Kef LS60 and Room correction?

[Purité Audio]

One of the advantages of the LS60 is they really don't need subs.

Well…
Keith

 

[Ron Texas]

Well…
Keith

Lack of delay for integrating subs was one of the criticisms mentioned by the OP.

 

[Ron Texas]

If they just added distance and delay/settings ? I forse problems integrating with subs that have quite a processing delay like arendals that have 8ms probably dues the peq built into thier sub

Built in delay in subs runs from a minimal 3 ms in Rhythmic subs, 6 ms in SVS and you quote 8 ms. The subs. may be further away from the LP than the mains although this is not a rule. Playing the odds, delaying the mains is a lot more likely.

 

[kemmler3D]

delaying the mains is a lot more likely.

That's what I thought too.

I think delay in subs integration is probably only worthwhile if you have things 20 feet apart or thereabouts. Phase adjustment is probably more useful for most people.

 

[Mnyb]

Built in delay in subs runs from a minimal 3 ms in Rhythmic subs, 6 ms in SVS and you quote 8 ms. The subs. may be further away from the LP than the mains although this is not a rule. Playing the odds, delaying the mains is a lot more likely.

I’ve seen 0,4 ms quoted for kef subs and 0,3 ish for subs with hypex plate amps ?

 

[kemmler3D]

I’ve seen 0,4 ms quoted for kef subs and 0,3 ish for subs with hypex plate amps ?

I guess the question here is - how much do you need to delay your sub outputs (or main speakers?) and why?

 

[Mnyb]

Yes sub delay should go in both directions ie actually delaying the mains vs the sub ?

 

[Ron Texas]

I’ve seen 0,4 ms quoted for kef subs and 0,3 ish for subs with hypex plate amps ?

That seems low. It takes 1ms to get a big subwoofer driver moving.

 

[Mnyb]

I guess the question here is - how much do you need to delay your sub outputs (or main speakers?) and why?

You are rigth that much can be done with phase control, but somewhat similar arrival time to listening position can’t be wrong hence why most AV products I owned had distance settings

 

[Mnyb]

That seems low. It takes 1ms to get a big subwoofer driver moving.

It migth be only the amp all such data one can find is not in a reliable context , one should probably not assume that it’s a single number for the whole system it’s probably frequency dependent to .

 

[Salt]

One of the advantages of the LS60 is they really don't need subs.

Except one needs really high sound pressure in sub , and that is not the playground of LS60

 

[Kal Rubinson]

One of the advantages of the LS60 is they really don't need subs.

They don't really need subs but they do appreciate (with) them.

 

[kemmler3D]

arrival time to listening position can’t be wrong hence why most AV products I owned had distance settings

I think this is more important for stereo and surround effects, it could theoretically help with subs but the wavelengths are already several milliseconds on their own, so I am not sure how audible a few ms timing difference is going to be.

 

[phoenixdogfan]

If your material can all be played on a PC, I don't see a reason you can't run EQ, Bass Management, Dirac, etc on the PC, and send a two channel LPCM to the LS60's. How much further room correction you would need after running DLBC is up to you, but I imagine whatever is being done on the Kef app could be minimized. Don't know if the KEF receives USB, but no problem anyway, just run the USB stream to a Topping D10, and output Coax to the Kef's. If you want to run subs, same difference, except substitute a miniDSP Flex with all digital outputs for the D10. Would not need a PC that way if you had a streamer, but you would lose DLBC.

 

[srrxr71]

Soo if you place a low latency sub close to the speakers , you can manage it good enough with an external solution ?

Yes because hopefully they can also add delay/phase correction as needed. I don’t know the difference aren’t they kind of the same thing?

 

[Mnyb]

Yes because hopefully they can also add delay/phase correction as needed. I don’t know the difference aren’t they kind of the same thing?

Not exactly , for example phase -180 flips the signals polarity. That you have in kef LS60 and the KC92 .
So this is covered by KEF what’s missing is a distance/delay setting like in your AVR where you set the distances from your speakers to you listening position in the case of for example an sub with 8ms processing delay in it’s DSP then you have to add approximately 2,7 metres extra to the subs distance ( in reality your AVR is delaying the main speakers in cases where the subs distance is longer than the main speaker distance ).

It migth not be such an huge deal in practice as base wave length are long and it all becomes an EQ problem in the end. But to much of delay of the sub signal relative to the mains speakers can’t be good for the impulse response even if the phase adjustment can get continuous sine waves to align and the fr response can be eased flat ?
But I don’t know how much is to much 1 second later would be a problem but would 4 ms be a problem ?

 

[srrxr71]

Not exactly , for example phase -180 flips the signals polarity. That you have in kef LS60 and the KC92 .
So this is covered by KEF what’s missing is a distance/delay setting like in your AVR where you set the distances from your speakers to you listening position in the case of for example an sub with 8ms processing delay in it’s DSP then you have to add approximately 2,7 metres extra to the subs distance ( in reality your AVR is delaying the main speakers in cases where the subs distance is longer than the main speaker distance ).

It migth not be such an huge deal in practice as base wave length are long and it all becomes an EQ problem in the end. But to much of delay of the sub signal relative to the mains speakers can’t be good for the impulse response even if the phase adjustment can get continuous sine waves to align and the fr response can be eased flat ?
But I don’t know how much is to much 1 second later would be a problem but would 4 ms be a problem ?

I don’t know but I agree with everyone who says it’s more likely the mains would be delayed more.

Also some subs have a knob to adjust between 0-180 degrees. That is sort of phase adjustment.

I think a good dsp sub integration would figure out phase/delay for you. So as to match the impulse response as best as possible.

 

[kemmler3D]

Not exactly , for example phase -180 flips the signals polarity. That you have in kef LS60 and the KC92 .
So this is covered by KEF what’s missing is a distance/delay setting like in your AVR where you set the distances from your speakers to you listening position in the case of for example an sub with 8ms processing delay in it’s DSP then you have to add approximately 2,7 metres extra to the subs distance ( in reality your AVR is delaying the main speakers in cases where the subs distance is longer than the main speaker distance ).

It migth not be such an huge deal in practice as base wave length are long and it all becomes an EQ problem in the end. But to much of delay of the sub signal relative to the mains speakers can’t be good for the impulse response even if the phase adjustment can get continuous sine waves to align and the fr response can be eased flat ?
But I don’t know how much is to much 1 second later would be a problem but would 4 ms be a problem ?

As a rule of thumb, 20ms is considered the transition point between continuous sound and audibly separable sounds. Like when a series of clicks becomes a tone.

So I would say if you have speakers more than 20ft apart you'll probably want to do some delay to get the IR back together.

Any closer together than that, and the effect of delay will tend to be pretty subtle.

 

[terryforsythe]

I have implemented room correction with my LS60s using two different methods. In both methods I measured my speakers using REW and generated a .wav convolution file following this tutorial:

1. For the first method I generated a .wav convolution file that includes both Left and Right channels, and uploaded that to SqueezeDSP, which is a plugin for Logitech Media Server (LMS). I then used LMS to stream music to my LS60s. LMS and SqueezeDSP are free and open source projects, so there is no cost other than the hardware you use to host LMS. I use my NAS, which I already had. Unfortunately, the Tidal plugin for LMS no longer is supported (it was a little clunky anyway), which led me to method 2.

2. The second method uses separate convolution files for the left and right channels, and requires additional hardware. Here is a very thorough tutorial provided by mdsimon2: https://www.audiosciencereview.com/forum/index.php?threads/rpi4-camilladsp-tutorial.29656/.

I purchased a WiiM streamer, a Raspberry Pi and a miniDSP MCHStreamer kit. The MCHStreamer connects to the Raspberry Pi via USB, and handles digital input and output for the Raspberry Pi via TosLink or coaxial SPDIF.

I installed Ubuntu Server and Camilla DSP (both free and open source) on the Raspberry Pi, and uploaded the left and right channel convolution files to Camilla DSP. The WiiM feeds the Raspberry Pi via the MCHStreamer input, where Camilla DSP applies the convolution files. Chamilla DSP uses the MCHStreamer output to feed a digital input on the primary LS60.

The MCHStreamer kit includes TosLink input and output jacks, but not RCA jacks for the coaxial SPDIF. Instead, the board has two connectors where the RCA jacks are to be connected. I will add the RCA jacks after I 3D print a case for the MCHStreamer. Until then I am using TosLink, and it sounds fantastic. My LS60s definitely benefit from room correction.

 

[srrxr71]

I have implemented room correction with my LS60s using two different methods. In both methods I measured my speakers using REW and generated a .wav convolution file following this tutorial:

1. For the first method I generated a .wav convolution file that includes both Left and Right channels, and uploaded that to SqueezeDSP, which is a plugin for Logitech Media Server (LMS). I then used LMS to stream music to my LS60s. LMS and SqueezeDSP are free and open source projects, so there is no cost other than the hardware you use to host LMS. I use my NAS, which I already had. Unfortunately, the Tidal plugin for LMS no longer is supported (it was a little clunky anyway), which led me to method 2.

2. The second method uses separate convolution files for the left and right channels, and requires additional hardware. Here is a very thorough tutorial provided by mdsimon2: https://www.audiosciencereview.com/forum/index.php?threads/rpi4-camilladsp-tutorial.29656/.

I purchased a WiiM streamer, a Raspberry Pi and a miniDSP MCHStreamer kit. The MCHStreamer connects to the Raspberry Pi via USB, and handles digital input and output for the Raspberry Pi via TosLink or coaxial SPDIF.

I installed Ubuntu Server and Camilla DSP (both free and open source) on the Raspberry Pi, and uploaded the left and right channel convolution files to Camilla DSP. The WiiM feeds the Raspberry Pi via the MCHStreamer input, where Camilla DSP applies the convolution files. Chamilla DSP uses the MCHStreamer output to feed a digital input on the primary LS60.

The MCHStreamer kit includes TosLink input and output jacks, but not RCA jacks for the coaxial SPDIF. Instead, the board has two connectors where the RCA jacks are to be connected. I will add the RCA jacks after I 3D print a case for the MCHStreamer. Until then I am using TosLink, and it sounds fantastic. My LS60s definitely benefit from room correction.

That’s pretty much end game in my book.