More details for KEF Audio

[Bilbao-Emira]

My KEF set up is 13 channel 7-2-6 driven by Denon X8500 and Marantz MM7025 Stereo Power Amplifier.

Sounds amazing!

​

Oops wrong forum

Wow!
How much HP and 0-60 you get from that?
Does it have cupholders and hydraulic steering?

 

[frazzer]

I read most of this thread and I have no idea what any of you are talking about . But I do have a question, can I play my tapes in the Emira?

 

[Bilbao-Emira]

I read most of this thread and I have no idea what any of you are talking about . But I do have a question, can I play my tapes in the Emira?

Cool! Radio-cassete in the back next to my fridge!

We'll have to wait for the Bespoke Edition though! 😂

 

[FederGigant]

Here's a relevant view of the scenarios from a great little company called MiniDSP, they make external digital-domain crossover tech. Their products aren't directly relevant here but the diagrams 100% are.

I suspect Lotus/KEF will be leveraging the 3rd scenario, with digital crossover and split to discrete channels as part of the DSP processing.

https://www.minidsp.com/applications/digital-crossovers/digital-crossover-basics

Yeah, I have a few of the miniDSP products flying around here. Great to work with them. A completely active design with DSPs is for sure the most obvious way to go for. However we can‘t tell because there are too many variables still.

 

[Porter]

I think that there is no extra amp for the sub. Like I said in another tread, you don't have to use two channels per Uni-Q, it would just be the preferred choice. If you don't have enough channels, you could build a passive 2way-crossover for the Uni-Qs. I already built such hybrid systems where the woofers are actively controlled and the mid/high section passively for home use and it can work great. You could then really see the Uni-Q as one driver, couple it to the amp/DSP unit, apply equalizing, time and output correction, and the required high-pass filter. A completely active setup for the Uni-Q would of course enable the Engineers to do all this even better. If you require only 3 channels for the Uni-Qs, you could drive all the other drivers with separate channels, which makes 7 channels used in total.
Another way is to go completely active, use 6 channels for the Uni-Qs and then drive the 2 woofers parallel with one channel and the 2 drivers in the fresh air subwoofer with one channel. This is possible because low frequencies are spread omni-directional and you could therefore not tell where they come from anyways. My concern with that would be that the low end requires the most power, therefore it would be better to drive them all independently. That would also lead to more possibilities for DSP correction on the low end, but because I don't know the amp, I can't really tell if that's a problem. Maybe it has two channels that have more power than the other six, maybe it has more than enough power for each channel anyways. In terms of sound quality that's not that relevant anyways. More power just leads to a higher decibel output and has nothing to do with sound quality, except maybe for a better dynamic range due to higher power reserves.

I totally missed this before, great post.

My instinct is that they are using the two amps together because one of them is more than just an amp, it likely has digital signal input (optical or SPDIF), and a DSP on board with programmable digital-domain crossover. The HK-sourced amp unit in my 2014 BMW is designed that way, so it's an approach that has been around for a while.

In this scenario it may use onboard amplification for some channels, and/or pass to a second amp for some other channels. It doesn't have to be a "booster" scenario necessarily, no matter what partial explanation may be translated through customer service.

 

[FederGigant]

I totally missed this before, great post.

My instinct is that they are using the two amps together because one of them is more than just an amp, it likely has digital signal input (optical or SPDIF), and a DSP on board with programmable digital-domain crossover. The HK-sourced amp unit in my 2014 BMW is designed that way, so it's an approach that has been around for a while.

In this scenario it may use onboard amplification for some channels, and/or pass to a second amp for some other channels. It doesn't have to be a "booster" scenario necessarily, no matter what partial explanation may be translated through customer service.

In that scenario they should have more than enough channels and could control everything via DSP. Seems like the most obvious solution then.

 

[Nova]

I believe this may be the amp or one very close to it in design. It offers 8 channels, DSP capability with the necessary delay, level, and filter settings.

JBL DSP Amplifier DSP4086

JBL DSP Amplifier is a DSP audio processor for fine-tuning of aftermarket audio systems. It is a self-contained audio system with 6-channel audio inputs and 8-channel powered audio outputs.

www.harmanaudio.com

It's possible that the Harman amp is being used for the 8 non-sub speakers while the DCY11 drives the two subs. But then the amp power may be different from this JBL unit - maybe 30watts per channel giving 240, and DCY11 is delivering another 100 watts to the two subs? I just don't see a case where the amplifier duty for the non-sub speakers is split between the DCY11 and the Harman 8-channel amp.

 

[SlavaUkraini]

340W is lower than I was expecting for the cabin size tbh. That’s not necessarily the end all be all to sound quality but I would be lying if I said I wasn’t a little disappointed

For comparison, in the cayman, the stock audio is around 305w, the upgraded Bose is 505w and then the premium Burmester is 820w

 

[FederGigant]

340W is lower than I was expecting for the cabin size tbh. That’s not necessarily the end all be all to sound quality but I would be lying if I said I wasn’t a little disappointed

For comparison, in the cayman, the stock audio is around 305w, the upgraded Bose is 505w and then the premium Burmester is 820w

If I consider that those 820W is RMS (which I doubt) and that there is a different setup with more speakers, you might get 3Db more from the Burmester system (if the drivers have roughly the same efficiency then the KEFs). The power figures just mean nothing. You require double the power to gain 3Db. The KEF system could technically be louder than the Burmester, if the used drivers are more efficient. 30W of power normally is more than enough to get a 25mm tweeter clipping like crazy.

 

[SlavaUkraini]

If I consider that those 820W is RMS and that there is a different setup with more speakers, you might get 3Db more from the Burmester system (if the drivers have roughly the same efficiency then the KEFs). The power figures just mean nothing. You require double the power to gain 3Db. The KEF system could technically be louder than the Burmester, if the used drivers are more efficient. 30W of power normally is more than enough to get a 25mm tweeter clipping like crazy.

You sound like you know more about this than I do so I’m going to take your word for it. That assuages my worries

 

[Porter]

340W is lower than I was expecting for the cabin size tbh. That’s not necessarily the end all be all to sound quality but I would be lying if I said I wasn’t a little disappointed

For comparison, in the cayman, the stock audio is around 305w, the upgraded Bose is 505w and then the premium Burmester is 820w

FederGigant totally covered it but I would add that "wattage" is a nearly meaningless figure nowadays. It used to be more relevant (or baseline-comparable) when most or all "car" speakers were 4Ω, nearly all amplifiers were class AB, and all crossovers were passive and located between the amp and the speaker as an analog load. A passive crossover network converts unwanted frequencies to heat and throws away a big portion of the input power in the process.

Literally none of those things are true today. In a scenario where you only need to amplify the specific frequencies appropriate to each driver, and the driver can be carefully impedance-matched to the electrical sweet spot of a particular amp (or vice versa), the specs appropriate to the desired result often end up looking radically different from the old model. You can get the same (or greater) specific output with far lower distortion with a lower wattage amp circuit because you aren't throwing half of it away as heat.

To go even deeper... a smaller amp circuit for a particular channel sometimes ends up being chosen based on its min/max current output and linearity, so that volume stepping is matched between channels with different power output - i.e. a tweeter with a 20w amp channel and a midbass driver with an 80W amp channel that operate together (from a signal perspective) as a single stereo audio channel, like "front left". This might be because you want both 5% and 40% volume to match exactly on each component so that it doesn't get warmer or brighter (nonlinear frequency response) as volume changes.

 

[SlavaUkraini]

FederGigant totally covered it but I would add that "wattage" is a nearly meaningless figure nowadays. It used to be more relevant (or baseline-comparable) when most or all "car" speakers were 4Ω, nearly all amplifiers were class AB, and all crossovers were passive and located between the amp and the speaker as an analog load. A passive crossover network converts unwanted frequencies to heat and throws away a big portion of the input power in the process.

Literally none of those things are true today. In a scenario where you only need to amplify the specific frequencies appropriate to each driver, and the driver can be carefully impedance-matched to the electrical sweet spot of a particular amp (or vice versa), the specs appropriate to the desired result often end up looking radically different from the old model. You can get the same (or greater) specific output with far lower distortion with a lower wattage amp circuit because you aren't throwing half of it away as heat.

To go even deeper... a smaller amp circuit for a particular channel sometimes ends up being chosen based on its min/max current output and linearity, so that volume stepping is matched between channels with different power output - i.e. a tweeter with a 20w amp channel and a midbass driver with an 80W amp channel that operate together (from a signal perspective) as a single stereo audio channel, like "front left". This might be because you want both 5% and 40% volume to match exactly on each component so that it doesn't get warmer or brighter (nonlinear frequency response) as volume changes.

Great explanation which is easy to understand. You learn something new every day!

 

[Johnson]

Righto. So we think the “Harmon 8ch booster” is receiving digital input from the head unit, and using onboard processing to deliver analogue to the 8 surround speakers. Calling it a booster seems a bit dumb, but there you go.

So that suggests the “DCY11/CX11 EU Amplifier“ is driving the subwoofer, likely bridged. Although I don’t understand why there’s two sub drivers — the Bluacs units look to have only one driver per enclosure. Could this be a custom unit that has opposing subs? Then it would work like a typical infinite bass box, porting into the cabin.

No one’s mentioned the woofers behind the seats… low-midrange fill? I guess that frees up the Uni-Qs for directional mid-highs. I wonder why there are no rear Uni-Qs for proper surround?

 

[Porter]

Righto. So we think the “Harmon 8ch booster” is receiving digital input from the head unit, and using onboard processing to deliver analogue to the 8 surround speakers. Calling it a booster seems a bit dumb, but there you go.

I mean, it could be the other way around, with the Lynk & Co amp doing the DSP and the HK amp just doing analog stage, or it could be some interesting combination of things where both are fully in play for different channels. The KEF project team was probably given a budget and a pile of component options (with cost attached to each) and told to figure out a combination that would be excellent. Let's wait and see what they were able to do.

The whole infotainment system upstream from the speakers in the Emira seems likely to be similar to that in the Lynk & Co 01 (formerly the Geely CX11 - as referenced in the name of the amp on the diagram). Take a look at this image of the Lynk 01 interior... there are tons of parts shared. We can play a game of "count the Emira parts bin items". A bunch of these are also on the new Volvo CX40, so there's a ton of cross-pollination going on.

 

[Nova]

No one’s mentioned the woofers behind the seats… low-midrange fill? I guess that frees up the Uni-Qs for directional mid-highs. I wonder why there are no rear Uni-Qs for proper surround?

There is no need for a 2-seater car stereo rear speakers to be full range. The speakers are there to produce a sense of "ambiance" and possibly for mid-bass fill as you've mentioned since they look to be about 6-inches. It wasn't really until true multi-channel surround sound formats like Dolby AC-3 that true full rage rear/surround speakers became a thing. It's rare to see tweeter drivers behind a listener in a car environment - unless they are mounted in the rear doors for the rear passengers.

 

[Johnson]

It's rare to see tweeter drivers behind a listener in a car environment - unless they are mounted in the rear doors for the rear passengers.

I wired some Rockford Fosgate tweeters into the headrests of my Mazda MX-5 (Miata)... the cutouts were already there (oddly). It was a weird thing, hearing directional sound behind your head while driving...

 

[Billyidoliser]

They are likely doing a DSP-based, digital-domain crossover for each discrete channel. So the amp is only getting a constrained set of frequencies on each amplification channel, which maximizes its electrical performance. This is the "magic sauce" of modern powered-monitor speaker technology for home and studio applications, where amplification stages post-crossover can be optimized and phase controlled in ways that were impossible in a passive (post-amplification) crossover design. Efficiencies are much better, in-room (or in-cabin) phase-related distortion is controllable in new and better ways, and the system can be fine-tuned in the digital domain in a very rapid fashion prior to production.

Bonus, it's easier to execute as a (partner-led) development project and easier to process-engineer for manufacturing due to the reduced physical complexity and BOM count.

F××* me there's some clever people on here!

 

[Maurice R]

F××* me there's some clever people on here!

Too funny. I obviously don’t fit into the clever category. I got lost after the initial post. Must be very interesting for those who understand all this sound stuff
Don’t get me wrong you are all pretty smart. From what I can gather from everyone is that the sound system will be fantastic. I’m good with that

 

[Samps]

F××* me there's some clever people on here!

I didn’t understand a word of it either but glad it appears a quality system that has our experts excited.
slightly on/off topic and apologies but is anyone worried what the i4 might sound like?

If it’s anything like the v6 the audio will be off mostly anyway …

 

[digilotus]

I didn’t understand a word of it either but glad it appears a quality system that has our experts excited.
slightly on/off topic and apologies but is anyone worried what the i4 might sound like?

If it’s anything like the v6 the audio will be off mostly anyway …

I'd say the i4 will sound a bit like this without all the pops and farts. Lotus will of course make it more grown up.
A Lotus engineer on the New Dawn Lotus Documentary described the i4 as sounding like a "jet fighter". There's a little audio clip of the startup sound there too. (At 2:44 you can see the i4 engine in a mule and the startup sound is at 3:14 onwards)

Here I am going off-topic in my own thread!