New KEF Studio monitors - any good?

[steve_1979]

Uh, an op-amp (operational amplifier) is an amplifier with differential inputs, not a filter. There is no way that "just an op-amp" can exhibit any kind of filtering properties like what's needed for a voice-range crossover.

I've read in academic hifi engineering and speaker design books that an active crossover is just an op-amp. Ashley James has also said the same thing several times on the AVI forum.

An active crossover contains active components (i.e., those with gain) in its filters. In recent years, the most commonly used active device is an op-amp

http://en.wikipedia.org/wiki/Audio_crossover

how many passive speakers use a "typical 2 way 2nd order LR crossover", you think?

Does the phrase "a typical 2 way" give you a clue to the answer?

 

[oldric_naubhoff]

I've read in academic hifi engineering and speaker design books that an active crossover is just an op-amp. Ashley James has also said the same thing several times on the AVI forum.

this is utter BS. active xover is op-amp based but that doesn't mean it's an op-amp. there's caps and resistors too. possibly even more goodies. just as a point of reference here's first active xover schematic I came across in google. 2-way, 3rd order, 2kHz xover point. you can clearly see loads of caps and resistors. BTW, what books were you reading too? will have to note never to buy them. :rofl:

An active crossover contains active components (i.e., those with gain) in its filters. In recent years, the most commonly used active device is an op-amp

http://en.wikipedia.org/wiki/Audio_crossover

[/quote]

well, yes. that's about right. you could always use tubes instead of transistors. but in recent years transistor based op-amps rule.

 

[steve_1979]

how can active speakers be more accurate?

1. Active speakers use amplifiers that are optimised to work with the individual drivers.

2. The passive crossover adds a considerable abount of audiable distortion but an active crossover does not.

3. With an active crossover it's possible to use a steeper roll off slope which sounds better than a shallow roll off slope.

4. A passive crossover also effects the way in which a amplifier controls the movement of a driver. With a passive crossover in between the amp and the driver it's like trying to push the speaker cone in and out using a spring. But when the amp is connected directly to the driver it has a much more direct and effective control over the movement of the speaker cone.

 

[oldric_naubhoff]

2. The passive crossover adds a considerable abount of audiable distortion but an active crossover does not.

sigh... go to page 3 in this tread and read my post #7 if you still haven't done so. it's about quoted harmonic distortion from new KEFs and Event Opals...

3. With an active crossover it's possible to use a steeper roll off slope which sounds better than a shallow roll off slope.

well, if you have properly designed speakers you won't hear a difference between xovers of various slopes. in fact 1st order xovers (shallow slope) have some advantages over steep slope xovers, like perfect transient reproduction and in-phase operation of drivers.

anyway, if you have well aligned drivers, so that all frequencies arrive at your listening point in the same moment and in phase you will not be able to distinguish between different xovers. mind you most speakers are not time coherent designs, therefore you might get an impression that a steeper slopes sounds better. and that's because smaller portion of frequencies interfere with each other. but that doesn't mean steeper xover sounds better. it only means you've only heard poorly designed speakers with shallow xover slopes before.

 

[steve_1979]

If you want to be pedantic, yes I should have used the phrase "op-amp based".

You are still only looking at about 0.001% distortion when using an op-amp based active crossover compared to about 1% distortion with a passive crossover.

 

[steve_1979]

2. The passive crossover adds a considerable abount of audiable distortion but an active crossover does not.

sigh... go to page 3 in this tread and read my post #7 if you still haven't done so. it's about quoted harmonic distortion from new KEFs and Event Opals...

Please read this carefully to understand why passive crossovers have more distortion than an active crossovers: [LINK REMOVED - house rules]

 

[jackocleebrown]

Hello All,

I thought that it might be useful to add some information to this discussion.

Active loudspeakers generally, if not DSP controlled, will have crossover filters which are based on op-amps. Modern op-amps are extremely linear. I believe that, with the very best op-amps, you are talking even lower figures that Steve_1979 has quoted (you can easily find datasheets from TI and AD which will give you figures for achievable THD) However, the distortion from passive components, mentioned above, is hugely overstated. For example, if you look in Colloms "High Performance Loudspeakers" he presents some measurements of inductor distortion (the worst source of distortion in a typical passive crossover). There is quite a variation in the performance, but with the right choice of component, low levels of distortion are achievable.

In the overwhelming majority of loudspeakers, active or passive, the major source of non-linear distortion (e.g. harmonic distortion and intermodulation distortion) is due to the drivers themselves. The measurements of harmonic distortion which are published with KEF products are full system measurements including the drivers and the crossover.

The passive crossover does have an effect on the interaction between the amplifier and the loudspeaker but, with modern low output impedance amplifiers, the effect is deterministic. In other words we take this into account when we design the loudspeakers: the drivers and crossover are designed to work together to give the desired output when connected to the amplifier.

Active or passive is just one choice among a plethora of design decisions. There are plenty of other areas where the designer can get it right or get it wrong.

All the best, Jack.

 

[the record spot]

Hello All,

I thought that it might be useful to add some information to this discussion.

Active loudspeakers generally, if not DSP controlled, will have crossover filters which are based on op-amps. Modern op-amps are extremely linear. I believe that, with the very best op-amps, you are talking even lower figures that Steve_1979 has quoted (you can easily find datasheets from TI and AD which will give you figures for achievable THD) However, the distortion from passive components, mentioned above, is hugely overstated. For example, if you look in Colloms "High Performance Loudspeakers" he presents some measurements of inductor distortion (the worst source of distortion in a typical passive crossover). There is quite a variation in the performance, but with the right choice of component, low levels of distortion are achievable.

In the overwhelming majority of loudspeakers, active or passive, the major source of non-linear distortion (e.g. harmonic distortion and intermodulation distortion) is due to the drivers themselves. The measurements of harmonic distortion which are published with KEF products are full system measurements including the drivers and the crossover.

The passive crossover does have an effect on the interaction between the amplifier and the loudspeaker but, with modern low output impedance amplifiers, the effect is deterministic. In other words we take this into account when we design the loudspeakers: the drivers and crossover are designed to work together to give the desired output when connected to the amplifier.

Active or passive is just one choice among a plethora of design decisions. There are plenty of other areas where the designer can get it right or get it wrong.

All the best, Jack.

Welcome to the Forum Jack...nice to have a speaker designer contributing. Voices from the coal face as it were.

 

[steve_1979]

3. With an active crossover it's possible to use a steeper roll off slope which sounds better than a shallow roll off slope.

When using a 6dB/octave (1st order) crossover it will only be 12dB quieter 2 octaves below the crossover point. This means that the tweeter will still be audible and thus will be distorting because it's at too low a frequency. A 2nd order crossover is better but will still suffer from the same problems.

With an active crossover a 4th or 8th order filter is possible allowing the drivers to reproduce only the frequencies at which they work best. This means that they will distort less and give better clarity.

 

[BenLaw]

If you want to be pedantic, yes I should have used the phrase "op-amp based".

An electronic crossover is just an op-amp

I've read in academic hifi engineering and speaker design books that an active crossover is just an op-amp. Ashley James has also said the same thing several times on the AVI forum.

Quite an important difference tho? Not mere pedantry. If people are going to give technical 'facts' then surely accuracy is key? Otherwise better just to express one's opinion based on experience.

I've responded to a few of your posts recently Steve, and none of it is personal. I am also a fan of actives, having listened to ATC 100As quite a bit. I'm just finding the recent 'active' proselytising very tiresome, and frankly think it works against the aims of the authors.

 

[steve_1979]

sigh... go to page 3 in this tread and read my post #7 if you still haven't done so. it's about quoted harmonic distortion from new KEFs and Event Opals...

It's futile trying to compare the measurements of the Kefs to the Opals as both have been measured in isolation. For the measurements to mean anything they would've had to be measured at the same place using the same equipment.

 

[oldric_naubhoff]

In the overwhelming majority of loudspeakers, active or passive, the major source of non-linear distortion (e.g. harmonic distortion and intermodulation distortion) is due to the drivers themselves. The measurements of harmonic distortion which are published with KEF products are full system measurements including the drivers and the crossover.

Active or passive is just one choice among a plethora of design decisions. There are plenty of other areas where the designer can get it right or get it wrong.

:clap:

thanks Jack.

that's basically what I was trying to convey over here for some time. it's nice to have backing from a guy from the industry.

 

[oldric_naubhoff]

2. The passive crossover adds a considerable abount of audiable distortion but an active crossover does not.

sigh... go to page 3 in this tread and read my post #7 if you still haven't done so. it's about quoted harmonic distortion from new KEFs and Event Opals...

Please read this carefully to understand why passive crossovers have more distortion than an active crossovers: [LINK REMOVED - house rules]

I actually did. a few weeks ago. the guy is using the worst inductors possible. that's why he's getting poor results. why don't you ask him to put an air core ribbon coil instead of those he used and then compare the results. you'd be surprised. but of course he's not going to do that because it's going to ruin he's marketing argumentation.

 

[oldric_naubhoff]

sigh... go to page 3 in this tread and read my post #7 if you still haven't done so. it's about quoted harmonic distortion from new KEFs and Event Opals...

It's futile trying to compare the measurements of the Kefs to the Opals as both have been measured in isolation. For the measurements to mean anything they would've had to be measured at the same place using the same equipment.

well, this is just lovely!

when you don't have any more arguments left why don't you just suggest that measurements are inconsistent. well Steve, I'm pretty sure that both tests were performed on calibrated, specialised equipment. you could argue the way you do if one of those test was made in your bed room, through some cheepo mic and of-the-shelf soft. but that didn't happen, did it?

congratulation Steve!

 

[oldric_naubhoff]

3. With an active crossover it's possible to use a steeper roll off slope which sounds better than a shallow roll off slope.

When using a 6dB/octave (1st order) crossover it will only be 12dB quieter 2 octaves below the crossover point. This means that the tweeter will still be audible and thus will be distorting because it's at too low a frequency. A 2nd order crossover is better but will still suffer from the same problems.

With an active crossover a 4th or 8th order filter is possible allowing the drivers to reproduce only the frequencies at which they work best. This means that they will distort less and give better clarity.

that's why you've got drivers and drivers. you need quality, linear drivers for 1st order xover. EDIT: and you need to know what your drivers are capable of. when you set the xover point with caution you can't be wrong. you don't need to go all the way down to freq response threshold of the driver to set the xover point up.

for instance Dynaudio is using exclusively 1st order in their passive speakers. and they cut pretty low too, below 2kHz mark. do you think they get it wrong?

also there are B&W, Vandersteen, Thiel, Analysis Audio and many more who use 1st order xover. I guess they too know what they are doing.

 

[Frank Harvey]

Has anyone got any questions for Jack? :?

 

[steve_1979]

sigh... go to page 3 in this tread and read my post #7 if you still haven't done so. it's about quoted harmonic distortion from new KEFs and Event Opals...

It's futile trying to compare the measurements of the Kefs to the Opals as both have been measured in isolation. For the measurements to mean anything they would've had to be measured at the same place using the same equipment.

well, this is just lovely!

when you don't have any more arguments left why don't you just suggest that measurements are inconsistent. well Steve, I'm pretty sure that both tests were performed on calibrated, specialised equipment. you could argue the way you do if one of those test was made in your bed room, through some cheepo mic and of-the-shelf soft. but that didn't happen, did it?

congratulation Steve!

The speakers were measured in two different rooms with different acoustics, under different condititions using two different pieces of equiptment.

Even if you measured two identical speakers under these conditions the results would differ.

 

[steve_1979]

for instance Dynaudio is using exclusively 1st order in their passive speakers. and they cut pretty low too, below 2kHz mark. do you think they get it wrong?

They're limited to using either 1st or 2nd order crossovers with passive speakers but their active speakers use a 4th order crossover which is better because they have a steeper filter slope.

 

[Bobbyhifi]

Wow, this got a long conversation started! :rofl:

Right, well I went and had my demo yesterday and got to listen to a lot of options! Without meaning to inflame any more argument the passive options were quickly dismissed compared to the actives, as much for practicality as for the fact I felt they sounded far "cleaner" and more natural in general. This is just my relatively uninformed or trained opinion.

With actives chosen as the desired route I can confidently say that all I listened to sounded different from each other. The ones that made my final shortlist were:

Adam A7x, Acoustic Energy AE22, Genelec 8030a, Yamaha HS80M.

Adam: Very good but large ports on the desktop seemed to make bass a little overblown at times, on stands they may well be better. Treble a little too sharp at times though great detail.

Acoustic Energy: Very good again but this time with far more natural bass, at first seemed too subtle compared to Adam's but on longer listen it's all there but in good proportion to everything else in the mix, only downside is that they're a little large but my desk is wide so not a massive problem.

Genelec: Again a nice speaker but for me seemed to struggle a little compared to the larger models, the strain made them sound less natural.

Yamaha: Good price but for me far too midrange forward. I found them uncomfortable on a longer listen because of this.

The outcome: For me the Acoustic Energy AE22's offered the best balance of everything and despite having quite a large "footprint" they will be fine for me in my room. The shop told me they were designed to be used on a mixing desk so this may have helped them in position on the desk! In short, next payday I'll be picking up my pair. I probably had a slight soft-spot for them already as since I've been back I've been really enjoying my AV system from the same company! :rofl:

With regards to pre-amp, this shouldn't be a problem as my soundcard has all this built-in I think - I will double check though!

Many thanks for all the great advice and debate, it's been an interesting read!

Rob.

 

[fr0g]

It would be interesting to know what amp/passive combinations you tried for comparison...

 

[Bobbyhifi]

It would be interesting to know what amp/passive combinations you tried for comparison...

Hi fr0g,

I tried some JBL Control 5's and the passive version of the Acoustic Energy AE22's. The other passive speakers were either very cheap or very expensive, those I tested were by no means bad but on both I felt the easy detail and bass definition especially was not as good as the active models. There was far less choice overall with the passives though. The amplifier was an ART one but I didn't see close enough as to what model, I think they said it was 140 watts so should have been plenty to power the speakers well.

Thanks,

Rob.

 

[fr0g]

It would be interesting to know what amp/passive combinations you tried for comparison...

Hi fr0g,

I tried some JBL Control 5's and the passive version of the Acoustic Energy AE22's. The other passive speakers were either very cheap or very expensive, those I tested were by no means bad but on both I felt the easy detail and bass definition especially was not as good as the active models. There was far less choice overall with the passives though. The amplifier was an ART one but I didn't see close enough as to what model, I think they said it was 140 watts so should have been plenty to power the speakers well.

Thanks,

Rob.

Ok, thanks for the reply. But it would be interesting to know which passive speakers at a similar price point you tried. The JBLs are pretty cheap monitors compared to those AE.

The ones you say were very expensive, what were they, and did the AE22 sound better? Or was it a case of the expensive passive ones being better, but not worth the difference?

 

[Frank Harvey]

After starting this thread regarding the KEF's Rob, I find it odd that you're not going to try them!

 

[steve_1979]

Wow, this got a long conversation started! :rofl:

I apologise for participating in the somewhat derailing of your thread.

I'm glad to hear that you've found some speakers that you like. I considered the AE22's myself, they're great sounding and fantastic value speakers.

 

[alchemist 1]

After starting this thread regarding the KEF's Rob, I find it odd that you're not going to try them!

Yes, quite agree. Did i miss something ? :?