Smsl a300 upgrade to drive kef r3

[maty]

Any specific recommendations withing $1500 and small form factor?

Later

Spotify? nope if the recording is good/very good.

Last night, I listened to

The Charlie Mingus Jazz Workshop - Pithecanthropus Erectus (1956) Vinyl, mono, Speakers Corner Records 2018, Germany

https://twitter.com/i/web/status/1650952600894726144

The sound via Spotify is unbearable for me with this album. If the original recording is demanding, with physical instruments, natural voices (without Autotune), without studio effects... Spotify does not measure up in my cheap second audio system.

You have two bottlenecks: amplifier and Spotify.

 

[HarmonicTHD]

Yes of course there are better amps and more powerful ones (and yes my usage of SOTA was not quite right, as I only quickly looked at the manufacturer specs eg 2x165w, THDN 0.004%, SNR 103dB, but independent measures, as @JSmith linked, indicate differently).

But as @VintageFlanker also said, in your circumstance (small room, not too loud), a better amp won’t solve your problem. And those who claim it would, should provide you with hard facts not anecdotes.

So good that you have ordered the UMIK1.

However, if you are interested in better amps look at the review index in this forum. Here the Hypex or Purify based amps deliver very clean (high SINAD) sound and plenty of power to easily drive a R3. (Yes yours is a bit weak, but again not causing the problem you perceive). Manufacturers like Boxem, Audiophonics come to mind which might also fit your budget.

Edit. Forgot Buckeye Amps and Apollon

 

[raindance]

It's a square room which is an acoustic nightmare. Proper room correction DSP with bass management will help and the cheapest way to do it might be trying an AV receiver (much less technical than using REW, etc).

 

[b7676]

If One doesn't want all the power of a NCx500, GAN, or other class d chip, LA90d mono blocks seem to be the thing to get at 1600$.

 

[maty]

Modern commercial AV receiver is a bad idea with KEF R3 (3.2 Ohms) and to play good recordings.

Small and square room, KEF R3 and you want subs? You need to improve others things before.

-> Before spending more money I would focus on improving the room, without spending too much.

My second system consists of some KEF Q100s that have been improved by me and that sound much better than the originals (Q100 driver is very good, the rest is not up to par, it is still a better driver than Q150). In December I released a hybrid, Magnat MA 900 (130/200 watts), which now sounds great (after an external improvement and... ). The other finalists: Arcam SA20 and Rotel A14 MKII (they say that the sound is different from the traditional Rotel, to be avoided with KEF and others with the type of music I like).

 

[maty]

SMSL A300 -> MA5332MS

[PDF] https://www.infineon.com/dgdl/Infin...N.pdf?fileId=5546d4627a0b0c7b017a3f402ad026a2

It should not have a problem with that minimum impedance (3.2 Ohms). Then, improve the room's acoustic.

 

[raindance]

I've used the Marantz slimline receivers and Denon full size receivers with 4 ohm speakers for years without issue. If he's not going to listen at ear splitting levels it'll be absolutely fine and barely even get warm (unless it is placed in an unventilated place). This whole minimum impedance concern is overblown. In all my years I've never had an amp blow due to load.

I'm offering a budget way to the OP to see what is possible.

 

[HarmonicTHD]

Modern commercial AV receiver is a bad idea with KEF R3 (3.2 Ohms) and to play good recordings.

Small and square room, KEF R3 and you want subs? You need to improve others things before.

-> Before spending more money I would focus on improving the room, without spending too much.

My second system consists of some KEF Q100s that have been improved by me and that sound much better than the originals (Q100 driver is very good, the rest is not up to par, it is still a better driver than Q150). In December I released a hybrid, Magnat MA 900 (130/200 watts), which now sounds great (after an external improvement and... ). The other finalists: Arcam SA20 and Rotel A14 MKII (they say that the sound is different from the traditional Rotel, to be avoided with KEF and others with the type of music I like).

All anecdotes and no facts as usual and not related to the OP situation.

 

[HarmonicTHD]

Some rough estimate when listening closely at decent levels. Adjust to your own will here:

Peak SPL Calculator

Yeah even when you conservatively add 20dB for crest factor of music (although Jazz usually doesn’t even come close) it is not outrageously high to explain the muffled sound coming from the amp.

Note: this is for 8ohm so multiply the power required by 8/3.2 for lowest impedance of R3 as a rough estimate.

 

[Kross]

Update.
Got NAD C700 to give it a shot.

Here is subjective impressions.

I like it way better than smsl a300. Now the issue with recessed vocals is gone. The whole setup sounds more natural, balanced and less fatiguing. Worth mentioning though, that on smsl I used E1 – Tone mode, not Direct (as direct really sounds dull). On the NAD I left treble and bass at default.

I'll play it for a couple of weeks before making a final decision but love it so far.

Also, subwoofer sound way better integrated. It gives a punch without booming. Smsl crossover mod didn't give any noticeable difference (could be my ears or amp issues, I had to return one smsl a300 due to hight noise, so quality control is questionable)

Extra benefit is just all in one box without streamer, dac, amp, remotes and cables all over.

Call it psychoacoustics, wow factor, but after 2 hours of listening to the same tracks that didn't click with me on smsl, I enjoy em all now.

I understand that Dirac could give even more benefits but so far happy the way it is. May consider NAD M10 in the furute as I'm a fan of minimalistic setups (call me crazy, but I have an OCD over having things visually organized in my home office with limited space)

 

[maty]

With the old Hypex UcD modules

NAD C 700
[Polish] https://audio.com.pl/testy/stereo/wzmacniacze-stereo/3462-c700

To English: https://audio-com-pl.translate.goog...l=auto&_x_tr_tl=en&_x_tr_hl=es&_x_tr_pto=wapp

 

[Kross]

With the old Hypex UcD modules

NAD C 700
[Polish] https://audio.com.pl/testy/stereo/wzmacniacze-stereo/3462-c700

To English: https://audio-com-pl.translate.goog...l=auto&_x_tr_tl=en&_x_tr_hl=es&_x_tr_pto=wapp

Doesn't bother me at all.

 

[Kross]

So I got to play with UMIK-1 and REW today. Quite confusing at first but this is what I got out of it.
The blue graph with dip at 100hz is with sub crossover at 80hz and phase on it set to 0.
The red one is with 100hz crossover and phase set to 180 on the sub which seem to eliminate the dip. I checked if bumping up crossover helped but it turned out that only changing phase to 180 helped.

I'm confused as speakers and sub are positioned at the front wall and both facing same direction.
I thought phase should be flipped to 180 if the sub is on the opposite wall.

Is my interpretation of the graph correct?
Anything else you spot on the red graph that requires attention?
Is the dip at 200hz normal?

 

[raindance]

Don't worry about what the phase says, the setting that removes the dip is correct. Your sub is at least 10dB too loud.

 

[ampguy]

Isn't E1 just enabling T and B settings? So not sure if E0 (flat, direct) is any different than E1 with t=0 and b=0, though someone might know different?

 

[TonyJZX]

good lord $2k speakers on a $200 amp

however saying that i reckon since this known science i reckon that even if you do want to spend $1k on an appropriate amp, say a Audiofonics 150w ncore thing and a digital preamp, you arent going to get leaps and bounds over the $200 a300.

I'd love to see how you go with a more expensive amp.

going by the specs:

NOMINAL IMPEDANCE
4 Ω (min. 3.2 Ω)
SENSITIVITY (2.83V/1m)
87 dB

the min. ohm at 3.2 bothers me... i feel like you need some stoutness there

i'm truly uncertain if ANY of these chinese amps really does it at 3 ohm

 

[Talisman]

What? S.M.S.L A300 is nowhere near a SOTA amp... I'm pretty sure it's the weak link in his system, especially if he's seeking for high SPLs.

Hi @VintageFlanker , this statement sounds strange coming from you, not because it can't be true, but because, as far as I know, we don't have measurements of smsl A300 or chip Infineon MA5332MS, and you generally make well-considered statements.
Do you know of any metrics I could use to evaluate this amp/chip?
I would also appreciate it because I bought a pair of Sabaj a20a 2022 which have the same chip and which I would like to use as monoblocks.
Thank you

 

[Talisman]

good lord $2k speakers on a $200 amp

This paradigm does not exist.
the audiophonics Ncore 252 hypex module costs about 400 euros and I would have no qualms about letting it drive speakers of any price as long as the power is adequate.

 

[VintageFlanker]

Hi @VintageFlanker , this statement sounds strange coming from you, not because it can't be true, but because, as far as I know, we don't have measurements of smsl A300 or chip Infineon MA5332MS, and you generally make well-considered statements.
Do you know of any metrics I could use to evaluate this amp/chip?

So far, I stand behind the AP measurements (best case scenario) provided by SMSL:

A300

Foshan ShuangMuSanLin electronics Co., LTD. founded in 2009 which located in Shenzhen, China. And our branch company VMV Technology(HK) Company Limited registered in HK in 2013. Our company is specialized in the audio DAC, stereo headphone amplifier, power amplifier, and with own R&D...

www.smsl-audio.com

 

[Talisman]

So far, I stands behind the AP measurements (best case scenario) provided by SMSL:

A300

Foshan ShuangMuSanLin electronics Co., LTD. founded in 2009 which located in Shenzhen, China. And our branch company VMV Technology(HK) Company Limited registered in HK in 2013. Our company is specialized in the audio DAC, stereo headphone amplifier, power amplifier, and with own R&D...

www.smsl-audio.com

View attachment 287282
View attachment 287283View attachment 287284

I hadn't seen these graphs, the only thing that really makes me turn up my nose is the dependence on the load, we should be used to super pumped power data by now.
Thank you.

ps
Is it possible that the load dependence is reduced with bridging use? or that a different implementation of the chip and the feedback loop could significantly reduce it?