Hello,
I was reading your post and I am, too, fixing a pair of 105.I that used to belong to my father.
I recently had the woofer and tweeter of one of ‘em fixed but the diffuser sounds remarkably muffled and softer than the other one (seems a lack of power in mids and lows).
I was wondering whether it could depend on the crossover and whether you had any success with your replacements.
thank you
I was reading your post and I am, too, fixing a pair of 105.I that used to belong to my father.
I recently had the woofer and tweeter of one of ‘em fixed but the diffuser sounds remarkably muffled and softer than the other one (seems a lack of power in mids and lows).
I was wondering whether it could depend on the crossover and whether you had any success with your replacements.
thank you
Hi, sorry can you be more specific, why you say woofer do you mean the 12" b300 bass driver? And when you say diffuser, do you mean the b110 midrange driver? I am interested to know what about the woofer and tweeter was "fixed" in your pair, the 105(.1) has rubber surrounds and no ferrofluid so there isn't too much there to go wrong unless you blow them completely.
I identified my issues by playing isolated tones from a signal generator website, and compared one speaker to the other. "Luckily" the issue turned out to be obvious when I hit a band of mid range frequencies that were not sounded from one of the speakers at all.
I then identified the fault was with the crossover and not the driver by disassembling the back of the "heads" of the speakers, removing the cross over circuit and directly hooking up my amp to the midrange driver and playing that same mid range tone through it again. This time it did sound, and at a similar level to the other speaker.
It was then that I noticed that the crossover had been quite heavily modified from the original design and instead of trying to correct this I just bought the parts kit for the 105 from Falcon Acoustics and just completely restores the cross over to it original design (see schematic on first page).
Obviously the modified design "worked" for one of the two speakers and I put the failure to sound down to some bad soldering on the other. I can't say whether the sound quality has improved because I am not longer using the modified design or because both speakers are now firing on all drivers whereas one of the mid ranges was not before... but I can say for sure that the end result is a much improved sound quality.
I identified my issues by playing isolated tones from a signal generator website, and compared one speaker to the other. "Luckily" the issue turned out to be obvious when I hit a band of mid range frequencies that were not sounded from one of the speakers at all.
I then identified the fault was with the crossover and not the driver by disassembling the back of the "heads" of the speakers, removing the cross over circuit and directly hooking up my amp to the midrange driver and playing that same mid range tone through it again. This time it did sound, and at a similar level to the other speaker.
It was then that I noticed that the crossover had been quite heavily modified from the original design and instead of trying to correct this I just bought the parts kit for the 105 from Falcon Acoustics and just completely restores the cross over to it original design (see schematic on first page).
Obviously the modified design "worked" for one of the two speakers and I put the failure to sound down to some bad soldering on the other. I can't say whether the sound quality has improved because I am not longer using the modified design or because both speakers are now firing on all drivers whereas one of the mid ranges was not before... but I can say for sure that the end result is a much improved sound quality.
Hi,
with diffuser I meant one of the two full cabs (hence with 3 speakers each - bass, mid and tweeter).
the 105s had not been used gor about 30 years. especially one of the b300 had probably received a frontal hit (recessed dustcap) so the coil has been repaired.
next to that, two of the tweeters were malfunctioning (have found that the ones in this model were notoriously delicate - one of mine had, as often seemed to happen, loose coil wires)
would you have the link to that site with the tones..?
thanks in advance
with diffuser I meant one of the two full cabs (hence with 3 speakers each - bass, mid and tweeter).
the 105s had not been used gor about 30 years. especially one of the b300 had probably received a frontal hit (recessed dustcap) so the coil has been repaired.
next to that, two of the tweeters were malfunctioning (have found that the ones in this model were notoriously delicate - one of mine had, as often seemed to happen, loose coil wires)
would you have the link to that site with the tones..?
thanks in advance
https://onlinetonegenerator.com/
It's not super scientific by any means, luckily the issue with my speakers manifested in a very obvious way.
If I were you, I'd put the two speakers physically next to each in the middle of the room, so you can minimise the impact of the room, and play something fairly full range through them and try and listen out for specific differences, also around with swapping which channel of the amp each are connect to.
Removing the back off the tops isn't too difficult although can be fiddly, you could take out a look at the crossover and make sure that there isn't anything very obviously wrong with the components and solder joints. If joints are dry and cracked or caps are bulging then you've got a contender for an issue.
After that I am not sure how much more help I can be over the internet, I don't want to suggest you fork out 100quid for a new capacitor set if there is no need to replacement.
Hi,
I find fairly suspicious the behavior of either the midrange or the ‘repaired’ b300.
I’m fairly familiar with the innards of the ‘heads’, tool em apart several times.
and, yes, did swap the channels on the amp and, as a matter of fact, the heads themselves too (from one base to the other).
if it comes down to poor b300 performance I am suspicious of some cap in the low range crossover sibfilter (the one located just under each of the 300s).
there are four caps on that, they wouldn’t cost much less than 40 quid for decent quality ones (5%..?) so … the kit is not awfully priced
thanks
I find fairly suspicious the behavior of either the midrange or the ‘repaired’ b300.
I’m fairly familiar with the innards of the ‘heads’, tool em apart several times.
and, yes, did swap the channels on the amp and, as a matter of fact, the heads themselves too (from one base to the other).
if it comes down to poor b300 performance I am suspicious of some cap in the low range crossover sibfilter (the one located just under each of the 300s).
there are four caps on that, they wouldn’t cost much less than 40 quid for decent quality ones (5%..?) so … the kit is not awfully priced
thanks
kit comes with the caps for the bass crossover.
It's high effort, not as high effort as recapping the bass crossover, but you could swap the B300 between the two cabs + crossover and see whether the muffled sound "follows" the driver or cab. Would also be worth a quick visual inspect and see if all of the solder joints and caps look in decent shape.
That'd narrow down whether it's the crossover that needs attention or you need to get your b300 looked at.
It'd worth doing. When they're up and running they sound fantastic.
It's high effort, not as high effort as recapping the bass crossover, but you could swap the B300 between the two cabs + crossover and see whether the muffled sound "follows" the driver or cab. Would also be worth a quick visual inspect and see if all of the solder joints and caps look in decent shape.
That'd narrow down whether it's the crossover that needs attention or you need to get your b300 looked at.
It'd worth doing. When they're up and running they sound fantastic.
I did a recap of my pair of first series 105 about two years ago with the Falcon kit. Afterwards I did not really find the sound balance right.
My pair before was sounding OK, I just thought it might sound better when recapped and actually it sounded worse, fresher but not with the right balance. To trebly.
So I hand-sorted all the caps by doing some long-term trials. I ended up not using the polyester cap at the Twitter filter input, for example. I used a Mundorf plain bipolar.
To my experience, also with other vintage KEF crossovers, it is a delicate job to substitute the original Alcaps — Falcon has found substitutes that do not always work in the same way as the originals did.
KEF obviously very carefully tuned the crossovers, and the type of caps, LL (low loss aka plain) or ordinary, their maximum voltage rating, the ESR and I think also just the type of electrolyte used do make a difference.
So it became somewhat of a nightmare to get the KEFs back to sounding right particularly problematic was the parallel 10uF cap in the midrange filter. This needed to be an original blue non-LL Alcap, 100v I think.
In my equally restored Calindas, Falcon set again, I added a small value resistor in series with the parallel cap of the bass circuit. Obviously the new cap has a too low ESR. Before the speaker sounded bright, and not really locked into the right balance just adding this resistor brought everything back to life.
My pair before was sounding OK, I just thought it might sound better when recapped and actually it sounded worse, fresher but not with the right balance. To trebly.
So I hand-sorted all the caps by doing some long-term trials. I ended up not using the polyester cap at the Twitter filter input, for example. I used a Mundorf plain bipolar.
To my experience, also with other vintage KEF crossovers, it is a delicate job to substitute the original Alcaps — Falcon has found substitutes that do not always work in the same way as the originals did.
KEF obviously very carefully tuned the crossovers, and the type of caps, LL (low loss aka plain) or ordinary, their maximum voltage rating, the ESR and I think also just the type of electrolyte used do make a difference.
So it became somewhat of a nightmare to get the KEFs back to sounding right particularly problematic was the parallel 10uF cap in the midrange filter. This needed to be an original blue non-LL Alcap, 100v I think.
In my equally restored Calindas, Falcon set again, I added a small value resistor in series with the parallel cap of the bass circuit. Obviously the new cap has a too low ESR. Before the speaker sounded bright, and not really locked into the right balance just adding this resistor brought everything back to life.
Yes,
that’s correct. as a matter of fact I think they’d match the components of every single unit (at least for the reference series) and did so with their own specific equipment too, so … most drop-in replacements might yield different results
that’s correct. as a matter of fact I think they’d match the components of every single unit (at least for the reference series) and did so with their own specific equipment too, so … most drop-in replacements might yield different results
It is not only about cap values, about exactly matching the flat frequency response by for example adding a small value capacitor on top of another one, as I have seen in factory made KEF crossovers.
What I was talking about is that the ALCAP and ELCAP capacitors KEF used in the past have brand- specific properties which make it near to impossible to replace them with something broadly similar, even if the replacement is very high-quality.
It is not only about quality, it is about replicating the electrical properties of the original component as closely as possible.
Therefore rebuilding a crossover means finetune the selection of every capacitor, at least of those in critical positions, like the cap at the input of the tweeter filter and the parallel cap in the midrange filter for three-way speakers — or in the bass filter for two ways.
That is at least my experience with vintage KEFs.
What I was talking about is that the ALCAP and ELCAP capacitors KEF used in the past have brand- specific properties which make it near to impossible to replace them with something broadly similar, even if the replacement is very high-quality.
It is not only about quality, it is about replicating the electrical properties of the original component as closely as possible.
Therefore rebuilding a crossover means finetune the selection of every capacitor, at least of those in critical positions, like the cap at the input of the tweeter filter and the parallel cap in the midrange filter for three-way speakers — or in the bass filter for two ways.
That is at least my experience with vintage KEFs.
Yes,
in fact I remember, as a kid, a ‘technician’ swapping two of the caps in the highs or mids filters of these units (quote: ‘listen to these, these are japanese, not that british crap’ …) and they’re still there and, as a matter of fact, I find them too bright.
thanks
in fact I remember, as a kid, a ‘technician’ swapping two of the caps in the highs or mids filters of these units (quote: ‘listen to these, these are japanese, not that british crap’ …) and they’re still there and, as a matter of fact, I find them too bright.
thanks
It’s interesting that KEF uses a resistor (R1) to burn off sensitivity in the woofer. Normally, the woofer provides precious bass (with baffle step loss etc) - the limiting factor for setting SPL and then all other drivers are padded down to match the woofer. I don’t see any padding on the mid or tweeter which suggests the woofer was really sensitive and more so than the mid or tweeter. Very odd. I am wondering if the speaker sounds bright because of this bass burner approach?
I agree that factory XO’s by a reputable company like KEF shouldn’t be messed with unless you really know what you are doing and if it’s not right because of someone’s over-confident XO commando exploits, the course of action is to return to base.
I recently got this meter and it works great for all XO parts. Very economical. I have a fancier one that has more resolution and measures ESR, phase and allows you to set test freq from 100Hz, 1kHz, 10kHz, 100kHz. But find this simple one does what I need 95% of the time.
I agree that factory XO’s by a reputable company like KEF shouldn’t be messed with unless you really know what you are doing and if it’s not right because of someone’s over-confident XO commando exploits, the course of action is to return to base.
I recently got this meter and it works great for all XO parts. Very economical. I have a fancier one that has more resolution and measures ESR, phase and allows you to set test freq from 100Hz, 1kHz, 10kHz, 100kHz. But find this simple one does what I need 95% of the time.
@luomotigre 's particular speakers sound bright because somebody messed with the tweeter filters.
The speakers, KEF Reference 105.1, as original do not sound bright. They sound right 🙂
R1, a parallel resistor, in my opinion does not have the function to lower the sensitivity of the bass circuit.
Maybe someone deep in the matter like @AllenB would like to enlighten us on that? 🙂
The speakers, KEF Reference 105.1, as original do not sound bright. They sound right 🙂
R1, a parallel resistor, in my opinion does not have the function to lower the sensitivity of the bass circuit.
Maybe someone deep in the matter like @AllenB would like to enlighten us on that? 🙂
Good to hear that they sound right with R1. Anytime I see a direct resistor to GND - that is burning off SPL. The ones with a capacitor in series are different as they are tuned to a particular frequency.
I've mentioned before that KEF is known to strive for flat impedance looking into their speaker terminals. So they "burn" power to achieve that. Their theory is that will result in an easier to drive speaker, which a wide range of amplifiers will tolerate well. They know audiophiles, and all customers are going to be using something different for amplification. They want to avoid someone saying "SLS" with my amplifier...
This theory is reflected in FR speakers, which have a flat impedance except at resonance. If I'm correct, those too present an easier to drive load for the amplifier. That's a tradeoff some are willing to make, to get that benefit. Or, not only tune the crossover for the speakers, tune it for the amplifier's task to drive as well.
This theory is reflected in FR speakers, which have a flat impedance except at resonance. If I'm correct, those too present an easier to drive load for the amplifier. That's a tradeoff some are willing to make, to get that benefit. Or, not only tune the crossover for the speakers, tune it for the amplifier's task to drive as well.
I think it is a low loss/ plain bipolar. Like Mundorf plain. Need to find the documentation (or unscrew the back of the assembly…) to be sure.
This is correct, it's not in a position to affect the woofer as a whole and vary it's sensitivity. In this position it can affect the damping of resonant parts of the circuit. There may be other ways to achieve the same result so it's not immediately clear why they used this resistor.