Shenzhen/SMSL RAW-HA1 Amplifier Review

[amirm]

This is a review and detailed measurements of the Shenzhenaudio SMSL RAW-HA1 integrated amplifier with DAC and headphone output. It was sent to me by the Shenzhenaudio and is on sale for US $212.49.

The user interface is much updated and more elegant than the older SMSL UI. It is easy to use even without the included remote control. Back panel shows the rich functionality of this amp:

For starters, we have included power supply which is not common in this class product. HDMI ARC is included which I unfortunately can't yet test. Bulti-in CS DAC has the usual inputs. The amplifier is based on Infineon IC. Let's see how they all perform together.

SMSL RAW-HA1 Amplifier Measurements
Let's start with using Line In (called AUX in the UI):

Switching to Toslink digital input shows no difference, indicating that the DAC is superior to the amplifier (as it should be):

Performance is good enough to beat the average for all amplifiers tested:

Noise test shows the same:

Here is how it rates at max power:

Frequency response unfortunately shows some amount of load dependency:

Strange to see a flat line for crosstalk, indicating some kind of resistive connection between channels:

Multitone test shows the class rise in distortion at higher frequencies:

Which then reflects into 19 and 20 kHz IMD test:

Let's see the power level at 4 ohm:

And at 8 ohm:

"FTC" like power sweep caused the amp to shutdown below 30 Hz so I am showing everything above that:

That is reflected in our power test vs frequency:

The unit has impressive ability -- at least in short term -- to drive even 2 ohm load:

Finally, the amp needs a bit of time to get its best performance:

It actually improved a bit more towards the end of my tests, causing me to go back and re-run the dashboard test.

The top of the unit had a small warm area but nothing of concern so I did not shoot my thermal image.

Pop noise is impressively low:

Headphone output is anemic for a desktop product:

Conclusions
The RAW-HA1 has impressive functionality with compact packaging with a lot of features and built-in power supply. The user interface is attractive. Considering the price, it is a very compelling offering. Objective performance is "good." Wish list for next version is lack of load variability and higher power headphone output.

I am going to recommend the RAW-HA1 integrated amplifier and DAC.
-----------
As always, questions, comments, recommendations, etc. are welcome.

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/

 

[amirm]

Reserved for @AdamG to kindly post the specs.

Manufacturer Specifications:

 

[alexanderino]

Many thanks for an informative review as usual, @amirm.

If you still have the unit, would you mind testing it for occasional and regular low-level popping when used via USB without the XMOS driver (that plagues some SMSL products and may indicate incomplete UAC2 compliance) as seen in my video?

Some notes:

• Only affects USB input (on Windows, Linux, Mac)
• After powering on, two pops are heard. Starting playback results in two more pops and a loss of the start of that track. Stopping and waiting a few seconds results in two pops, after which there will be two more pops upon playback
• The 'fix' exists for Windows only, by installing the XMOS driver, editing the XML config file to expose more Control Panel tabs, then changing the Streaming option from 'On when Needed' to 'Always On'
• No Mac or Linux XMOS driver & Control Panel, hence no such fix for those OSes
• This setting is not stored inside the unit
• Audio captured with a MOTU UltraLite mk5. The only processing performed is normalisation.

 

[amirm]

I assume it is audible over headphone as well?

 

[alexanderino]

Yes.

 

[JSmith]

US $212.49

Certainly well priced, so one can't expect SOTA... cheers for the testing.

Pics;

JSmith

 

[restorer-john]

Great review @amirm.

It looks to be a good value, reasonable performance, featured and compact little unit. It also looks well made and would probably do well in a 2nd system or less critical games/bedroom system.

I think you'd get your money's worth out of it.

 

[dtaylo1066]

For 212 clams it is pretty hard to complain about this one. Damn nice little unit for a dorm room or study. Can't go wrong with it.

I think I would spend the extra money and get a Wiim.

 

[Bruce Morgen]

The UI and feature set are impressive. That said, the amp section is not significantly better at 8 ohms than the now-elderly and still dirt-cheap Aiyima A07, and it has similar load dependence. In 2024, one would think PFFB -- not to mention a respectably beefy headphone output -- would be mandatory for such products, but what do I know?....

 

[Doodski]

@amirm. In the power test for 4 Ohms and 8 Ohms how is clipping determined? I see it around -60dB but they are both different. Is it a slope reference? How do you or the AP determine clipping?

 

[amirm]

@amirm. In the power test for 4 Ohms and 8 Ohms how is clipping determined? I see it around -60dB but they are both different. Is it a slope reference? How do you or the AP determine clipping?

I try to find the point where the hockey stick shoots up. There is not a ton of resolution (points) there so the assessment is approximate. This is why I follow up with the 1% THD test. In that graph, I could have picked the point below it.

 

[Doodski]

I try to find the point where the hockey stick shoots up. There is not a ton of resolution (points) there so the assessment is approximate. This is why I follow up with the 1% THD test. In that graph, I could have picked the point below it.

In the 1% THD graph are you seeking a specific power reduction point on the graph or is it the AP determining this? I'm accustomed to manually using a O-scope and frequency sweeping for hard clipping and so this AP metering method and estimating the hard clipping point is new to me as per the rational and methodology used. I'm trying to wrap my head around it.

 

[GWolfman]

Thanks Amir!

The popping reported here (https://www.audiosciencereview.com/...l-raw-ha1-amplifier-review.58162/post-2126940) as seen in other SMSL units would drive me nuts! Is this found in other brands as well?

 

[restorer-john]

In the 1% THD graph are you seeking a specific power reduction point on the graph or is it the AP determining this? I'm accustomed to manually using a O-scope and frequency sweeping for hard clipping and so this AP metering method and estimating the hard clipping point is new to me as per the rational and methodology used. I'm trying to wrap my head around it.

Amir has picked the 1kHz 1% power and then swept it, to see how it holds up across the spectrum.

It's basically a full power bandwidth test or similar.

 

[Doodski]

Amir has picked the 1kHz 1% power and then swept it, to see how it holds up across the spectrum.

It's basically a full power bandwidth test or similar.

IC. I'm not familiar with the automation of the AP and have no idea what the operator input is or if manual operation is possible.

 

[pma]

So we may expect exactly this output impedance plot:

https://www.audiosciencereview.com/forum/index.php?attachments/lc_outimp-png.105654/

 

[WOLFAGRAM]

Performance is good enough to beat the average for all amplifiers tested:

It probably won‘t beat the mighty YAQIN in the number of „funny“ comments it generates.

Cheers.

 

[kemmler3D]

Lots of features for the money but none of them rate higher than "alright" compared to alternatives. For me this would be on the list as an option... if I needed a full DAC/AMP setup, but not as my primary system.

 

[amirm]

In the 1% THD graph are you seeking a specific power reduction point on the graph or is it the AP determining this?

That is a special mode in AP where it automatically sets the generator to parameters I give it, namely 1% THD at 1 kHz. Once there, I use that to do a frequency response graph. The voltage value is computed iteratively by AP much like it is used for the two power levels at 1% THD. It searches up and down from parameters you give it until it hones in on the target value. You could do the same manually but it of course is much faster this way.

 

[pma]

Is USB input isolated? (it is a class I safety device)

A test result with grounded digital source (PC) would be of much interest of potential customers to prevent a surprise from clicks and buzz.