Computer Audio Blasphemy, nige tries Another SD Card Player

[nige2000]

The code for the screen is hardly too taxing?
Kinda like the idea of having the ability to hook up a screen to help with debugging if ever required

If starting anew with a pcb design buy the chips caps resistors etc
A blank board could be 5 to 10 usd depending on quantities
Components could be 15 usd plus
Plus programming
Plus pcb assembly

Seems like a bit of hassle
Maybe if there was extra functionality night be a runner

Difference between flac and wav is small
Not a deal breaker by any means but it's also free and easy to implement

[nige2000]

Or there's this offering for $91,181 (yes 91 thousand dollars)
https://www.lessloss.com/laminar-stream ... p-207.html

Here's what they say:

The all-new Laminar Streamer finally solves the problem of digital audio harshness by using only a single clock source for the formation of the pristine digital audio stream. Although the device plays .wav and .aiff files off of SD cards, LessLoss did not incorporate the use of Windows, MAC OS, or Linux for its operating system. Instead, from ground up, a dedicated audiophile operating system was built, streamlined for the specialized task of serving the topmost quality data while avoiding jitter. Unlike any other computer audio playback system, the Laminar Streamer's operating system runs synchronously at the audio sampling rate via the very same high quality audio clock used for the formation of the audio data stream. Thus, a new level of quality is achieved in digital playback, previously unattainable due to the inherent complexity of running an OS clock and an audio clock separately. The intricacies of asynchronous re-clocking, buffering, and jitter reduction have been solved at their very source, a world first for digital audio!

Obviously what costs $91,143 more than the $38 dollar card is this :)

"Microvibration Control
Superlative microvibration control through precision milled solid steel and Panzerholz construction.

I think it's great if they can sell them
Can't imagine they sell many though

[jkeny]

The code for the screen is hardly too taxing?
Kinda like the idea of having the ability to hook up a screen to help with debugging if ever required

If starting anew with a pcb design buy the chips caps resistors etc
A blank board could be 5 to 10 usd depending on quantities
Components could be 15 usd plus
Plus programming
Plus pcb assembly

Seems like a bit of hassle
Maybe if there was extra functionality night be a runner

Yea, like a lot of stuff - when you add up the time & materials cost for doing it yourself, it's hard to beat the prices of the Chines offerings, if they are properly done

Difference between flac and wav is small
Not a deal breaker by any means but it's also free and easy to implement

Sure!

[randytsuch]

the ams1117 reg says it takes up to 15v
so getting going shouldnt be difficult

Except I noticed this morning that the cap on the ams1117's input is only rated for 6v. Safest thing would be to remove that, and install a higher voltage cap. Not sure if its really required, sometimes regulators can be unstable without the caps, and depends on the quality of what you're feeding the reg.
I'll look in my junk drawer, and see what I have that I can use.

Randy

[nige2000]

the ams1117 reg says it takes up to 15v
so getting going shouldnt be difficult

Except I noticed this morning that the cap on the ams1117's input is only rated for 6v. Safest thing would be to remove that, and install a higher voltage cap. Not sure if its really required, sometimes regulators can be unstable without the caps, and depends on the quality of what you're feeding the reg.
I'll look in my junk drawer, and see what I have that I can use.

Randy

Oops I seen that after... meant to post it
U could put a cell on a switch in parallel with the 3.3v rail of the sd player
The powered on board reg would keep it charged
As long as you don't forget to turn it off or connect a discharged cell to overwhelm it

[randytsuch]

The switches on the front panel are 3.3v input
Would a ss signal relay work better?
Maybe we don't need any isolation?

I'm just figuring the ESP's are probably noisy, so it would be better to isolate. But maybe we don't? No idea if the noise will come though from the ESP.
But I already have the opto's, so it pretty easy for me to add them. So I prefer to do it, and not worry about it later lol.

The GPIO pins on an ESP can source 12ma and sink 20ma.
If required, you can also add a gpio expander chip that can handle higher current
I've used PCF8574's and they can sink 50ma.

So you can use any relay, ssr or opto you want as long as sinking 20ma will drive it closed.

I was having trouble finding relays that would work with 20ma at 3v. They want closer to 50ma at 3v.

I just was looking at someone's SSR, but it was basically same as an opto.

Randy

[Fran]

Mine arrived today, but I won't get to play with it for at a week....

[nige2000]

Ah you got a remote, that be handy for a while till the ui is finished

[Beresford]

Is this card the same as one mentioned previously? About 54usd.
https://item.taobao.com/item.htm?spm=20 ... 87e11f6f7f

(Fascinated by this thread, I suspect like many others who have not yet learnt to solder properly)

[jkeny]

Is this card the same as one mentioned previously? About 54usd.
https://item.taobao.com/item.htm?spm=20 ... 87e11f6f7f

(Fascinated by this thread, I suspect like many others who have not yet learnt to solder properly)

Good spotting, Beresford
It may well be the same but what concerns me is that there are 3 clocks on the board - the secret sauce in the original SD card is that it uses the audio clocks as the master/system clock for the ARM processor - this cuts out the issues inherent in crossing clock domains.

In this pic of the board there is a AK4118 chip which seems to take the I2S output from the ARM & converts it to SPDIF output (the RCA connector & Pulse transformer are part of this)

Now the question is - are the two audio clocks used by the ARM & the 3rd clock used by the AK4118 or is it the other way around (one system clock used by the ARM & the two audio clocks used by the AK4118)? I suspect from the physical placement on the board that it is the later - the two audio clocks are close to the AK4118 chip