9 ms·

> Consumers use batteries that are a specific voltage. Nowadays with sealed, non-serviceable batteries, it is not always abundantly clear what voltage the devi

by samtho 3y ago

> Consumers use batteries that are a specific voltage. Nowadays with sealed, non-serviceable batteries, it is not always abundantly clear what voltage the device takes for power (which can be different from its charging voltage) unless it’s in your hand and you can read the specifications that are printed on the back (if it is at all). Breaking out these two datapoints into voltage and amp hour no longer makes sense given that most batteries where consumer-advertised capacity are product specific. Our relationship with battery powered devices has changed since the days of 1.5V cells that we tossed in at varying amounts (2x = 3V, 4x = 6V, 6x = 9v, etc). Pre-“USB-PD”, it still kind of made sense to use mAh because we knew it was always 5v but that too has changed. The mistake is not changing to mWh once we stopped describing 5v devices.

amluto 3y ago
> Pre-“USB-PD”, it still kind of made sense to use mAh because we knew it was always 5v but that too has changed. Really? How many of those devices actually had 5V batteries inside? (Are there even credible 5V Li-poly battery banks?) Original USB (based on a quick search) specified 5V +- 5%. You’re not getting that by direct connection to a battery, and you also can’t charge the internal battery by direct connection to any common supply. Somewhere in the power brick is a switching converter. So maybe a 5000mAh USB-A power brick can supply 5000mAh output before conking out, but I doubt it.
- samtho 3y ago
  > How many of those devices actually had 5V batteries inside? (Are there even credible 5V Li-poly battery banks)? That’s a good point and revealed that I conflated battery capacity with source-able/usable power. You are correct (as far as I am aware as well) that you cannot use a lithium ion/polymer based battery chemistry and have between 4.75 and 5.25 volts output. Thinking about it some more, the battery packs seem to represent a special case where the function of the device is to provide power and, as per your point about squeezing out maximum power, we probably need to have two separate power ratings: one for the battery and one for the usable output.
  - vel0city 3y ago
    But most consumers probably don't care about the internal pack voltages or whatever, all of that is just implementation details. All they care about is how much more time that battery is going to give their device power. Which if they know how many watts their device uses and how many watthours their battery is good for, they'll roughly know the answer. For a lot of people, all they really need to know is the watts and watthours. Everything else will just line up in the implementation. If I know my phone uses ~1W on average, and my battery is 18Wh, then I know I'll get roughly 18 hours of life on that battery. If I want to charge the 18Wh battery in two hours, I'll need one that can supply at least 10W after charging losses.
- numpad0 3y ago
  There is no 5V Li-po batteries, LiFePO4 is 3.2V. Nominal per-cell voltage of a battery depends on anode/cathode pairs. Can be tweaked, state of charge dependent, but isn't an adjustable parameter.

by samtho 3y ago

amluto 3y ago

> Pre-“USB-PD”, it still kind of made sense to use mAh because we knew it was always 5v but that too has changed. Really? How many of those devices actually had 5V batteries inside? (Are there even credible 5V Li-poly battery banks?) Original USB (based on a quick search) specified 5V +- 5%. You’re not getting that by direct connection to a battery, and you also can’t charge the internal battery by direct connection to any common supply. Somewhere in the power brick is a switching converter. So maybe a 5000mAh USB-A power brick can supply 5000mAh output before conking out, but I doubt it.

samtho 3y ago
> How many of those devices actually had 5V batteries inside? (Are there even credible 5V Li-poly battery banks)? That’s a good point and revealed that I conflated battery capacity with source-able/usable power. You are correct (as far as I am aware as well) that you cannot use a lithium ion/polymer based battery chemistry and have between 4.75 and 5.25 volts output. Thinking about it some more, the battery packs seem to represent a special case where the function of the device is to provide power and, as per your point about squeezing out maximum power, we probably need to have two separate power ratings: one for the battery and one for the usable output.
- vel0city 3y ago
  But most consumers probably don't care about the internal pack voltages or whatever, all of that is just implementation details. All they care about is how much more time that battery is going to give their device power. Which if they know how many watts their device uses and how many watthours their battery is good for, they'll roughly know the answer. For a lot of people, all they really need to know is the watts and watthours. Everything else will just line up in the implementation. If I know my phone uses ~1W on average, and my battery is 18Wh, then I know I'll get roughly 18 hours of life on that battery. If I want to charge the 18Wh battery in two hours, I'll need one that can supply at least 10W after charging losses.
numpad0 3y ago
There is no 5V Li-po batteries, LiFePO4 is 3.2V. Nominal per-cell voltage of a battery depends on anode/cathode pairs. Can be tweaked, state of charge dependent, but isn't an adjustable parameter.