Kali Audio IN-8 Round 2, How Should I Measure Hiss? Hiss Videos? (Anyone else notice less hiss in later purchased pairs?)

[detlev24]

Thankfully these 'little' Genelecs play noticeably higher than 105dB peak, and if you have the data showing what you're insinuating with your last line I'd love to see it. [...]

That is basically what shows on the dB/kHz grid of the magazine's measurements. To be fair, at maximum acceptable distortion levels that is 105 dB SPL around 40 Hz for the 8260A (instead of @Absolute's good guess of ~50 Hz).

Max. 3% THD is considered for the blue curve and max. 10% THD for the red curve [8260A; reversed colors on the M2 grid], until the amplifier (RMS) limiters activate [on the M2, that is reached at 400 W for the woofer and 50 W for the tweeter; see editorial notes to the linked measurements - no specific information is given for the 8260A]. The 10% THD measurement was conducted until 250 Hz for the 8260A and until 300 Hz for the M2; all measurements taken at 1 meter [= 3.28 feet] distance and with a 185 ms Sinus sweep (information found on the 8260A review). Considering psychoacoustics, I would not want more than 10% THD in the bass region vs. 3% from the 'Low mid' upwards, as this clearly becomes audible (and much more damaging the listening experience, than some of the previously mentioned electronic hiss). Regarding "hiss", they measured a 26.8 dB(A) noise floor at 10 cm [~4 inches] distance for the 8260A vs. 30 dB(A) for the M2 [with 'Crown I-Tech 4x3500HD' amplifier at maximum input sensitivity]. However, I am not sure how much of a realistic idea this gives.

As of the chart you posted, the threshold of our hearing lies roughly at 75 dB for 20 Hz; equivalent to 0 dB at ~2 kHz, where our hearing is most sensitive. // At 100 dB SPL in 1-meter distance [theoretical equivalent of 92.25 dB SPL at 'your' 8 feet = 2.44 meters], the 8260A already shows 10% THD on the grid's low end. Those are quite good results for a loudspeaker of this size! But not comparable to the clean bass and headroom an M2 delivers.



[EDIT]
Unfortunately, they cut off the M2 grid at 40 Hz, it seems. My previously linked in-room measurements of the M2 show worst ~5% THD at ~15 Hz; @ 90 dB SPL and ~10 feet listening distance. // Out of interest, would you be able to provide similar measurements of your 8260A, without application of GLM or similar? Easiest way would be using a UMIK-1 with REW.

 

[stevenswall]

, would you be able to provide similar measurements of your 8260A, without application of GLM or similar? Easiest way would be using a UMIK-1 with REW.

Thanks for the charts, that's exactly what I wanted to see. I'll reel it back a little as it sounds like I'd need subs to be on par with the M2 in a living room. Now I'm curious if they used a larger internal volume like Genelec how much of a difference that would make. Genelec to compete with the M2 as designed could perhaps seal the cabinet and add additional drivers on each side of the cabinet and give them more excursion to be like a massive Phantom Reactor.

Measurement wise, my room is currently untreated, but if that doesn't matter, send a link to a guide for the measurements you want and as long as it won't destroy the speaker I can try it. I have a umik and miniDSP, REW, etc. and can switch GLM off.

 

[Trouble Maker]

Also odd: The hiss seeming lower allows me to hear really strange radio tuning noises coming from the tweeter if I put my ear right next to it. need to see if there is a time roommates/tenants are all gone where I can unplug all the wireless things in the house.

"What's the frequency, Kenneth?"

 

[Absolute]

Thanks for the charts, that's exactly what I wanted to see. I'll reel it back a little as it sounds like I'd need subs to be on par with the M2 in a living room. Now I'm curious if they used a larger internal volume like Genelec how much of a difference that would make. Genelec to compete with the M2 as designed could perhaps seal the cabinet and add additional drivers on each side of the cabinet and give them more excursion to be like a massive Phantom Reactor.

Measurement wise, my room is currently untreated, but if that doesn't matter, send a link to a guide for the measurements you want and as long as it won't destroy the speaker I can try it. I have a umik and miniDSP, REW, etc. and can switch GLM off.

Genelec have speakers to compete with the M2 for SPL levels and even bigger stuff best classified as completely bonkers.
https://www.genelec.com/1236a#section-technical-specifications

Even the M2 is designed to play along with 18 inch subs for larger rooms, so it's all about context. In most living rooms you don't need a ton of capacity at 20-30 hz and can instead focus on achieving the most even response from smaller subs placed appropriately. I think that's a better proposition for pure sound quality.

 

[detlev24]

[...] Measurement wise, my room is currently untreated, but if that doesn't matter, send a link to a guide for the measurements you want and as long as it won't destroy the speaker I can try it. I have a umik and miniDSP, REW, etc. and can switch GLM off.

Actually, Genelec does provide great competition for the M2; see their Main Monitors. // @Absolute: You have been quicker!

As of the operating manual, the 8260A does have a driver protection circuitry for long term output and is rated at a maximum short-term sine wave SPL output of ≥ 113 dB SPL [1 meter on axis in half space; and from 100 Hz to 3 kHz]. 113 dB SPL at 1 meter [3.28 feet] are equivalent to 105.25 dB SPL at 2.44 meters = 8 feet. Furthermore, REW has 'Protection' settings on the "Measure" window. You should be fine if on the "Generator", you calibrate for 'Pink random' noise, use 'Speaker Cal (500 to 2000 Hz)' and '-12,00 dBFS', as per the previously linked miniDSP instructions. Just note, that the windows on the latest REW versions look different to what the tutorial shows - but similar enough for being able to follow it. If you want to start clean with REW back to its default settings, just select "Preferences -> Delete preferences and shut down". On your measurements, you can adjust the graphs at any time by changing the "Limits".

I would suggest you start with calibration to standard 75 dB(Z) F, measure, see "Distortion" and then increase by 5 dB steps; so a second measurement at 80 dB(Z) F would follow, check the distortion graphs again and so on, until you reach 90 dB(Z) F - if you feel comfortable to do so. [You might want to protect your ears, though, as you seem to be sensitive against noise.] If you want to stop earlier, no problem, I can always provide measurements at a lower SPL for comparison purposes. This by no means is meant as a scientific approach, but fun enough to have an idea of what happens to the 8260A in your room vs. the M2 in my room.

The application of room EQ, GLM in your case, might lower or increase distortion levels; depending on whether the software needs to reduce or boost frequencies - in accordance with the desired target curve.

PS: Should you wonder what a proper treatment, specifically of early reflections would do; have a look at Ethan Winer's explanation.

 

[tifune]

Assumptions:
-The Genelec 8260 will have more bass extension.
-The Kali will have no port chuffing. (As tested before. Have a few more songs this time too.)
-The Genelec will hiss less.
-The Kali will his about on par with a JBL LSR308.
-The Kali will handily outperform my HiVi DIY 3.1
-The Kali, if okay with some hiss or not used to close, will be a better value because you can afford subs to get the extension of the 8260
-Grounding vs Ungrounding will not make a difference for hiss. (In the video, I removed a grounding plug from the orange splitter. The outlet I tested with that was supposed to be grounded was not though.)

Am I totally missing it, or did this thread get so far derailed into Genelec vs M2 that we never found out how your tests went? I stumbled on this thread because I recently got the IN-8 and, with gain at 0, hiss is noticeable from 2m+.

To be fair, my setup is unusual - IN-8's connected to a passive channel selector via phono-1/4" adapters. Using the channel selector to create 5 "zones" throughout the house, some of which are active and some not. Not at all a great representation of how things should be done, but $. I was shocked at the hiss compared to the LSR305 attached to same selector.

 

[stevenswall]

Am I totally missing it, or did this thread get so far derailed into Genelec vs M2 that we never found out how your tests went? I stumbled on this thread because I recently got the IN-8 and, with gain at 0, hiss is noticeable from 2m+.

To be fair, my setup is unusual - IN-8's connected to a passive channel selector via phono-1/4" adapters. Using the channel selector to create 5 "zones" throughout the house, some of which are active and some not. Not at all a great representation of how things should be done, but $. I was shocked at the hiss compared to the LSR305 attached to same selector.

The hiss is still noticeable, but slightly lessened and less "ragged" on the pair Kali sent to me. Haven't heard a way to reliably measure hiss since I don't have a calibrated SPL mic.

Definitely more than the LSR3xx series.

Most places will allow you to return them, though there aren't many competitors in this price range with a coincident driver like the Kalis and a separate woofer.

I'd either wait for Kali to release another model and lower the noise floor, or double/triple the budget and try an active KEF, Elac Navis, or Devialet Phantom.

The latter two in my home are silent. The KEF LS50w I heard a demo of seemed to be quiet too, but there was background noise.

 

[tifune]

.

Most places will allow you to return them, though there aren't many competitors in this price range with a coincident driver like the Kalis and a separate woofer.

That was the big selling point for me, my speakers are mounted about 10' in the air so the benefits of a coaxial's vertical dispersion offer a better value than in a typical setup. The room is also quite large, so a 8" woofer helps. Not sure of anyone else fitting that bill except the higher end Genelecs and Presonus S8. FXaudio has something but the sparse reviews I could find weren't promising.

I was hoping not to spend the extra $ because this isn't intended to be a critical listening space but the S8 cost is looking more justified each day. How did you like the Navis?

 

[ernestcarl]

... my speakers are mounted about 10' in the air so the benefits of a coaxial's vertical dispersion offer a better value than in a typical setup.

Wow that is high, high up!

Here's a little handy triangle calculator you might find useful: https://www.omnicalculator.com/math/triangle-45-45-90

The vertical dispersion control of the S8's horn is +-45 degrees

So if you're not planning on angling the speaker down, you'd need to sit/stand at least ~10 feet away as well to get the max phase-linear or time-coherent direct sound.

Even without any additional external EQ, you should still get a relatively flattish response (dark green).

My 'basic' EQ

But remember, much of the reflections coming back summing in the ears will give us the perception of a tonally flat response.

And if you add a little bit of EQ, you can make the overall tonality at every angle even more neutral.

If hiss is the concern... well, your speakers being that high up should negate the issue altogether. Another thing to look at is to make sure the output voltage of your preamp is at least 2v (3-4v better) so you don't need to crank up the input gain of the speaker's internal amp.

 

[ernestcarl]

How did you like the Navis?

Found some vertical directivity curves for the smaller Navis, but it's quite limited so not enough info for your use case:

Complete review here.

Doesn't look particularly efficient power-wise, though.

 

[thewas]

Doesn't look particularly efficient power-wise, though.

What do you mean by that?

 

[ernestcarl]

What do you mean by that?

Oops! I didn't read that it had a BASH amp and assumed it used class AB for all drivers. I only skimmed and saw 85dB sens. and class AB from the spec.

But I'm guessing given the small size, the 5-inch bass driver and 4-inch mid woofer (260w total peak?) of the Navis will require a heck of a lot more power compared to the single 8" driver (100w peak?) of the S8 previously mentioned.

 

[ernestcarl]

Now this is very interesting:

Both the tower and the bookshelf use the same amp -- manufacturing/design cost saving? Now presuming the bookshelf uses 1/3 the power around its max (sensible continuous SPL -- not specified) limits that would mean realistically the bookshelf will not likely use well over 100w for both mid and bass drivers. Also given the advanced electronics inside, I would suspect that limiters are included in the system.

110 dBA SNR is emphasized in the marketing, which is great, but not that essential to know...

It does not make any sense why they do not just publish their recommended/max SPL limits for their active speakers. One could put a thousand watt amplifier in these boxes for all I care, and it will still not give me the information I actually want to know -- i.e. how loud will they play before crapping out.

 

[stevenswall]

Both the tower and the bookshelf use the same amp -- manufacturing/design cost saving? Now presuming the bookshelf uses 1/3 the power around its max (sensible continuous SPL -- not specified) limits that would mean realistically the bookshelf will not likely use well over 100w for both mid and bass drivers. Also given the advanced electronics inside, I would suspect that limiters are included in the system.

110 dBA SNR is emphasized in the marketing, which is great, but not that essential to know...

It does not make any sense why they do not just publish their recommended/max SPL limits for their active speakers. One could put a thousand watt amplifier in these boxes for all I care, and it will still not give me the information I actually want to know -- i.e. how loud will they play before crapping out.

I think they are doing something similar to Kali: More or larger speakers are more sensitive, so they may not need additional power. I predict the single bass driver in the bookshelf version is using more than 1/3 of the power to keep up with the tweeter and midrange unit.

Interesting on the newly announced IN-8v2 that they use a lower powered woofer amp on the IN-8 vs the IN-5.