After dealing with all this I now lastly understand why a lot of the audio gear has previous-faculty transformers to energy them: it makes it very easy to make a dual-rail provide. The other part that wanted figuring out is the ability part. The top left a part of the board generates the voltages for the opamp circuits and the top right has the regulators for the codecs (and the Pc input jack). Then the next bytes are 0x01 and 0x01. Now the hard half is determining what meaning. 12V and -12V with a switching regulator after which feed these into an LDO to filter out the switching noise from the switching regulator. Specifically I had a capacitor within the feedback path for the switching regulator that was two orders of magnitude too massive. This can be a far more trendy switching regulator operating at 1.3Mhz as a substitute. This chip is a bit more expensive but it surely generates each the positive rail and destructive rail with a single chip and as a result of order of magnitude larger switching frequency the inductor and capacitor may be way smaller.
It's a switching regulator that runs at 100Khz max. The ability-save pins on the regulator connected to GND instead of VCC and in the bottom left corner an attractive stack of 3 capacitors and a 1k resistor on the output of the regulator. Luckily in any case this the regulator began working reliably. Sadly there is not a "getting started with designing audio interfaces" e book however I found the good webpage from Elliot Sound Products that has too much of knowledge on these circuits. Then after powering it on a second time it started making a screeching coil-whine sound and inside a second the 12V generator made the bottom of the PCB too hot to touch. Back then the cost of those mixers was extremely far out of attain for house use, into the thousands of euros. For cost optimization I skipped battery safety on the board. Luckily the additional constraints massively simplify the design which is great for price optimization as well. Hardware design is a complete other can of worms. So I did the most harmful factor a software program developer can do: I launched Kicad. There were additionally just a few small software program issues to deal with because of the hardware.
While the conversion is "clear" the software program have to be in a position to communicate with the expanded community options. Adding a community port for the audio mixer would mean it needs more change ports in the box and for the reason that network swap ports are on the skin it would want an ugly cable popping out to the front of the box to attach it up. The adapter includes a 3 foot 'type A' cable connection that plugs into any Pc or MAC USB port. To ensure that this type of implementation to work correctly, properly wired RS485 connectors are required on the Pc running the NMS software in addition to all of the linked subsystems. Circuits could also be terminated on screw terminals, D-subminiature connectors, or other types of connectors. After soldering on the connectors I powered it on and checked the voltage rails, it all appeared high quality. The precise voltage for the opamps does not matter much however it has to be excessive enough that they are always above the analog audio enter ranges and because these are low-cost opamps it wants a couple of volts further as a result of the TL072 can not course of indicators near the provision voltage which ends up in distortion. 60Khz however this ends in needing a big capacitor and inductor on the board.
This revision of the power supply was scrapped as a result of with the right inductor the power provide simply turned too giant for the board and that i didn't wish to make the board any larger since the inside of the FOSDEM box is very area constrained. This model of the ability provide was not full but because it was missing a number of capacitors and that i discovered that the inductor I chosen was way too small to function accurately. The codec chips need two 3.3v rails and one 1.8v rail to perform. The opamp circuits are even more complicated because they need a positive and unfavorable voltage rail to function. The voltage drop by way of the diodes I picked might be too much for the serial communication to work. It incorporates too much of various voltage regulators to make the varied rails. To make it a bit nicer on my desk I additionally 3d printed a simple entrance panel for the mixer. So as a substitute of constructing the precise mixer I want I made the opensource base to make any digital mixer. Setup is similar when multiple meters are interconnected via cables CBL03, besides that every one meters now should be set to a different digital tackle by way of their entrance panel.