I wonder if in this way my DAW would see the Zynthian straight away, as a marked external device for Midi output, or I would have to tinker with CLI instructions or webconf Midi configuration…
Thanks @ToFF, for pointing me at a previous explanation by @riban.
I will have a try, since on Pi5 I am on Oram. I do not have a spare usb-C connector on my enclosure, because the only available one is for power.
I don’t know if the same Midi OTG / gadget mode from an external PC works with the Pi5 usb-A ports, converted to usb-B female through an hardware adapter.
What I aim at is having the Pi5 with Zynthian exposed as a Midi output device to an external computer, not the other way around.
No! That won’t work. The Raspbery Pi has four USB type-A connectors to which you can connect USB peripherals (mouse, keyboard, MIDI, etc.) It can also present as a USB peripheral itself by connecting via its USB-C connector. If you are using the USB-C connector to power the RPi then it is not available to connect to your PC.
You reminded me that there was a thread some years ago, about coupling power and data when V4 was still powered through USB-C.
I have recovered this link which you published in that discussion:
Would this USB-C splitter allow me to plug both the specific PSU for my enclosure and an A to C cable with midi data from my PC?
And if no, what alternatives would there be for my Rpi5 kit, that only has one USB-C connector? (apart from my current solution, which is midi over 5-din on external USB interface).
That adapter may work although the physical construction may cause problems, e.g. cables at right angles to the connector and obscuring the HDMI sockets.
Zynthian powers the RPi via its 40-pin header, leaving the USB-C connector free to be extended to the Type -B connector on the case.
It claims to support power delivery up to 60 W and data stream up to 480 Mb/s (60 Kb/s), both of which should be more than adequate for powering even the Raspi5 (max official PSU wattage 27 W - 1.8 A, 15 V) and transmitting the relatively modest flow of the Midi protocol. I don’t think that it should introduce any detectable added latency. Will let you know when it arrives.
Thanks. Unfortunately, it remains a question how it would deal with the non-standard requirement of the Pi5 to get 5A (at 5V). I know too little about the PD standard and even less about its implementation on the Pi5, but PD 60W is commonly achieved by 20V at just 3A. The 5A requirement is beyond most common PD specifications.
The cable looks passive so one might expect / hope that PD is handled by the PSU / Charger and device (RPi). Of course a substantial factor for reliable power delivery is cable size and a lower voltage will require higher current for the same power delivery which requires thicker cables. RPi has previously been particularly sensitive to high impedance (thin / long / poor quality) power supplier cables. This cable (at £10) should be okay but price is not necessarily an indication of quality. You may have to pay for what you get but you don’t always get what you pay for!!!
Both your arguments are true, and the Raspberry PD specifications at 5A are notoriously out of common standards.
Since there use not to be any detailed data sheets, for this kind of general-purpose items, I went indeed for what appears to be a relatively upper range product, with a good quality and sufficient section of cablings.
The fact that it is advertised for also charging Macbooks, whose PSU delivers up to 3.65 A at about 60 W, seemed a promising starting point.
I too surmised that this Y adapter behaves passively, as far as current is concerned, leaving to the official Raspberry PSU to push the top absorption of 5A.
Conversely, if it’s not passive it might furnish no more than 60 W / 15 V = 4 A, which is unfortunately not enough for Pi5 and peripherals, if I’m doing the electrical maths correctly.
Let’s hope that I will be lucky with this attempt: I will keep you updated!
I can confirm that the posted splitter cable works correctly on Rpi 5, and that, at least for the moment, I can power the unit and send midi data in OTG mode to the Zynthian system from Cubase 13, without any apparent issues.
This is a nice step forward! Raspi5 with Oram is exposed in my DAW, plainly and clearly, as ZYNTHIAN, among the available midi output ports.
I haven’t attempted medium overclocking for now, but I will, and thus will also realise if the drawn current from the power leg of the Y cable still sustain the strain.
Great to hear! However, load and overclocking won’t test the cable. Pi5 itself should be quite happy with 3A at any load, which is what the cable can do. The 5A are only required to raise the amount of power available on the USB ports of the Pi5 from 600mA to 1600mA. I have currently no exact idea how the Pi detects whether the power source can supply 5A or not. And I have no idea what happens if the power supply IS capable of 5A, but the cable not…
If rpi 5 does not successfully complete USB PD negotiation with power supply for 5V/5A mode (e.g. power supply does not provide PD protocol for negotiation of available outputs) it will set maximum USB output to 600mA for connected devices. You can override usb port power limit of 600mA by setting usb_max_current_enable=1
I have a 5V/30A power supply for a few stripes of WS2812b RGB LEDs as well…
OK. So 5V/5A is still a part of the PD specs, despite not being supported by many supplies. But the wires and connections probably cannot report anything, so if you DIY from cheap parts, you probably should better monitor their temperatures
BTW, I suppose the Pi system should indicate the results somewhere in procfs or sysfs…?
Thanks @stojos for this useful bit of info, about raising the available current on Raspi5 USB ports, with:
usb_max_current_enable=1
By the way, where should I insert this command in an Oram on Pi5? Maybe webconf terminal (if it’s still there) or possibly via SSH?
At the moment, I have just an USB touch cable for a (separately powered) touchscreen and a Presonus Studio 26c audio interface, connected to the Pi5 USB hub.
The data sheets of audio interfaces are mostly elusive about drawn current, but I have been able to gather the following information from Presonus.
Maybe, the last part on voltage rails might be of some interest, but I need the knowledge of some engineering guy to really understand it.
Presonus Studio 26c
Microphone Inputs
Input Impedance 1.6 kΩ
Phantom power +48 VDC (10 mA total)
Line Inputs
Input Impedance 10 kΩ
Instrument Inputs
Input Impedance 1 MΩ
Headphone Outputs
Maximum Power 30 mW / channel (56Ω load)
Impedance Working Range 32Ω to 300Ω
Since it is a bus-powered device, the Studio 26c has lower voltage rails than other Studio-series interfaces, resulting in reduced control-voltage range (positive to negative voltage)