• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Kef R series dual concentric

Zvu

Addicted to Fun and Learning
Joined
May 1, 2020
Messages
830
Likes
1,416
Location
Serbia
Here are few pics and measurements for start. They've arrived today CHN/UK/FR/SWE/SRB so quite a journey. I'll be doing more measurements (fr and dist.) but not too soon. I've aquired them through Setelec shop from France.

Tweeter and mid impedance

ВИСОКОТОНАЦ.pngНИСКОТОНАЦ.png

Few closeups

1.jpg2.jpg

3.jpg4.jpg

5.jpg6.jpg
 

Draki

Active Member
Joined
Jan 22, 2020
Messages
112
Likes
148
Location
Macedonia
Fs 173 Hz ???
OK, just saw this is a unit from the R series - thought it was LS50
 
OP
Zvu

Zvu

Addicted to Fun and Learning
Joined
May 1, 2020
Messages
830
Likes
1,416
Location
Serbia
No Metamaterial in the back?! Return that ASAP. ;-)

If they work at least as old R series dual concentrics from 2011, i'll die before i return them :)

Fs 173 Hz ???
OK, just saw this is a unit from the R series - thought it was LS50

I have big plans for this little fellow. Few woofers and in few configurations so we'll see.
 
Last edited:
OP
Zvu

Zvu

Addicted to Fun and Learning
Joined
May 1, 2020
Messages
830
Likes
1,416
Location
Serbia
I plan to use Omnes Audio Exclusive 6 woofers. Those are 182mm diameter woofers with 50mm voice coils and honeycomb sandwich cones that are used in Type 20, Type 30 and TM 80 module at HEDD Audio. They were about 200€/piece for a long time and then were on sale at 50€ /piece. I've managed to get my hands on 10 pieces. My plan is to use 1, 2 or 4 per side - how much bass power, will depend on subjective performance but i'll make test cabinets and try them in that order.

Here are some measurements of it.

Омнес Аудио Ексклузив 6.png

It will be passive crossover between midrange and tweeter and DSP active between woofers and midrange.

A friend of mine uses IcePower 125ASX2. Amir measured it HERE. I listened to it in his LXmini clone and it sounds good enough up to about 1kHz so that will probably be used for woofers since the crossover point will probably fall somewhere between 300 and 500Hz.

Shadow flare is being prepared for 3D printing:

Звуковод за средњетонац.png
 
Last edited:
OP
Zvu

Zvu

Addicted to Fun and Learning
Joined
May 1, 2020
Messages
830
Likes
1,416
Location
Serbia
@Nusse printed a flare and sent it to me for testing. I'll be making some measurements soon . Plan is to make a test cabinet out of chipboard and measure horizontal and vertical response 0-90 in 10 degree increments and to do some distortion testing.

Р3 коакс са звуководом.jpg

Other thing is to figure out how to mount the flare. Should i glue it on to a front baffle and mount the coax from behind or some other way. I guess that press fitting for flare is out of the question since R series shows problems in that area (flare movement due to continuous usage). Also, a mounting ring must be made for midrange. In 2011 R series coax is mounted directly to the front baffle which is quite thick and it isn't beveled from behind to let the midrange breathe. 2018 R series is made better in that regard:

Монтажни прстен.jpg

In old R series there was an aberration in frequency response at around 1100Hz. How much of it is because of poor driver mounting i really couldn't tell. We'll se how this little fellow measures. So, pace is slow but steady.

It would be interesting to see how large flare could be made and how low can we control the directivity with it. Maybe using a bigger woofer bellow ? Exciting stuff :)
 
Last edited:

bo_knows

Addicted to Fun and Learning
Joined
Aug 17, 2020
Messages
797
Likes
788
Location
Dallas, Texas USA
@Nusse printed a flare and sent it to me for testing. I'll be making some measurements soon . Plan is to make a test cabinet out of chipboard and measure horizontal and vertical response 0-90 in 10 degree increments and to do some distortion testing.

View attachment 125770

Other thing is to figure out how to mount the flare. Should i glue it on to a front baffle and mount the coax from behind or some other way. I guess that press fitting for flare it is out of the question since R series shows problems in that area (flare movement due to continuous usage). Also, a mounting ring must be made for midrange. In 2011 R series coax is mounted directly to the front baffle which is quite thick and it isn't beveled from behind to let the midrange breathe. 2018 R series is made better in that regard:

View attachment 125772

In old R series there was an aberration in frequency response at around 1100Hz. How much of it is because of poor driver mounting i really couldn't tell. We'll se how this little fellow measures. So, pace is slow but steady.

It would be interesting to see how large flare could be made and how low can we control the directivity with it. Maybe using a bigger woofer bellow ? Exciting stuff :)
I guess my next speaker purchase will be from you and not from KEF ;). Keep up the good work!
 

Ron Texas

Master Contributor
Joined
Jun 10, 2018
Messages
6,074
Likes
8,908
The R series concentric driver does not have the long travel surround found in the LS50/Meta. It's designed to go down to around 400 Hz.
 
OP
Zvu

Zvu

Addicted to Fun and Learning
Joined
May 1, 2020
Messages
830
Likes
1,416
Location
Serbia
I guess my next speaker purchase will be from you and not from KEF ;). Keep up the good work!

Now you're making me blush :)

The R series concentric driver does not have the long travel surround found in the LS50/Meta. It's designed to go down to around 400 Hz.

I'm aware of that. It has other benefits because it is a dedicated midrange. I'll try to make it down to at least 300Hz. Measurements from Kef R series whitepaper shows very good distortion down to 200Hz at 5.6V applied. That would probably be a no-no but 300Hz is doable. I did it before with R300 and it sounded very good.
 

Ron Texas

Master Contributor
Joined
Jun 10, 2018
Messages
6,074
Likes
8,908
Now you're making me blush :)



I'm aware of that. It has other benefits because it is a dedicated midrange. I'll try to make it down to at least 300Hz. Measurements from Kef R series whitepaper shows very good distortion down to 200Hz at 5.6V applied. That would probably be a no-no but 300Hz is doable. I did it before with R300 and it sounded very good.

Good luck on this project.
 

Snoochers

Active Member
Joined
Mar 28, 2020
Messages
187
Likes
70
Are they releasing a new R series or is this from the existing R3, R7 Etc line?
 
OP
Zvu

Zvu

Addicted to Fun and Learning
Joined
May 1, 2020
Messages
830
Likes
1,416
Location
Serbia
^This is from existing R series.

@Ron Texas Thanks.

Any model or drawing of what the final speaker is supposed to look like?

This is supposed to be an exploratory project. I'm tempted to jump to dual woofer W-DC-W first, similar to Reference 3. Since woofers i chose don't need large volume to get down to low 40's i might get away with standmount size. So nothing new here. I am doing measurements with >10ms gate quasi-anechoic, so measurement resolution will be very good. I also intend to try constant radius ports developed by JBL which should hopefully keep the distortion low at bass frequencies.
 

q3cpma

Major Contributor
Joined
May 22, 2019
Messages
3,060
Likes
4,416
Location
France
^This is from existing R series.

@Ron Texas Thanks.



This is supposed to be an exploratory project. I'm tempted to jump to dual woofer W-DC-W first, similar to Reference 3. Since woofers i chose don't need large volume to get down to low 40's i might get away with standmount size. So nothing new here. I am doing measurements with >10ms gate quasi-anechoic, so measurement resolution will be very good. I also intend to try constant radius ports developed by JBL which should hopefully keep the distortion low at bass frequencies.
I see, hope you get a nice result.
 
  • Like
Reactions: Zvu
OP
Zvu

Zvu

Addicted to Fun and Learning
Joined
May 1, 2020
Messages
830
Likes
1,416
Location
Serbia
Thinking about it, maybe we could try something different after all.

Using two woofers in dual opposing configuration but instead mounting them on sides, those would be mounted on front and back. Something like R3 but with aditional woofer behind and dsp active. This would probably necessitate 3 amplifier chanels per speaker. One IcePower 125ASX2 per speaker for woofers and something a little cleaner for dual concentric. Still maintaining passive crossover between midrange and tweeter. It would remain a threeway though. No point of trying 3.5 way since crossover frequency between woofer and midrange will be around 300Hz, so much lower than baffle step effect. It would be interesting to see how delaying back woofer influences dispersion.
 
Last edited:

Draki

Active Member
Joined
Jan 22, 2020
Messages
112
Likes
148
Location
Macedonia
If you choose dual woofers, DSP and at least two amp channels per box, maybe you can try the cardioid dispersion.
It can work with one front and back each, but note the back needs a cascaded xo branch over the front one, not a delay as such since it is already physically distant.
Something like this (simulated as a concept, not actually built, but with actual drivers measurements):
Zvu.PNG
Zvu 1.png
 
OP
Zvu

Zvu

Addicted to Fun and Learning
Joined
May 1, 2020
Messages
830
Likes
1,416
Location
Serbia
This looks quite nice. I'm guessing your example is with 4 woofers per cabinet ?
 

q3cpma

Major Contributor
Joined
May 22, 2019
Messages
3,060
Likes
4,416
Location
France
If you choose dual woofers, DSP and at least two amp channels per box, maybe you can try the cardioid dispersion.
It can work with one front and back each, but note the back needs a cascaded xo branch over the front one, not a delay as such since it is already physically distant.
Something like this (simulated as a concept, not actually built, but with actual drivers measurements):
View attachment 126196View attachment 126197
Ehhh, isn't the goal to be CD down to at least the Schroender frequency, not actually cardioid? Since 90° still has a lot of energy, you don't get the benefit of reduced early reflections, only diminished SBIR.
 
Top Bottom