Modding SMSL Sanskrit 10th MKII/III

Just like the title says. How far did you go and what caused worthwhile differences?

All changes were by listening and checking with an oscilloscope as I don’t have any other equipment anymore. I could not detect a noticable improvement between buspower or external power with one of those popular phone chargers (SMPS).

Then I removed the best battery system so both diodes. That means this one no longer has buspower. Replacing the external power series diode for wire made the DAC getting 5V instead of 4.7V. This together with a linear 5V PSU connected via the Lemo connector made the most significant improvement till now. Power is clearly cleaner.

Replacing the original unmarked opamp for OPA1642 resulted in a small but noticable improvement.
 

Attachments

  • FF1F3A9A-D977-45EF-BFA4-E9A8BAC9EEB6.jpeg
    FF1F3A9A-D977-45EF-BFA4-E9A8BAC9EEB6.jpeg
    303.6 KB · Views: 726
  • 83F6E5D0-847D-46C3-841F-CA011D7F6976.jpeg
    83F6E5D0-847D-46C3-841F-CA011D7F6976.jpeg
    417.8 KB · Views: 624
Last edited:
Just like the title says. How far did you go and what caused worthwhile differences?

All changes were by listening and checking with an oscilloscope as I don’t have any other equipment anymore. I could not detect a noticable improvement between buspower or external power with one of those popular phone chargers (SMPS).

Then I removed the best battery system so both diodes. That means this one no longer has buspower. Replacing the external power series diode for wire made the DAC getting 5V instead of 4.7V. This together with a linear 5V PSU connected via the Lemo connector made the most significant improvement till now. Power is clearly cleaner.

Replacing the original unmarked opamp for OPA1642 resulted in a small but noticable improvement.
Last week I compared OPA1641 and ADA4610 in Walt Jung regulators and definitely ADA is a better op amp for that purpose, it is possible that it is the same with audio signals. I think that test would definitely be worth it.
 
That was in Jung regulators which seems a different application but I will check ADA4610. Did you see table 1 in the datasheet? It contains a few old friends 🙂

https://www.analog.com/media/en/technical-documentation/data-sheets/ada4610-1_4610-2_4610-4.pdf

I lack the time this week to do further research on the Sanskrit....all I can say is that the microUSB connector is 666. Why oh why did they choose this old fashioned and unreliable connector? I haven't seen a MKIII yet but that one has USB C which is way better. Unfortunately MKIII seems to have the infamous muting IC that breaks down often in other Topping/SMSL products.
 
Last edited:
I saw table 1 in DS long time before, certainly an interesting company.
Yesterday I received the ADA4625, which I intend to test in the next few days as an I/V stage, I doubt that it will be better than the AD811, but if I don't try it I won't know. It will probably end up in the regulator afterwards🙂

They had to save a little money, that's why they put an irritating USB connector, nowadays it's normal in everything. Since the EU passed the new USB law, only USB-C is allowed, so there is no more banter😆
 
Just disassembled a defective 10th MKII (not mine). It had a defective AK4493 with crackling left channel probably because of static/too high voltage caused by an external linear PSU that had floating PE and no reference of Audio GND to PE. There was around 30V AC both on + and -. Yes I know I am supposed to speak in terms of leakage current but as it was a goner anyway I leave it at that.

The SPDIF receiver is a LC89091. As I now dissected the device I noticed there is a stepup PSU compensating the Uf of the “best battery” diodes that are between buspower and aux power. It does not care much about either 4.7 or 5V input voltage and produces 5.6V. The 5.6V is made 5V again. Right … The miracles of ChiFi are mysterious.

Also see the front cover construction which is typical for SMSL. Once one pulls the flexiprint connecting the front PCB to the main PCB there is no easy way to assemble it again.
 

Attachments

  • image 3.jpg
    image 3.jpg
    299.6 KB · Views: 302
Last edited:
  • Like
Reactions: Siberia
That was in Jung regulators which seems a different application but I will check ADA4610. Did you see table 1 in the datasheet? It contains a few old friends 🙂

https://www.analog.com/media/en/technical-documentation/data-sheets/ada4610-1_4610-2_4610-4.pdf

I lack the time this week to do further research on the Sanskrit....all I can say is that the microUSB connector is 666. Why oh why did they choose this old fashioned and unreliable connector? I haven't seen a MKIII yet but that one has USB C which is way better. Unfortunately MKIII seems to have the infamous muting IC that breaks down often in other Topping/SMSL products.
Yikes. I just ordered a MKIII. I hadn’t read anything about an “infamous muting IC that breaks down often”. Could you provide some details? I’m a newbie, and just getting interested in doing mods.
 
Hi no one noticed I noticed 🙂 but it is one of the differences between MKII and MKIII. For some reason these manufacturers choose this IC over muting transistors and it is sensitive. I know muting is done with an IC but for now it is an assumption it is the one breaking down (hence the "seems"). It breaks down in other SMSL/Topping products.

Regarding MKII: meanwhile I experimented and found out that 5V instead of 4.7V makes a difference in character of the DAC. So the protection/separation diodes for aux/USB power better are removed. Aux/USB power is a feature not an enhancement. The diodes also make that the large caps can only discharge to the circuits after the diode which might be an explanation for defective DAC chips. So I think that could be a thinking error and will probably also change the 2 x 4700 µF caps to 2 x 1000 µF ones.
 
Last edited:
Meanwhile I was sent a MKIII and it is true that it has a muting IC and of course the newer AK4493SEQ. Otherwise it does not differ much from an MKII. I did not make pictures but I replaced the LME49720 opamp for OPA1642 and found it was an improvement. It did not sound as good as the modded MKII however likely because that one has a clean linear PSU. I don't have a linear PSU with USB C plug yet and using an USB B to USB C cable did not work out.

MKIII has the texts "coa", "opt" and "usb" instead of the the "input 1" etc.
 
Yeah put a famous name on it and people will like it better. That’s how it works even if devices measure and sound excellent.

Some even prefer mediocre sounding stuff as long as they see the brand name they know and that is associated with a quality (or a feeling) in their minds.

Audio is a bizarre hobby as many listen with eyes and emotions 🙂
 
Last edited:
Well, it was a bit of shock to me how that small thingy with few mods sounds. Maybe I'll put normal USB B on it to be able to use it with good USB cable without USB B to Micro adapter. I didn't check how it measures, bought it because it was cheap enough to try and sell with small loss if it's s*it. It stayed 🙂