I would like to try with my 9038 a transformer output...any advice/experience on that front ? I saw quiet a bit 9018 in combination with transformers...and I guess the much stronger output stage of the 9038 might drive a transformer even better ? Target is not necessarily to directly drive a power amp, but maybe a SE-tube-output stage...
By the way: Any plans on the dual monos ?
By the way: Any plans on the dual monos ?
TPA USB iface discussion could be moved to a separate thread.
Actually, upgrading XMOS FW from Linux is easier than from Windows. They provide module_dfu/host/xmos_dfu_osx/xmosdfu.cpp tool that can either upgrade fw or revert it to a factory one using the USB connection.
I'd prefer if TPA USB could be used also as a DAC controller, with possibility to draw on a display or to control power or trigger relays.
It is far easier to have a control in one place, even if programming XMOS is pain in the ... head
Actually, upgrading XMOS FW from Linux is easier than from Windows. They provide module_dfu/host/xmos_dfu_osx/xmosdfu.cpp tool that can either upgrade fw or revert it to a factory one using the USB connection.
I'd prefer if TPA USB could be used also as a DAC controller, with possibility to draw on a display or to control power or trigger relays.
It is far easier to have a control in one place, even if programming XMOS is pain in the ... head
As I read it, the first level of MQA decoding is done in playback software, typically by the TIDAL Windows or Mac desktop applications (there's no MQA support in the TIDAL Android or iOS apps) - this will "unfold" a 44.1kHz MQA-encoded file to 88.2kHz, or a 48kHz file to 96kHz. So if you want to stream TIDAL with this basic level of MQA decoding, then the solution is clear - use the TIDAL desktop application.
But if the source file has a higher samplerate, the second "unfold" is done in hardware, typically available as an option in XMOS firmware, and this requires a licence, as miero mentioned.
If this full level of MQA decoding is important to you, and you want to use it in conjunction with a Buffalo III-SE-Pro, then I think you would need to modify an MQA-capable DAC to break out the I2S lines from the MQA-DAC's USB decoder, to redirect into your Buffalo DAC.
The cheapest DAC's I can find with MQA support are the AudioQuest DragonFly Black US$100, DragonFly Red US$200, and Meridian Explorer2 US$200.
Can any of these be suitably hacked? I don't know.
But first it may be worth reading John Darko's article about why Schiit Audio have decided not to support MQA -
Schiitting on MQA | Darko.Audio
It looks like you are not setting sufficient shunt current for the Placids. Did you check leonbv's guide on the TPA site for setting the Placids up?
Apologies, wasn't meaning to be stating the obvious, just that some people have had issues setting up the currents on the Placid boards. Glad you have the issue figured out.
This is has nothing to do with the DAC
It will always play DSD exactly as it is input per the DS.Some sources (USB or DoP etc) don't necessarily agree on what DSD channel 1 and channel 2 should be - left or right.
Last edited:
Our firmware is open source - and is an excellent starting point for any custom idea.
GitHub - twistedpearaudio/Buffalo-III-SE-Pro-On-Board-Firmware
GitHub - twistedpearaudio/Buffalo-III-SE-Pro-On-Board-Firmware
Finally debugged my Arduino code, some code errors plus a number of plain stupid errors, but the DAC lives. The Arduino switches the B3SEPRO SW1 and SW2 banks for input, IIR Bandwidth, DPLL and Filter type. Additionally it ties in to a 4:1 MUX, with 3 inputs, one being TOSLINK. Display is a simple 4 row LCD and control is via remote.
The build is fairly straightforward, separate 100VA transformers for the analog and digital TPA boards and a small 10VA supply for the Arduino.
Awaiting a custom fascia and control wiring clean up and I will post some images.
Impressions - very impressed, everything is more 'open' with greater depth and clarity (compared to the older ES9018 BII). A great accomplishment by Russ and Brian, thanks guys.
One remaining question. We have 4 switches for DPLL bandwidth, with the following settings for SW2 switches 5-8 (with ON = low)
5 6 7 8
on, on, on , on - default
off, on, on, on - lowest (narrowest)
off, off, off, off - highest (widest)
For intermediate settings I am somewhat (actually totally) confused as to how the switch settings are derived.
The build is fairly straightforward, separate 100VA transformers for the analog and digital TPA boards and a small 10VA supply for the Arduino.
Awaiting a custom fascia and control wiring clean up and I will post some images.
Impressions - very impressed, everything is more 'open' with greater depth and clarity (compared to the older ES9018 BII). A great accomplishment by Russ and Brian, thanks guys.
One remaining question. We have 4 switches for DPLL bandwidth, with the following settings for SW2 switches 5-8 (with ON = low)
5 6 7 8
on, on, on , on - default
off, on, on, on - lowest (narrowest)
off, off, off, off - highest (widest)
For intermediate settings I am somewhat (actually totally) confused as to how the switch settings are derived.
I received my unit as well a couple of weeks ago and plan to power kt up now...sofar I got two dual mono 9018 running with two placidhd...
Instead of powering up the placids: Could I use as well a nice passive raw supply like a LCLC PSU ? How sensitive are the on board regs regarding power supply fluctuation or input voltage fluctuation as the dacs seems to take different current levels from the power supply depending which resolution it is playing ?
I played around with a lot of PSU designs for my tube output stage and the raw supply of my coleman DHT heater regs and found that LCLC sounds much better, less fuzzy than CRC and on HV+ a passive PSU with good amorpous chokes can sound more open than a shunt reg, no matter if that was a silicon or a a tube shunt.
Instead of powering up the placids: Could I use as well a nice passive raw supply like a LCLC PSU ? How sensitive are the on board regs regarding power supply fluctuation or input voltage fluctuation as the dacs seems to take different current levels from the power supply depending which resolution it is playing ?
I played around with a lot of PSU designs for my tube output stage and the raw supply of my coleman DHT heater regs and found that LCLC sounds much better, less fuzzy than CRC and on HV+ a passive PSU with good amorpous chokes can sound more open than a shunt reg, no matter if that was a silicon or a a tube shunt.
It is a binary number: on = 0, off = 1
pos 8 is most significant bit - pos 5 is least.
8765
0000 ( 0) default
0001 ( 1) Lowest
0010 ( 2)
0011 ( 3)
...
1111 (15) highest
Cheers!
Russ
Accurate supply is pretty critical for the safety and longevity of your modules. For instance you really don't want to exceed 5V for tridents because of the high currents and the max voltage of the regulator ICs.
The best idea is - don't exceed 5V - ever.
Thanks Russ, I was thinking that it might be a Hex number, binary makes total sense, I'll just space the intervals out evenly.
Regards
Just had my BIII-SE upgraded to Pro (9028) and had to set 5-8 on SW2 to off for DSD256 to play w/o static or skips. The default setting was ok for DSD128 but no way for DSD256. With 5-7 set to off played ok but frequent skips (drop outs). So far everything is good with 5-8 set to off.