4 ms·

> then wouldn't need voltages to be fixed in the comparison. > How long is it going to run my 5W load? 10 hours? Voltages are fixed. Fixed, passive. Amperage i

> then wouldn't need voltages to be fixed in the comparison. > How long is it going to run my 5W load? 10 hours? Voltages are fixed. Fixed, passive. Amperage is designer's discretion. You can't make a computer run at 7V because you want to. The computer draws 12V/1A, and you have bunch of 3.7V/2600mAh cells around, so you join 4 in series to get 14.4V, shave off 2.4V through LDO, and get 2.6h runtime(is that 2.4W excess heat there? ew). Wh? Not handy at all. I don't control what many Watts my device draws. I know the CPU runs at precise 3.3V +/- 200mV and draws up to 20mA at full speed. Not "50 mW at up to 3.3V", no. I know my phone charges at up to 5V/2A or 10W, so what. An airline approved XL sized battery should roughly run the phone for 10 hours give or take few with just the battery and that's accurate as it gets from advertised data. > Is that mAh rated at the 5V for USB? It's practically ALWAYS rated at 3.7V because every "Chinese" brands - not always from there or with associated ethnicity - wants figures to be based on maximum indisputable advertisable numbers. That had already eliminated ambiguities years ago. I've seen batteries rated for a mAh capacity on 3.6V advertised nominal voltages as well as those in 5V, those don't exist anymore. All in all, battery capacity in Wh, not so practical. Maybe it could make some sense for consumer variable voltage power supplies. Definitely not for cells.

vel0city 3y ago
> Voltages are fixed. Fixed, passive. And yet this thing has some internal pack voltage, its putting it out at something like 5V, and then my phone is then turning it into something other than 5V for whatever its internal pack voltage is, which then once again gets changed for whatever its internal circuitry runs at. Doesn't seem very fixed to me. USB-PD has tons of different voltages it can operate at, all vastly different from whatever pack voltage my phone or laptop run at, vastly different from what the CPU runs at. Sure, when you're building something and you control all the variables, its easy to say the voltage is fixed. But most consumers aren't in that environment, they're plugging their phone into whatever charger supporting a myriad of voltages and amperages. And circling around, you can't be certain how many hours my 10,000mAh battery pack will run a 10W load. The only voltage assumed here is USB, 5V, so why would anyone assume 3.7? Why not 2.5? Lets just make up whatever numbers for the internal pack voltage. And you can't tell me about how many times that 10,000mAh battery pack can recharge my phone's battery pack, since we don't know either pack's voltage. But if we knew how many watthours each pack was, we could reason that.
- numpad0 3y ago
  > when you're building something and you control all the variables, its easy to say the voltage is fixed No, my point is I don't control that variable. I can't get 14.4V microcontrollers, nor 6.5V batteries. Neither can you. I am obliged to power it with 3.280V/0.1-150mA regulated power from 4.2/3.7/2.8 max/typ/min aluminized pouches of wet wipes. Wattage figures therefore won't help me. USB-PD sure runs at 5/9/12/15/20V. Off of 3.7V li-ion cells, to charge 3.7V li-ion cells. So it's clear how many times more capacity a 20100mAh power bank battery has over 5000mAh phone battery, or a series 6-cell 18650 laptop battery pack(with nominal 2600mAh capacity per cell: 9x2600mAh of 3.7V equivalent capacity). > And circling around, you can't be certain how many hours my 10,000mAh battery pack will run a 10W load. 37Wh? 3.7 hours, _ish_. Give or take one full hour. I don't even know what capacity of an advertised 10000mAh battery at ~0.3C might be. Battery capacity depends on discharge current. And here again it's current, not discharge energy.
  - vel0city 3y ago
    > 3.7 hours, _ish_. Give or take one full hour. So after assuming a value that's not even printed on it we're not even sure within +-25%. What a great unit of measure for battery storage. When if it was just rated in Wh we'd actually know. I'd much rather a battery say how many Wh its good for, its nominal voltage, and its peak amps. Its incredibly easy to reason how many watts a phone uses, a laptop uses, a game console uses, a TV uses, a car uses, a blender uses, a kettle uses, etc. Then its easy for me to know how long can this battery potentially run X.
    - numpad0 3y ago
      > When if it was just rated in Wh we'd actually know. No. The capacity literally changes depending on how fast you draw. This is especially a problem on lead acids, so they often have couple "n-hour rate capacity" figures to make estimates for your particular use cases. It's less of a problem for li-ion, where C-rates(not coulomb but alphabet cee) is used to plan for cell counts, runtime, current limits and required wire thickness(because wire thickness depends on current, yet again, not total energy through it). It's a non-issue for dry cells because it's among equipment designer's responsibilities to account for. And that has little to do with Wh vs mAh. It'll be easier to guesstimate more precisely from mAh using experiences and implications, if anything.

vel0city 3y ago

> Voltages are fixed. Fixed, passive. And yet this thing has some internal pack voltage, its putting it out at something like 5V, and then my phone is then turning it into something other than 5V for whatever its internal pack voltage is, which then once again gets changed for whatever its internal circuitry runs at. Doesn't seem very fixed to me. USB-PD has tons of different voltages it can operate at, all vastly different from whatever pack voltage my phone or laptop run at, vastly different from what the CPU runs at. Sure, when you're building something and you control all the variables, its easy to say the voltage is fixed. But most consumers aren't in that environment, they're plugging their phone into whatever charger supporting a myriad of voltages and amperages. And circling around, you can't be certain how many hours my 10,000mAh battery pack will run a 10W load. The only voltage assumed here is USB, 5V, so why would anyone assume 3.7? Why not 2.5? Lets just make up whatever numbers for the internal pack voltage. And you can't tell me about how many times that 10,000mAh battery pack can recharge my phone's battery pack, since we don't know either pack's voltage. But if we knew how many watthours each pack was, we could reason that.

numpad0 3y ago
> when you're building something and you control all the variables, its easy to say the voltage is fixed No, my point is I don't control that variable. I can't get 14.4V microcontrollers, nor 6.5V batteries. Neither can you. I am obliged to power it with 3.280V/0.1-150mA regulated power from 4.2/3.7/2.8 max/typ/min aluminized pouches of wet wipes. Wattage figures therefore won't help me. USB-PD sure runs at 5/9/12/15/20V. Off of 3.7V li-ion cells, to charge 3.7V li-ion cells. So it's clear how many times more capacity a 20100mAh power bank battery has over 5000mAh phone battery, or a series 6-cell 18650 laptop battery pack(with nominal 2600mAh capacity per cell: 9x2600mAh of 3.7V equivalent capacity). > And circling around, you can't be certain how many hours my 10,000mAh battery pack will run a 10W load. 37Wh? 3.7 hours, _ish_. Give or take one full hour. I don't even know what capacity of an advertised 10000mAh battery at ~0.3C might be. Battery capacity depends on discharge current. And here again it's current, not discharge energy.
- vel0city 3y ago
  > 3.7 hours, _ish_. Give or take one full hour. So after assuming a value that's not even printed on it we're not even sure within +-25%. What a great unit of measure for battery storage. When if it was just rated in Wh we'd actually know. I'd much rather a battery say how many Wh its good for, its nominal voltage, and its peak amps. Its incredibly easy to reason how many watts a phone uses, a laptop uses, a game console uses, a TV uses, a car uses, a blender uses, a kettle uses, etc. Then its easy for me to know how long can this battery potentially run X.
  - numpad0 3y ago
    > When if it was just rated in Wh we'd actually know. No. The capacity literally changes depending on how fast you draw. This is especially a problem on lead acids, so they often have couple "n-hour rate capacity" figures to make estimates for your particular use cases. It's less of a problem for li-ion, where C-rates(not coulomb but alphabet cee) is used to plan for cell counts, runtime, current limits and required wire thickness(because wire thickness depends on current, yet again, not total energy through it). It's a non-issue for dry cells because it's among equipment designer's responsibilities to account for. And that has little to do with Wh vs mAh. It'll be easier to guesstimate more precisely from mAh using experiences and implications, if anything.