• 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!

PecanPi - Next Generation Raspberry Pi DAC and Streamer

orchardaudio

Addicted to Fun and Learning
Audio Company
Joined
Aug 24, 2018
Messages
853
Likes
1,242
Location
Succasunna, NJ
Meet the PecanPi DAC and Streamer

IMG_0351.jpg


IMG_0350.jpg


IMG_0356.jpg


IMG_0355.jpg
Improvements over previous generation DAC include:

  • Fully linear ultra low noise power supplies
  • Re-clocking with Crystek femtosecond jitter oscillator
  • Built in HW volume control (no Volume-Clocker needed)
  • Built in high performance headphone amplifier
  • Full size XLR connectors
  • Ability to back-power the Raspberry Pi
  • Improved thermal design
  • Aluminum case available

The PecanPi Streamer is a ready to go plug and play streamer, that uses the PecanPi DAC, and will be available with the following software options:
Specifications:
  • Balanced output voltage (XLR): 5 Vrms
  • Balanced output impedance: 200 ohms
  • Balanced Dyanmic Range: 125dB
  • Single ended output voltage (RCA): 2.5 Vrms
  • Single ended output impedance: 0.050 ohm
  • Single ended Dynamic Rance: 121dB
  • Supported sample rates: 44.1, 48, 88.2, 96, 176.4 and 192kHz
  • Supported bit rates: 16 and 24 bit
  • Headphone output power into 32 ohms: 390mW
  • Headphone output power into 16 ohms: 780mW
PecanPi Implementation Details:
  • DAC chips:
    • Dual flagship Burr-Brown PCM1794As in monaural mode
  • Digital HW Volume Control and Re-clocking:
    • Crystek CCHD-575 oscillator -- ultra low clock jitter of 82fSec
    • SRC4193 for volume control
  • Output stage: True balanced dual different output stages
    • Uses OPA1612s
    • Low Noise Panasonic Resistors
    • Proprietary filtering topology
  • Ultra low noise analog power supplies:
    • TPS7A4700 (4.17uV noise) for positive op-amp power supply.
    • TPS7A3301 (16uV noise) for negative op-amp power supply
    • TPS7A4901 (15.4uV noise) for DAC Chips
  • Headphone driver:

Preorders now avaialble:
PecanPi DAC -- 224.99 -- 25% off
https://orchardaudio.com/shop?olsPage=products/pcnp-dac

PecanPi Streamer -- 399.99 -- 20% off
https://orchardaudio.com/shop?olsPage=products/pcnp-strmr

With every preorder you will be entered into a raffle to receive our next product for free. It will be a 100W Class-D amplifier.

Will be shipping in 3 to 6 months.
 
Last edited:
OP
orchardaudio

orchardaudio

Addicted to Fun and Learning
Audio Company
Joined
Aug 24, 2018
Messages
853
Likes
1,242
Location
Succasunna, NJ
Dashboard View for Balanced (XLR) Outputs
Dashboard (XLR).JPG


Dashboard View for Single Ended (RCA) Outputs
Dashboard (RCA).JPG


XLR jitter noise and spectrum 256k point FFT, 16 averages
Jitter (XLR).JPG

RCA jitter noise and spectrum 256k point FFT, 16 averages

Jitter (RCA).JPG


Dynamic Range (XLR), no weighting
Dynamic Range (XLR).JPG


Dynamic Range (RCA), no weighting
Dynamic Range (RCA).JPG


THD+N vs Frequency (XLR)
THD+N (XLR).JPG


THD+N vs Frequency (RCA)
THD+N (RCA).JPG


IMD vs Level (XLR)
IMD (XLR).JPG


IMD vs Level (RCA)
IMD (RCA).JPG


THD+N vs Level (XLR)
THD+N vs Level (XLR).JPG


THD+N vs Level (RCA)
1551839016843.png


Frequency Response (XLR)
Frequency Response (XLR).JPG


Frequency Response (RCA)
Frequency Response (RCA).JPG


Headphone Amplifier Measurements:

Dynamic Range (32 ohm load)
Dynamic_Range_32ohm.jpg


Frequency Response (32 ohm load)
frequency_response_32ohm.jpg


IMD vs Level (32 ohm load)
IMD_vs_level_32ohm.jpg


THD+N vs Frequency @ 0dBFS (32 ohm load)
THD+N_vs_freq_32ohm(0dBFS).jpg


THD+N vs Level @ 1kHz (32 ohm load)
THD+N_vs_lvel_32ohm.jpg


Please post any measurements that you would like to see and I will try to accommodate, as you know it is not easy to interface the APx525 with the Rpi.
 
Last edited:

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
This looks very nice!

SINAD, IMD, output levels on balanced and RCA.

Is price info available?
 

la2ygoo

Member
Forum Donor
Joined
Dec 19, 2018
Messages
56
Likes
41
it is great.:D

I want to see square waves measurements in 44.1k.
 
Last edited:

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
Dashboard View for Balanced (XLR) Outputs
View attachment 23091

Dashboard View for Single Ended (RCA) Outputs
View attachment 23092

XLR jitter noise and spectrum 256k point FFT, 16 averages
View attachment 23093
RCA jitter noise and spectrum 256k point FFT, 16 averages

View attachment 23094

Dynamic Range (XLR), no weighting
View attachment 23095

Dynamic Range (RCA), no weighting
View attachment 23096

THD+N vs Frequency (XLR)
View attachment 23097

THD+N vs Frequency (RCA)
View attachment 23098

IMD vs Level (XLR)
View attachment 23101

IMD vs Level (RCA)
View attachment 23100

THD+N vs Level (XLR)
View attachment 23102

THD+N vs Level (RCA)
View attachment 23103


Frequency Response (XLR)
View attachment 23105

Frequency Response (RCA)
View attachment 23104

Please post any measurements that you would like to see and I will try to accommodate, as you know it is not easy to interface the APx525 with the Rpi.

One word is enough to describe this - impressive!
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
it is great.:D

I want to see square waves measurements in 44.1k.

What would be the point of that?

But maybe if you can show how the cut-off looks after 22kHz. Btw, can the filters be choosed or there's only one?
 

pos

Addicted to Fun and Learning
Forum Donor
Joined
Feb 13, 2018
Messages
572
Likes
718
The filter that is used has a sharp roll off.
I must be missing something but I fail to understand why most DAC chips do not target maximum attenuation at fs/2.
Here -130dB is reached at 0.546 x fs (eg 24kHz for 44.1kHz sampling rate), and looking at the graph it seems like it will be somewhere between -35dB and -13dB at fs/2
 
Last edited:

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
I must be missing something but I fail to understand why most DAC chips do not target maximum attenuation at fs/2.
Here -130dB is reached at 0.546 x fs (eg 24kHz for 44.1kHz sampling rate), and looking at the graph it seems like it will be somewhere between -35dB and -13dB at fs/2

Well, you can't have both - if you target max attenuation at 0.5fs than you don't really cover range up to 0.5fs. This filter is having -3db at 0.49fs and -130dB at 0.546fs which seems like a good compromise to me.
 

pos

Addicted to Fun and Learning
Forum Donor
Joined
Feb 13, 2018
Messages
572
Likes
718
You can have both, but it requires a longer FIR, which was probably not possible at the time the PCM1794 was produced.

0.49 x fs is 21.6kHz for the -3dB point.
If you look at the frequency response posted above I think there would still be some margin to get a flat(ish) passband up to 20kHz and reach maximum attenuation at fs/2 with the same type of filter.
Anyway, this cannot be helped as it is part of the chip :)
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
You can have both, but it requires a longer FIR, which was probably not possible at the time the PCM1794 was produced.

0.49 x fs is 21.6kHz for the -3dB point.
If you look at the frequency response posted above I think there would still be some margin to get a flat(ish) passband up to 20kHz and reach maximum attenuation at fs/2 with the same type of filter.
Anyway, this cannot be helped as it is part of the chip :)

Chip designers probably aim for flat response up to 20kHz for marketing reasons. :)
 
Top Bottom