Sound Quality Vs. Measurements

Status
Not open for further replies.
Well, a valiant German member, very active here, has been banned. I have no idea why.

Perhaps he sinned terribly against the powers that be, or something, the point is, I never saw the message which caused him to be banned.

However, knowing him for some years now, I doubt it was such to merit a ban. But that's just me.

In general, Germans tend to stay on German forums, and it's not because of any language barrier. I believe, right or wrong, that it's because it was the Germans who started adding subjective tests to objective ones way back in the early 70ies, I must say, quite some time before anyone else, along with the Italians. To them, some of the discussions here are really passe, they went through them many years ago.

But this would be best explained by our German members.

Well, I'm german.

Up until I moved to the UK in '87 I had not seen a single subjective test.
All tests by the two major german HiFi publications until then were blind group tests and whoever did the measuring and writing up of the article was not one of the listeners.

On the other hand almost all tests in the UK press back then were purely subjective.
 
For the record, in my view, Italy's "Suono" was probably the most balanced of them all, until it went down the drain like most paper mags.

They had a deal with Italy's Instituto Alta Fedelita to do all the measurements for them, so we had waterfall diagrams of speakers WAY back before anyone else, just as we had 3D diagrams of amplifier behevior from -3dB to +3dB. About two full pages of it. A half page discussion. Then a 2 page subjective test, with clearly named sources, both hardware and software.

Not unlike today's Stereophile, but in the early 70ies and using the technology of the day.

They also had very memorable projects by people like Bartolomeo Aloia and others, in their time, very state of the art, about 99% of all commercial units didn't have some of those goodies done like that.

But, in the mid-80ies, the old crew broke up, and that was the end of a legend. Today, it's just something to weep over.
 
All tests by the two major German HiFi publications

Audio, Stereoplay, and Stereo magazine.
Audio was particularly good with measured data during the days of Bernhard Fuß, before he and his brother Andreas went the commercial design route.

Absolutely subjective was Hifi Vision : "Absoluten Spitzenklasse" for everything gold, shiny and expensive.
For some odd reason, in particular when it was German manufacture under test, which was all too often. Very nice pictures though, unfortunately.

(me still have all issues of all four in storage, from the early '80s till beyond 2K, magaholic tendencies i think)
 
First, do these Technics turntables have Velocity or Phase lock servos?
PLL.
Second, what is the 'bandwidth' of these servos?
Exact current of charge pump is not known (tens of microamperes?), since it is Matsushita's custom IC, but externel filter components are as follows, if that helps:
C303 = 1uF
C304 = 10uF
R302 = 470 Ohm
Nothing fancy, as you can see. C304/C303 ratio is 10:1, so it is pretty much textbook design. Both capacitors are electrolytics (ouch!)
Third, how much inertia does the turntable have. Is it enough to fill in where the servos can't correct anymore?
Not enough for my taste, but opinions may vary.

Best,
 

Attachments

  • SL1200_PLL.jpg
    SL1200_PLL.jpg
    149.8 KB · Views: 145
diyAudio Member RIP
Joined 2005
The Technics I have (in storage, don't remember the model) had a rather poor servo, as judged from the tachometer on the side of the platter illuminated by the neon bulb. It may have been that the potentiometer was getting noisy.

I found the turntable mass to be pretty light, not very reassuring. I now have two Heybrook TT2 on loan from a friend. One is fitted with an SME V and sounds quite decent to my ears with a Grado moving iron cartridge (again on loan and I don't know the model number).

The Heybrooks both required repair of their crystal-controlled drive electronics upon receipt :mad: The engineer also made some questionable decisions by switching crystals with the speed selector switch, which besides making things more iffy with the relatively long runs of cable, also makes the switch likely to develop dry-switching problems (sometimes the oscillator doesn't start). And the motor phase shift cap is optimized for 33.33 and not 45 r.p.m. And as mentioned before, the plastic dustcover is prone to acquiring a large static charge when dusted, which can be enough to lift the tonearm right off the record. And the back hinges all bust out after a while.

But hey. It does sound pretty nice when it all works.
 
Administrator
Joined 2004
Paid Member
I think we may have a winner! Whoever said IM might cause the irritation on my EL34 amp could be right. (Anthony?) There is a little ringing on resistive and reactive loads, but it doesn't look like much more than 1/2 a dB. That might still be annoying.

But IM is bad and gets worse on a reactive load. Interestingly, harmonic distortion is a little lower on a speaker load than on a resistor. Maybe an impedance thing. I'll post charts in a bit.
 
Haven't been here for a while, but just thought I'd report my very first impression of an active crossover setup, before I get used to it. Using a PC and my own software, a pair of sacrificial Mission two-way speakers and some secondhand Japanese amps I can honestly say that I can hear a major difference! - and I'm the biggest sceptic of them all.

I haven't tuned anything yet, but I am blown away by the 'solidity' of the stereo image. I keep on being startled by sounds within recordings that I am mistaking for being in the room with me. I really do wonder if all the other audiophile stuff is very, very minor in comparison to getting rid of the passive crossover/over-worked-amplifier combination. Before, I was hearing two separate speakers that sounded pretty fruity in combination at times, but no solid 'soundstage' to use the jargon. (The same was true for my much better speakers and amplifier.)

Having heard this system, I can now imagine that much listening fatigue stems from your ears and brain attempting to keep track of sounds which, because of the primitiveness of passive crossovers, are wavering about and blending in complex and arbitrary ways. (It could also be that this provides a superficial 'aural excitement' at first, that gives an illusion of actually sounding better than the original recording - for a short while.)

The active crossover is clearly the way of the future! And there is still ample scope for mixing, matching and tweaking (even more so than sticking with the old steam-powered methods).

Edit: I am listening to this setup and regretting that I didn't do this long ago. It really is wonderful; not a subtle improvement at all. I've been listening to a complete audio muddle for years and imagining it was stereo.
 
Last edited:
Administrator
Joined 2004
Paid Member
Welcome to the world of active crossovers! They can be quite startling when you first hear them - they do a lot of things right.

I think active crossovers shine because, in part, they are so much easier to get right. Great passive crossovers are a b**** to design. It can be done, it's just harder. Active crossover offer a world of tweaks and possibilities.
 
diyAudio Member RIP
Joined 2005
Haven't been here for a while, but just thought I'd report my very first impression of an active crossover setup, before I get used to it. Using a PC and my own software, a pair of sacrificial Mission two-way speakers and some secondhand Japanese amps I can honestly say that I can hear a major difference! - and I'm the biggest sceptic of them all.

I haven't tuned anything yet, but I am blown away by the 'solidity' of the stereo image. I keep on being startled by sounds within recordings that I am mistaking for being in the room with me. I really do wonder if all the other audiophile stuff is very, very minor in comparison to getting rid of the passive crossover/over-worked-amplifier combination. Before, I was hearing two separate speakers that sounded pretty fruity in combination at times, but no solid 'soundstage' to use the jargon. (The same was true for my much better speakers and amplifier.)

Having heard this system, I can now imagine that much listening fatigue stems from your ears and brain attempting to keep track of sounds which, because of the primitiveness of passive crossovers, are wavering about and blending in complex and arbitrary ways. (It could also be that this provides a superficial 'aural excitement' at first, that gives an illusion of actually sounding better than the original recording - for a short while.)

The active crossover is clearly the way of the future! And there is still ample scope for mixing, matching and tweaking (even more so than sticking with the old steam-powered methods).

Edit: I am listening to this setup and regretting that I didn't do this long ago. It really is wonderful; not a subtle improvement at all. I've been listening to a complete audio muddle for years and imagining it was stereo.

That's interesting. I wonder (since the audiophile community is so reluctant to embrace powered speakers per se) that we might persuade some speaker manufacturers to bring out all driver connections, and have their passive crossovers as an option. Of course disclaimers will abound, about too much LF energy to the tweeters, etc.
 
Recent days I was looking for some objective justification of my preference of using UHF parts at the output stages. One respectable Russian designer, wrote the following about why he does not use switching MOSFETs:

“Spice-models of MOSFET transistors are adequate only for lateral MOSFET, operating mainly at the mode of saturation velocity of carriers.
Power MOSFETs, especially the "vertical" switching MOSFETs, are qualitatively different from the "laterals" in almost all properties.
In particular, the initial part of the drain-gate characteristics of practically all power MOSFETs is exponential, exactly the same as that of the bipolar transistors (only with slightly smaller and different for p-and n-channel transconductance).
This is well illustrated in the semi-logarithmic datasheet graphs (when the gate-source voltage is given on a linear scale, and the drain current - a logarithmic).
Power (switching) MOSFETs are designed for a relatively small current density in the channel, to minimize resistive losses. Therefore, the "initial segment" of the input characteristics (exponential for any electronic devices, including vacuum), is extended to a sufficiently large (in absolute terms) currents.
This is the main difference between the "switching" DMOS FETs from those for UHF applications. For the last ones, most important are the minimum input and output capacitances, and the channel current density is significantly higher for them.
In addition, switching MOSFETs usually have a polysilicon gate, and, accordingly, the "distributed" array of R gate x C inp.

The physical effect, which explains these properties, is very simple - the subthreshold current is exponentially dependent on the voltage at the control electrode. And only with increasing control voltage (and current density) other factors come into play - first, the estimated square-law characteristic is reached, and then, if the current density is sufficiently high ( several amperes per millimeter width of the channel) - saturation of the drift velocity is achieved with subsequent fall of the steepness.

For the "old" switching MOSFETs (such as IRF540) exponential characteristic of currents is valid up to about 0.3 ... 0.5 A, for a more modern (with a smaller R channels) - up to 1 ... 2A (the last is worse for audio). It is just this range of sub-treshold currents is not included in the standard SPICE-model.
The graphs in the datasheet are usually show clearly visible region of high currents, where the dependence is close to square-low.

From practical experience, with typical quiescent currents in the output stage of an amplifier of 50-100 mA with IRF540 (and 30 ... 70 mA with IRF640), we have almost exponential drain-gate characteristics.
Exponential subthreshold current range is a good control criterion for the uniformity of the threshold voltage over the structure area, it is sometimes even used in controlling the production of crystals.”
 
Motional Feedback .

Active speakers .
I have a very limited experience of this and strictly a DIYer . However what I have found out suggests to me that it is much easier to do active crossovers than conventional passive crossovers in speaker design . An amplifier can be fashioned to do exactly the task required . This allows design to be optimized ( class of output and power ) . From the miniscule understanding I have 7 kHz is an ideal crossover frequency if the drive unit is able to do it . If the bass end 150 Hz . Filter type is up to the one trying it to decide . Building speakers where time alignment is possible simplifies the design . In the one I am helping to build a Fostex unit sits on top of a larger bass cabinet . The Fostex has been moved many times .

The program I use is below for designing filters . An amplifier that resembles an op-amp can be configured as a filter . As a minimalist that appeals to me and have done exactly that .

Simple first order filters can be the input capacitors of the amplifier . Second order the output capacitor in addition . Third order the feedback cap . I often find output caps less of a problem than thought sonically . Passive bass and active top is possible . Technically the simplified version I suggest is not active . However I feel it lives within the spirit of an active concept .

The system I have helped with has 300W transistor amps driving old Leak Bass 12 inch . Fostex FE108 ( ? ) top . Hopefully an Air Motion Transformer tweeter one day . The owner prefers a rolled off treble so Fostex is fine for him without a super tweeter . The Fostex is powered by a modern design of UL SE tube amp with KT88 ( ECC 83 voltage amp ECC 82 SRPP ) .

Negative output impedance is possible with amplifiers ( a concept which even RCA tube amps offered ) . Belgium Professor Korn made products in the 1970's called Servo Sound which were approximately using that idea ( Philips also in Motional Feedback speaker were looking at bass control servo's ) . This can get better bass performance and comes virtually as a no cost advantage . Meridian perhaps use it ? Some car speakers have dual coils . Perhaps that could be used ?

AES E-Library Application of Negative Impedance Amplifiers to Loudspeaker Systems

Sub-Woofer Controller

http://www.ti.com/lit/an/sbfa003/sbfa003.pdf

https://www.google.co.uk/search?q=m...ZC4TU8gO_6p3wDw&ved=0CHEQsAQ&biw=1272&bih=849

BTW . Always have 2 sub-woofers .
 
Last edited:
Who? I thought you meant Thorsten but he is not marked as banned unlike a few others.

You thought right, Scott.

His hard time is done, but I don't think he'll be back soon. It's not the actual banning, but he feels he has been wronged, and that doesn't just go away. "Creative differences" will only go so far, I believe he thinks that the local powers that be have overstepped their authority by an unacceptable margin.

Hence, his "Mr Pottymouth" power amp project is evolving elsewhere. I think this is a pity, but there's nothing I can do about it.
 
For decades, I have been saying and I say it again - NOTHING beats a well designed and made active system. On the other hand, active systems are like all other systems, a few great ones, a few solid ones and a lot of so-so ones.

My first contact with them was in TV editing room, where they had a pair of Klein & Hummel big three way active speakers. Upon hearing it first, I asked for a gun to shoot them, because they camke screeching at me with a vengeance. Upon inspection, I discoveverd that the mid pot was about +3 dB up and the tweeter level was all the way up. I asked for a calibration, a tech came around with mikes and equipment, spend an hour or so, and after that, it became what it was supposed to be.

My entire loudspeaker project is revolving aroud the idea that my currently passive speakers should eventually become active, with the electronics outside the case. Hence my envolvment in a power amp project, a version for the bass, and there should be another two evolved from it, a version for the midrange and a thrd for the tweeters.

What has given them a boost lately is the appearance of single chip power amps, which is a quick, easy and rather economical solution. This can make an insiginificant speaker be noticed, but if you want really great, you still need good drivers - no shortcuts there.

As Nige says, the main benefits are the electronic XO and the fact that one can optimize an amp for a given function. Simply using electronic XOs will do much for efficency, I beleive you gain no less than 2 dB efficiency there, passive XOs are power gobbling phase shifters.

Sure, it's quite a bit more expensive that a classic system (all else being equal), but I believe the audible reward is just as great.

A version of my amp could be made to deliver pure class A of up to 50W/8 Ohms, which is more than enough for my tweeters. In its current state, it will deliver 100W/8 Ohms for the bass no problemo, and it will not begin to saturate until just over 170W/8 Ohms, which should be enough for the bass driver, which is rated at 150 Watts "music power" (120W RMS), but also "up to 300 Watts non repetative music program", whatever all that means.
 
...

BTW . Always have 2 sub-woofers .

Odd you should say that, Nige, you are one of the very, very few who think so.

I remember a text published in a UK mag, I forget which, probably HFN&RR, where somebody also said the same thing.

This flies in the face of the theory that below 200 Hz we essenetially have a mono information and thus one sub is quite enough. Well, that may well be som but just try the same system once with two of the same subs, taking care to integrated them well, and yu will realize that there is a world of a difference.

I did once, thank God I can do things like that simply because I've been around for several decades and know all the local dealers. I borrowed a JBL sub just like the one I have, 10" driver backed up with a 100 Watt chip amp inside, and man oh man, it really took off. You get bass which can literally tickle your kidnies if need be, while staying pretty clean.

Obviously, I agree with you completely.
 
DVV . Although not required one can use zero bias class B up to 500 Hz . I am not saying one should . However it simplifies design and makes power delivery safer ( OR emitter resistors are even possible ) . I would say do it for a day just to say your opinion is based on truth . Some say I didn't do it because it is obviously wrong . As 5 minutes work to turn the bias off how lazy is that ? I have built zero bias amps which have been fabulous . They are even better when bias applied . That's a thought worth pursuing ?

I suspect eventually active is cheaper if a total concept and not like Linn Naim active of the past .

The Chip amps can even be pushed slightly into class A with a pull down resistor to the negative rail from output . This mimics D Self power amp patent for peanuts ( Cambridge 840 ) . It has been in the public domain for years now in small op amps .

Have 2 sub-woofers as we have two ears . If not stereo images pulls to the sub woofer . Not always so . Easier my way and nicer looking .

Sub Sub is interesting and may even make the treble sound sweeter . The same effect is heard with big church organs that have subsonic ability . It is also dangerous and might cause heart problems .

http://www.google.co.uk/url?sa=t&rc...Or66wN&usg=AFQjCNFDkXFMKXzgMjELDMOhgP8qWangwg
 
Last edited:
Ah, pure class B!

Unfortunately, old habits die hard, and my preoccupation has always been how to push it as much as I can into class A by raising the bias, not lowering it.

But you are right, it doesn't cost much time or effort to try that as well. And I love to experiment.

On the other hand, the versions of the amp for the mid and high range I see as very much pushed into pure class A for the mid and completely class A for the treble.
 
Audio, Stereoplay, and Stereo magazine.
Audio was particularly good with measured data during the days of Bernhard Fuß, before he and his brother Andreas went the commercial design route.

Absolutely subjective was Hifi Vision : "Absoluten Spitzenklasse" for everything gold, shiny and expensive.
For some odd reason, in particular when it was German manufacture under test, which was all too often. Very nice pictures though, unfortunately.

(me still have all issues of all four in storage, from the early '80s till beyond 2K, magaholic tendencies i think)

Well, in the mid 1980 when the term "High End" started we had a German magazine called "Das Ohr - The Ear" run by Klaus Renner (an engieer who worked for Texas Instruments and his hobby was Audio) He was a rviewer for HiFi-Exclusiv as well and "Das Ohr" was the first magazine with don't use measurements - in the pattern of "Absolute Sound".

Nowadays we have "Fidelity" (some guys from the "Image HiFI" - team - and doing emasurements now), the before named "Image HiFi" (still without measurement test), "Hörerlebnis" (no measurements), "HiFi Stars", "LP" (specialized for analog playback - and more) to mention some mags.
 
VladimirK

That is fascinating about FET's . Especially the information about common cheap FET's . That is already a useful range . My experiences with them has not been especially good . I use Exicon which were designed for audio use .

Do you have experience of UHF op amps ? Some even give audio specifications and are cheap , they even can work in conventional voltage feedback configuration if I remember correctly ? Normally 5 V operating so have to be used with imagination . Seeing as you know so much about UHF your opinion even if that you have none is useful .

Thank you in anticipation .

I am designing a Gu 50 amp for fun , what a nice tube and from Russia ( was Telefunken LS 50 ) .

D V V . 10 watt class A amp will have Gu 50 to replace KT 88 soon . No choice in this as my friend wants tubes . Me too now I have made it work . This is in the active set up .
 
Last edited:
Status
Not open for further replies.