I suspect some AC driven devices could operate on DC. The rectifier acts like a set of one-way valves, so the DC would just pass continuously through one pair instead of alternately through the different pairs of diodes. And there wouldn't be anything to filter, so the filter caps would be effectively inert (or charged and never discharging).

alebro, post: 464945, member: 51993 wrote: This question pertains to something mentioned in the thread over here https://recording.org/threads/in-need-of-a-power-supply-for-a-alesis-multimix-8-usb-18v-will-pay.54380/ - but I thought it was worth a thread of it's own.

Alesis Multimix mixers (and probably others), take an AC power supply that is internally converted to DC.

Boswell mentions this

The mixer unit would probably work on 18V d.c. input, but there would be no phantom power.

Can anyone explain how that works?

It depends on the design of the power supply and the required output voltages.

A steady floating voltage applied to a bridge rectifier would indeed pass a slightly smaller voltage to the output, with the polarity at the output being independent of the polarity of the input. The voltage loss is about 1.5V with standard silicon diodes. By using a centre-tap on the input supply, bipolar outputs each of half the input voltage can be obtained (full-wave configuration).

If you have an alternating input voltage available, you can build higher output voltages by passing the alternating input through capacitors to add progressively to the steady voltage. There is no theoretical limit to the number of ladder stages you could use, but the efficiency of the process falls off rapidly as you add stages. This configuration was devised by a Swiss called Greinacher around 100 years ago, but developed by the physicists Crockroft and Walton, and is often called a Crockroft-Walton Multiplier. They managed to get tens of KV this way for their charged-particle experiments.

It is the Cockroft-Walton multiplier (usually with 3 rungs on the ladder) that some small mixers use for their 48V phantom power generation given an a.c. input in the region of 18V r.m.s. It clearly would not work with a d.c. input, as that does not pass through capacitors.

Boswell, post: 464951, member: 29034 wrote: Crockroft-Walton Multiplier.

It's a bit like a block and tackle. You trade away one property for another.

I think of the C-W multiplier stages as being like a tower of circus acrobats, each below the top one standing on the shoulders of the one below. The performers get smaller and thinner as they go up from the ground.

Really great information thank you both!

Boswell, post: 464961, member: 29034 wrote: I think of the C-W multiplier stages as being like a tower of circus acrobats, each below the top one standing on the shoulders of the one below. The performers get smaller and thinner as they go up from the ground.

That's freaking Awesome! Thanks for that Boswell

Wish I'd thought of that one when I was teaching! Best one yet!