Bitcoin Forum

https://en.bitcoin.it/w/images/en/thumb/c/c1/Avalon-pdu-v1.2.JPG/800px-Avalon-pdu-v1.2.JPG

Always wondered.
Looking at the photo it appears the caps are for the 12V distribution and the ATX PSU provide very good regulation why does the PDU need caps?

Well, they would smooth out current spikes, by-passing one connector and cable-run.

Not sure if that would be strictly needed though.

Edit: did PDU v1.1 have those caps?

I don't know if there ever was a v1.2 released publicly; this is the only photo of the PDU I have seen.

I get that the capacitor smooths out voltage but this board is being connected to a highly regulated power supply.  Are the tiny capacitors here having any meaningful effect compared to the PSU itself.  ATX PSU should be able to handle pretty significant changes in current and keep the voltage stable.

Anyone know if there is a board diagram for the PDU available?

I think the idea is that all of the PSU wires and Molex pins can end up adding a fair bit of resistance to the circuit, and that a sudden spike in current demand can cause the supplied voltage to drop at the load.

By locating caps on the PDU board, they can help smooth out some of those big spikes and ease the load on the PSU wires/pins.  Then again there's caps on the modules, and I think on the backplane too.  I don't know why those couldn't do the job just as well/better.

They don't look like very big caps.  My fancy Antec 1200W "PowerCache" PSU has 2200uf caps on the ends of its cables that are way bigger.

I have some blank Avalon PDU PCBs here that I obtained from BlackArrow.  I'll try and take some nice photos of one sometime.

http://en.wikipedia.org/wiki/Decoupling_capacitor

In case of Avalon it helps prevent noise from one module affecting the other modules. I believe the main load on each module has a inductance component that is used for energy storage with switching. So they could push back voltage spikes into the main power supply. The hashing modules are identical, so in the worst case they could go into synchronized oscillations in their buck voltage regulators. Better be safe then sorry. Together with the contact resistance of the connectors they are actually forming nice RC low-pass filters.

Hop on ebay and pick up a cheap oscilloscope.  If you live in a big city, you can pick up a shitty old one for cheap, and even if you don't ever use it as a service tool, the educational value can be amazing.

Basically, computer guys like to pretend that their machines use digital signals.  Sadly, there is no such thing.  Real lines ring, bounce, wiggle and squirm.  That applies just as much to "supply" lines as it does to "signal" lines.  I put them in quotes because the circuits don't know that we call them by different names.

If you are lucky, a digital circuit acts like an AC line with all of the fun that comes with complex phase relationships.  If you are unlucky, the damn things spew radio interference all over the insides of your gizmo.

Most real digital devices have bushels of capacitors strewn about.  In a DC (imaginary) circuit, they eat voltage swings and spikes.  In an AC (real) circuit, they change the resonant properties of the circuits.  Every time you have resistance and capacitance in the same place, they act like a filter.  These filters can limit circuits to low frequency oscillation, preventing RFI.