Linn Kan story, I loaned a friend a pair of Linn Kan MK1 speakers. He decided to emigrate and asked if I wanted them back. I went and collected them and found one of the drivers had been damaged beyond repair. I looked on the the internet and after a few months the only option was to go with a retro make. Anyway found a single driver dating back to 1979. Could not resist, the temptation was to much. Got it last week and installed it today. Did not expect them to sound great, but was I wrong, a big YES!!!.
Installed new driver and connected it up to my trusty Naim Nait 3, chord speaker cable and interconnects. Marantz cd 63 ki and a WiiM.
And it sounds amazing, has anyone ever had this experience. Just streaming Dido and it is so good. It just proves that the old equipment can sound good with new kit.
Thank you for reading.
Sitting here listening to Natale Merchant MTV Unplugged and Dido Acoustic with my CB and Linn Kan’s. Life is good.
I sent my crossovers to England for service and I’m curious to see whether they will get any better, they run with a Nait 1 and Bluesound Node 2i and I was very satisfied 
Hi have you got any pictures of your x overs … both front and back…cheers
A very nice and simple crossover.
DG…
My x overs is already finished, I’m very curious whether it makes a difference
When Linn made Speakers. 
That’s considerably simplified from the original Mk1 Kan crossover which has a CR across the B110 of 10uF and 10R, plus a 24 or 27ohm resistor to ground in parallel with the 0.25mH tweeter inductor.
If those were my Kans - and that Hiquphon tweeter plus the SP1057 version of the B110 are the drive units to have - I would wait for a pair of earlier crossovers to come up on eBay with the blue Alcap capacitors and Ferrite cored inductors. The changeover to iron dust cores was a hugely retrograde step as while they saturate at a higher level (not one you can generally achieve on Kans without them squealing in pain for other reasons) their distortion is substantially higher at lower level. (They kinda hold the same level of distortion whatever the current going through them.) I did a significant amount of work on this subject almost 20 years ago and it later came up in a discussion with Roy George who borrowed some large iron dust inductors from me to measure himself. On probably better kit than I had at the time (Naim had an AP Cascade vs. my CLIO system from Audiomatica - not that CLIO isn’t great but it was less than a tenth of the price of the then SOTA AP machine) he came to the same conclusion. I think he’d probably also agree that those blue Alcaps are very underrated and are in fact one of the best sounding capacitors that one can buy. They even outperform some film caps because they have a substantially lower inductance and this is noticeable within the audio band; well before you reach 20kHz.
Christian did you measure them with your CLIO and get any Spectral decay measures?
I have a pair I’d like to measure and have a crack at improving the crossovers - and if they measure well enough just replicate the crossover with aid cored inductors and good PP Film caps and wire wound resistors.
Very much in line with Danny Richie’s approach from GR-Research
Hi Steve, Are you still talking about measuring the inductors? The distortion and spectral decay measurements are derived separately, with the distortion being done with a sinusoidal signal and the waterfall plots coming from a MLSA chirp. One could do both, though the waterfall is really just a function of the frequency response so will vary depending on what circuit the inductor is placed in. By all means send me a PM and I’ll give you my number. I’d be happy to help.
I’d be interested if one can improve the Kans. One has to remember that they’re really a complete fluke, so one has to be wary of killing the magic. For instance, the DCR of that main inductor will probably be important (It’s about 0.25 ohm from memory). You’ve also got a bit of luck on your side as the main capacitor values are all quite close to each other at 4.5uF, 4.7uF and 5uF, so it’s conceiveable you could get all six caps out of a small batch of 4.7uFs. The more you buy, the more likely you are to get close to the values. If you’re good at statistics, you could work out how much you’re prepared to pay versus how accurate you want it. 
The capacitors are probably a good place to start. The one I’d start with is the main C to ground on the bass circuit, 4.7u, then after that you can have a go with the two tweeter ones.
No I was talking about spectral decay of the overall speaker and the frequency response measures with Clio. I’d be surprised if small variations in measured component values made much difference.
I was thinking more in terms of the smearing effect of electrolytic caps and non air cored inductors
Sadly you won’t really be able to see that in the waterfall plots. I know they are usually front and centre on all the marketing blurb but they are more for show than utility. From a technical angle, they are really just a derivation from the frequency response, so if you change the frequency response you will also change the waterfall. What happens is that if you have an issue somewhere, say a peak and a dip in the response somewhere, that part will extend for longer in the time slices and give you one of those ridges. In this respect they are not quite what they appear to be. For instance, if you had a perfectly flat response there would be no decay at all and all you would see is a single frequency response.
If you got the DCR of an air cored inductor the same as the ferrite one it’s replacing I doubt you’d see any change in response. Similarly with changing from electrolytic caps, but you will hear a difference. The ESR of electrolytics is overstated in my view and the ESL of film caps is almost ignored, even though that can show up in the frequency response. One place where films actually do more of what the circuit intended is that first capacitor to ground after the main inductor in the bass-mid circuit. Here the ESR puts a floor under how much the capacitor can roll off the response, so at some frequency it will literally stop working. A film will keep going further, doing more of the rolling off job. However, even that tends to be limited as the crossover ground is at the speaker end and you’ve got cable resistances back to the amplifier. What you then get is the crossover working to about 40dB down (which is pretty low down at 1%) and then flatlining there, never being able to get any further.
I think these changes are more in low level information than in frequency response changes. So from things like losses in the cores and additional 2nd and 3rd harmonic distortion from non linearities in the core’s permeability - especially if they approach saturation (where the whole value of the inductor will change). Speakers aren’t great on the distortion front anyway but ferrite cores do make them noticeably worse.
You will find the improvements worthwhile but doing a wholesale change won’t necessarily give you a speaker you like more. Also there are places where it’s hardly worth making a change, such as on that 10uF in series with 10R in parallel with the bass unit. First of all a 10uF PP is expensive and you’re getting rid of a tiny bit of ESR that might even be wanted, or is irrelevantly small in series with 10 ohms. Changing the resistor there, OTOH, to something that can take the power and doesn’t change its value with temperature, will probably give you quite a lot more than you expect. I really would do this piecemeal and listen in between, ideally with a bit of living with the changes before changing the next bit.
PS. One thing that really will improve the Kans (and should have no possible downside, unless they are even more witchcraft than I thought) is getting the crossovers out of the cabinet. You’ve got almost the same amount of noise going into the cabinet as is going into the room, so just getting them out of that environment should be a substantial improvement. In fact more goes into the cabinet than into the room because the radiation resistance is higher. If you were to put a microphone inside the cabinet you’d see a rising response at low frequencies, which will keep going up as the frequency goes down until you hit the driver’s LF resonance, at which point it will flatten off and stay at that level, probably down to DC, but for various amplifier rolloffs.
I’m relying on Wilmslow Audio on this one, hope they know what they’re doing to improve it. They said to me they are looking at whether it is even necessary to do a summary and write a few days later that they are finished, £120 for labor and parts.