For me, downsampling from 24 to 16 bits with Volumio sounds worse than simply playing my highres files at their native bitrate and allowing 1387 to truncate. Its easy to try both and see which you prefer. I think minimizing CPU load on RPi is a benefit. Not sure how it is with other transports. You will definitely want to leave the bitrate up at the native (high) rate. TDA1387 handles 192K beautifully.
1387 sounds excellent playing highres material, in my opinion. I was surprised that it bettered some other DACs I tried with true 24-bit decoding. The treble droop correction will be over-correcting, but it should be very subtle.
Last edited:
Thanks Kevin! I will just push the play button and forget evereything!
Hi George - good to see you again!
In regards to the cap on pin7, I haven't found a cap that makes it sound better than none. Either I hear no change (with 'lytics and film caps) or it seems to get worse (using classII ceramics like X5R). Hence parsimony requires no caps are included for the time being. But by all means experiment and report back.
How I wired my JLsounds I2SoverUSB v.III board to my PhiHex dac
(thanks to help from member Rick https://www.diyaudio.com/forums/members/r8833.html )
From all the v.III board jumpers I bridged only B1 and B3
On v.III board there is a 20 pin header (H3)
(H3) pin 11 (BCLK_I2S) goes to CLK on PhiDAC
(H3) pin 13(DATA_I2S) goes to DATA on PhiDAC
(H3) pin 15(LR_CLK_I2S) goes to WS on PhiDAC
(H3) pins 12, 14, 16 (GND) goes to GND on PhiDAC
George
(thanks to help from member Rick https://www.diyaudio.com/forums/members/r8833.html )
From all the v.III board jumpers I bridged only B1 and B3
On v.III board there is a 20 pin header (H3)
(H3) pin 11 (BCLK_I2S) goes to CLK on PhiDAC
(H3) pin 13(DATA_I2S) goes to DATA on PhiDAC
(H3) pin 15(LR_CLK_I2S) goes to WS on PhiDAC
(H3) pins 12, 14, 16 (GND) goes to GND on PhiDAC
George
Thank you Richard.
It’s an interesting exercise. If I will do it I will report here.
George
>Edit. I wonder what is the real function of pin 7. Is it to reference the internal 'REFERENCE SOURCE' to ground?
Data sheet says Vref (output reference voltage) is ~ 1/6 of VDD and Rref (output resistance at pin 7) is btn 7.6kOhm and 14.8kOhm
Hi Bjarne
Oh, it’s now I noticed that you are using an external oscillator while I use the on board oscillator.
I use the settings as in 1st attachment (page 4 of manual)
I am of the opinion that you would have to use the settings as in the second attachment (page 5 of manual)
Try it and see what you get (you won’t brake anything). Report back
George
Oh, it’s now I noticed that you are using an external oscillator while I use the on board oscillator.
I use the settings as in 1st attachment (page 4 of manual)
I am of the opinion that you would have to use the settings as in the second attachment (page 5 of manual)
Try it and see what you get (you won’t brake anything). Report back
George
Attachments
I recently compared many different caps for Pin 7 on my 1387 DAC (Matt Garman’s v1.6 HAT). This cap (or lack therof) has a significant impact on the sound. I tried everything I could think of, from 3300uF electrolytics (which made output unstable, and bloated the bass) all the way down to 0.1uF polypropylene (which gave a subtle lift in the treble that ultimately annoyed me).
I eventually decided that the Philips engineers got it right when they chose the datasheet value of 1uF. I still think that is best to my ears, but I will have to listen again with no cap in there at all. I did try that several times, and from memory, I found it sounded very clear and open that way, but I think I was just biased against leaving an empty space on my board!
I guess I also had a hard time accepting that leaving it floating could be the best option. That being said, I often felt like every cap I tried in there just seemed to make things worse in one way or another. Forget about polyester, polyphenylene sulphide, and the newfangled acrylics (I tried them all!). But I did find that when I used very good film caps on Pin 7, I could get a slightly smoother/warmer tonal balance while retaining the clarity and focus of no cap. For a long while I was convinced that I could do no better than a 1uF Jantzen Standard (mounted like this to minimize lead length).
Then just this month, on a friend’s recommendation, I tried a 1uF WIMA MKP2. These are the smallest 1uF PP caps I have ever seen, with a 5mm pitch, and impossible to buy in the USA, so I ordered from this Taiwanese seller. It is very neutral—actually very hard to distinguish from no cap at all. But I think it lends a little bit of dynamic authority and healthy tonal warmth that sounds natural to me and works well in my system. WIMA makes them with steel leads. I haven’t tried this tweak, but I might.
Again, I need to listen once more with no cap and an open mind. Richard, now that I know that you like it that way, I think I'll feel less sheepish with an empty spot on my board.
None of those caps are too big in electrical terms.
Kevin - your eBay link wouldn't work for me, even when I searched eBay for 1uF MKP2 I couldn't find anything approaching 1uF in 5mm pitch with PP dielectric. So count me intrigued as to the provenance of your caps Over on Taobao PP 1uF seems to come only in 22.5mm size. I shall continue looking...
Kevin - your eBay link wouldn't work for me, even when I searched eBay for 1uF MKP2 I couldn't find anything approaching 1uF in 5mm pitch with PP dielectric. So count me intrigued as to the provenance of your caps
Over on Taobao PP 1uF seems to come only in 22.5mm size. I shall continue looking...
At long last! She sings!
EDIT: I have to mention, this is the DECA with 7th order filter. I have heard the hexa too but with 3rd order filter. They couldn't be further apart.
I'm part of the group buy from India by the way.
I'm using this product as my input.
W-Input – Compact digital input board – nihtila.com
It gets Coax in from a Gustard U12
Man.. Once you pair this thing with a DDC, it sounds nothing like it's USB counterpart.
USB version had an evident lower end emphasis and some of us even claimed it sounded "smooth".
With this, that R2R treble haze isn't there, none of the low end emphasis either.
The differences slowly become evident.
Normally I'd be able to imagine a soundstage in front of me between the speakers with my previous DACs.
That's gone.
Speakers actually vanished.. it's taking me time to understand and appreciate this product.
This DAC is surprisingly neutral.
It's not trying to wow with detail retrieval or texture or any of that.. Everything becomes so seamless and effortless that you might take it for granted.
It makes everything seem so easy that it's hard to be impressed. More impressions to follow soon
EDIT: I have to mention, this is the DECA with 7th order filter. I have heard the hexa too but with 3rd order filter. They couldn't be further apart.
I'm part of the group buy from India by the way.
I'm using this product as my input.
W-Input – Compact digital input board – nihtila.com
It gets Coax in from a Gustard U12
An externally hosted image should be here but it was not working when we last tested it.
Man.. Once you pair this thing with a DDC, it sounds nothing like it's USB counterpart.
USB version had an evident lower end emphasis and some of us even claimed it sounded "smooth".
With this, that R2R treble haze isn't there, none of the low end emphasis either.
The differences slowly become evident.
Normally I'd be able to imagine a soundstage in front of me between the speakers with my previous DACs.
That's gone.
Speakers actually vanished.. it's taking me time to understand and appreciate this product.
This DAC is surprisingly neutral.
It's not trying to wow with detail retrieval or texture or any of that.. Everything becomes so seamless and effortless that you might take it for granted.
It makes everything seem so easy that it's hard to be impressed. More impressions to follow soon