? 56 ? AA ??? PC ?? ????? 5 ??????????? - ????
HKEPC
<https://www.hkepc.com/25171/>

How Many AA Batteries Does it Take to Power A PC Setup? - YouTube <https://www.youtube.com/watch?v=U5lskFXDbWs>

--
@~@ Simplicity is Beauty! Remain silent! Drink, Blink, Stretch!
/ v \ May the Force and farces be with you! Live long and prosper!!
/( _ )\ https://sites.google.com/site/changmw/
^ ^ https://github.com/changmw/changmw


--- PyGate Linux v1.5.12
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On Tue, 3/3/2026 12:17 PM, Mr. Man-wai Chang wrote:

? 56 ? AA ??? PC ?? ????? 5 ??????????? - ???? HKEPC <https://www.hkepc.com/25171/>

How Many AA Batteries Does it Take to Power A PC Setup? - YouTube <https://www.youtube.com/watch?v=U5lskFXDbWs>

This style adapter is used for mini-itx lower power systems.

This runs off +12VDC but does not regulate the +12V feeding
the motherboard and disk drive. It makes 3.3V and 5V, some -12V
perhaps.

The total power draw of that adapter is 12V @ 10A (if the
motherboard draws that much in total). That means 10 amps
would be flowing through an Adaptaplug M barrel connector.
That's a bit too much current for the most popular
barrel connector type. Even the fatter barrel connector
is not likely to be rated at 10A, maybe a bit less.
The machine I'm typing on right now, runs at 36W (I have the
power meter connected to it). It used to run at 33W when I had
a small monitor, it runs at 36W driving a 4K monitor.

https://c1.neweggimages.com/productimage/nb640/AK97_13201846999615423495CeNLx40h.jpg

You could use the dry cells as (7) sets of 8 cells each (8*1.5V).

An automotive 12V lead acid battery would be 12.6V fully charged, and
if it was slightly run down, that would be a better voltage
value for running a PC. 12.5V feeding the ITX adapter would
be about as high as I'd want to go (if you were using a
3.5" hard drive). If there is no 3.5" hard drive, you could
go a bit higher on the 12V. You would not take the battery
off its charger, and use it while the terminal voltage on it
was still 14.4V. You have to wait a day for the battery to
settle to 12.6V .

Other battery chemistries are better (lithium), as the cells can
produce more amperes. With the dry cells, the old ones
would "swell" at 2 amps load or so (which could be gas
evolution in the battery jacket). That's one reason you
have to be careful when playing around with battery power.
You have to know what the failure mode of the batteries is,
and what the consequences could be.

But otherwise, that's a pretty expensive way to power a PC, as
the batteries aren't going to last all that long.

He could have demonstrated his design prowess, by running the
PC off a large solar panel.

*******

If you want to see some desperate science at work, the people in this
area have trouble finding chargers for their BEV cars. So someone
is building "fast chargers" which are powered from solar cells
plus a battery bank. The battery bank only has 200kWh in it.
You can tell from that, how many cars they can fill before the
"station" is "out of gas" :-) These are fast chargers without
mains power. This approach is more practical in summer, as the
solar output in winter is 10x lower. In a couple more years,
a competing solution will come along that can also charge cars.
But this will suffice to make "heroic" trips possible on that road.
Pack a lunch.

https://www.ctvnews.ca/winnipeg/article/new-charging-stations-open-northern-gateway-for-electric-vehicle-travel/

Some of the indigenous towns have a solar farm, which aids power
generation for homes so it does not all come from the
consumption of diesel fuel and diesel generators. Apparently
the solar farm is large enough, they can turn off the diesel
generators from 10AM to 2PM... if it is a cloudless day
(which is not too likely). That style of solar farm has
no lithium battery bank. In the news article above, the
design uses lithium storage so the station can charge
a BEV later in the day. To build a real charging station
that way, would take a lot more storage than 200kWh. But
since the project is unlikely to be subsidized, the dimensions
of the project must be kept small.

Paul

--- PyGate Linux v1.5.12
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On 3/4/2026 2:15 AM, Paul wrote:

An automotive 12V lead acid battery would be 12.6V fully charged, and
if it was slightly run down, that would be a better voltage
....
But otherwise, that's a pretty expensive way to power a PC, as
the batteries aren't going to last all that long.

He could have demonstrated his design prowess, by running the
PC off a large solar panel.

He shuold have forget about a regular PC motherboard, but Raspberry Pi.
I think Pi 4 could run Windows 8? I dunno. Never tried. :)

--
@~@ Simplicity is Beauty! Remain silent! Drink, Blink, Stretch!
/ v \ May the Force and farces be with you! Live long and prosper!!
/( _ )\ https://sites.google.com/site/changmw/
^ ^ https://github.com/changmw/changmw

--- PyGate Linux v1.5.12
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On 3/4/2026 2:15 AM, Paul wrote:

But otherwise, that's a pretty expensive way to power a PC, as
the batteries aren't going to last all that long.

He could have demonstrated his design prowess, by running the
PC off a large solar panel.

BTW, he should have tried some old motherboards. Might be a lot easier, simpler and use less electricity. Modern CPUs are power-hungry.

--
@~@ Simplicity is Beauty! Remain silent! Drink, Blink, Stretch!
/ v \ May the Force and farces be with you! Live long and prosper!!
/( _ )\ https://sites.google.com/site/changmw/
^ ^ https://github.com/changmw/changmw

--- PyGate Linux v1.5.12
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On Wed, 3/4/2026 5:47 AM, Mr. Man-wai Chang wrote:

On 3/4/2026 2:15 AM, Paul wrote:

But otherwise, that's a pretty expensive way to power a PC, as
the batteries aren't going to last all that long.

He could have demonstrated his design prowess, by running the
PC off a large solar panel.

BTW, he should have tried some old motherboards. Might be a lot easier, simpler and use less electricity. Modern CPUs are power-hungry.

It's the opposite. Modern systems have good idle power ratings.
Like mine draws 33W at idle, when driving a smaller-than-HD
resolution screen. And that's a system with a low end NVidia
card in it as well. The NVidia card was added because of a
crashing problem when using the iGPU.

My first PC compatible, drew 150W at idle and 153W when
the single core CPU was railed. There was no power
control to speak of. And at 150W, it is unlikely
the PC could boot before the batteries were flat (or overheated).
The CPU draws about 35W out of the 150W number (that was
a 440BX with a Tualatin adapter in it and a Tualatin 1.4GHz
was the fastest CPU I had for that setup).

Some of the mini-PC boxes, they have 7W processors, so
their power consumption is similarly limited.

If you are talking about running a 14900K in such a project,
that is unlikely to be a good choice. The CPU can turbo to 6GHz
on the P cores, and when that happens, those batteries will be
flat in no time. Similar to my computer on the other side
of the room, they draw 200+ watts when running 7ZIP compression,
and that isn't a good choice for a battery project. But if you
had an N150 processor (4 cores), those have quite low power
consumption and would last a bit longer.

Paul

--- PyGate Linux v1.5.12
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On 3/5/2026 7:03 AM, Paul wrote:

On Wed, 3/4/2026 5:47 AM, Mr. Man-wai Chang wrote:

It's the opposite. Modern systems have good idle power ratings.
Like mine draws 33W at idle, when driving a smaller-than-HD
resolution screen. And that's a system with a low end NVidia
card in it as well. The NVidia card was added because of a
crashing problem when using the iGPU.

My first PC compatible, drew 150W at idle and 153W when
the single core CPU was railed. There was no power
....

I surely have no data to support my wild guess about power requirement
of old motherboards and CPUs. :)

--
@~@ Simplicity is Beauty! Remain silent! Drink, Blink, Stretch!
/ v \ May the Force and farces be with you! Live long and prosper!!
/( _ )\ https://sites.google.com/site/changmw/
^ ^ https://github.com/changmw/changmw

--- PyGate Linux v1.5.12
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On Thu, 3/5/2026 1:32 AM, Mr. Man-wai Chang wrote:

On 3/5/2026 7:03 AM, Paul wrote:

On Wed, 3/4/2026 5:47 AM, Mr. Man-wai Chang wrote:

It's the opposite. Modern systems have good idle power ratings.
Like mine draws 33W at idle, when driving a smaller-than-HD
resolution screen. And that's a system with a low end NVidia
card in it as well. The NVidia card was added because of a
crashing problem when using the iGPU.

My first PC compatible, drew 150W at idle and 153W when
the single core CPU was railed. There was no power
....

I surely have no data to support my wild guess about power requirement of old motherboards and CPUs. :)

I only got the power meter, maybe 10 years ago, and to measure
the 440BX, that was done with a clamp-on AC ammeter.

The 150W measurement is not very accurate.

The power meter, uses a chip similar to the one that does
whole house digital power metering, about 1% accurate or so.
The measurement is done at the wall plug, the chip integrates
the area under curves and is capable of determine W and VA.
It can measure real and reactive power, and work out the
power factor PF. A power factor of 1.0 exists on ATX supplies
that have "active power factor correction". What that means
is, the power drawn is purely resistive (like an incandescent
bulb), there is no reactive (inductive or capacitive) power drawn.

For example, when the PC power supply makes +5VSB and the
computer is not really doing anything at all (this is NOT sleep),
the power meter measures this as 1W and 7VA. The local power
billing, charges for the 1 Watt. The 7VA is "reactive power"
and the power factor of the PSU is poor. The +5VSB portion
of the power supply, may not be using power factor correction,
which is why the reactive component (7VA or seven volt amperes)
is that high.

If I were to measure the 440BX computer again, I could use
that power meter, instead of the ammeter I used previously.
And then I would get a much more accurate measurement. The
ammeter has poor poor accuracy for this work.

But the message remains the same, that old computers are
generally bad citizens when it comes to wasting power.

The old computer, cannot even drive my 1920x1080 monitor. The
DACs on the video card, don't go that high. Which means if
I booted the 440BX into Windows 98, the screen would be
distorted and circles would be rendered as ellipses.
And back in the old days, the video card had two output
connectors, one might do 1152x864 , the other could do
1024x768 , so if you had dual-monitors, it could not
even do both of them at 1152. The second output used
a "cheaper" DAC.

Paul


--- PyGate Linux v1.5.12
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On 3/6/2026 1:57 AM, Paul wrote:

I only got the power meter, maybe 10 years ago, and to measure
the 440BX, that was done with a clamp-on AC ammeter.

The 150W measurement is not very accurate.
....
But the message remains the same, that old computers are
generally bad citizens when it comes to wasting power.

I understand.

Too bad nobody would manufacture Intel 440BX era hardware stuffs using
modern fab. :)

It's still an interesting experiement to try powering modern
motherboards using AA batteries.

--
@~@ Simplicity is Beauty! Remain silent! Drink, Blink, Stretch!
/ v \ May the Force and farces be with you! Live long and prosper!!
/( _ )\ https://sites.google.com/site/changmw/
^ ^ https://github.com/changmw/changmw

--- PyGate Linux v1.5.12
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)