KEF LS50 - crossover?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
How do you figgure stealing?


I am sure there is some one out there who moded their tweeter section and is willing to share the plan :)


The same thing the German did with the woofer.


E: And just out of curiosity: What difference does it make, if I open one and then copy the original plan as you suggested it?
 
Last edited:
I bought a pair of those drivers on ebay a while ago and finally wanted to get started (I thought the German guy uploaded the full schematic otherwise I woulnd't have bought them in the first place).


There is no way that my design will be better than Jack's. Especially since I haven't developed any crossovers in the past. But since I own these wonderful drivers over a year now, I am not going to sell them.


I am sure, that there is someone out there who has opened theirs and can help me out :)


E: And it still is DIY. I am going to build my own enclosure and add a woofer. It is going to be two way active (3 way is to expensive for me).
 
Last edited:
Are you sure they are LS50 drivers? They aren't quite the same as R100 drivers which look very similar. Since you are adding a woofer you won't need to baffle step correct the coaxial which changes everything anyway. The LS50 crossover is not the place to start. You are starting from scratch.

Audioexcite concetric

Have a look at Goran's design
 
Last edited:
Yeah I am pretty sure it is the LS50 (the guy I bought them from said he is 100% sure back when I bought it, because I was aware of the R100).


I chose the same width for the enclosure, so the baffle step is still relevant.


Do you happen to know how different the drivers are?
Thx for the link :)
 
Last edited:
You approach the crossover for a three way entirely differently from a two way otherwise you are just throwing efficiency away. The woofer in a three way usually takes care of the baffle step as the baffle step is within its passband. The mid (ie your coax woofer) remains unequalised in this region as its outside its passband. That's one of the major reasons for going three way.

Differences here KEF LS50
 
I was planning to use 200 Hz as a cut off frequency (that is what the German guy, who modified the crossover, recommended). My enclosure is 220 mm wide with a 40 mm 30° phase.


200 Hz is the value, where the KEF starts loosing output (in a 7 liter CB design).



That means I would have to use 1700 Hz as a cut off frequency. What advantages, if I do this?


I always thought, 3-way is used to get more/deeper bass?


But your link is exactly what I was looking for. The tweeter seems to be the same, which means I can use the crossover you posted earlier :)
A thousand thanks.
 
Last edited:
I was planning to use 200 Hz as a cut off frequency (that is what the German guy, who modified the crossover, recommended). My enclosure is 220 mm wide with a 40 mm 30° phase.


200 Hz is the value, where the KEF starts loosing output (in a 7 liter CB design).



That means I would have to use 1700 Hz as a cut off frequency. What advantages, if I do this?


I always thought, 3-way is used to get more/deeper bass?


But your link is exactly what I was looking for. The tweeter seems to be the same, which means I can use the crossover you posted earlier :)
A thousand thanks.
I don't get what is forcing you to use 1700Hz crossover at all. With a 5" you can go much higher and keep the tweeter in its comfort zone.

The advantage of a three way, apart from deeper bass and improved efficiency is the 5" can be used as a mid and spared excursion, which reduces distortion and improves power handling.

That SEAS Loki circuit looks almost exactly what Goran got after a 4 ohm conversion:

690416d1530871068-kef-ls50-crossover-seas-loki-crossover-png


Goran's 20uF doesn't really do much, but helps things along a bit on frequency response. A waveguide tweeter gains about 5dB below about 5kHz, so the input capacitor on the tweeter filter is smaller than you would use with a regular dome.

www.audioexcite.com >> Concentric Three – M5

A two way LS50 style two way would get about, er, 1mH and 15uF on the bass section. Goran's treble section with some level adjust. And positive polarity as I thought. But it won't go loud. Laws of physics.

A happy conclusion anyway. Good luck with this.
 
Last edited:
I was planning to use 200 Hz as a cut off frequency

200 Hz is the value, where the KEF starts loosing output (in a 7 liter CB design).

That seems OK to me,you would definitely not need to correct the coax for baffle step, the woofer at the right level would take care of it. The LS50 crossover would not work if that is the case.

That means I would have to use 1700 Hz as a cut off frequency. What advantages, if I do this?

Very few in my opinion as well as Steves! The lower you crossover the tweeter the harder time you give it, the more strained it will sound, Distortion rises, excursion rises. I'd stick to what KEF decided as optimum. Around 3kHz. The Coax woofer is so well behaved its not an issue.


I always thought, 3-way is used to get more/deeper bass?

You gain lots of efficiency, better power handling and maybe more extension in a 3 way. It does depend on the woofer you use of course. The LS50 certainly isn't short of bass as it stands....
 
Last edited:
This thread has turned out well. Why? Because you have Goran's 3 way circuit. 2 way or 3 way, you now only have to adjust for anything you change from his idea. Fundamentally just the bass section if you have a different woofer setup.

The theory of three ways (or two ways) says once any section is defined, you just adjust the other two to fit around it. Bafflestep is an added issue, but match the acoustic slope of the new bass to the original one by adjusting the bass filter, and it works. A simple problem for most simulators which can optimise.

I could do this in Visaton Boxsim with similar drivers. In fact I did.
Standcouplet WG – Boxsim Projektdatenbank
 

Attachments

  • Aequal4_Steen_Duelund_filter (2).JPG
    Aequal4_Steen_Duelund_filter (2).JPG
    36.1 KB · Views: 348
  • Aequal_root2_Steen_Duelund_filter (2).JPG
    Aequal_root2_Steen_Duelund_filter (2).JPG
    34.6 KB · Views: 359
Last edited:
Thank you both for your input, I really appreciate it.


Due to a misunderstanding (my fault) my intentions might have been misunderstood.


In the end I want a 3-way crossover:
- tweeter (KEF coax): 20 kHz - 2700 Hz (audioexcite)

- mid woofer (KEF coax): 2700 Hz (audioexcite) - 200 Hz

- woofer (t.b.d): 200 Hz - 30 Hz



The baffle step appears at 1100 Hz (343/0,31) (1700 was typing error).


@simon: You wrote
The woofer in a three way usually takes care of the baffle step as the baffle step is within its passband.
So if the woofer had to deal with the baffle step, I would have to cut it off above 1100 Hz, otherwise the mid-woofer (of the Koax) would be affected.


Up to this point, I would use the cutoff frequencies above.


Because you have Goran's 3 way circuit. 2 way or 3 way, you now only have to adjust for anything you change from his idea. Fundamentally just the bass section if you have a different woofer setup.
That is, why I was so happy :)
 
Last edited:
In the end I want a 3-way crossover:
- tweeter (KEF coax): 20 kHz - 2700 Hz (audioexcite)

- mid woofer (KEF coax): 2700 Hz (audioexcite) - 200 Hz

- woofer (t.b.d): 200 Hz - 30 Hz
What is the part number of your drivers? The KEF coaxial used in Goran's speaker is a midrange+tweeter and it is unable to cross lower than about 400 Hz. The KEF coaxial used in the LS50 is a midwoofer+tweeter and will be able to crossover at 200 Hz. They look and measure differently and so you cannot simply take crossover parts for one and use it with the other. You can compare them here and here.
 
On a second thought: You are right. I simulated the TSP and they are different.


Also the crossover for the woofer from the German guy of the LS50 is a 2nd order and the tweeter crossover from Goran is third order.


Well, looks like I celebrated too early....
If you have the midrange from the R series then there have been a number of projects like Goran's that you can follow. I am working on one now but it will use an active rather than a passive crossover. If you have the midwoofer from the LS50 then I do not recall seeing any DIY projects that used it. Most KEF coaxial drivers knocking around are from the entry level Q series which are all midwoofers and lack one or two refinements compared to the R series as one would expect. The Reference series midrange drivers are more refined again.

I don't know if the tweeters are the same in the R series and the LS50. The sensitivity of the midwoofer in the LS50 is close to that of the midrange in the R series (a midwoofer would usually be a few dB less sensitive than the equivalent midrange) suggesting that it might be. The KEF crossover between mid and tweeter in the LS50 is different to that in the R series as can be seen in the measurements and heard when listening.

You are aware that a baffle will usually put wiggles of a few dB in the response and a good crossover will compensate/work with it? Even if the drivers are identical, a well designed crossover will not work as intended in a different baffle.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.