U31 and U27 definitely as a minimum; 1V reg should not be impacted but worth checking before installing the new U27 obviously. Clock gen and SPI flash , to my knowledge, are the only other things that could in theory suffer from a shorted 3v3 rail but are usually OK ?. I understand MP1605 and MP2141 is essentially the same silicon, maybe some packaging differences. I'd be really curious to talk to focusrite and MPS engineers to see what's up with all this...

I'm ordering the parts for mine now from Focusrite and they have these available for the 4i4: Part Number: SEMS002022 (U27) Description: STEP DOWN CONVERTER SYNCHRONOUS 1A MP2141NGTF ADJUSTABLE "MPS" Part Number: SEMS002049 (U31) Description: XU216 QFP128 CPU 256K RAM 2-TILE 16-CORE 1000MIPS ENCRYPTED Part Number: DDES001001 (U33) Description: DIODE SOT23-6 ESD PROTECTION "SEMTECH" RCLAMP0504S.TCT Notice that 2 are the ones you replaced. They also mentioned a service bulletin for the 4i4 3rd gen so trying to get a copy of that as well which might provide more insights.

Oh wow that's the first I hear about a service bulletin, let me know if you get a copy of it !

Try a few different cable(s) , computer / hub / port combinations.

@@fenugrec5697 tried everything. unfortunately, no luck. Is there a way you could show me what to measure exactly to identify the exact problem? I have a multimeter and would be more that happy to learn about this topic :). Went to your page but no further explanation on the subject :(. Hope you can help me.

Hi, unfortunately what I have on my website is about the maximum amount of detail I had energy to explain. I reasoned that replacing the CPU is a challenging task that would realistically only be undertaken by somebody with the knowledge to probe around and troubleshoot; as I may have stated in the video definitely not an easy repair, although the troubleshoot itself is not terrribly complex. More that can fit in a comment reply, anyhow. If you still want to pursue this, go to that page on my website, look up the datasheet for the problematic components, and try to figure out which voltages are where. Probing voltages (carefully) in a defective 4i4 is essentially harmless at this point

Try downloading the drivers for the interface

@@Middas no luck :(

A single zener isn't ideal for this : you would want near-zero conduction at 3.35V, and "full" at 3.5-3.6V. Low voltage zeners have a very soft knee, look at datasheet curves for example BZT52* or 1N53* . Also, power rating would be a challenge, since a shorted buck regulator means the zener is conducting between VUSB rail and ground.

well, in my measurements, it's a relatively short peak, and on average it is definitely below 900mA. I expect high inrush scenarios like this are handled quite differently depending on the hub. Also, with 3A peaks, voltage drop over the cable and contact resistances is also noticeable. I could've gone into more detail on my current measurement (resistive shunt) which also affects the peak amplitude and duration.