I borrowed the latest PS Audio Stellar DAC from a dealer - my thinking was that it is hard to find a DAC which better complies to PS Audio HDMI standard than a PS Audio DAC.
I have all S1 switches all in the OFF position (default - PS Audio standard), J7 removed (I connected two separate 3.3V PSUs to HDMI Pro II and Q7II boards - after switching ON all 4 PSUs, all lights power up on both HDMI and Q7 boards, so my understanding is they all get the correct power).
And those are the LEDs with all power connected to the transport (2x 5V + 2x 3.3V = 4 PSUs in total):
The problem I have is that the D1 Power LED on the HDMI board lights up when you connect the DAC but the transport itself is completely off. The power must be literally flowing backwards via the HDMI cable from the DAC to the transport to power this LED.
Transport: OFF
DAC: ON
HDMI cable: connected
D1 power LED ... mysteriously lit.
I have all S1 switches all in the OFF position (default - PS Audio standard), J7 removed (I connected two separate 3.3V PSUs to HDMI Pro II and Q7II boards - after switching ON all 4 PSUs, all lights power up on both HDMI and Q7 boards, so my understanding is they all get the correct power).
And those are the LEDs with all power connected to the transport (2x 5V + 2x 3.3V = 4 PSUs in total):
The problem I have is that the D1 Power LED on the HDMI board lights up when you connect the DAC but the transport itself is completely off. The power must be literally flowing backwards via the HDMI cable from the DAC to the transport to power this LED.
Transport: OFF
DAC: ON
HDMI cable: connected
D1 power LED ... mysteriously lit.
Last edited:
I've had the same thing happen but my understanding is it's basically harmless. When I reached out to the maker of the pi2aes I believe he said as much. Never caused any damage to that or the hdmiPi for me
I have asked Ian 2 questions regarding the R15 resistor removal:
1. if left “as is”, with the R15 resistor in place, can it be harmful to the DAC or HDMI Pro board?
Ian: 5V power from your DAC may pollute the clean 3.3V of FifoPi
2. will removing the R15 resistor limit the funcionality / compatibility of the HDMI Pro board?
Ian: No
I hope this info will make way into the next revision of HDMI Pro II manual.
BTW - which resistor have you removed? The big one circled red? There are markings next to multiple resistors, but sadly - no R15 in sight Ian is reffering to.
Board in high res:
I never removed a resistor. When I had a PS Audio DAC I noticed the same behavior as you, but everything sounded fine. I was/am using the hdmiPi mk ii (non-pro version).
OK, I left the resistor in place (for now) - it must be doing something, since it is there. Since the PS Audio DAC is just a loaner, I will make the decision up once I connect the transport to my own Lampizator Horizon DAC, which is currently getting an upgrade @ Lampizator factory (so don't have it on hand).
After crossing my fingers ... I powered everything up, ignorring the resistor issue and got the transport & PS Audio DAC working.
One caveat tho: on the recommended Audiophonics I-sabre 9028q2m driver, it works on all sample rates up to DSD128 (in DoP mode), but 44.1kHz. When playing 44.1/16 files the signal gets converted (by the transport or Volumio, as it leaves Roon as 44.1/16) to 44.1/32 and I hear nothing, but static noise. All other freq / formats - 48/32, 88.2/32 etc - work perfect. Too bad the one doesn't play is the one which is the most popular.
That said, I have quickly tried the Generic I2S DAC driver in the Volumio settings and voila - now everything works perfect!
Thanks for support guys.
Next upgrade, coming in the next 1-2 weeks is UcPure 3.3V PSU for the Q7II and HDMI boards. I'm waiting for the 3D printed brackets for the big ultra capacitors and a custom 80VA, 1x12V transformer from Toroidy.pl.
The goal of the whole project is to see how this transport compares to the Taiko Extreme server on the Lampizator Horizon DAC (Taiko connected via USB and Ian's transport via HDMI).
After crossing my fingers ... I powered everything up, ignorring the resistor issue and got the transport & PS Audio DAC working.
One caveat tho: on the recommended Audiophonics I-sabre 9028q2m driver, it works on all sample rates up to DSD128 (in DoP mode), but 44.1kHz. When playing 44.1/16 files the signal gets converted (by the transport or Volumio, as it leaves Roon as 44.1/16) to 44.1/32 and I hear nothing, but static noise. All other freq / formats - 48/32, 88.2/32 etc - work perfect. Too bad the one doesn't play is the one which is the most popular.
That said, I have quickly tried the Generic I2S DAC driver in the Volumio settings and voila - now everything works perfect!
Thanks for support guys.
Next upgrade, coming in the next 1-2 weeks is UcPure 3.3V PSU for the Q7II and HDMI boards. I'm waiting for the 3D printed brackets for the big ultra capacitors and a custom 80VA, 1x12V transformer from Toroidy.pl.
The goal of the whole project is to see how this transport compares to the Taiko Extreme server on the Lampizator Horizon DAC (Taiko connected via USB and Ian's transport via HDMI).
Last edited:
Hmm, I can't recall if I ever experienced that. Are you saying you have the Q7 converting everything to 32 bit, or that volumio is set to do that? I remember having to mess with the Q7 via the monitor pi at some point to get everything working.
It is either the Ian's transport or the Volumio - the files leave Roon as 44.1/16 and yet - the PS Audio DAC shows 44.1/32.
On the Generic I2S driver the PS Audio DAC shows the correct 44.1/16 and music is playing as it should (no static noise).
On the Generic I2S driver the PS Audio DAC shows the correct 44.1/16 and music is playing as it should (no static noise).
Sorry for the mess ..
So , if your DAC (receiver) sending +5V by the pin18 of Hdmi to the HdmiPiPro (source) it can mess with your superclean battery / supercap power supply and 22r0 resistor MUST be removed for any kind of "critical" listening.... because PS Audio don't follow their own "standard" of i2s pinout...
You can safely remove this 22R0 resistor . it is for lowering current on the pin 18 of rhe hdmi socket - BTW it is proof that PS Audio "standard" it is not standard even for PS Audio - because "normally" on pin18 +5V should be used on the "source" side only (for powering i2s hdmi receiver ...) and PS Audio using +5V pin18 on the receiver side - an I;ve checked in my HdmiPiPro - it is connected (resistor) to the pin 18 of Hdmi socket and the clean 3.3V power input.
So , if your DAC (receiver) sending +5V by the pin18 of Hdmi to the HdmiPiPro (source) it can mess with your superclean battery / supercap power supply and 22r0 resistor MUST be removed for any kind of "critical" listening.... because PS Audio don't follow their own "standard" of i2s pinout...
It seems the problem with playing 44.1/16-bit files (Red Book) still persists.
It plays 44.1/24 and ALL other freq up to DSD128, but no pure redbook 44.1/16-bit files.
I have tried that on 3 different DACs: Topping D90, PS Audio and LampizatOr Horizon. Same result on all 3.
A friend at Lampizator factory (they are local to me) has measured the Word Clock freq and says that during playback, instead of 44100 * 2 * 16 = 1.411200MHz the transport outputs ... double that! (2.8Mhz)
It kind of matches what PS Audio DAC shows, as it locks to ... 44.1/32 signal (which suggests it gets 2.8Mhz Word Clock from the transport) trying to play music (but only plays distorted signal).
I have tried Volumio in Roon Bridge mode, but then changed to Volumio internal player – with same result.
Any ideas why the transport outputs wrong word clock freq playing 44.1/16 files?
It plays 44.1/24 and ALL other freq up to DSD128, but no pure redbook 44.1/16-bit files.
I have tried that on 3 different DACs: Topping D90, PS Audio and LampizatOr Horizon. Same result on all 3.
A friend at Lampizator factory (they are local to me) has measured the Word Clock freq and says that during playback, instead of 44100 * 2 * 16 = 1.411200MHz the transport outputs ... double that! (2.8Mhz)
It kind of matches what PS Audio DAC shows, as it locks to ... 44.1/32 signal (which suggests it gets 2.8Mhz Word Clock from the transport) trying to play music (but only plays distorted signal).
I have tried Volumio in Roon Bridge mode, but then changed to Volumio internal player – with same result.
Any ideas why the transport outputs wrong word clock freq playing 44.1/16 files?
OK, it seems this seems to be a common problem, described in the manual (silly me).
You need to enable a feature in the Q7 to extend a bit depth from 16 to 32-bit (this is a lossless process, as it basicly adds 0s).
You need a MonitorPi for that - thankfully, I have one on order with UcPure parts - should have them tomorrow or a day after tomorrow. Will report back.
You need to enable a feature in the Q7 to extend a bit depth from 16 to 32-bit (this is a lossless process, as it basicly adds 0s).
You need a MonitorPi for that - thankfully, I have one on order with UcPure parts - should have them tomorrow or a day after tomorrow. Will report back.
MonitorPi arrived - switched on the feature extending the bit depth from 16 to 32-bit and the transport is now working at all freq. and bit depths, no prob.
It seems there is a dip switch under the Q7 to enable that feature without the MonitorPi beeing installed? If so, I may go that route as I don't need a display and it certainly generates some extra noise.
Meanwhile, got the UcPure up and running as well:
Had the support brackets 3D printed - it is still missing the screws to clamp both halves together (ordered M4x60, but were a few mm too short, so had to order M4x70 set).
Also still waiting for a dedicated 12VAC 80VA transformer from Toroidy - the transport has to do with a linear 12VDC supply for now.
It seems there is a dip switch under the Q7 to enable that feature without the MonitorPi beeing installed? If so, I may go that route as I don't need a display and it certainly generates some extra noise.
Meanwhile, got the UcPure up and running as well:
Had the support brackets 3D printed - it is still missing the screws to clamp both halves together (ordered M4x60, but were a few mm too short, so had to order M4x70 set).
Also still waiting for a dedicated 12VAC 80VA transformer from Toroidy - the transport has to do with a linear 12VDC supply for now.
Last edited:
Even with the screen turned off? If memory serves, using the dip switch is not recommended, though that might be a mild suggestion. Also I can't recall if it's the monitorPi itself that stores the settings or the FifoPi. If it's the FifoPi it may be possible to make the setting on the monitorPi and then disconnect it until more settings changes are needed.
From the Q7II manual:
I. Jumper settings (not recommended)
Jumper switch S1 is located at bottom side of PCB. It can be an alternative to the MonitorPi for the FifoPi Q7 settings. To use this hardware jumper settings, all the MonitorPi settings have to be kept as default. MonitorPi would be a better solution for FifoPi Q7 settings. Jumper switch S1 is not recommended to use.
I wonder, why it is not recommended if it is there.
I would rather switch the dip than install noisy OLED screen so close to sensitive clocks.
I. Jumper settings (not recommended)
Jumper switch S1 is located at bottom side of PCB. It can be an alternative to the MonitorPi for the FifoPi Q7 settings. To use this hardware jumper settings, all the MonitorPi settings have to be kept as default. MonitorPi would be a better solution for FifoPi Q7 settings. Jumper switch S1 is not recommended to use.
I wonder, why it is not recommended if it is there.
I would rather switch the dip than install noisy OLED screen so close to sensitive clocks.
Hello friends,
I recently introduced my raspberry player on SW Volumio
Now there is a modernized updated version Mk2!
I recently introduced my raspberry player on SW Volumio
Now there is a modernized updated version Mk2!
- Added IAN CANADA LINEARPI PRO SOLO Linear Power Supply Module 5A/5V for raspberry Pi 4 which is powered via IAN CANADA UCPI Universal Ultracapacitor Power Supply Board for Raspberry Pi
- Original Power supply: IAN CANADA LINEARPI MKII Ultra-Low Noise Linear Power Supply Module 5V 2.5A is set to 3.3V and powers the new sound card
- - IAN CANADA TRANSPORTPI DIGI II Ultra Low Jitter Digital Audio Interface Module I use the sound output exclusively via Optical Toslink. And at the same time powers the:
- - IAN CANADA ISOLATORPI III Galvanic Insulation Module.
With the addition of the shieldPi pro mkiii between the bridgePi and Q7, I think my DDC fed by a rpi CM5 is finally better in all respects than the rpi 5 with the same Q7 and hdmiPi mkii HATs connected directly.
The biggest test seems to be getting dsd64 right, as it seems to be a format that's very sensitive to EMI noise. The rpi 5 i2s bus seems to excel at this, but it doesn't seem to handle dsd128 or PCM 352.8/384 PCM very well, and of course can't do anything above those rates.
With the DDC approach everything I throw at it, including quad rate DSD sounds fantastic.
The biggest test seems to be getting dsd64 right, as it seems to be a format that's very sensitive to EMI noise. The rpi 5 i2s bus seems to excel at this, but it doesn't seem to handle dsd128 or PCM 352.8/384 PCM very well, and of course can't do anything above those rates.
With the DDC approach everything I throw at it, including quad rate DSD sounds fantastic.