SpeakerTalk

[iso]

101 and CS1 model crossovers seems to be quite similar. Most visible difference is 400uf series capacitor and different eq for B110. Does SG or David remember if series cap was used only as protection device for DC or does it has something to do with LF alignment.

It would be also nice to know why B110 eq in low pass section different for ea model even CS1 was based on 101 model.

Best Regards

Kimmo

[audiolabtower]

Guru will know for certain, but the 101 came from the era of using the series bass capacitor as a function of the crossover alignment to eg increase the efficiency compared to a normal closed box, the dc protection was a useful side effect. I remember at the time some people used to short out the cap because single supply amplifiers used to have a large series cap in the output and "all capacitors are therefore bad"! However doing this will not give the flat response Kef originally designed, and I prefer to believe the people who had the expertise and equipment to know what they were doing.''

[speakerguru]

I've just sent a pdf of KEFTOPICS Vol 3 No 1 Model 105 to Terry. Doubtless he will post it on this site in due course. It partly explains the series capacitor 3rd order high pass alignment used in a lot of KEF designs of that time.

I thought I had answered this one before, but a search has come up blank. I'll elaborate on the technique when I get more time.

[speakerguru]

Extract from KEFTOPICS on Model 105 (pending website upload)

“The method of achieving efficient transfer of power from the amplifier to the low-frequency drive unit of the Model 105 is a further example of the KEF system-engineering approach, in which the unit’s enclosures and electrical networks are all developed together. A new bass loading technique, involving the interaction between the electrical, mechanical and acoustic characteristics of the system, has been devised.

Normally, the rise in electrical impedance in the region of the low-frequency resonance reduces the current drawn by the drive unit from the amplifier, and as a result, the inherently high efficiency of the unit cannot be realised in this part of the frequency range. By re-matching the drive unit via a special electrical network, a better transfer of power has been achieved. This technique makes it possible to use a closed box (infinite baffle) without the loss of efficiency usually associated with this type of enclosure, and to extend the frequency response down to 38 Hz with ample power handling capacity on programme while avoiding subsonic disturbances arising, for example, from record warp or amplifier d.c. offset.”

[speakerguru]

The above is a bit cryptic as KEF, at that time, while describing their novel and advantageous technique, did not want to give away too much to competitors. How exactly was this “re-matching” done?

When designing a three or four way system with a low crossover frequency between midrange and woofer, it is well known that the series inductor in the woofer low pass filter can react with the capacitor which represents the driver mass in the woofer equivalent circuit. This creates a series resonant circuit which reduces the impedance thereby taking more current from the amplifier. This produces a peak in the response just above the driver-box resonance. Normally unwanted, this is avoided by raising the crossover frequency or conjugating the capacitative part of the woofer impedance at resonance.

Could this effect be used to advantage? What happens when there is a series capacitor in the woofer crossover? In a similar way, the capacitor can react with the inductor in the woofer equivalent circuit representing the box/driver stiffness to produce a series resonant circuit below the driver-box resonance. Again, this produces a rise in response by lowering the impedance seen by the amplifier. If the driver, box and series capacitor are correctly aligned, the result is a decrease in the system LF cut-off frequency at the expense of a change from 2nd order to 3rd order cut-off slope.

[iso]

101 seems to have close to closed box 12dB/oct cut off slope below resonance. If I did understand it correctly, loading used in 105 will produce 18dB/oct cut off slope below resonance?

Best Regards

Kimmo

[speakerguru]

Yes 3rd order is 18dB/Oct.

[iso]

You can find 101 response here http://kef.com/uploads/files/en/museum_pdf/70s/Reference_Series_Model_101.pdf

101 LF response is very similar to CS1 here http://kef.com/uploads/files/en/museum_pdf/80s/Constructor_%20Series_Model_CS1.pdf

101 brochure talks about tonal system design but says very little exactly, like you say a bit cryptic. This made me place my original question about 400uf series capacitor and different eq (different series chokes) for B110.

Best Regard

Kimmo

PS marketing material has pissed me off for long time. Today it seems that there is usually not too much useful information in brochures apart color options. Even then you can be pretty sure that walnut means some other dark wood or even walnut grained vinyl. Things seems to get worse every day in this respect... but term like "tonal system design" was not too useful for average customer even in the 70´s, as broadest band with with optimum efficiency is most likely goal of quite many speaker design.

[speakerguru]

I would not believe either of those curves below 100Hz. They were not meant to mislead, but it was very difficult and time consuming to make accurate measurements at LF. With limited time available, those would probably have been considered good enough for Marketing leaflets at that time.

Re your OP: the final LF cut-off slope of a closed box is 12dB/Oct, discounting leakage errors. If you add a series capacitor to the LF XO then the cut-off slope increases to 18dB/Oct.