Hacker News Reader

Comment by mschuster91 | Hacker News Reader

This design does lack a configurable dc/dc voltage regulator, I've seen laptops with anything between 12-24V... any idea for a circuit?

View on HN

Due to component tolerances, laptops aren't that picky about what voltage they need; capacitors with 16V and 25V max ratings have basically created two ranges, 9-12V and the far more common 18-20V:

https://news.ycombinator.com/item?id=12618406

I find the other comment about a programmable output voltage in 50mV increments a little amusing, since 50mV of 5V is 1% and of 20V is 0.25%. To maintain the former level of regulation over changing loads would already require a very expensive and powerful supply with remote sense, while the latter is basically voltage reference territory.

Voltage refs are more typically 2 orders of magnitude better than that.

Better references, yes, but 0.25% is typical of a cheap reference like a '431.

I stand corrected; thank you. I am genuinely surprised to learn that those are called voltage references, but that is an accepted industry term.

> I find the other comment about a programmable output voltage in 50mV increments a little amusing

To clarify, when you program the chip, you set a voltage with 50mV resolution and a lower and upper percentage bounds that you'll accept. The minimum percentage is 5%. So for instance, at 5V, you can't narrow down the range to less than 0.5V (4.75-5.25v).

The STUSB4500 supports output voltages of 5V to 20V in 50mV increments and is programmed over I2C, so it should be pretty easy to add a small micro controller and some input interface to make the output voltage configurable.

you could hang one off the output, not the most elegant but at least modular then

Reply on news.ycombinator.com