• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

SMSL SU-1 Stereo DAC Review

Rate this DAC:

  • 1. Poor (headless panther)

    Votes: 11 2.1%
  • 2. Not terrible (postman panther)

    Votes: 11 2.1%
  • 3. Fine (happy panther)

    Votes: 64 12.5%
  • 4. Great (golfing panther)

    Votes: 427 83.2%

  • Total voters
    513
did you add that muses 02 op amp?
I can see the opamp change, plus dac chip reference voltage filter caps changes. I'd recommend opa adsitional 100-470uf power filter caps as well, probably these would be the biggest improvememnt overall.
 
I can see the opamp change, plus dac chip reference voltage filter caps changes. I'd recommend opa adsitional 100-470uf power filter caps as well, probably these would be the biggest improvememnt overall.
You can replace the capacitors at the op amp, but a large value of 100-470. µF is too much, 0.5 - 10 µF is enough.

100-470µF was a good option in the 80s, when stages (signal and channels) were separated by resistors.
 
You can replace the capacitors at the op amp, but a large value of 100-470. µF is too much, 0.5 - 10 µF is enough.

100-470µF was a good option in the 80s, when stages (signal and channels) were separated by resistors.
Sorry, I forgot to mention I mean the small +-12V SMPS module output ripple filtering, increase power clarity on OPA not about channel separation purposes. PSRR is decreasing rapidly w/ frequency even for high-grade OPAs.
 
Sorry, I forgot to mention I mean the small +-12V SMPS module output ripple filtering, increase power clarity on OPA not about channel separation purposes. PSRR is decreasing rapidly w/ frequency even for high-grade OPAs.
Which module on the board is the +-12V SMPS? Can you mark where to solder the additional capacitors? Aren't there already 2 pieces of 220uF 16V?
 
Thanks. Soldering two additional capacitors in this place with the soic->dip8 base already soldered is quite a challenge. I would have to desolder it with a hot air station and then try to solder the capacitors.

But I have another great idea… to solder two capacitors between the +-12V measuring fields visible on the left and ground. There is a lot of free space there.
 
Thanks. Soldering two additional capacitors in this place with the soic->dip8 base already soldered is quite a challenge. I would have to desolder it with a hot air station and then try to solder the capacitors.

But I have another great idea… to solder two capacitors between the +-12V measuring fields visible on the left and ground. There is a lot of free space there.
Yep, no space w/ dip8 adaptor indeed. Giod idea to find another place, as well use good quality small factor low esr caps (because smps high switching freq).
 
Yep, no space w/ dip8 adaptor indeed. Giod idea to find another place, as well use good quality small factor low esr caps (because smps high switching freq).
I have free 270uF 16V oscon's. They will be good in this place.

When I wrote earlier that there are 2 capacitors 220uF 16V I confused them with SMSL PS200. SMSL put these capacitors in it on the +-12V line. In SU-1 there are only those at the operational amplifier itself.
IMG_9886.jpeg
 
I have free 270uF 16V oscon's. They will be good in this place.

When I wrote earlier that there are 2 capacitors 220uF 16V I confused them with SMSL PS200. SMSL put these capacitors in it on the +-12V line. In SU-1 there are only those at the operational amplifier itself.
View attachment 418256
Same smps module imho, but strange why these ripple filtering caps missing - or using smaller ones in from su-1.
 
Last edited:
Same smps module imho, but strange why these ripple filtering caps missing - or using smaller ones in from su-1.
Mass taken from the pad from the bottom of the board. It's not the best solution but it works.
IMG_9987.jpeg
 
It's not the best solution but it works.

What do you mean by that? That the device is still functioning or that there is an improvement? If the latter, can you show a before/after comparison? What parameter did improve?
 
What do you mean by that? That the device is still functioning or that there is an improvement? If the latter, can you show a before/after comparison? What parameter did improve?
Everything you wrote is correct. The ground should be connected as close as possible, but my solution also works and there is progress in SQ. Do you know that the comparison you are writing about is not feasible? I need to have another identical device from before the modification and perform an A/B test in the same conditions. Do you expect an improvement in sound measurements? I doubt that anything will be visible in the measurements. At most, you can measure the voltage ripple on the +-12V line with an oscilloscope. Why does no linear power supply manufacturer measure how its power supply affects sound measurements.
 
Do you know that the comparison you are writing about is not feasible? I need to have another identical device from before the modification and perform an A/B test in the same conditions.

You could measure the noise level at the DAC output. Expect something in the low microvolts. It would be interesting to see whether the noise level changes. If you have an analyser of sorts (or an oscilloscope with FFT) you could also look at the noise spectrum. Perhaps your caps manage to knock down that litte mains hum spike:

1736239948819.png


I doubt you'd be able to hear that, though. With very sensitive headphones perhaps...
 
You could measure the noise level at the DAC output. Expect something in the low microvolts. It would be interesting to see whether the noise level changes. If you have an analyser of sorts (or an oscilloscope with FFT) you could also look at the noise spectrum. Perhaps your caps manage to knock down that litte mains hum spike:

View attachment 419291

I doubt you'd be able to hear that, though. With very sensitive headphones perhaps...
Sure, I have a Fnirsi 1014D oscilloscope with some kind of FFT analysis, which is useless. On the other hand, a Focusrite Scarlet 2i2 3rd generation audio interface, which has AD sampling parameters worse than SU-1, so I won't make any measurements or they won't be reliable.
 
Sorry if this has been addressed already but I've had an SU-1 for a few months now and was running it with an older USB driver from my SMSL do-100 dac. It seemed to work fine. I found after I had uninstalled that driver that Windows 10 no longer needed a driver for full us of the DAC to 32/384, so I used it that way. After a while I found myself complaining about how terrible many recordings were. Very distorted in the high frequencies especially when pushed hard.
Today I decided to download the latest SMSL driver to give it a try and wow, what a difference! Anyway, the reason I'm posting is to let people know that even though Windows 10 doesn't need a driver for the SU-1, you should download it from SMSL and give it a try. I think this may be the reason some people have said that the DAC sounds much better using the SPDIF input instead of USB. Perhaps they didn't install the driver, because I noticed a big difference.
 
What are the driver’s settings you’re using?

I tried out that S.M.S.L USB driver a while ago and wasn’t impressed much.
 
I tried many driver settings but once I changed to the Windows 10 driver, there were no settings. With the new SMSL driver (5.70), its excellent but I haven't gone in to edit the file for the XMOS yet, so it's stock.
 
After a while I found myself complaining about how terrible many recordings were. Very distorted in the high frequencies especially when pushed hard.
Today I decided to download the latest SMSL driver to give it a try and wow, what a difference!
TBH, this was likely confirmation bias.

All of Amir's measurements were performed with the default Windows driver, and match SMSL's official best-case measurements of the SU-1.
 
Back
Top Bottom