Tought on this q concerto mod ?

[voodooless]

I agree with your sentiment regarding the capacitor and resistor. I disagree, however, with the inductor. Air core inductors eliminate core hysterises and eddy current losses, which are not insignificant when operating in the upper midrange frequencies. The issue, however, is ensuring that heavy enough gauge magnet wire is used for the air core inductor so as not to increase the DCR. The other issue is whether the prototypes were tested and evaluated with the specific inductor design used in the final production crossovers. Oftentimes that is not the case, but KEF seems to be more thorough in their engineering than other speaker companies I have seen, so perhaps they were. If so, the design may be optimized for the expected core losses.

Sure, an inductor can make a difference in some cases. But here, an air core was replaced with another air core with a heavier gauge. I very much doubt that it makes a difference. Possibly the midrange will play a bit louder. Maybe you can hear that? On the other hand, you can use EQ to do the same. Costs nothing.

 

[terryforsythe]

Sure, an inductor can make a difference in some cases. But here, an air core was replaced with another air core with a heavier gauge. I very much doubt that it makes a difference. Possibly the midrange will play a bit louder. Maybe you can hear that? On the other hand, you can use EQ to do the same. Costs nothing.

I doubt replacing the air core for another air core will make much of a difference, but one of the cored inductors, which was in series with the midrange, also was replaced with an air core inductor. That will make a difference. See the April 2025 edit and the below pictures copied from the article:

Original crossover:

Modified crossover:

 

[Ktacos]

I doubt replacing the air core for another air core will make much of a difference, but one of the cored inductors, which was in series with the midrange, also was replaced with an air core inductor. That will make a difference.

Maybe, maybe not. Seems that core saturation tends to happen at voltages that far exceed what people will be putting into the KEF's, at least on all but the last few tested here. One would need to measure to know but I'd wager it won't make much of a difference.

MTG Designs - Inductor Testing Part2

Part 2 Inductor Saturation Current: For these tests I'm increasing the drive voltage until the THD from the inductor reaches approximately 0.5%. I originally planned on using using 1% as the distortion threshold but ran out of amplifier power before hitting 1% THD on several of the inductors,

www.mtg-designs.com

The changes to the mid cabinets stuffing might change things. Surprised it's just a bare wall behind the driver.

 

[terryforsythe]

Maybe, maybe not. Seems that core saturation tends to happen at voltages that far exceed what people will be putting into the KEF's, at least on all but the last few tested here. One would need to measure to know but I'd wager it won't make much of a difference.

You are confusing core saturation with core losses. Although they both are related to hysterises, they are different concepts.

 

[Ktacos]

You are confusing core saturation with core losses.

I'm just going off what people are talking about. Losses are really not that big of a deal, go ahead and load of vcad and play with the DCR on inductors, it's a really small change.

 

[Mnyb]

I agree with your sentiment regarding the capacitor and resistor. I disagree, however, with the inductor. Air core inductors eliminate core hysterises and eddy current losses, which are not insignificant when operating in the upper midrange frequencies. The issue, however, is ensuring that heavy enough gauge magnet wire is used for the air core inductor so as not to increase the DCR. The other issue is whether the prototypes were tested and evaluated with the specific inductor design used in the final production crossovers. Oftentimes that is not the case, but KEF seems to be more thorough in their engineering than other speaker companies I have seen, so perhaps they were. If so, the design may be optimized for the expected core losses.

In one of their whitepapers KEF even described how they actually measured the performance of capacitors and chose accordingly. Of course no absurd audiophile brands involved. However the Q series may have more budget constraints than the R series.

In this particular mod changing component values definitely change the tuning of the speakers ( the resistor ) you better do a full spinorama measurement of that before calling it a success .

 

[Mnyb]

I mean you can’t just change stuff in any xover without verifying the acoustics performance of the speakers.
People who sells kit better put upp the numbers and prove what they done at the same technical level/rigor as the original manufacturer and motivate the changes.

I mean basement tinker vs KEF engineers who takes bets …

 

[terryforsythe]

I'm just going off what people are talking about. Losses are really not that big of a deal, go ahead and load of vcad and play with the DCR on inductors, it's a really small change.

Something is incorrect in your interpretation or people have fed you incorrect informatoin. Core losses are not related to the DCR of the magnet wire. They are related to the frequency, magnetizing force, flux densify and core material used. The impact of changing out a cored inductor to an air core inductor may be small, but it may not be, especially if the inductor being replaced is using cold rolled steel for the core and it is in series with the midrange. In some cases the difference is way, way more significant than changing out resistors and/or capacitors for better quality ones of equal value.

 

[Ktacos]

Something is incorrect in your interpretation or people have fed you incorrect informatoin. Core losses are not related to the DCR of the magnet wire. They are related to the frequency, magnetizing force, flux densify and core material used. The impact of changing out a cored inductor to an air core inductor may be small, but it may not be, especially if the inductor being replaced is using cold rolled steel for the core and it is in series with the midrange. In some cases the difference is way, way more significant than changing out resistors and/or capacitors for better quality ones of equal value.

I never made any comment on where the losses from inductors originate from. I simply stated you can model the DCR in vcad.

 

[Ktacos]

I mean basement tinker vs KEF engineers who takes bets …

To be fair some of the basement tinkerer's out there do make some exceptionally good speakers.

 

[terryforsythe]

I never made any comment on where the losses from inductors originate from. I simply stated you can model the DCR in vcad.

Then the model does not account for the core losses if ferromagnetically cored inductors are used.

 

[Ktacos]

Then the model does not account for the core losses if ferromagnetically cored inductors are used.

Doesn't matter, inductors come as a whole, thus DCR specific by manufacturer/or what is measured is more than adequate to design and predict filter behavior.

You've used vcad before right?

 

[terryforsythe]

Yes, I've used VituixCAD. It is good, but not perfect. The issue is that inductor hysterises losses are dependent on frequency and magnetizing force. Inductor core eddy current losses are dependent on frequency and flux density. It is difficult to model those in VituixCAD.

You've designed inductors before right?

 

[Ktacos]

You've designed inductors before right?

yes I invented the coil of wire.

 

[terryforsythe]

yes I invented the coil of wire.

Good. Keeping the inductor limited to a coil of wire is the best approach for midranges and tweeters.

 

[Paulkouhan]

Hi,

I just get a pair of this Q concerto meta 2 days ago.
I think that it sounds really great.

My wife, who is a musician, said after the first seconds of listening, that it lacks someting in the middle.

I did a little digging and find the post of tweakradje.

I introduced the +1.7 dB at 2.9 KHz (Q 0.7) on my Wiim ultra amp DSP and that changed the sound for the better.

I made some pictures based on Erin's measurements which seems to confirm the effectiveness of this modification.

I'll take measurements in my room later.

My guess is that the Q Concerto Meta and the R3 meta were playing on the same league with the right value resistor for the tweeter.

Thanks tweakradje to point that out.

 

[beagleman]

Hi everyone,


I recently stumbled across an interesting series of internal tweaks made by "Tweakradje" to the new KEF Q Concerto Meta. His mod focuses on refined crossover parts and improved cabinet damping.

tweakradje - KEF Q Concerto meta

The result To be able to hear space, placement, delays and clarity as never before. Singers in your room, hearing their vocal cords vibrate, hammer strikes of the piano, etc. Addictive. Sometimes even scary with some sounds because they are so realistic you think someone makes them in the room

sites.google.com

Summary of the changes:

• Replaced the original 3.9 Ω inline resistor in the tweeter’s path with a 2.7 Ω Jantzen Superes resistor (~+1.7 dB gain above 2.9 kHz)
• Swapped the stock 39 µF electrolytic midcap for a 39 µF Jantzen Cross-Cap
• Added an upgraded 1.2 mH Jantzen air-core coil in the midrange path (added in April 2025)
• Increased internal Sonofil damping—two rolls rearranged and a portion added into the mid‑box

These mods cost about €70 + some effort, with no driver substitutions or DSP involved—just component and damping improvement. The result, according to Tweakradje and one listener in Barcelona, was a more open, detailed sound “lifting the veil,” with enhanced clarity and staging .

I am actually making a home theater, i have 2x q150 for rear, 4x kef ci13ER for atmos height, and two kube 12b.
I have also ls50 wireless 2 that i love but planning to sell them because they not integrate well in a AVR system.
Room is 4x4 meter.
i was struggling between getting 3 LS50 meta to keep the same sound i was wondering about getting Q concerto META with Q6 META, but the sound with the q concerto meta feel much darker , and soundstage less wider that the ls50.

Do you think these king of mod can improve the linstening to especially in the treble region or i just have to stick with ls50? Especially in my room size.
Use will be 50/50 music/movie.

Best regards.

Not to bash on his "Work", but I think he is relying too much on the "On axis" response Erin got and not the in room that is more realistic for real listening not on a desk.

He is basically boosting the treble about 2 db to make up for the downward slope on axis, when in reality, they are fairly flat "in room" and that will make them almost too strong treble wise.


And have to add, when Erin shows "in room" his slope line he draws is just an estimation of what he thinks it sounds like, not some super accurate exact line, but more a estimation of the speaker in room sound. Its not set in concrete. So another might hear it a bit more flat and so on.