I have replaced 4 caps and 3 inductors on the YJ 2.1 and it has worked - noise is almost as low as the YJ2.0 - nothing short of a miracle. And I reckon the two will match if I also replace the AMS1117 3.3V reg with something like an ADP7140 - it has around 50dB ripple rejection at 50kHz and 9uV noise. So, it's clear that the 3.3V supply to the STA326 must be low noise and it's possible to do this even with an LM2575.
I couldn't run the C2 cap back to the ground pin of the LM2575 - it refused to take any solder and I didn't want to have to persuade it, so I used the ground point on the reg's input decoupling ceramic cap. Good enough.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
I couldn't run the C2 cap back to the ground pin of the LM2575 - it refused to take any solder and I didn't want to have to persuade it, so I used the ground point on the reg's input decoupling ceramic cap. Good enough.
I tried to replace the AMS1117 with an ADP7160 - but the tiny little itty bit popped out of its packaging. Lost and gone forever. Thankfully, I have an almost identical part - ADP151 - so I fitted that. And that has dropped the YJ 2.1 noise down to the same level as the YJ 2.0 but the 2.1 still doesn't have an auto mute working so with no signal lock, it's the same noise as before the mods. With signal lock, I'm satisfied it's at the noise floor for the STA. And the "schlock" of getting signal lock is still there. More investigation needed to get it working at its best.
The AD151 looks like a blob of solder under a second PLE cap but that's the catenary wire connections hiding the miniscule part.
The AD151 looks like a blob of solder under a second PLE cap but that's the catenary wire connections hiding the miniscule part.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
This function is enabled or disabled by some register inside STA326 so maybe it is just turned off by the firmware.
Thanks Audeo. I don't have the tools, skills, or knowledge to edit firmware so perhaps I won't try ;-) I can live with it anyway.
Nanoloop & others using D802, D8 or STA326
Do you have EQ setting control on the amp or the remote control? what do you config to bypasses Bypass Equalizer (EQ) allows Digital Direct path (DDX) path?
STA326 manual talks about EQ ("Tone", "Normal", or "Flat", etcc)
Personal conclusion is: only "Tone" bypasses EQ allows DDX path
Any thoughts on your side?
Do you have EQ setting control on the amp or the remote control? what do you config to bypasses Bypass Equalizer (EQ) allows Digital Direct path (DDX) path?
STA326 manual talks about EQ ("Tone", "Normal", or "Flat", etcc)
Personal conclusion is: only "Tone" bypasses EQ allows DDX path
Any thoughts on your side?
The YJ has a gazillion different preset EQs such as rock, jazz, etc and also bass and treble controls and these are accessed from the remote control only. I set everything flat. There does not appear to be anyway to set a bypass. I assume that with everything flat, the data signal passes through with no audible effect but I don't know if that's the case.
Normal goes through the EQ.
Flat bypasses EQ but goes through the bass and treble control circuitry.
Tone is straight through.
It's a bit confusing as, at least here in the UK, we would intuitively understand tone and flat to be the opposite of how they are in the STA326.
STA326 manual states "flat" is actually 1 of the EQ modes; So i would say 99% it is "digitally modified", non-DDX.
For D8 there is a "normal", and "Tone" .... and my guess of "tone" is now affirmed with page 7 D8 Manual , which reads:
"D8 support 14 EQ, include Normal, Flat, Rock, Soft-rock, Jazz, Classical, Dance, Pop, Soft, Hard, Party, Vocal, Hip-hop and Dialog..........when the mode is Normal, EQ effect is ignored, but the Bass and Treble is effective; Other EQ mode, EQ, Bass and Treble are effective...........
In the Tone mode, EQ, Bass and Treble are invalid, and the sound is
soundtrack."
Thk gwingn... confusion cleared
Felt fortunate bought D8; as not just remote, also on D8 menu can directly config to "Tone" 😀
For D8 there is a "normal", and "Tone" .... and my guess of "tone" is now affirmed with page 7 D8 Manual , which reads:
"D8 support 14 EQ, include Normal, Flat, Rock, Soft-rock, Jazz, Classical, Dance, Pop, Soft, Hard, Party, Vocal, Hip-hop and Dialog..........when the mode is Normal, EQ effect is ignored, but the Bass and Treble is effective; Other EQ mode, EQ, Bass and Treble are effective...........
In the Tone mode, EQ, Bass and Treble are invalid, and the sound is
soundtrack."
Thk gwingn... confusion cleared
Felt fortunate bought D8; as not just remote, also on D8 menu can directly config to "Tone" 😀
Last edited:
Yes, the D8 is more sophisticated 🙂 I think the YJs do not have a normal or tone setting but I should fully investigate. I've mostly been just listening and tinkering and just found yesterday it can receive 192kHz 24 bit when before it wouldn't do over 48kHz into spdif. Odd.
After some experimenting, I found for sure the YJ does not have a tone or normal setting.
I tried 3 different psu's - AMB Sigma11 32V 4.5A, LT1083 30V 4A, and a cheap smps brick 24V 5A. They made no difference to the background noise level with signal lock - on 96dB/W Klipsch, the noise vanishes around 30-40cm from the speaker. They do make a direct change to noise level with no signal lock on the YJ 2.1 that does not have auto mute. Similarly, gut punching bass is more gut punchy with the AMB so the main voltage supply benefits from lower impedance. Finally, it seems the 2.1 is more sensitive to its power supply than the 2.0. The AMB is definitely the best partner for the 2.1.
I also tested the roll off created by the 2.1 crossover filter. As per the datasheet, the sub channel seems to have a decent 12dB filter which can be set by one button on the remote - every press is 20Hz higher until 360Hz goes back down to 80Hz.
The satellite channels also have a filter that becomes noticeable at around 200Hz and becomes more significant around 100Hz. This is with the filter set to 80Hz into 4 ohm speakers. I assume this behaviour is because the output of the SE channels goes to ground through a 220uF cap and that forms a filter with a pole at 180Hz and 4 ohms.
Unfortunately, the speakers are more efficient than the sub so the sound is a little too bass light. I will need to set this up when I finalise the speakers I will use with this amp.
I still have more mod plans for the YJ 2.1 - it works well with a Bluetooth APTX-HD spdif source so I will fit one inside the case and dump all the usb stuff which I won't ever use on this amp. Removing the tcxo and PCM270x will reduce current demand to accommodate the CSR8675 so that shouldn't create any issues at all and I can feed the CSR's spdif into the appropriate big blue cap. Piece of cake. 🙂
I tried 3 different psu's - AMB Sigma11 32V 4.5A, LT1083 30V 4A, and a cheap smps brick 24V 5A. They made no difference to the background noise level with signal lock - on 96dB/W Klipsch, the noise vanishes around 30-40cm from the speaker. They do make a direct change to noise level with no signal lock on the YJ 2.1 that does not have auto mute. Similarly, gut punching bass is more gut punchy with the AMB so the main voltage supply benefits from lower impedance. Finally, it seems the 2.1 is more sensitive to its power supply than the 2.0. The AMB is definitely the best partner for the 2.1.
I also tested the roll off created by the 2.1 crossover filter. As per the datasheet, the sub channel seems to have a decent 12dB filter which can be set by one button on the remote - every press is 20Hz higher until 360Hz goes back down to 80Hz.
The satellite channels also have a filter that becomes noticeable at around 200Hz and becomes more significant around 100Hz. This is with the filter set to 80Hz into 4 ohm speakers. I assume this behaviour is because the output of the SE channels goes to ground through a 220uF cap and that forms a filter with a pole at 180Hz and 4 ohms.
Unfortunately, the speakers are more efficient than the sub so the sound is a little too bass light. I will need to set this up when I finalise the speakers I will use with this amp.
I still have more mod plans for the YJ 2.1 - it works well with a Bluetooth APTX-HD spdif source so I will fit one inside the case and dump all the usb stuff which I won't ever use on this amp. Removing the tcxo and PCM270x will reduce current demand to accommodate the CSR8675 so that shouldn't create any issues at all and I can feed the CSR's spdif into the appropriate big blue cap. Piece of cake. 🙂
- Status
- Not open for further replies.