KEF R3 Speaker Review

[stevenswall]

I've been trying to work out the implications (for me!) of the sensitivity measurements that Amir made in this review. He wrote:



I'm assuming that the test involved sending a consistent 2.83V rms signal to the speakers across a range of frequencies. The impedance of the R3 at 40Hz is 3.2Ohm so if my assumption is correct, this means that it takes 2.5W of power to generate 80dB at that frequency.

I tend to listen pretty quietly by the standards of this forum as far as I can tell. According to an iphone app, typical volume would be around 65dB at 1m from the speakers. Nevertheless I am sure it makes sense to build in headroom for dynamic range. But how much headroom would you build in? How much power into 4Ohm (despite the nominal 8 Ohm impedance of these speakers) would you want, for this kind of situation? It seems to me that 25W ought to be enough for my purposes, and any half-decent amp can do that or more, so I have little to worry about... But "anplification requirements will be significant" seems to push in the other direction.

Many recorded songs have between 10 and 16 dB of dynamic range.

It takes 10x more power to gain 10dB if I remember correctly.

So if you wanted 90dB peaks it sounds like you might need a 25w amp. For 100dB peaks that would be a 250 watt amp.

 

[holla]

This analysis agrees with my own as far as I can see -- there are no "significant" amplification requirements for what I would call reasonable listening levels.

 

[mike70]

We can't forget about speakers ... amplifiers doesn't drive oscilloscopes. We need to "add" speaker properties to music properties (dynamic range, frequency spectrum, etc).

Speakers can have nasty impedance drops at certain frequencies, making the amplifiers to double the power instantly to not distort.

Audible? Can be .
Important? as always, we generally try to "justify" our beliefs / shopping decisions, so one will say "nah" and other will say "oh yes".
Human psychology isn't easy

 

[tecnogadget]

You can get all geeky and technical you want. But what impress me about this speakers is that it’s low sensitivity and impedance isn’t that important in “real life” as long the amp is “decent” (power and design).

I’ve owned in the past 93dB speakers, and I’ve used this R3 paired with a cheap SMSL SA-60 (2x55w class D) that costed me 44€ on sale, just for fun, and damm it can drive the speakers if you just want to enjoy music, no need to go bonkers.

Of couse thats not the way I use them, I got them paired with a SOTA Yamaha DSP-Z9 with more power than I’ve ever needed, but I’m aware this isn’t a strict necessity. On the other hand, please don’t have unreal pretentions of reference quality reproduction by the means of scarse 25w or 50w with this kind of sensitivity.

 

[Streamc]

You can get all geeky and technical you want. But what impress me about this speakers is that it’s low sensitivity and impedance isn’t that important in “real life” as long the amp is “decent” (power and design).

I’ve owned in the past 93dB speakers, and I’ve used this R3 paired with a cheap SMSL SA-60 (2x55w class D) that costed me 44€ on sale, just for fun, and damm it can drive the speakers if you just want to enjoy music, no need to go bonkers.

Of couse thats not the way I use them, I got them paired with a SOTA Yamaha DSP-Z9 with more power than I’ve ever needed, but I’m aware this isn’t a strict necessity. On the other hand, please don’t have unreal pretentions of reference quality reproduction by the means of scarse 25w or 50w with this kind of sensitivity.

I used then with Ayima 07. And with Purifi EigenTakt with no buffer I feel more volumetric bass, more serious from lows. This is not only about power. And I enjoy R3 with EigenTakt. Not would say this about Ayima 07 at least because of bass.

 

[holla]

You can get all geeky and technical you want.

...

On the other hand, please don’t have unreal pretentions of reference quality reproduction by the means of scarse 25w or 50w with this kind of sensitivity.

I'd like to get geeky and technical about the last part. I am very intrigued here.

What is it about a 50W amp that limits its ability to deliver reference quality reproduction? I can think of a few possibilities but could be way off:
- the amp may simply not be able to deliver enough current to reproduce without distortion; but in terms of the power spec alone, I think we ruled that out above
- as @mike70 pointed out, the amp may struggle to handle the sudden variations in current (power) requirements with the impedance variations of the speaker across frequencies; I suppose this is something like transient response. Again this is not purely about the power spec of the amp but about its performance more generally
- maybe "reference quality" requires a certain amount of dynamic range, which in turn demands a certain maximum SPL over and above some normal room noise level. For example if the room background noise is 30dB and you want 80dB of range, then you'd need 110dB max SPL which is certainly above what a 50W amp can produce with these speakers and also louder than I'd want to listen without hearing protection, but that's me

Only the last of these has any direct connection with the power spec of the amp taken in isolation i.e. if we assume the amp is a perfect voltage source up to its current limit.

What did I miss?

 

[mike70]

I only want to add something ... remember that the specifications are made by the fabricant itself. I mean, the same person that wants to sell as much as possible, says to you "my speakers are high / decent sensitive" or "my amplifier is able to ..."

So, unless you have trusty third part measurements ... we're talking in the air.
Only in a very few cases I saw sensitivity measurements near the specifications.
And with power, I need to say something?

So, is better to have a "positive gap". My opinion is not much about science ... but based on facts.

 

[tecnogadget]

@holla All your points are valid. In this case it’s just that specific amp isn’t “stellar” by all metrics, not exactly the cleanest distortion numbers but still a really good amp considering the low price and versatility. You could even carry it on a regular coat pocket

 

[SiamXIII]

Currently use R3 with VMV A1 amplifier, which has not so much power, but sounds very could. The thing that concerns me is clipping. Can it be avoided by bi-amping or with some class D amp on bottom end?

Can we assume from this chart, that midrange-high section has higher impedance? Does speaker impedance affect clipping?

 

[buz]

Can anyone recommend affordable stands for these (available in Europe)?

I currently have them on the sideboard next to the TV but it seems to generate resonance (not terribly surprised, the sideboard is made from steel). Also need stands for the surrounds (also R3).


Thanks a lot for a all pointers,

 

[mglobe]

Can anyone recommend affordable stands for these (available in Europe)?

I currently have them on the sideboard next to the TV but it seems to generate resonance (not terribly surprised, the sideboard is made from steel). Also need stands for the surrounds (also R3).


Thanks a lot for a all pointers,

I recently purchased the Kanto SX 26 stands. They were on sale at Crutchfield for $285 at the time. They come with two different top plates. The larger of the two have holes drilled that are at the correct width to mate up with the mounting holes on the R3’s but slightly too close to each other depth wise. I was able to drill out and enlarge the holes so that I could mount the speakers to the stands. Good build quality and dimensions.

 

[ROOSKIE]

I've been trying to work out the implications (for me!) of the sensitivity measurements that Amir made in this review. He wrote:



I'm assuming that the test involved sending a consistent 2.83V rms signal to the speakers across a range of frequencies. The impedance of the R3 at 40Hz is 3.2Ohm so if my assumption is correct, this means that it takes 2.5W of power to generate 80dB at that frequency.

I tend to listen pretty quietly by the standards of this forum as far as I can tell. According to an iphone app, typical volume would be around 65dB at 1m from the speakers. Nevertheless I am sure it makes sense to build in headroom for dynamic range. But how much headroom would you build in? How much power into 4Ohm (despite the nominal 8 Ohm impedance of these speakers) would you want, for this kind of situation? It seems to me that 25W ought to be enough for my purposes, and any half-decent amp can do that or more, so I have little to worry about... But "anplification requirements will be significant" seems to push in the other direction.

I realize this question is a few weeks old. In your situation you do not need a high powered amplifier at all. Even a 10watt amp will be perfectly fine.
Low SPL listening is simply not using much power. 65db averages with 70-80db peaks is low SPL by the standards and norms of HiFi. (when I sit down for a session it is common for averages to be at least 80-85db/90-95db peaks or higher, even 90db averages/100+db peaks which is very loud by the meter level but in reality with low distortion speakers and high powered amps sounds about right for that realistic experience.)
Many people are in your camp which is why so many low powered amps satisfy so many.

If you ever plan to crank it up and sit 4 or 5 meters away you will want that 100+watt amp.

Subjectively in my experience the R3 did not require more power than any typical monitor I have tried. It is a generally 'normal' contemporary era speaker.(The Phase angle and impedance around 40hrz is a bit rough though so bass heavy stuff will suffer there at SPL with weak amps)

View attachment 240203

Can we assume from this chart, that midrange-high section has higher impedance? Does speaker impedance affect clipping?

No need to assume, that is what that chart is.

Yes, speaker impedance affects clipping. When the speaker is playing high energy notes/sections in the ranges below 4ohms some amps will clip far before others. (some amps even have trouble with 4ohms but that is uncommon on new/contemporary amps) There may also be a limiter built into the amp.

You also have to look at the phase angles in conjunction with the impedances. The R3 is a generally a normal speaker except around 35-45hrz where the impedance is low and the phase angle high. If you listen at SPL to tunes with high powered energy releases there then use a beefy/well designed amp. (note how easy it is drive a ported speaker at the tuning frequency where impedance rises and phase is zero, of course a very high rise in impedance can also be an issue.)

Otherwise your amp power level capability and listening distances and musical tastes will affect clipping. All clipping is not equal either. Some clipping is perceived as euphonic, some horrid and some not perceived at all, as I have read even 6-10db of clipping can be said as inaudible in common situations for many listeners.

 

[Triliza]

Can anyone recommend affordable stands for these (available in Europe)?

I currently have them on the sideboard next to the TV but it seems to generate resonance (not terribly surprised, the sideboard is made from steel). Also need stands for the surrounds (also R3).


Thanks a lot for a all pointers,

Monoprice.eu has some, discussed on page 76.

 

[buz]

Monoprice.eu has some, discussed on page 76.

They didn't show any when I went there yesterday?

 

[mglobe]

For me the key to the Kanto’s is that the top plate is big enough to allow you to bolt the speakers to them. I’m not comfortable with these speakers just sitting on top of most of the available stands as the top plates are typically pretty small.

 

[Triliza]

They didn't show any when I went there yesterday?

Indeed. They had a month or so ago, a couple of them at least, send them an email to ask if they will be in stock any time soon.

 

[bloodshoteyed]

not really sure if they're seizing ops in the EU and selling everything off, or if there's new batches comming in - there's ever less articles available and not coming back in stock, it seems

 

[iNetRunner]

Can anyone recommend affordable stands for these (available in Europe)?

I currently have them on the sideboard next to the TV but it seems to generate resonance (not terribly surprised, the sideboard is made from steel). Also need stands for the surrounds (also R3).


Thanks a lot for a all pointers,

In Europe, look for NorStone stands. E.g. the Stylum 2 should be good.

 

[buz]

In Europe, look for NorStone stands. E.g. the Stylum 2 should be good.

I was looking at those myself but I have my doubts if the top plate is big enough to securely hold the R3 (ideally I would indeed like to screw them down)

Guess emailing monoprice is worth a shot

 

[mglobe]

I was looking at those myself but I have my doubts if the top plate is big enough to securely hold the R3 (ideally I would indeed like to screw them down)

Guess emailing monoprice is worth a shot

Compare the top plate to the large top plate that comes with the Kanto and the dimensions of the hole locations on the R3. It works with the Kanto large top plate , so if the top plate is as big or bigger than that it will work. Will likely require drilling or modifying holes.