SpeakerTalk

[SaSi]

I just added the KUBE to the amplification chain (removing the quick and dirty solution of the parametric equalizer I used) and it makes a huge difference. But it really strains the amplifier (2x110W/4Ohms) if I crank up the volume.

Of course I could look into finding a more powerful amplifier, but that's too simple.

I'm considering using multiple amplifiers to actively drive the drivers (4 per enclosure).

I had a look at the crossover schematic (to try and decipher how to configure filters) but I don't know how to handle the "conjugation" parts that make the loudspeaker a resistive 4Ohm load. Do I ignore these? Will it matter?

As I see it, apart from the slopes at the crossover frequencies, there's a notch filter at the ~180Hz peak generated by the 4th order bandpass enclosure configuration that obviously needs to be part of the active crossover.

Is that all there is to it?

[speakerguru]

[SaSi]

So you are saying, tri-amp, in effect, by splitting the crossover and drive each section alone, and avoid the active crossover stage.

That's an idea, however, if IIRC, I've read that the R107 main crosover includes a small section that affects the midrange. I could be wrong, and remember wrong that the KUBE slopes extend to the midrange affecting the B110.

I have traced the main crossover and have the schematic somewhere. Will look it up and check it out.

[audiolabtower]

I think it depended on the exact series. Some Kubes equalised the 107 bass only, so could be left out of mid and treble sections if tri-amped, but others did the shaping through the mid and higher to get maximum efficiency and volume out of the system, so would need something in front of the mid and treble amp to give the same shelving effect, otherwise the output will not be flat (although some of the cheaper magazines said the sound was better without the kube, undoubtedly because the "reviewers" had no idea about natural sound and liked some kind of JBL peaky impact).

[SaSi]

If the KUBE is extending it's slopes to the midrange, it can be set before the bass and midrange power amps and take care of both ranges, leaving the treble alone to feed directly from the preamp to the power amp, and in effect (perhaps small) reduce IMD perhaps caused by the equalization in the bass affecting the tweeters.

The easy thing to try first is unhook the heads from the main crossover and drive them directly from a different amp pair and use my ears as a first test.

Something should be wrong there, as checking the main crossover schematic shows two 100uF in parallel running the signal to the mid/high plug with a 6,4mH inductor across. So there's a second order high pass filter in the main crossover tuned to 140Hz where there's more space for the big inductor and capacitors.

[speakerguru]

[audiolabtower]

[audiolabtower]

By the way, here is an example of the Kube shaping from Mk1
http://www.stereophile.com/content/kef-r107-loudspeaker-1987-measurements#rGwm0Ajm3vlQjdpc.97
as you can see without the Kube the high crossover will be peaky and treble too strong.

For the Mk2
http://www.stereophile.com/content/kef-r107-loudspeaker-1991-measurements#fwXDcvWPKLl97wkc.97

I would be inclined to stick with the Kube, and if you reckon you can hear hiss or something try more modern opamps, some of which have spectacular performance, just being careful to limit bandwidth on high spec ones to avoid radio freq oscillation.

[SaSi]

Thanks for the KUBE curves. I have mk1.

I had that issue with bright treble before adding the KUBE. The parametric equalizer had been set for -3dB beyond 5kHz and that tamed it enough. Thought it was something wrong with the tweeters, like dried out ferrofluid, but then I thought that should make them sound dull instead.

I don't intend to remove the KUBE from the equation. Since it extends its influence until the high end of the response, it obviously needs to drive all three amps.

One thing that needs to be done to the low pass crossover is to isolate the first part of the highpass filter and drive it separately. That leads to a succesful bi-amping (low and mid-high separated).

[SaSi]

It has been some time since I posted this and I thought I should bring in an update.

I finally decided to go the easy way and get a more powerful amplifier. It was not really a plan, but when I stumbled across a dbx BX-3 MkII, being able to deliver more than 400WRMS/4Ohm, for $350, I knew I should not pass.

Does it make a diference? Well, yes. How much? Let me give you an analogy. If you drive in traffic, a 100HP car or a 300HP car doesn't make much difference, except for a WOW effect.

But if you go to an open road, pressing the throttle does make a difference you can feel. Something like that.

In low volumes, not really much difference. Not even anything better in the low end, where I was expecting a more powerful amplifier to be able to deliver current that a less powerful amp would starve (thinking of the KUBE equalization making high demands for current below 100Hz.).

But as the volume is increased, it gives a feeling that the dynamic range is increased as well. And distortion I thought was coming from the loudspeakers is practically gone. You can turn up the volume and listen to really high sound levels without realizing it.

I had the same happen to me when I got a pair of Sennheiser HD650 headphones. They have so little distortion and can handle so loud music that I found myself listening to very loud music without realizing it, since the hint for loudness used to be distortion.

I really need to fulfill my dream of going active with a decent loudspeaker. Since the 107s are going to stay as they are and are used in the permanent primary set, I am now considering my 105 MkI. They are long overdue for a crossover rebuilt and I'm now contemplating on what modifications I need to do both recap the crossovers and also make them switchable between passive and active.

[Mind_Drive]

Hi,
That sounds great- active 105s! I'm certain they would sound even better driven this way. You'd also be able extract better low-end depth by not having the same LF cutoff, which I'm sure was mainly designed into the bass crossover to get rid of turntable rumble. Not that they don't go quite deep already....

Hopefully my reply will show that there are still some people out there, reading this forum. Activity seems to have dried up of late.

[audiolabtower]

Actually any rumble filtering was a secondary effect. The purpose of the capacitor was to interact with the inductance of the coil and other components to achieve the efficiency of a reflex design (maybe 3dB higher than the same closed box) without the 4th order slope after cutoff. So taking it out of circuit (which some misinformed people used to recommend) was not a good idea, and it complicates how to do an active filter for the same result.

[Mind_Drive]

Thank you for correcting my misunderstanding! You've just shown why this forum is so great.
The efficiency of a reflex from a sealed box- best of both worlds (I cannot get on with reflex/ported designs, and mourn the near extinction of the sealed box speaker in the modern hifi scene).

[audiolabtower]