscilloscope.

I believe that business was the store whose cranky and outspoken owner later became famous on the internet, Sunshine Stereo, but during its very brief incarnation in California before moving to Florida.

Anyway, as a pimply faced high school sophmore I wandered into one of the back auditioning rooms, there on a shelf was the 20B.  It blew my mind.  It may have been playing and sounding elegant as I recall, or just quiet, but sitting there, with scope and blue face and white lights, it was so cool I thought.  The owner wandered in, quizzed me about what stations I wanted to listen to, decided I needed his most sensitive tuner, and showed me a Sansui, which kinda looked cheap I thought.  He also had an SAE, which he didn't think performed as well as the Sansui for picking up weak stations.  But I still lusted for the Marantz.

Later I saw a 10B, and was disappointed that it was not the same tuner.  I didn't warm to the appearance of the 10B at first.  Now, however, I admit I lust for the 10B more than the 20B.  I've bid on 10B's and lost.  I hope to get a good 10B eventually, but they are incredibly expensive and unpredictable.  They are the most legendary "good sounding" tuner in history, many people have reported fabulous sound from them, though just as often you find negative reviews of them.  They appear to need a lot of service only by apprentices of the original designers whose charges are astronomical, though sellers just as often claim their unit has never needed service and still works perfectly.   Advertised as never-needing-alignment, the actual history seems to read never-stops-needing-alignment.

Quite a number of people who as Marantz employees or audio salesman lived with both 10B and 20/20B like the 20/20B better. The 20B is said to be a transistorized version of the 10B, based on the same precision toroidal filters (or something very similar), which are often incorrectly described as Butterworth.  This tuner uses no ceramic filters which are used ubiquitously in other tuners, but some think are inherently inferior for sonics.  Both tuners were designed by the original genius Marantz designers Sid Smith and Dick Sequerra who also went on to design the original Sequerra tuner.  It is said to be one of the best sounding tuners, if not one of the best performing tuners (among online reviewers you can find many opinions including these: some think the 10B performs better than anything but sounds worse than the 20B, others think the 20B is better in all ways to 10B as well as being best sounding tuner but not as good performing as others like McIntosh MR78, many tuner fans think both these early Marantz tuners are a joke except in their mythological impact).

The 20B version is different from preceding 20 and 10B in an important way for people who use indoor antennas.  It has a dual gate MOSFET RF amplifier with AGC, and changes the first IF transistor from bipolar to FET.  That makes it more sensitive.  But also, it makes it potentially more subject to overload.  There is a switchable attenuator for that.  A modern design would use a more modern silicon device (they have gotten extremely good) as RF amplfier for better performance without needing bandaid attenuator.  But anyway, unless you have outdoor antenna, the 20B is clearly the one-of-these to get.  It should be more sensitive than even the legendary Marantz 10B, which feeds the incoming signal directly to a diode mixer with no RF amplifier.

But despite the boost in sensitivity, this is still not noted for weak station pulling by just about anyone.  It is noted for nice sound, interesting built-in scope, and cool looks.  A bargain solid state version of the legandary Marantz 10B, and notably more reliable than its predecessor.

Anyway, more fun, I hope, on the way.  I bought this one for a relatively high price because it appeared to be in excellent physical and working condition.  What happens on ebay is that some mint unit will sell for a huge price, and from then on all the rust bucket barely working units will try to be sold for the same high price.  Whatever the primo unit costs extra is often worth paying, but only if it is the primo unit and not some copycat seller.  This is about the best looking unit (with numerous high resolution photos) I've ever seen and is described as fully working by the long time audio guru who is selling it.

Symphony, 101dB

The San Antonio Symphony had a stirring rendition of Pines of Rome on Saturday night.  This is a big audiophile favorite going way back, but however many times I've heard it reproduced it was even more impressive live last night.  It was mind blowing!  Perhaps wrecking the mood, at the final climax I discretely pulled out my mobile device to measure peak dB.  I did see a 101dB average A weighted reading, pretty impressive in itself, but after I switched to peak I couldn't reproduce that again.  Immediately thereafter I was the first to rise in the unanimous standing ovation.  [Note: I tried to be discrete, and didn't disturbe anyone around me because there was none.  But I think I'll avoid doing this in the future, the symphony hall is my church.]

Wednesday, October 20, 2010

Cleaner Power, Lower Noise

I switched my line level components over to the dedicated amplifier circuit.  As a result, wideband amplifier noise dropped from 1.8mV to 1.7mV, and there was a similar small drop in A weighted noise.

More impressive was comparing the AC lines directly with my AudioPrism Noise Sniffer.  I can now give these reading quantitative values by holding an SPL meter directly above the widest part of the sniffer speaker.  For SPL meter, I used a very accurate mobile device SPL app.  The dedicated line got a measurement of 78dB.  The old household line got a measurement of 100dB (!) as it was buzzing like hell. I got that measurement after turning off the dimmer for the kitchen fluorescent lamp.  Strangely, with the dimmer on at 50% brightness level, the line noise level was lower.  The old household circuit is crazy, it goes from the the outside light in front of the garage to the outside light in back of the house which are both fluorescent with photosensors.  In between it goes to the entry light and outlet and the kitchen light/fan which is on a special fan-compatible dimmer.

It's hard to tell exactly how much the sound improved since I've been making so many changes that have improved signal-to-noise ratio over the past month or so.  The overall result is that the sound is very warm and imaging is very stable.

I think I'll keep the line level components connected to this circuit in the future.  In the past, I had used this line exclusively for the Krell FPB 300 amplifier, which draws current mightily sometimes.  But I think now it's worth keeping the line level components on this line also, and if the Krell lowers the line voltage significantly the new UPS will restore it, but I don't expect that to happen often.  If it turns out this line causes the UPS to make too much noise when the Krell is on it, then I'll have to move back to the old circuit.

The dedicated amplifier circuit has these specifications: 120V, 20A Breaker, 10ga wire, wired straight to main power entry box, insulated ground, oyade outlet.  Cost the better part of $1300 to install.

The reduction in noise could come from less noise on the hot, neutral, and ground wires.  But it could also come from putting both line level components and amplifier on the same AC circuit.  For the purposes of deciding how to do my connections, this answer to this question is unimportant.  But for audio science, the question is important, because reducing amplifier noise by using a quieter AC might be considered tweaky and not part of standard audio engineering.

To have a controlled experiment, I would have to try one or more additional comparisons.  For example, having all components, including the main amplifier (currently a Parasound HCA-1500A because the Krell is waiting for repair) on the noisy line.  Then I could compare the two cases:

1.  All components on noisy line
2.  All components on quiet line

which would single out the issue of noisy or quiet line, and remove the additional issue of having components on one or two lines.

Actually, because I had been operating on two lines before, I had previously ground-lifted the Parasound.  For some reason, every Parasound amplifier I've had has needed to be ground-lifted or connected to an otherwise ground-lifted system.  Perhaps now, with everything on the same line, I could operate the amplifier without being ground lifted, but I made the change internally.

But since I already ground lifted the amplifier, the issue regarding the previous operation of components across two different AC lines should have been relatively small.

I am interested in the scientific question, so I expect to go back and do some additional experiments, but I think that operating both line level components and amplifier from the dedicated circuit makes the best sense.

When doing the noise sniffing, I also tested the outlets of the Belkin PureAV AVU1500 when plugged into the quiet circuit.  In that condition, they measured the exact same "noise sniffer level" as the circuit itself, 78dB.  My interpretation is that 78dB is already as quiet as AC gets.  The other interpretation might be that the PureAV filtering doesn't do anything.  But I recall (no numbers) that when the PureAV was plugged into the noisy circuit, it was still quiet.  So that supports the idea that the Belkin filters power pretty well, but when you are already starting from fairly clean power, it may not do much.  And clean power may mean simply a dedicated circuit with insulated ground.

Tuesday, October 19, 2010

Air has noise, and other measurements

Unplugging my Parasound HCA-1500A amplifier from the Meguro noise meter, the meter immediately jumps up (not down) to a higher reading: 10.5 mV.  That is apparently the EMI/RF voltage in the air across a pair of banana jacks, in my room at this time.  (I've noticed a little variation over time, the range seems to be 9-11mV so far.)  The Meguro has a bandwidth of something like 1 Mhz, and an impedance of 1 Megohm.

It's particularly funny to remove an amplifier and see the voltage reading go up IN THIN AIR.  It's like the amplifier cables are shorting out the air around the banana jack.  Fortunately the air has a rather high impedance so the cables win.  But one can wonder what affect this has on amplifier circuitry.

BTW, with 6dB's of attenuation dialed in on its level control, but otherwise being fed by the noisy Behringer DCX 2496, I get 1.8mV wideband and 0.3mV "A" weighted noise from the Parasound amplifier.  The majority of that may still be from the Behringer itself.

Sunday, October 17, 2010

The total noise reduction was achieved through various means NOT

Update and Edit: Additional testing has cast doubt on the original claims of this post below.  It now does appear that 6dB of noise reduction in the supertweeter amplifier output was achieved by using a 6dB attenuator.  I now measure 21mV, 0.64mV (A weighted) with no attenuator and 9.7mV, 0.33mV (A) with attenuator.  Specifically re-measuring the change of putting most equipment (plugged into power conditioner/UPS) on dedicated amplifier circuit instead of regular (crazy!) household circuit, I have found that change makes zero measureable difference in noise level.  Surprise Surprise.  I have now switched from using a 3dB attenuator on midrange amplifier Parasound HCA-1500A to using its built-in level control to achieve 6dB attenuation.  The Harrison Labs 3dB attenuator I was using seemed to make the sound slightly funny sounding in the midrange probably because it presents 3K ohm load to the Behringer.  But the Harrison Labs 6dB attenuator on the supertweeter amplifier, which has no level controls, does not seem to have that effect, probably because ears can't hear it at such high frequencies.  I watched the voltage output on the tweeter amplifier for awhile, it seems it stays below 1 volt (!) which is much less than a watt output (on a 250 watt amplifier).  Based on that test, there should be no problem with 12dB attenuator which will be ordered today.  [End edit, now back to original erroneous post.]

Yesterday's noise reduction exercise (the supertweeter which used to make a rough fizz now only makes a barely audible background hiss as close as you can get to it) mainly focused on use of interstage attenuation.  Actually, only about half of the noise reduction came from attenuation.  The other half came from switching to short shielded equipment cords on 3 main pieces (UPS, Tact, DCX), which definitely made a measureable but small difference, switching the power conditioner to the dedicated amplifier circuit, and using some better shielded cables.  Unfortunately, my measurements of all this are incomplete, but the total noise reduction was from 21mV at the output of the tweeter amp (pretty hefty noise level) to 9mV at the amp, a greater than 6dB reduction.   The use of a 6dB attenuator didn't quite reduce noise 6dB, more like 4dB according to my measurements, likely due to other noise sources not yet fixed that are not being attenuated.

These noise measurements are wideband, made with my nice Meguro noise meter shown above.  I didn't do A weighted measurements consistently unfortunately.  The audible noise reduction seemed much greater than 6dB actually, which is possible because it's possible the audible noise has been reduced more than the inaudible noise.  In fact, that often happens.  The A weighted noise levels are much much lower.  I figured out that is how Behringer can meet their noise specifications for the DCX 2496.  The "A" weighting ignores a huge amount of high frequency whine from the Behringer's switching power supply.  A weighting is not a completely accurate weighting according to audibility, either.  I understand that the B weighting is better but nobody uses it.  I think noise is bad whether it's considered "audible" or not.  Even the "inaudible" noise modulates audibly with lower frequencies.  And it may well be that our hearing system is affected by tones we don't actually perceive; there is at least one peer reviewed research paper which suggests that.

One funny thing in all this was that I didn't intend to go this way.  I had long figured the supertweeter noise was ground loop noise with all the lower harmonics filtered by the supertweeter's build-in crossover.  However, when I tested for ground loop noise by temporarily ground lifting the tweeter amp (using a grounding adapter) I discovered that made no difference.  To be sure I was hearing everything, I put a full range speaker on the amp instead of the supertweeter.

So then I got involved switching the line cords to the nice new ones (SJT shielded 14 gauge, much nicer than freebie SVT), etc., which wasn't making much difference, so I asked myself "Just where is this noise coming from."

I then put my meter on the Behringer itself, with no actual digital input, and the wideband noise level was an astoundingly high 0.7 mV.  So that's where the noise is coming from!  Then I immediately realized an attenuator would be a quick fix.

Now I'm a little worried about the 3dB attenuator on my midrange amp.  The voltage levels there are pretty high, and the Behringer may be getting distorted into the 10K load or less of the attenuator.  Now I'm thinking I may just use those volume controls on the back of my Parasound HCA-1500 after all.  They maintain a relatively high load impedance for the Behringer, and may sound better for that reason if not others.  In the end, I'd like to replace those volume controls with fixed resistor networks of the lowest noise resistors.

Another day, another twofold (6dB) reduction in noise

Actually this doesn't happen often, that I am able to improve my audio system by a factor of two in one day.  If it did, my system would already be astronomically good, two to the power of over ten thousand. But it is often possible to find improvements that can be made which are real, provably real because they are also measureable, and having an ability to do audio measurements can help find them.

The above picture shows a Harrison Labs 3dB attenuator like the one I am now using now on the output of my Behringer DCX 2496 digital crossover that feeds my midrange amplifier.  I'm using a similar 6dB attenuator on Behringer output that feeds my supertweeter amplifer, where I have achieved more than 6dB in total measureable noise reduction (also including some smaller noise reductions from other factors I'll discuss in future posts).  [Correction: In the following post, I give measurements showing that possibly the entire noise reduction came from attenuation, not from cord changes.]  I think the 3dB may be the optimal limit in improvement for the midrange amp, but I'm planing  to try a 12dB attenuator, when I can get one, on the supertweeter amplifier.

For some time, I have noticed a whiney high frequency noise from the supertweeters (like the right one shown above) when I put my ear right up to them.  I had figured the problem was ground looping (to be discussed in future posts) but ultimately found to my surprise and horror that the source of the problem was that the Behringer DCX is rather noisy for a hifi component.  (And this is after having spent months doing bench tests on the Behringer, which convinced me that it was fairly quiet.)   Since the Behringer both has excess noise and excess output capability for most hifi components, it makes sense to attenuate it for them.

I don't really notice any noise coming from my midrange speakers, the Acoustats.  But anyway I put a 3dB attenuator on the input to the midrange amplifier, so there is a 3dB noise reduction through the midrange as well, even if requires a listening session to hear, not just a quick listen.  As well as being less immediately obvious, the benefit in the midrange is uncertain.  In addition to being objectively quieter, there might also be a slight rise in distortion, which is much more noticeable in the midrange.  I sometimes think I hear a slight hardening of the sound, but it is superceded by a tremendous improvement in depth.  But because it may already be a mixed blessing, I'm not certain I'll precede to using a 6dB attenuator for the midrange once another one becomes available when I replace the tweeter attenuator with a 12dB one.  3dB attenuation in the midrange may be the best compromise option.

Actually, I have argued against using attenuators with the Behringer in a DIYAudio forum.  But that was when I was using a Krell FPB 300 as my midrange amplifier.  No attenuator is desireable for that combination, because the Krell can use the extended high output of the Behringer, and also because the Krell is less sensitive and already suppresses the excess noise of the Behringer.

But while I was making these arguments, I was overlooking the need for attenuation of my supertweeter amplifier.  Attenuation was clearly called for there.  And now that I am using a different midrange amplifier, a Parasound HCA-1500A, some attenuation is useful there also.

Under my loading conditions, the Krell requires 2.6dB more signal input to output the same level as the Parasound.  This means the Krell, relatively speaking, suppresses the noise of it's source by 2.6 dB compared to any THX approved amplifier.  So long as the preamp or source component can provide enough level (and the Krell requires about 5-8V input max) this can be an overall benefit.

Now why doesn't everyone play this game, make their amplifiers less sensitive to make better sounding systems (and sell more amplifiers)?  The answer is standards.  Currently, there is a defacto standard for audio amplifier sensitivity.  It is a proprietary standard, THX.  THX determines the "correct" level of amplifier sensitivity.  They have chose a comparatively high sensitivity level that tends to exaggerate the noise level of preamplifiers.  But anyone like Parasound who wants the THX seal of approval needs to use the THX sensitivity level (can't remember what it is offhand).  John Curl has complained about this, though his amplifiers do often (like my 1500A) have a volume pot on the back.  If I wanted to, I could lower the sensitivity using the volume pot.  I simply hate volume pots because of noise and other factors and will always use a fixed attenuator instead.

So anyway, the context of my old system was the Behringer crossover (with max output around 10V) and Krell amplifier (max input similar to that).  In that context, it makes no sense to attenuate the output of the Behringer, because if you do, you are limiting how much of the peak power of the Krell you can access.  And that peak power is a lot higher then the specified power of 300 watts, more like 550W into 8 ohms (and more into lower impedances).  So therefore don't calculate the max level from your preamp based on the amplifier power spec.  Based on that calculation, the Krell would need only about 4 volts to reach max output.  But if the preamp runs out of steam at 4V it will never access the ultimate peak power of the Krell.  It needs more like 8 volts for that.

In this context, not only was attenuation not good, it was not needed, because of the 2.6dB noise suppression from lower sensitivity.  Thus the Behringer and Krell are a good match without any attenuation.  Attenuating the Behringer in this context either reduces peak power or increases distortion.  In addition to being rather noisy, the Behringer gets slightly more distorted at the higher voltage levels above 4V output.

But the Behringer and Parasound are not a good match.  The Behringer puts out way more voltage than the Parasound will ever use, so that extra voltage is wasted.  And being more sensitive, the Parasound exposes the noise of Behringer much more loudly.  In this context, some level of attenuation is desireable, possibly 3-9dB.  I'm using 3dB now, and 6dB on my supertweeter amplifier, an Acurus A250, which has similar sensitivity.  The supertweeter generally only needs a few watts, so it's unnecessary to access the upper power ranges of the A250, so using lots of attenuation is not a problem.

Why use the noisy/distorted Behringer?  Because it is a miracle box, letting me build a fantastically complex speaker system that could not have worked 20 years ago.  It was designed to cheap semi professional standards, not audiophile standards, but the capabilities of the digital processing it offers is without peer (at least under $5K).  I plan to get my Behringer modified with new audiophile grade circuitry from Europe soon.  That will make it an audiophile grade unit.

I actually blame the THX standards somewhat also.  Given current technologies, the sensitivity and output levels of the Behringer and Krell work out better.  There is no sensible reason why amplifiers need to reach maximum output level with less than 2 volts input.  But historically, lower levels like that were common. In fact, for many years, amplifier sensitivies were around 1 volt.  Why?  I don't know, maybe just because it's an obvious number.

OK this is getting too long and I need to get to yard work before I post pictures, but there is more to come, and other tricks helped get lower noise levels also.

Friday, October 15, 2010

Ethan Winer

I stumbled across Ethan Winer's website while looking for info about power conditioners.  Interesing guy, mostly on the audio objectivist side of things, owned a studio but sold it to go with Cakewalk digital production exclusively, written lots of magazine articles in popular and recording engineer magazines.

He has a take on the use of various kinds of distortion to sweeten recordings.  He says the way recording engineers use tubes, transformers, etc., these days is not because of higher fidelity or the inherent superiority of analog.  It is because the addition of a tiny bit of distortion often makes things sound better.  This is not a stupid idea at all, and he proved it in a cello recording project.  He calls for the use of digital plug-ins to get these effects instead of often extremely expensive hardware.

Here is his article.

I think this is probably true, at least largely true.

And here is his discussion of power filtering (and dimmers)

he recommends the use of the Corcom Series R (inductor-capacitor) filters, which has got to be a more cost effective solution than using Monster, Belkin, Furman, or even APC filtering strips.

I note that he also has delt with ground loops in a way not recommended by electricians and institutions like UL.  He says:

In my studio I have ground lifters on everything except my main mixer. The entire system is grounded through the audio wire shields that connect to the mixer, but only the mixer is connected to the power line's ground through the third grounding pin of its power cord
BTW, for liability reasons, I won't recommend that anyone do this, it could cause death through electrical shock, it's against all electrical codes and may even be illegal.  However, this is similar to approaches I have used in some of my systems.  The idea is to "lift" all grounds except one (either power amp or preamp) using grounding plug adapters that are not sold to be used this way.  I know lots of audiophiles who do this, and I can see why, it's almost impossible to get a complex system hum-free any other way.

Now I have done lots of reading and research on this subject.  If equipment had been designed properly, this would not be required, but hardly anything actually is, it turns out.  There is a particular way of isolating signal and chassis ground that can be done with a diode bridge, but hardly any manufacturer actually takes the trouble and expense to do this.  They just connect signal ground to the chassis at some point, and this will cause hum if there is more than one piece of equipment connected.  I will discuss the grounding and hum issues more in a future post, but here I just wanted to show what somebody else says.  Here we have a well informed professional actually suggesting something against electrical codes and potentially deadly.

The First Remote Extender modifications

Already having a remote extender "transmitter" (which picks up infrared from the remote and transmits it on radio frequencies) in the bedroom, to add remote control to the modified Sony XDR-F1HD tuner in the living room I first put a remote extender "receiver" (which picks up the RF and converts it back to IR) in the living room next to the tuner.  That basically didn't work at all.

About the same time, I got my first RF spectrum analyzer, a used boat anchor, from an eBay seller.  This was a useful test.  I used a small antenna to probe the room background noise and signals at 413Mhz, the frequency that remote extenders use.

I discovered to my horror that my beloved Tact 2.0 Room Correction Preamp is for some reason pumping out as much noise at 413Mhz as the remote extenders produce signal at that frequency.  So obviously no remote extenders are going to work in the vicinity of the Tact.  And I have two Tact's, one in the living room and another in the bedroom.  (Somehow the one in the bedroom doesn't seem to interfere with remote extenders.)

So, obviously remote extenders aren't going to work in the vicinity of the Tact, which is located very close to the Sony tuner.  I tried putting the receiver in nearby locations and using the 6 foot extension wire that comes with them with an LED at the end.  That way I could locate the receiver as much as 6 feet away from the tuner (and the nearby Tact).

Well, it seemed to work.  Several times I got it working by finding some clever location for the receiver where it didn't get as much interference from the Tact.  I tended to get it working in the night time when I do most of my listening.  Then I tended to discover that it didn't work in the morning.  I still don't quite understand why it doesn't work in the morning.  Obviously there is more radio frequency interference in the morning.  But where is it coming from?  Is there some way I can block it?

Finally I got really fed up with this approach when the clever receiver locations I found at night didn't work on the following night.  It was almost OK if it didn't work during the day.  But not working on some nights was not acceptable and before long it seemed not to be working on any nights.

About this time I also tried a bunch of different brands of wireless remote control extenders.  I found they are all very similar, with similar sensitivity.  Actually at first I thought one new brand was better.  Then I found it was worse.

Ultimately, to make this reliable, I decided to make my own extension wire with an LED at the end.  I made it long enough so that I could plug it into the known working receiver in the Kitchen.  That took 50 feet of wire, and conveniently I had a 50 foot roll of cheap 18 gauge speaker wire.  The hardest part of this was running a wire through the kitchen doorway.  I ran the wire up and around the doorway as inconspicuously as possible (but you can imagine my non-WAF house already) using AV staples (one of the most useful things to have).

Then I discovered that while the receiver in the Kitchen continued to receive signals from the transmitter in the bedroom, it would not simultaneously send them down my new wire and control the components in the kitchen.  Either plugging in the wire switched off the internal LED's, or it simply soaked up so much power that the internal LED's were no longer bright enough.

So I had to put a second receiver in the kitchen just to power the 50 foot extension wire with LED going into the kitchen, and leave the first receiver, which controls my kitchen video components, untouched.

Then while all this seemed to be working for awhile, eventually it wasn't working any more again.  I tried moving the receiver in the kitchen around a bit, putting it in different locations to see if it could pick up the bedroom signal better.  This sometimes seemed to work, but only for awhile.

Finally I decided something even more radical was required.  I would need to take either the transmitter or receiver apart to see if I could somehow juice them up a bit.  I had read stories online about other people doing it.  It sounded like something I could do, I even have equipment to test that my modifications are working properly.

Remote Extenders and RFI

The topic of remote extenders is more "home theater" than "audiophile" according to audio stereotype.  The purist audiophile does not use remote controls, but prefers to get up and clean records before putting them on the 'table.  I've known some audiophiles who even eschewed volume and balance controls.  Their system was set up for the one correct volume that makes everything sound best, they said.  The only control is the power switch on the main amplifier.

I have a record cleaner and occasionally use it, but generally I AM the kind of guy who does like to use remote controls, especially with digital sources.  I have lots and lots of remote controls.  I like to use the factory remote for each gizmo, and I have mountains of gizmos that have remotes.  It always seems to me that "Universal" remotes leave out some of the obscure functionality that makes each unit special.  So I have big wood boxes for remote controls in every room.  And I have a whole house audio system called Sonos which lets me play music from my hard drive and internet services like Rhapsody and Pandora in any room in the house.  Sonos also lets me listen to non-digital sources.  Every Sonos box has an analog input as well digital and analog outputs.  In every room there is a preamp or other source selector which feeds the Sonos input in that room.  So I can listen to any source from any room in any other room or in all the rooms in my house at once.  My Sonos system is set not to use any form of audio compression, so it digitizes all analog inputs in 16 bit 44.1Khz digital ("CD Quality").  And Sonos has its own remote controls, but thankfully they're radio frequency wireless and in my experience provide 100% reliable transmission unless they lose their memories from power failure and need rebooting.

Contra audio stereotype, I like this kind of audio distribution system for automation and convenience.  Sometimes I set my system just to play music selections randomly.  That has actually allowed me to listen much more to the obscure pieces in my collection than I would otherwise do.  In fact, mainly it allows me simply to listen much more than I would otherwise do.  Not that I am always listening to audio, in fact it's less than half of the time.  I like silence some of the time also.  But if I didn't have an automated music system, I'd be getting more silence than I'd like, since it's actually quite a hassle to select music from your record collection, clean the record, clean the stylus, and put it on the turntable, etc.  Not much better with a CD.  Much easier to press a button and load songs from your hard drive (as it seems everyone but purist audiophiles has discovered).

Anyway, I started using remote control extenders long long ago.  Back when I had both Super Beta and VHS vcr's (hey, I still do!).  I had the Sony 900 Super Beta vcr in the living room, and the Panasonic VHS vcr in the bedroom.  But it always seemed I wanted to watch beta videos in the bedroom, and VHS videos in the living room.  Both video systems were wired on the same RF line from the cable company.  (Things were so simple back then, there weren't 10 different kinds of video interconnection to hassle with.  But then the quality wasn't as good either.)  With a remote extender, I could control the Super Beta box from the living room, etc., just as I wanted to.

Now as time has advanced, I got more and more remote extenders to bridge more and more needs for functional control across the house.  Now I have 4 video systems which interchange digital HDMI video through OWLink fiber optic cables.  I have 3 remote control extender "transmitters" in strategic locations, and I've forgotten how many receivers, they're all over.  I aim my remotes at one of the "transmitters" and the infrared remote signal gets transmitted to everything, so I have total control, everywhere, BWAHAHA.

But there is a fundamental problem with this approach to home entertainment.  The problem is that these wireless remote extenders use radio frequencies, and our home space is getting more and more saturated with radio frequencies.  So I'm finding that the range of wireless remote extenders is getting less and less as they are forced to compete against this background.  They advertise 50 or 100 foot range, but if you search online you can find people complaining that they don't go as far as 8 feet.  I've learned a lot from these postings how the units work and can be modified.

But for me, it seemed like all my remote control extenders were working perfectly until I needed to add just one new one for the modified Sony XDR-F1HD tuner which is now my "whole house tuner" that I listen to everywhere in the house through Sonos.  I like having it always available, but I can't listen to FM without having the ability to change the station at whim, otherwise, before too long, I simply have to turn the radio off.  The tuner is actually located in the living room where I have my best indoor FM antenna.  But it was hard, hard, hard to get the remote control extender working to there from the farthest location, the bedroom.  I struggled with this for months.  I even bought an RF Spectrum Analyzer to see if I could determine the nature of the problems (and for other reasons, mainly I just wanted it).  Wierdly, the remote extender tended to work in the night, but not during the daytime.  I figure that is because there is far more background radio frequency "noise" (that's someone else's signals) during the daytime.

This week I made one more upgrade to the remote control extender system which I hope will finally fix this problem,  so I can control the FM radio from the bedroom any time of the day.  I'll describe this upgrade and how I got there in a future post.

Thursday, October 14, 2010

Source for shielded cords in different lengths

Here is an online storewhere you can buy shielded power cords (US) in different lengths:

Useful if you just want shorter power cords without the expensive audiophile cords.

From Krell to Lavry

The last month has been harrowing in my audio explorations.  It began with me tackling on of the bigger annoyances of my main system, the tendency of my Krell FPB 300 amplifier (which I love dearly for its natural sound) to make fairly loud mechanical noises like klinks (or pings) from time to time.  I could now write at least 10,000 words on this subject.  I think the problem can be solved on my unit (eventually the problem was solved in later production I hear) but I haven't gotten there yet.  I read on Audiogon that the klink problem can be solved by loosening all the screws on the case just a little.  So I tried that, didn't help.  I even got a precision set of Wiha torque control drivers ($200) and precision torqued one whole side to an exact setting.  That seemed to make matters worse, actually.  Before that, I had moved the Krell into the corner and out-of-the-way from the periodic A/C blasts.  That didn't help either.

After all this messing around with the Krell, moving it in and out of the corner several times, working all the vast numbers of Torx screws it has many times (I had to overcome great trepidation before loosing screws on the heatsinks, I was afraid that might destroy the amp), I noticed something bad.  One side was getting very hot, too hot to touch for more than a couple seconds.  Even just idling it would do that.  After a bunch of torque tests, but just before that discovery, I had left the amp idling for a few days thinking that might help the chassis metal plates find their natural position with the new lower torque.  That was when I noticed how hot the amp was getting on one side only.  Actually, I think I might have noticed one side being hotter than the other for a long time, but I hadn't realized it got that way just by idling.  I put a Kill-a-Watt meter to see how much power it was using.  But before I put the meter on, I shut the amp down for a couple minutes.  Then I turned it on, and noticed that the power consumption just kept on increasing.  Finally, in a minute when I wasn't watching, the amp got so hot it shut itself off.  OK, it has to go back to Krell again to fix the bad left channel.  Last year it had been back to Krell for a bad right channel.  Hopefully after enough things get fixed, it will stay fixed for awhile.  I bought the amp apparently working 3 years ago, but I think it had been just about ready to break from 10 years usage.

So then I had to switch to another amp in my amplifier collection.  Fortunately I had just purchased a nice Parasound HCA-1500A as an upgrade for the amplifier in my bedroom.  But with 300W at 4 ohms, that has enough power for the living room Acoustats as well.  When that Parasound amplifier had arrived in August, I was thinking about testing it on the Acoustats.  But I figured it wouldn't be as good as the Krell, so I didn't bother.  Well now I was forced to try it.  And it seemed to play OK without shutting down (I tried a lower power Parasound HCA-1000A once, and a moderately high level it shut down from overheating).  It sounded pretty good.  But it lacked the incredible natural depth that the Krell has always provided.

Just about then, the first Belkin PureAV power conditioner arrived.  At that time my living room audio wiring was a total mess and without massive reorganization I wouldn't be able to fit the Belkin in.  So I embarked on a large reorganization, moving all the line level equipment from the left side toward the right side for connection to the power conditioner without extension cords or power strips as had been required previously.  About two boxes of unnecessary or too-long cords were removed.  So the wiring had gotten much simpler and shorter and there was now a power conditioner.  A handful of long interconnects got replaced with shorter versions, but for this post I'm not mentioning all the specific changes made.

The result, after the conditioner had been installed, and the wiring layout cleaned up, was a considerable improvement in sound.  "The blacks got blacker" as they say.  This actually restored much of the depth that had been lost when I swapped out the Krell amplifier.  Playing classical FM radio, the sound was just about as good as it had been with the Krell.  But not quite.

So then I decided to try another upgrade that had been waiting to be installed.  My main system is based on digital signal processing for crossover and room equalization.  This works wonderful with digital sources because the signal is already digital, and digital signal processing is as close to perfect as it gets.  But analog sources, such as FM radio, must be converted to digital for me to even listen to them.  The Analog-to-Digital converter (ADC) in my Tact 2.0 RCS preamp is only so-so.  It's supposed to be 24 bit (from 1999) but only has 16 bit performance according to a review in Stereophile and my own testing.

I had already purchased a much better ADC, a Lavry AD10.  This has 24 bit operation with 120dB signal to noise ratio, about as good as it gets.  The Lavry (and it's even more expensive brother, the Lavry Gold) are highly praised by recording engineers.  BTW it is priced at $1480 and I ordered mine directly from Lavry in Washington State.  I always feel good when I can buy top quality equipment made in the USA.

Replacing the Tact's built-in ADC with the Lavry was a revelation, vast improvement in depth, deep bass, transparency, coherency, everything.  Now listening to FM radio is even better than it was when I had the Krell online, though for different reasons.  Hopefully it will get still better when the Krell is fixed and back online.

The rest of the Belkin PureAV AUR1500 photos

OK, here are the rest of the photos.  I've figured out that I can easily upload from my mobile device (love that term, heard Garrison K use it on Writer's Almanac a few days ago) to Picasa using a mobile app.  Once photos are in Picasa, it's a snap to move them over to Blogger (both owned by Google).  The mobile app needs wifi, it simply won't work (no messages or anything) over G3 cellular.

 The above picture shows the A/C filter board for the 4 different filter banks (I think the analog section gets two filters).  This looks like pretty serious stuff, equivalent to the big filtering conditioners like the Monster 5100.  The film capacitors don't look quite as pretty as the Monster ones, but they look to be UL rated.  There's some foam and board filler on the inside of the wirewound chokes.  That looks a bit tacky, I suppose it was done to control mechanical noise.  If you get one of the top industrial Corcom AC filters, it's may be filled with potting material to achieve the same end.  Speaking of which, I wonder how these filters compare with the top-of-the-line Corcom as well as Monster, Furman, and others.  I've begun to find filter "specifications".  For example, Monster claims 80dB line noise reduction, and a cheap but highly regarded APC power strip claims 70dB reduction.  The filter above must be more in the 80dB or higher category.  But it's not really possible to boil this down to a single number.  The advantage of the big complex filters is that they can filter down lower closer to the 60Hz line frequency.  A cheap filter may only be able to filter radio frequencies, not audio frequencies as well.  So you really need to know the filter attenuation at different frequencies, such as 120Hz, 1Khz, 100Khz, 1Mhz, 100Mhz.

In case it wasn't clear from earlier photos or descriptions, the above photo shows how the back of the Belkin has two board, the filter board and a much larger UPS sinewave inverter board underneath.  In the middle are the inverter transistors in a heatsink and a transformer.

The above is a closeup of one inductor and capacitor used in the line filter, and off to the side you can see the chokes that the incoming ground (!) and powerlines run through.  Good to see that the power is filtered right at the input.  But the convenience outlet on the front panel is not only not filtered it can reflect noise back into the unit, as I discovered by plugging a tensor lamp into it.  Plugged in there, the lamp created huge noises on stereo when being switched on and off.  However plugging lamp into outlet beneath power conditioner shows the conditioner provides nearly perfect filtering from the outside, there is only just barely detectable noise (ear up to speaker) switching lamp then.  So my advice is (ironically) only to use the front outlet for temporary convenience purposes like using a tensor lamp or soldering iron.

Above is the heatsink for the inverter transistors.  Is this adequate for providing 1000W power?  I suppose it must be, though it looks like the above heatsinks are only good for 100-200W actual dissipation at high temperature.  Probably the transistors operate in Class D switched digital or Class G rail switching (like Carver amps) for maximum efficiency.  I believe one of the most highly touted regenerators, the PS Audio Power Plant Premier also uses Class D.  With these classes, much smaller heatsinks are required than for Class AB or Class A.

Above is another picture showing complete filter board.

Above are two photos showing the internal layout.  Fortunately the heat sinks are in a mostly empty area of the unit.  The transformer is also quite nicely separated from everything else.

Above is a view of the heat sinks from the side of the battery.  Notice that on the left side of heat sink there are two medium size electrolytic capacitor within about a quarter of an inch.  Perhaps they needed to be that close for electrical reasons, but the heat from the heatsinks is likely to shorten the life of the capacitors if the inverter is run to hot on a daily basis.  Somebody might buy this unit thinking they can now run exclusively on battery sourced power...charge during the day and then disconnect and play at night, etc., powering amplifier and everything for 30 minutes at a time.  But I think the construction of this unit does not look adequate for that kind of heavy duty cycle usage, only for occasional use for mostly short durations.  With that kind of low duty cycle usage, it can probably last as long as typical electronics, about 20 years before needing repair.

The above is a picture of my living room where the previous pictures were taken, with my stereo and TV in the background.  Notice on the far right of my stereo you can see the blue screen of the Belkin I am currently using there.  The new unit being photographed is intended for the kitchen stereo.  I recently rearranged the stereo putting most components as close as possible to the power conditioner so I can use short power cords and a minimum number of additional power strips.  I think keeping AC wires as short as possible is useful because it helps to keep the EMI noise level down.

I bought a few short power cords recently (1 foot, 2.5 foot, and 4 foot) hoping to improve the current arrangement in which generic 6 foot cords are coiled somewhat in the corner.  But the new cords were not shielded, and I decided it was important to stick with shielded cords.  So I have now found a source of short shielded power cords and have ordered them.  I might also choose to upgrade particular cords, like the 3 foot cord that connects to the Belkin itself, to something super premium like Cardas.  I've spent some time recently looking at the designs of nice quality (e.g. Belden) and audiophile grade (Audioquest, Cardas, Valhalla).  I generally like the approaches taken and the materials used by George Cardas.  I'm strongly skeptical that you can actually hear the difference between using $200-$1000 power cord from Cardas and a standard 14 gauge shielded cord in a double blind test.  But it looks to me like the Cardas cords and speaker cables are designed and made about as well as can be.  In both cases of low lumped impedances, the chief performance issue is self-inductance, and Cardas deals with that through elegant Litz wire designs.  This is not fake snake oil, it is the real stuff, the only question is whether you actually need it or not.  (But if you're an audiophile, you know how that question is answered.)

Wednesday, October 13, 2010

Interesting blog on power conditioners

This is a really interesting blog on power conditioners:

Like me, this guy has an AudioPrism Noise Sniffer to determine the line noise on any AC outlet.  But he's figured out how to get repeatable measurements by measuring the audio output of the sniffer with a SPL meter.  So he's got numbers that show that sinewave regenerators (like PS Audio Power Plant) produce the quietest power, followed by transformers.  Noise filters in the more expensive outlet strips also filter pretty well.

I'm thinking that the noise filters in my Belkin PureAV AVU1500 are doing a pretty good job and that is why I got some of the depth back in my system.  The quieter backgrounds help with depth perception.

I just did a simple noise test comparing industrial grade Smart UPS 1500 (a 70 pound unit) with the Belkin.  Unfortunately it was not a good test, as these big heavy units were in different rooms plugged into different outlets.  But while the Belkin had basically clean power, the Smart UPS power (with normal A/C input) was buzzing away badly.

I'm a little unclear how the Noise Sniffer works and how best to use it.  It's my take that the volume control also changes the frequencies that the sniffer is testing, so you always have to run it through the whole range.

I have never used any power conditioning device with my Krell FPB 300 amplifier.  It gets power from dedicated circuit with isolated ground, through audio quality outlet (currently Oyaide), and that's it.  I'm afraid that anything might melt under the immense current draw of this amp which challenges the 20 amp circuit that it's on.  And anything might restrict that power also, obscuring some of the Krell magic.

The Krell has been offline now for three weeks, but I'm using the same philosophy with the temporary substitute amp, a Parasound HCA-1500A.  Which is actually one of the few analog  pieces in the system, and it might actually benefit from some kind of conditioning.

Belkin PureAV AVU1500 vs APC S15 continued

Just taking another look at APC S15, I see it weighs quite a lot less than the Belkin, only 47 pounds.  The Belkin weighs 77 pounds, 30 pounds more.  I think about 15 pounds of that is a lighter and lower power battery.  The remaining 15 pound difference probably just lighter weight stuff, including lighter weight transformer.  Although this contributes little to the weight, the Belkin claims 8000 joules of surge supression, whereas the APC claims only 4080 joules.

Anyway, given that the Belkin does not impress me with UPS overbuilding, it would be hard to believe the  even lighter weight APC would impress me more.  And I know it's hard to judge these things from appearances, though I still do.

Did I mention that not only is the AVU1500 being discontinued, Belkin announced it will be leaving the UPS market.  That does not sound good, though it might be that they simply couldn't compete against the marketing clout of Monster and APC.  Or they could have had too many product failures and decided to bail.  If you read online reviews of the AVU1500, it seems that one of the most common complaints is that the unit is essentially "dead" when first powered on.  You have to allow 12 hours of charging before you can get the blue information screen.  But people should know that, all my units came marked with a big warning label "CHARGE 12 HOURS BEFORE USE".  So if that's the only complaint, it's really a user error.

I've also gleaned that if you discharge the battery all the way that you are back to the inital "charge only" condition with no operation allowed until there is sufficient charge in the battery.  Hopefully that won't happen often, though it could also be considered a design flaw.

I've mentioned Monster in these comments without comparing specific UPS models.  Monster does have a number of audio and home theater UPS, but at least the ones I've looked at so far are only stepped-approximation-to-sinewave generators, which are generally considered inferior .  If Monster does have a sinewave generating UPS, you can bet it will cost more than similar products from APC and Belkin.

I already feel a big guilty about buying 3 of these UPS's.  It seemed like a no-brainer when they were selling for $349 at this summer.  But just after I got my first one, the price shot up to $500, so I got two units around $500, and now the price is up to $770, right about the same as the APC S15.

So I've put about $1349 into power conditioning and protection instead of music or more cutting edge audio equipment this month.  (But the month is not over yet...)

I need 3 because I need one for the main living room system, one for the kitchen system, and one for the computer room system.  I decided not to get one for the bedroom system, because the latter is only switched on when in use and is already protected by a nice surge/filter unit.

OTOH, I think there is audible improvement comparable or better than expenditures in other areas, equipment protection, and that warm feeling of being safe in your own home next to your stereo.

Inside a Belkin PureAV AVU1500

It's been very hard to get this posted, first time, from my mobile phone.  Ultimately I had to re-forward from  another email account.  I had a bunch of other pictures, but will get posted slowly until I can figure a better way.

Note that you're getting a pretty full box of stuff in your Belkin home theater UPS.  The most impressive bit to me was the set of linear filters in the back, one for each bank of outlets and perhaps two for the analog outlets.  With wirewound chokes and large film capacitors, it looks similar to the components inside a large Monster Power 5100.  In addition to the chokes at the AC input, each bank apparently sees a ferrite choke followed by a shunt cap, a series inductor, and a final shunt cap.  Then there are two flat MOV's and one cylindrical TSS before going to the outlet bank.

You are also getting an impressive 30 pound battery, not found in cheapo UPS's because it's worth more than many of them.

Now I want to believe the sine-wave backup generator is good, and it's much better to have sinewave generating UPS than "stepped-approximation-to-sinewave" in my opinion, who knows what that kind of distorted wave is going to do to equipment (though I guess it must be safe, because almost everyone does it).  In any case pay a large premium for pure sinewave in the UPS market.  APC, for example, has a cheap home theater UPS with stepped sinewaves (H15, $250 street), and one priced about the same as the Belkin (S15, $1500 list, $800 street) with sinewaves.  Hey I wonder if the APC S15 is a clone of the Belkin, though the wattage spec is 100W lower for the APC (900 vs 1000W).

But this particular sinewave inverter, while probably similar to comparable units from APC and Monster, does not impress me with over-building.  The heatsinks for the the transistors notably look fairly small for a unit supposed to be able to supply 1000W and 1500AV.  Also, the internal air path does not go directly through those heatsinks.  (I was actually expecting to see a wind tunnel inside this thing, but I guess nobody does that anymore now that transistors are so much better.)  Instead, the heatsinks are simply in the center of the unit, where there is at least some open area to *allow* airflow.  Also, notably, within half an inch of the heatsinks are some vulnerable electrolytic capacitors.

But I don't think your cheapo stepped wave UPS's are going to have that fairly hefty transformer.  It looks like the sinewave generator drives the transformer which steps up the voltage as required.  The transformer is apparently an important part of this, maybe does a lot of the smoothing, and is located right next to vent holes on the side of the unit and therefore in the direct airflow path.  That's probably because the transformer heats up as much as anything in sustained operation.  And that's probably because we are dealing with somewhat marginal parts and design here, not something built to extremist audiophile standards.  Something like this built to extremist standards wouldn't cost $500 or even $1500 but something more like $20,000.

In my mind so far this is not a "server grade" UPS that you would feel safe running your hedge fund on.  It's probably not up to long daily stretches on backup power without failing during warranty period either, but it's probably adequate for quasi-weekly power hiccups, which is really what it's intended for anyway.