Sure. But you can do all of that without the FPGA. There has to be some intelligence there, but there's already an MCU in the HPA-4.
That aside, I do agree with you. I would probably set up a test mode that ran the attenuator from min to max and back to min volume with some signal applied and look at the output waveform. Similar to INL/DNL tests for DACs. At that point it becomes easy to try different algorithms for relay control to see which one produces the lowest zipper noise.
Tom
We're getting there. Now draw at least one USB data line and also add the power connections for the I2C isolator.
I'm curious what you plan to use the USB connection for.
Tom
Don't understand. Is the ground on one side of the I2C coming from the MCU and the other from the USB connection?
Tom
Please draw: Power, ground, and one data line for each of the blocks. It'll make it much easier for me to explain to you where the currents flow.
Right now the I2C isolator is unpowered on both sides.
Tom
Right now the I2C isolator is unpowered on both sides.
Tom
i just checked
https://de.aliexpress.com/item/1005...t_main.41.fb815c5f3Xxqie&gatewayAdapt=glo2deu
both sides of the isolator need to be powered seperately, so i imagine it only isolates both i2c wires
as long there is no way to make this controllable over PC the usb connection is more of a gimmick for easy programming for the ControlMCU since i dont know what else todo with the usb connection
i just picked this cheap isolator: https://de.aliexpress.com/item/1005007593501630.html for the DC isolator, tho there are probably better standalone versions out there, i just dont know any
So now imagine the I2C signal from the MCU going high. That will create some current flow onto the relay board. The return current will flow in the ground line through the regulator and USB isolator back to the MCU board. That's a huge loop, so it'll radiate like crazy.
Tom
Tom
the dc/dc converter is part of the usb isolator, tho if you can recommend some standalone dc/dc converter i would also use this but from some reading basicly all dc/dc converters create some switching noise, im kind of wondering what would be better
1. a dc/dc converter to isolate the controlmcu + some passive filtering between dc/dc converter and power supply/regulator
2. just passive filtering between controlmcu and powersupply/regulator
are you sure this is the case with this kind of i2c isolator?
but what way i would need to draw it ?
to me the i2c isolator looks like its creating two loops, one with the local ground and it creates another loop with the actual isolated GND, i dont believe current will flow "over" the isolator back to the controlMCU
think about a optical isolator it has essentially two power supplys utilized and there is no current at all flowing, its creating two loops
Goostknight,
Your block drawing does not make absolute sense, you for instance have 3.3V derived from USB 5V in the MCU block back to the isollator block as well as a 3,3V regulator to the relay control block. Is there interconnection to 5V perhaps.
Your block drawing does not make absolute sense, you for instance have 3.3V derived from USB 5V in the MCU block back to the isollator block as well as a 3,3V regulator to the relay control block. Is there interconnection to 5V perhaps.
1. usually mcu's like the raspberry pico will have an 3.3V regulator on board you can use
2. the 3.3V regulator will take 5V and regulate down to 3.3V for the relay board to use, no need for 5V there
-----
i was also looking into the PCA9555 chip one of you recommended a few pages earlier, i might use this one over a MCU on the relay board, since i dont do much else anyway
it comes with 8 different selectable i2c adresses and a i2c filter which is nice to have..
Keep in mind that current flows in loops. So if you power both sides of an isolator with the same supply, even if the 'isolated' side is through a regulator or filter, any current drawn on the 'isolated' side of the isolator will have to flow back to the original supply, so you'd have to share grounds.
If you want isolation you need to provide different power supplies for the input and output side of the isolator. That means either a lot of SMPSes or a lot of transformer windings.
Tom
well i get what you wanna get to but isnt the DC-DC converter inside the usb isolator essentially "a different power supply" ? ground loops get broken by isolators, so there is no noise current flowing...