E-ink calendar and ToDo list

[Giangi72]

Hello everyone,
I'm a new user in this newsgroup.
I would like to create a personal google calendar and ToDo list using a,
e-ink display without power adapter: a frame with display in it on the wall.
The display has a power consume of 48mW during refresh and I would like
to refresh it once at hour (maybe 30 minutes), obviously it will be
connected by wi-fi.
Once a week/month/year I can take it from the wall, charge it an the put
it again on the wall.

I have some questions to people more expert than me:
1) It have to be without power adapter so I need a "battery", which is
the cheapest way to feed the raspberry? I googled for it but I see
battery hats for 3-4 € and UPS hats for 30-40 € but I don't undestand
the differences between them.
2) I would like to use Visual Studio Code to develop the software side.
For the power consumes is it better to install a full OS (Raspberry Pi
OS with desktop) or a limited one (Raspberry Pi OS Lite) and develop in
different ways?

These are the first dubts I have.
If you have more informations or dubts please don't be shy :-)
Many thanks in advance
Giangi

[Giangi72]

Il 16/12/2021 15:00, Giangi72 ha scritto:

> 1) It have to be without power adapter so I need a "battery", which is
> the cheapest way to feed the raspberry? I googled for it but I see
> battery hats for 3-4 € and UPS hats for 30-40 € but I don't undestand
> the differences between them.
> 2) I would like to use Visual Studio Code to develop the software side.
> For the power consumes is it better to install a full OS (Raspberry Pi
> OS with desktop) or a limited one (Raspberry Pi OS Lite) and develop in
> different ways?

3) Is there any way to suspend the raspberry after refresh and restart
it again before next refresh to save power?

thanks again
Giangi

[Giangi72]

Il 16/12/2021 17:21, Dennis Lee Bieber ha scritto:
> On Thu, 16 Dec 2021 15:00:37 +0100, Giangi72 <inv...@invalid.net>
> declaimed the following:

>
>> The display has a power consume of 48mW during refresh and I would like
>> to refresh it once at hour (maybe 30 minutes), obviously it will be
>> connected by wi-fi.
>

> What voltage/amperage? That may have a limitation on what type of power
> source you can use.

It SHOULD be powered by raspberry, the power consume of display during
NOT refresh should be 0 (it is an e-ink, the consume should be only
during refresh).
During refresh, I read on datasheet
Typical operating current 6.6 mA

>> Once a week/month/year I can take it from the wall, charge it an the put
>> it again on the wall.
>>
>> I have some questions to people more expert than me:
>> 1) It have to be without power adapter so I need a "battery", which is
>> the cheapest way to feed the raspberry? I googled for it but I see
>> battery hats for 3-4 € and UPS hats for 30-40 € but I don't undestand
>> the differences between them.
>

> UPS is "Uninterruptible Power Supply" -- a system that uses wall power
> most of the time, but can switch to battery backup when the wall power
> fails -- and can do this in a way that the attached device does not notice
> a power glitch. If it is a good one, when the wall power comes back up, it
> will recharge the battery. It may also (via monitoring software and GPIO or
> such) signal to the R-Pi if both wall power has been lost, AND the battery
> is running down -- so the R-Pi can perform a clean shutdown.
>
> Obviously needs a power adapter for normal operation.
>
> In contrast, a plain battery is just that... a battery that can run
> down. Linux systems do NOT like having the power pulled -- it can lose
> unwritten data (at best) requiring a journal rebuild of the file system on
> next start-up, up to corrupting the file system beyond recovery.

Perfect, but are there any battery hats (cheaper) that can tell to
raspberry how many remaning charge they have?
e.g.
https://thepihut.com/products/li-ion-battery-hat-for-raspberry-pi
but I found the same, in the past, a 3-4€

If I'm able to understand how many charge remains I can send a visible
warning: mail, notification, ...

>> 2) I would like to use Visual Studio Code to develop the software side.
>> For the power consumes is it better to install a full OS (Raspberry Pi
>> OS with desktop) or a limited one (Raspberry Pi OS Lite) and develop in
>> different ways?
>

> I'm tempted to suggest you develop on one device using a full desktop
> if you require any graphical editor/development environment, and then
> install the application on a device without a desktop system. Your
> application sounds like it could run using the smallest WiFi enabled R-Pi,
> but for development (especially if using C/C++ or other compiled language)
> might be better on a larger/more powerful R-Pi model.

This could be a good idea, or develop on a full OS sd card and the swap
to a liter SD card.
But in terms of power consume: what happens if I use a full desktop OS
and the switch off the desktop environvent?

> Does a library already exist for the display, or do you have to create
> that too?

No, no everything already exists
https://www.waveshare.com/wiki/7.5inch_e-Paper_HAT_(B)

> Actually, have you even done a search to see if this has been done
> already? The first page of a search on Google found two implementations
> (the first having way too many reviews/references on the page <G>)
>
> https://www.tomshardware.com/news/raspberry-pi-e-ink-google-candar
> https://github.com/speedyg0nz/MagInkCal

I know this project, but it is in pyton, some informations are very
useful but not the whole project

> https://hackaday.com/2019/02/11/get-organized-with-this-raspberry-pi-e-ink-calendar/

This is new to me. Also this is interesting, I think I'll steal the
button to refresh idea but ...pyton again.

My problem is now mainly on power and power duration: which cheap hat
use to power the raspberry.
And then my later question to reduce the power consume

3) Is there any way to suspend the raspberry after refresh and restart
it again before next refresh to save power?

Many thanks
Giangi

[Andy Burns]

Giangi72 wrote:

> My problem is now mainly on power and power duration: which cheap hat use to
> power the raspberry.
> And then my later question to reduce the power consume
>
> 3) Is there any way to suspend the raspberry after refresh and restart it again
> before next refresh to save power?

Maybe use an ESP32 instead of an RPi?

[Theo]

That was my first thought. Although it depends how complicated the Google
Calendar API is, and whether it's something the ESP software can handle. If
it's basic JSON or something it'll probably be fine, but if it needs lots of
stateful stuff (OAuth etc) that could be harder. Something like CalDav
looks quite complicated.

But maybe you could run that part on a server/a RPi and have the ESP
poll it with simpler requests?

I quite like the M5Stacks as neatly packaged ESP32 boards, with batteries
and displays:
https://shop.m5stack.com/products/m5paper-esp32-development-kit-960x540-4-7-eink-display-235-ppi?variant=37595977908396
https://shop.m5stack.com/products/m5stack-esp32-core-ink-development-kit1-54-elnk-display?variant=37404426174636

Theo

[Giangi72]

I don't know ESP32, what I see is that it is an arduino-like board.
From what I saw I'm not sure it is able to support complex software as
google calendar as mentioned in another message, but I'm not an expert
Some addtional informations here
https://stackoverflow.com/questions/19766912/how-do-i-authorise-an-app-web-or-installed-without-user-intervention/19766913#19766913

What are the main advantages to use ESP32 instead of Raspberry pi?
Thanks
Giangi

[Theo]

Giangi72 <inv...@invalid.net> wrote:
> I don't know ESP32, what I see is that it is an arduino-like board.
> From what I saw I'm not sure it is able to support complex software as
> google calendar as mentioned in another message, but I'm not an expert
> Some addtional informations here
> https://stackoverflow.com/questions/19766912/how-do-i-authorise-an-app-web-or-installed-without-user-intervention/19766913#19766913
>
> What are the main advantages to use ESP32 instead of Raspberry pi?

ESP32 is a wifi microcontroller. Being a microcontroller, low power
is part of its DNA. Microcontrollers can enter a low power state until
something happens - for example your device sleeping until the next update
is required. You can be sure that the device is actually going to idle for
that time.

An application class processor can't really do that, because there's too
much other stuff going on. Maybe you could if you ran bare-metal, but not
with a full Linux OS. Smartphones are a good example - you might get a week
of idle, but once you have wifi and notifications and and... the battery
life starts to drop.

If you're going to run an e-ink screen to save power, it would be
unfortunate if the CPU wasted most of it.

On the other hand, e-readers are good examples of basic application-class
cores (often comparable with the original Pi Zero) with e-ink, micro SD and
wifi, but which aggressively turn those things off when not needed. An
e-reader might be an interesting starting point for such a project - eg I
have a Nook Simple Touch that runs Android (2.1 I think - despite being
released in the Android 4 era, later Androids got more resource hungry so
they went with an older version).

Theo

[Giangi72]

Il 16/12/2021 22:51, Dennis Lee Bieber ha scritto:
> On Thu, 16 Dec 2021 17:58:30 +0100, Giangi72 <inv...@invalid.net>
> declaimed the following:
>

>> During refresh, I read on datasheet
>> Typical operating current 6.6 mA
>>
>

> Taking your previous statement of 48mW and this 6.6mA, that yields a
> 7.3V power source (W = V * A). Raspberry-Pi's provide 5V and 3.3V (and
> GPIOs can be killed if fed more than 3.3V; the 5V is from the input power
> supply and primarily provided to USB ports)

The display is directly connected to raspberry's GPIOs and use 3.3 V.
But you are right, I checked the datasheet and I found different values
than 48mW written on amazon.

There are 2 datasheets: V2 and V3 and I don't know the differences

V2 page 12:
https://www.waveshare.com/w/upload/4/44/7.5inch_e-Paper_B_V2_Specification.pdf
Image update current typical 8, max 12 mA
Power panel (update) typical 26.4, max 45 mW
=
typical 3.3, max 3.75 V

V3 page 9:
https://www.waveshare.com/w/upload/8/8c/7.5inch-e-paper-b-v3-specification.pdf
Typical power 21.78 mW

Typical operating current 6.6 mA

=
3.3V

>> Perfect, but are there any battery hats (cheaper) that can tell to
>> raspberry how many remaning charge they have?
>> e.g.
>> https://thepihut.com/products/li-ion-battery-hat-for-raspberry-pi
>> but I found the same, in the past, a 3-4€
>>
>

> Did you look at the documentation for that unit?

Sigh, no, just now.
It was under "View More" then "wiki ...", then manual

[CUT]
This means there is no way to power raspberry by battery and let it last
for at least a week?

>> I know this project, but it is in pyton, some informations are very
>> useful but not the whole project
>>
>

> So what is wrong with Python -- it IS the language the R-PI foundation
> targeted for users <G>

:-) you are absolutely right, simply it is not the language I usually use.
I have a very low knowledge in pyton and I don't want to loose time
learning it well.
Additionally I would like to also implement a sort of ToDo list so I
need to change the application.

> For an hourly/half-hourly (or whatever rate you want to refresh) you
> don't need the speed from using a compiled language. And the easiest way to
> set up that timing would be to have the application merely do one refresh
> and exit -- and invoke it using a crontab entry specifying how often to
> run.

This is absolutely correct and wonderful suggestion.

>> 3) Is there any way to suspend the raspberry after refresh and restart
>> it again before next refresh to save power?

[CUT]
> WiFi is likely the biggest power drain you have on an R-Pi.

Thanks for your detailed explanations, I surrender: no battery.
If I connect it to a power source all my problems are washed away.
Many thanks
Giangi

[Giangi72]

Il 17/12/2021 00:25, Theo ha scritto:
> Andy Burns <use...@andyburns.uk> wrote:
>> Giangi72 wrote:
>>
>>> My problem is now mainly on power and power duration: which cheap hat use to
>>> power the raspberry.
>>> And then my later question to reduce the power consume
>>>
>>> 3) Is there any way to suspend the raspberry after refresh and restart it again
>>> before next refresh to save power?
>>
>> Maybe use an ESP32 instead of an RPi?
>
> That was my first thought. Although it depends how complicated the Google
> Calendar API is, and whether it's something the ESP software can handle. If
> it's basic JSON or something it'll probably be fine, but if it needs lots of
> stateful stuff (OAuth etc) that could be harder. Something like CalDav
> looks quite complicated.
>
> But maybe you could run that part on a server/a RPi and have the ESP
> poll it with simpler requests?

This is a good idea, I can create a mini service on my web site and the
application should only read the result

Just to let you hate me better :-): what about power consumes?
Is it possible to power it by battery and let it last for one week or more?

> I quite like the M5Stacks as neatly packaged ESP32 boards, with batteries
> and displays:
> https://shop.m5stack.com/products/m5paper-esp32-development-kit-960x540-4-7-eink-display-235-ppi?variant=37595977908396
> https://shop.m5stack.com/products/m5stack-esp32-core-ink-development-kit1-54-elnk-display?variant=37404426174636

Thanks for your suggestions but the display is too small and I would
like to put it in a beautiful frame.
Thanks
Giangi

[Andy Burns]

Giangi72 wrote:

> What are the main advantages to use ESP32 instead of Raspberry pi?

putting it to deep sleep between updates, and boards with available LiPo battery
management.

[Giangi72]

Could you please give me more informations about the battery management.
Only links is enough
Thanks
Giangi

[Andy Burns]

Giangi72 wrote:

> Andy Burns ha scritto:
>
>> Giangi72 wrote:
>>
>>> What are the main advantages to use ESP32 instead of Raspberry pi?
>>
>> putting it to deep sleep between updates, and boards with available LiPo
>> battery management.
>
> Could you please give me more informations about the battery management.
> Only links is enough

e.g. <https://unexpectedmaker.com/tinys2>

[John Aldridge]

In article <j23att...@mid.individual.net>, inv...@invalid.net
says...

>
> This means there is no way to power raspberry by battery and let it last
> for at least a week?

I think your options are...

1. Use a Very Big battery. AFAIR, a PiZeroW draws around 250 mA on
average, when idling, so something around 40 Ah at 5V. Say 20 Ah at 12V,
after you've included the 12V->5V voltage converter. It'll be something
like a car battery in size, and correspondingly expensive.

2. Add some extra circuitry to completely power off the RPi between
refreshes (making sure you wait until the RPi has shut down cleanly
before cutting the power). I think you'd have to home-brew this
circuitry.

3. Use a different processor, e.g. Arduino, ESP32, RP2040. I don't have
any personal experience, but I think these all support a low power deep-
sleep mode from which you can wake on a timer.

--
John

[Martin Gregorie]

The RP2040, aka PICO, may be what the OP wants. Its cheap, low power, can
be made to sleep for a user-definable time, and is intended to be
programmed in C/C++ or PICO-assembler. It has no OS - just a set of
support code, including a C standard library. You link your application
code with C library etc and the hardware-specific support libraries to
get a binary blob. Load that into the PICO, connect power and it runs.
All the documentation is on the PICO section of the RaspberryPi
Foundation website so, if this sounds interesting, go and read the docs.

The process of developing code for it looks similar to the way you'd
program a Parallax STAMP or a PICAXE chip except that both of those are
programmed using their own flavours of compiled integer BASIC while the
PICO can be programmed in C/C++.

[Theo]

Martin Gregorie <mar...@mydomain.invalid> wrote:
> The RP2040, aka PICO, may be what the OP wants. Its cheap, low power, can
> be made to sleep for a user-definable time, and is intended to be
> programmed in C/C++ or PICO-assembler. It has no OS - just a set of
> support code, including a C standard library. You link your application
> code with C library etc and the hardware-specific support libraries to
> get a binary blob. Load that into the PICO, connect power and it runs.
> All the documentation is on the PICO section of the RaspberryPi
> Foundation website so, if this sounds interesting, go and read the docs.
>
> The process of developing code for it looks similar to the way you'd
> program a Parallax STAMP or a PICAXE chip except that both of those are
> programmed using their own flavours of compiled integer BASIC while the
> PICO can be programmed in C/C++.

The RP2040 and the RPi Pico don't have networking. You can connect it via a
wire to a USB port, but that's not what the OP wants.

It seems the Arduino people have an 'Nano RP2040 Connect' which has a u-blox
NINA WiFi module on it... that has an ESP32 inside.

So you might as well just use an ESP32 to begin with.

Theo

[Giangi72]

Il 17/12/2021 14:40, Andy Burns ha scritto:

>>>> What are the main advantages to use ESP32 instead of Raspberry pi?
>>>
>>> putting it to deep sleep between updates, and boards with available
>>> LiPo battery management.
>>
>> Could you please give me more informations about the battery management.
>> Only links is enough
>
> e.g. <https://unexpectedmaker.com/tinys2>

Thanks
Giangi

[Giangi72]

Il 17/12/2021 16:04, Theo ha scritto:

> The RP2040 and the RPi Pico don't have networking. You can connect it via a
> wire to a USB port, but that's not what the OP wants.
>
> It seems the Arduino people have an 'Nano RP2040 Connect' which has a u-blox
> NINA WiFi module on it... that has an ESP32 inside.
>
> So you might as well just use an ESP32 to begin with.

Thanks to everyone.
I think the ESP32 + web service is the best solution.
I have found this board with integrated battery management and socket
https://www.aliexpress.com/item/1005003416635072.html?spm=a2g0o.cart.0.0.31be3c00pggd6r&mp=1
and
https://www.aliexpress.com/item/1005002412550877.html?spm=a2g0o.cart.0.0.31be3c00pggd6r&mp=1
and I read
"Lithium battery interface(battery not include), 500mA Max charging
current."

1) What does it means? If I have a 700mA or 1000mA battery? Sorry for
this newbie question.
2) I don't understand how to charge the battery once it is connected to
the board. Using usb port or detaching it and charging it externally?

Many thanks
Giangi

[Andy Burns]

Giangi72 wrote:

> 500mA Max charging current."
>
> 1) What does it means? If I have a 700mA or 1000mA battery? Sorry for this
> newbie question.

I presume you mean 700mAh or 1000mAh capacity, rather than 500mA charging
current, just mean it'll take between 90 minutes and 2 hours to charge

> 2) I don't understand how to charge the battery once  it is connected to the
> board. Using usb port or detaching it and charging it externally?

whenever USB power is supplied, the board will charge the battery.

[Giangi72]

Il 17/12/2021 18:04, Andy Burns ha scritto:
> Giangi72 wrote:
>
>> 500mA Max charging current."
>>
>> 1) What does it means? If I have a 700mA or 1000mA battery? Sorry for
>> this newbie question.
>
> I presume you mean 700mAh or 1000mAh capacity, rather than 500mA
> charging current, just mean it'll take between 90 minutes and 2 hours to
> charge

Sorry, obviously my fault it was mAh, I read 500mA and wrote 700mA.
My newbie question is:
I read "500mA Max charging current", if I have a 700mAh battery will it
work? Obviously with longer charging times.

>> 2) I don't understand how to charge the battery once  it is connected
>> to the board. Using usb port or detaching it and charging it externally?
>
> whenever USB power is supplied, the board will charge the battery.

Perfect, thanks.
Giangi

[The Natural Philosopher]

On 17/12/2021 18:40, Dennis Lee Bieber wrote:
> Charging Li-Ion batteries is not simple, advanced charging circuits
> include temperature sensors to detect when the charge has peaked and avoid
> overheating the battery.

In cars perhaps. Not at lower capacities. Temperature sensors do not
detect 'when the battery has charged' on Li-Ion. Voltage detects that.
The temperature sensors on big vehicle packs are there for safety, in
case a cell goes bad and overheats.


The very simple 'charge each cell at less than the hourly rate to 4.2v
and no more' algorithm dominates hobby chargers.

The R/C model community has all the bits and pieces to charge packs and
the requisite circuitry to drop the voltage to 5V for avionics - or Pi -
usage.
e.g.

https://hobbyking.com/en_us/zippy-compact-8000mah-2s1p-30c.html
= 8 Ah 7.4v battery.

https://hobbyking.com/en_us/turnigy-5a-8-26v-sbec-for-lipo.html
= Switched mode 5V output regulator

https://hobbyking.com/en_us/turnigy-e3-compact-2s-3s-lipo-charger-100-240v-eu-plug.html
= 3 x 800mAh balancing charger.

Put that lot together and you have a system that will deliver ~5V/1A
average for as long as the mains is on, and for about 10 hours if the
mains goes out.


--
"Socialist governments traditionally do make a financial mess. They
always run out of other people's money. It's quite a characteristic of them"

Margaret Thatcher

[sc...@alfter.diespammersdie.us]

Giangi72 <inv...@invalid.net> wrote:
> [CUT]
> This means there is no way to power raspberry by battery and let it last
> for at least a week?

It's possible...it's just a matter of figuring out how big a battery pack
would do that. Someone else theorized that a pack built of 14500 Li-ion
cells (two in series, presumably, through a buck converter to get 5V) might
run a Raspberry Pi 3B+ for five hours. 18650s are more common and larger;
for sake of argument, let's say you get 3x the runtime (2400 mAh) from an
18650 as you'd get from a 14500. If two 18650s would run the aforementioned
RPi 3B+ for 15 hours, 24 18650s in a 2S12P configuration would provide 7.5
days of runtime.

Whether a pack made of two dozen 18650s would be too large for your
application is a question only you can answer. :)

--
_/_
/ v \ Scott Alfter (remove the obvious to send mail)
(IIGS( https://alfter.us/ Top-posting!
\_^_/ >What's the most annoying thing on Usenet?

[Folderol]

On Mon, 20 Dec 2021 23:05:36 GMT
sc...@alfter.diespammersdie.us wrote:

>Giangi72 <inv...@invalid.net> wrote:
>> [CUT]
>> This means there is no way to power raspberry by battery and let it last
>> for at least a week?
>
>It's possible...it's just a matter of figuring out how big a battery pack
>would do that. Someone else theorized that a pack built of 14500 Li-ion
>cells (two in series, presumably, through a buck converter to get 5V) might
>run a Raspberry Pi 3B+ for five hours. 18650s are more common and larger;
>for sake of argument, let's say you get 3x the runtime (2400 mAh) from an
>18650 as you'd get from a 14500. If two 18650s would run the aforementioned
>RPi 3B+ for 15 hours, 24 18650s in a 2S12P configuration would provide 7.5
>days of runtime.
>
>Whether a pack made of two dozen 18650s would be too large for your
>application is a question only you can answer. :)

I wouldn't advise it! Once you start stacking up large numbers of batteries all
sorts of load balancing issues become apparent - both for charge and discharge.

--
Basic

[Martin Gregorie]

On Tue, 21 Dec 2021 16:46:09 +0000, Folderol wrote:

>
> I wouldn't advise it! Once you start stacking up large numbers of
> batteries all sorts of load balancing issues become apparent - both for
> charge and discharge.
>

Quite, Unless you've got some cutting-edge chippery that can run for
months of a 2032 coin cell, there are only about two sensible ways to go:

1) use some sort of UPS with sufficient battery capacity to run your
calendar for 2-3 hours until the power comes back AND has the ability to
tell your system to do an orderly shutdown if the power hasn't come back
and its battery is nearly flat. If you choose a Pi, as least some of the
UPS HATs made for fitting to a Pi can't do that.

2) Forget about using Internet time and use either a GPS receiver or one
of the VLF frequency time broadcasts as your master time source. That
might use:

- the DCF77 German sime signal with Conrad part number 641138-92
as receiver

- http://www.buzzard.me.uk/jonathan/radioclock.html using the Rugbk (UK)
time signal - it can be interfaced to the standard Linux time system
using this software:
http://www.buzzard.me.uk/jonathan/downloads/radioclk-1.0.tar.gz

- one of the US time sources, WWVB or WWVH

I've used Jonathon Buzzard's receiver and Rugby in the past: its 'just
worked'

I've also used GPS time - older Puck-style receivers can be picked up in
FleaBay. They normally use an RS-232 serial connection as output, and
good, inexpensive RS-232->USB adapters are easy to find.

The benefit of of these time sources is that it doesn't matter if your
calendar does get killed by a storm or power cut: when power comes back
the system can re-boot and re-establish an accurate time in a few minutes
at most. also, it will still work if the power cut took out your section
of the internet.

[Theo]

Martin Gregorie <mar...@mydomain.invalid> wrote:
> The benefit of of these time sources is that it doesn't matter if your
> calendar does get killed by a storm or power cut: when power comes back
> the system can re-boot and re-establish an accurate time in a few minutes
> at most. also, it will still work if the power cut took out your section
> of the internet.

The OP wants to display their Google Calendar and todo list in a frame on
their wall. That requires an internet connection. You can't display the
appointments in their calendar without fetching it from the internet - you
could only have a generic day/date calendar which is not what they're asking
for. I suppose you could cache the calendar in case the network goes down,
and a third party time source would help with bring that back up, but that's
a fairly niche use case, and no help if the calendar is later changed.

And so the problem is that the OP needs an internet connection but the power
consumption is too high to run a Pi full time unless the battery is
implausibly large to mount on the wall. The key is thus turning things off
- either running a Pi on some kind of timeswitch, or using something that is
better at entering a low power sleep mode.

Finding out the time is not the problem, getting the calendar updates is.
The dilemma is that something that can handle the complexity of talking to
the Google servers is probably not low power enough to run from a battery in
a wall frame for a week/month/year. Which is what makes me think a two-box
solution is the way forward.

Theo

[Martin Gregorie]

On 21 Dec 2021 22:46:27 +0000 (GMT), Theo wrote:

> Finding out the time is not the problem, getting the calendar updates
> is. The dilemma is that something that can handle the complexity of
> talking to the Google servers is probably not low power enough to run
> from a battery in a wall frame for a week/month/year. Which is what
> makes me think a two-box solution is the way forward.
>

Yes, that or using a UPS with a smaller 1-2 hour battery, which probably
would keep it up indefinitely with mains power thats as reliable as here
or in most of the EU.

[Axel Berger]

Theo wrote:
> That requires an internet connection. You can't display the
> appointments in their calendar without fetching it from the internet - you
> could only have a generic day/date calendar which is not what they're asking
> for.

That's not true. All my calendars are local to my phone and not synced
to anything not fully under my control. I regularly save backups over my
LAN at home and could import that to any other local calendar.

In the present case you do have a point though, unless the OP is
satisfied with having his appointments on the wall at home and nowhere
else. But that too does not need the internet. Syncing between the
(presumably) phone and the wall box over the LAN when at home would
suffice.


--
/¯\ No | Dipl.-Ing. F. Axel Berger Tel: +49/ 221/ 7771 8067
\ / HTML | Roald-Amundsen-Straße 2a Fax: +49/ 221/ 7771 8069
 X in | D-50829 Köln-Ossendorf http://berger-odenthal.de
/ \ Mail | -- No unannounced, large, binary attachments, please! --

[Ahem A Rivet's Shot]

On Wed, 22 Dec 2021 09:16:19 +0100
Axel Berger <Sp...@Berger-Odenthal.De> wrote:

> Theo wrote:
> > That requires an internet connection. You can't display the
> > appointments in their calendar without fetching it from the internet -
> > you could only have a generic day/date calendar which is not what
> > they're asking for.
>
> That's not true. All my calendars are local to my phone and not synced

The OP did specify *Google* calendar - which I rather think does
require a public internet connection.

There are of course other solutions that do not but from the
perspective of a battery powered, wireless device on the wall it makes no
difference whether the data is on a local server, at Google or elsewhere it
has to be accessed using IP - in other words over an internet connection
and it makes no difference whether the scope of that connection is global
or local.

--
Steve O'Hara-Smith
Odds and Ends at http://www.sohara.org/

[Theo]

Axel Berger <Sp...@berger-odenthal.de> wrote:
> Theo wrote:
> > That requires an internet connection. You can't display the
> > appointments in their calendar without fetching it from the internet - you
> > could only have a generic day/date calendar which is not what they're asking
> > for.
>
> That's not true. All my calendars are local to my phone and not synced
> to anything not fully under my control. I regularly save backups over my
> LAN at home and could import that to any other local calendar.
>
> In the present case you do have a point though, unless the OP is
> satisfied with having his appointments on the wall at home and nowhere
> else. But that too does not need the internet. Syncing between the
> (presumably) phone and the wall box over the LAN when at home would
> suffice.

Like I said, the OP's calendar lives on Google Calendar. You can't fetch
that without internet, because that calendar lives in the cloud. You can
sync to something locally (a server/phone/another Pi), and then the wall
frame fetches from there, which is the two-box solution I suggested. If you
want a one-box solution you need internet on your one box, there's no
getting around that.

You can of course have a wall frame without an internet connection, to which
you enter calendar entries on a touch screen or SD card or whatever. But
that is an entirely different problem from the one the OP asked for, which
was something to show live updates from their Google Calendar. As is
worrying about getting the time from GPS, which is also a problem the OP
doesn't have.

Theo

[Ahem A Rivet's Shot]

On 22 Dec 2021 09:29:59 +0000 (GMT)

Theo <theom...@chiark.greenend.org.uk> wrote:

> Like I said, the OP's calendar lives on Google Calendar. You can't fetch
> that without internet, because that calendar lives in the cloud. You can
> sync to something locally (a server/phone/another Pi), and then the wall
> frame fetches from there, which is the two-box solution I suggested. If
> you want a one-box solution you need internet on your one box, there's no
> getting around that.

Quite so, and the easiest (but expensive) way to achieve a one box
solution would be to use an off the shelf e-ink tablet such as a Likebook,
but given that they have a display in hand to use and want to DIY the ESP32
option being explored elsethread seems to fit the requirements.

[The Natural Philosopher]

Not really. not for parelleling.

The charge/voltage curves for cells are all similar, what differs is
their capacity , so you might think it a problem as some cells charge up
sooner than others, but that doesn't matter except at very high charge
rates, because the same voltage represents the same state of charge for
all the paralleled cells.


But lithium batteries up to 10AH or even more are available. And at 14v
nominal that's 140 Watt hours, and a Pi is about what - 5 watts or so?

Or a car battery at 12v/70Ah nets you 160 hours

Forget voltages and architectures - that's all soluble, just pick a
battery that has the capacity you need and is happy to be trickle
charged at slightly more than the Pis power draw.

Pi draw is roughly 5W.
a week is 7x24 = 168 hours

168 hours times 5 W = 840 watt hours

At 12v nominal that is 70 A/h which is spot on for a car or caravan battery.

Looking at lithium cells I cant find a pack at less than £100 that comes
close.

440Wh at £156 and 22v is closest.

So at least you know what you will be dealing with. A car or caravan
battery.

And then the problem resolves into how to step the voltage down for api
- but that trivial - the amount of car to USB power converters is
enormous, and likewise mains to trickle charge a caravan battery is off
the self.


--
"Corbyn talks about equality, justice, opportunity, health care, peace,
community, compassion, investment, security, housing...."
"What kind of person is not interested in those things?"

"Jeremy Corbyn?"

[Andy Burns]

The Natural Philosopher wrote:

> just pick a battery that has the capacity you need and is happy to be trickle
> charged at slightly more than the Pis  power draw.

Once you reach the point where the battery is too large to fit within the frame
on the wall, you either have to re-think whether a dangling power cable is
acceptable, or use something other than a Pi ...

[The Natural Philosopher]

Well exactly. peole seem to think that engineering is how to make stuff work

Mostly the theoretic side of engineering is there to show you what stiff
can never work, so its crazy to build it.

Hasn't stopped renewable energy though, but then they never bothered to
analyse it, they just built it for profit. irrespective of whether it
worked or not.



--
I would rather have questions that cannot be answered...
...than to have answers that cannot be questioned

Richard Feynman

[Axel Berger]

Theo wrote:
> Like I said, the OP's calendar lives on Google Calendar.

Perhaps I misunderstood, or at least understood differently. For you a
Google calendar is a calendar maintained for you by Google in their
cloud. For me it's the calendar app in my phone and provided by Google.

[Martin Gregorie]

On Wed, 22 Dec 2021 12:21:39 +0000, The Natural Philosopher wrote:

> Hasn't stopped renewable energy though, but then they never bothered to
> analyse it, they just built it for profit. irrespective of whether it
> worked or not.
>

The only really crazy item on the table right now is so-called "Blue
Hydrogen": the energetics of making it mean that burning coal is less
environmentally damaging than making and using the 'Blue hydrogen'.

[The Natural Philosopher]

I beg to differ. Everything on the table right now is crazy.
I've spent many years analysing it, and the only thing that will
actually work, other than a 97% drop in population and a return to the
sort oft technology the Greens would understand. Presumably horses whose
methane emanations would be forgotten in their sheer organicity... is
nuclear power of some sort.

Its abundant, uranium is ubiquitous, cheap and easily stockpiled, and
comes already stored. The reactors are well understood known technology,
and it transpires that radiation isn't nearly as dangerous as we had
been led to believe, and we could easily achieve adequate safety at much
lower cost.

But nuclear isn't on the table. Every other permutation of artStudent™
thinking in terms of wacky schemes that were discarded generations ago
by sane engineers, are the substance of dinner party chatterati.
Hydrogen, carbon capture, Geothermal, tidal, wave, run of river,
spinning flywheels, grid scale batteries molten salt tanks (that one
might almost work with a nuke to drive it) heat pumps, electric cars.
All of this to avoid the basic solution that nuclear power means, at a
stroke, you don't need ANY of that crap.

At current electricity wholesale prices, a nuclear fed simple restive
heater would be one third the cost of a renewable fed heat pump in terms
of electricity, and one tenth the installation cost. Not very efficient,
but who cares? uranium is cheap as chips and there's tonnes and tonnes
of it around and you don't need very much. E=mC^2 etc etc

In short there is no shortage of cheap energy. Of all the problems
society has, lack of cheap energy is not one of them. Only our refusal
to embrace it is the problem

That doesn't sort out the use of fossil fuels as chemical feedstocks of
course, and as the only appropriate energy density energy source to fly
transatlantic airliners, but we will all be locked down anyway, wont we.?





--
You can get much farther with a kind word and a gun than you can with a
kind word alone.

Al Capone

[Martin Gregorie]

On Wed, 22 Dec 2021 13:18:14 +0000, The Natural Philosopher wrote:

> I beg to differ. Everything on the table right now is crazy.
> I've spent many years analysing it, and the only thing that will
> actually work, other than a 97% drop in population and a return to the
> sort oft technology the Greens would understand. Presumably horses whose
> methane emanations would be forgotten in their sheer organicity... is
> nuclear power of some sort.
>

You're certainly right about population. That seems to be such a bad
thing to say that its not even whispered about by anybody. Yet, any
attempt to prevent global warming is doomed to failure without reductions
in both the humam population and in individual (net) resource consumption.

> Its abundant, uranium is ubiquitous, cheap and easily stockpiled, and
> comes already stored. The reactors are well understood known technology,
> and it transpires that radiation isn't nearly as dangerous as we had
> been led to believe, and we could easily achieve adequate safety at much
> lower cost.
>

Maybe it can be used, presumably in small factory-produced, reactors
similar to those used in submarines, but studies I've seen point out that
in terms of global warming, nukes still produce around 30% of the CO2
from conventional thermal generation once mining and refining the stuff
is taken into account:
Storm van Leeuwen & Smith - http://www.stormsmith.nl/

This also fails to ignore the problem of dealing with fanatics who think
nuking someone or something would Be A Good Idea.

I'n not against nuclear, PROVIDED THAT the problems of disposing of the
radioactive waste from fuel preparation and the radioactive debris from
decommissioning old plant can be sorted out. To date the solutions have
mostly been to pile the junk in a corner and hope nobody notices it.

> That doesn't sort out the use of fossil fuels as chemical feedstocks of
> course, and as the only appropriate energy density energy source to fly
> transatlantic airliners,
>

True enough, do we really need so many transatlantic airliners? Also, I'm
seeing talk of running ships on Ammonia rather then heavy oil:

NH3 + O2 => N2 + H2O + energy

Is ammonia's energy content high enough to run an airliner? Gaseous
hydrogen's isn't and nor is liquid hydrogen's once to take density/volume/
container mass into account.

[The Natural Philosopher]

On 22/12/2021 14:01, Martin Gregorie wrote:
> On Wed, 22 Dec 2021 13:18:14 +0000, The Natural Philosopher wrote:
>
>> I beg to differ. Everything on the table right now is crazy.
>> I've spent many years analysing it, and the only thing that will
>> actually work, other than a 97% drop in population and a return to the
>> sort oft technology the Greens would understand. Presumably horses whose
>> methane emanations would be forgotten in their sheer organicity... is
>> nuclear power of some sort.
>>
> You're certainly right about population. That seems to be such a bad
> thing to say that its not even whispered about by anybody. Yet, any
> attempt to prevent global warming is doomed to failure without reductions
> in both the humam population and in individual (net) resource consumption.
>

Oh purlease. What 'global warming'?

That stopped 20 years ago before they started 'adjusting' historical
records to bring it back.

The winter I am looking out at with frost on the ground is as bleak as
it was 50 years ago.

High levels of population are a proble'm but not because of lack of
abundant cheap energy, or because of 'global warming

Water is a far more urgent problem.

>> Its abundant, uranium is ubiquitous, cheap and easily stockpiled, and
>> comes already stored. The reactors are well understood known technology,
>> and it transpires that radiation isn't nearly as dangerous as we had
>> been led to believe, and we could easily achieve adequate safety at much
>> lower cost.
>>
> Maybe it can be used, presumably in small factory-produced, reactors
> similar to those used in submarines, but studies I've seen point out that
> in terms of global warming, nukes still produce around 30% of the CO2
> from conventional thermal generation once mining and refining the stuff
> is taken into account:
> Storm van Leeuwen & Smith - http://www.stormsmith.nl/
>

Well its definitely in te renewable lobbies interst to pay for a study
like that.

It is of course nonsense. You need far more concretre per lifetime MWh
generated in a windmill than a nuke

> This also fails to ignore the problem of dealing with fanatics who think
> nuking someone or something would Be A Good Idea.
>

Anyone that stupid cannot build a bomb. They cant even build a bomb
properly out of fertiliser which the IRA were much better at. If - say -
Iran were to manange to deliver a nuke into Israel and set it off, can
you imagine what would happen to Iran?

> I'n not against nuclear, PROVIDED THAT the problems of disposing of the
> radioactive waste from fuel preparation and the radioactive debris from
> decommissioning old plant can be sorted out. To date the solutions have
> mostly been to pile the junk in a corner and hope nobody notices it.

No, all of the solutions that work perfectly well have been opposed by
greens.

The simplest and most obvious and cheapest solution would be to take it
all out to the Marianas and throw it overboard in concrete cans. To join
the 4 BILLION tonnes of uranium in et seas already

uber low level waste is merely landfilled in steel cases that will do te
100 years plus needed.

uber high level is tomorrows nuclear fuel anyway.

Intermediate - the sort of thousand year slightly radioactive and
biological active shit - ceasium and the like - simply needs putting out
of reach for 1000 years.
How old are the pyramids? Stonehenge? I maen really galssifying it and
stuffing it at the bottom of a disused coal mine in a sealed box is fine.

If you want to REALLY scare yourself go and take a geiger counter to
some disused radium/uranium mine in cornwall or dartmoor and exmoor.

hundreds of times worse than a block of internediate waste.

>
>> That doesn't sort out the use of fossil fuels as chemical feedstocks of
>> course, and as the only appropriate energy density energy source to fly
>> transatlantic airliners,
>>
> True enough, do we really need so many transatlantic airliners? Also, I'm
> seeing talk of running ships on Ammonia rather then heavy oil:
>
> NH3 + O2 => N2 + H2O + energy

Well ships of course will be nuclear powered - the weight of the
shielding is not an issue - but ammonia is an interesting one. It would
certainly be a way to replace natural gas to make fertiliser anyway,
which is one of the big non-energy uses of it.

Hydrogen my well be useful as a reducing agent for smelting metals,

But its hard to see anything to beat long chain hydrocarbons for
portable energy use, and its possible that dirt cheap nuclear poweer
would allow reasonable yiled in stynethsis to make it viwable ..

>
> Is ammonia's energy content high enough to run an airliner? Gaseous
> hydrogen's isn't and nor is liquid hydrogen's once to take density/volume/
> container mass into account.
>

The volume gets you with hydrogen, plus safety and containment.

"The energy density of ammonia is 22.5 MJ/kg at HHV, which is about half
of that for typical hydrocarbon fuels but higher than metal hydrides
(Zamfirescu and Dincer, 2008; Züttel et al., 2010). The raw energy
density of liquid ammonia is 11.5 MJ/L, which is higher than the 8.491
MJ/L for liquid hydrogen and the 4.5 MJ/L for compressed H2 at 690 bar
and 15°C1 . Ammonia is a good energy vector for on-board hydrogen
storage (Green, 1982; Klerke et al., 2008; Lan et al., 2012). However,
safety is regarded as the major drawback of using ammonia as the fuel.
Ammonia is toxic but it is detectable by humans in concentrations of
just 1 ppm (Reich et al., 2001). Anhydrous ammonia is lighter than air
then tends to disperse in the atmosphere. NH3 would be as safe as the
use of gasoline as a transportation fuel (Olson and Holbrook, 2007). The
ammonia released from an ammonia tank during a car accident may cause
potential safety problem but this can be solved through the application
of metal amines with low ammonia partial pressure (Klerke et al., 2008).

Compared to hydrogen, ammonia is easier to be transported. It is much
more energy efficient and much lower cost to produce, store, and deliver
hydrogen as NH3 than as compressed and/or cryogenic hydrogen (Figure 1)
(Olson and Holbrook, 2007). The infrastructure for ammonia already
exists while for hydrogen, new fueling stations have to be built, which
is a big investment (Lan et al., 2012)."

https://www.frontiersin.org/articles/10.3389/fenrg.2014.00035/full

The problem is that nitrogen don't oxidise and release energy, it
oxidises and releases nastyShit.™

You really want to reduce CO2 and water and make synthetic hydrocarbon.
Trouble is there is bugger all CO2 in the atmosphere.

So whereas it might be OK in a fuel cell in a car to go with ammonia, it
isnt really going to make a nice pollution free plane ride.

But what may in fact be the answer is bloody damned fast nuclear ships

The larger the ship the less drag at a given speed per unit weight, and
ships - containers and tankers - are conventionally operated to find a
local minimum to the cost equation. Too fast and you burn more fuel per
tonne mile, too slow and your cost of capital per tonne mile increases.

If your fuel cost is peanuts but your cost of capital is high, and
that's the case for a nukey ship, your accountant will tell you to push
it to the maximum speed it can handle, and your engineers will start
talking about lifting it out of the water on hydrofoils, or building an
ekranoplane as well.

If one wants to say, do southampton to new york in a day, before taking
a high speed electric train, that's probably 120-150mph Well within
ekranoplane range, but not sure about hydrofoils. its above the world
record for a hovercraft, but not by much.

A displacement nuclear ship should do it in 2-3 days nuclear subs have
broken 50mph as well

Anyway,. my point is that climate change and renewable energy are all
red herrings, there is no shortage of nuclear fuel and all the problems
in using it for land and sea based installations are eminently soluble.
What there is is a r9ising shortage of affordable fossil fuels.

All the problems with nuclear arise with portable power off grid. And
basic chemical feedstocks. I think the feedstock chemistry is soluble,
but portable off grid power is a real bugger.

You really want carbon based fuels, but without fossil sources, carbon
is in very short supply, and the only means of getting it out of the air
is really using biofuel of some sort - algae and the like. Possible
nuclear illuminated photosynthesis tanks to remove it from the air
might work, but we are already right on the lower limits of CO2 in the
air which allows plants to work efficiently - we really would need CO2
up around 800ppm to get that to be efficient.








--
The theory of Communism may be summed up in one sentence: Abolish all
private property.

Karl Marx

[Axel Berger]

Dennis Lee Bieber wrote:
> And that phone app typically duplicates appointments in Google's
> "cloud" to make them available to other devices

Typically yes. As with everything it's up to you to set it up right.

[David Higton]

In message <spt9f2$9na$1...@dont-email.me>

Martin Gregorie <mar...@mydomain.invalid> wrote:

> using the Rugbk (UK) time signal
>

>I've used Jonathon Buzzard's receiver and Rugby in the past

Just a reminder to everyone that the transmitter in question left
Rugby many years ago and is nowadays in Anthorn, Cumbria. Rather
more central to the British Isles.

Still the same time signal.

David

[Ahem A Rivet's Shot]

On Wed, 22 Dec 2021 13:42:46 +0100
Axel Berger <Sp...@Berger-Odenthal.De> wrote:

> Theo wrote:
> > Like I said, the OP's calendar lives on Google Calendar.
>
> Perhaps I misunderstood, or at least understood differently. For you a
> Google calendar is a calendar maintained for you by Google in their
> cloud. For me it's the calendar app in my phone and provided by Google.

It is of course both of these - in the context of displaying it on
a wall mounted device the cloud service seems to be the appropriate
interpretation - running the app may however be part of an implementation.

[Martin Gregorie]

Thanks for the preceeding analysis of ammonia as a general purpose
transport fuel.

I'd not previously heard anything about it, pro or con, until a piece on
Radio 4 mentioned it as a likely minimally polluting fuel for container ships,
so it seemed logical to consider if ot would also work for aircraft: the
answer seems to be 'yes'.

[The Natural Philosopher]

Well, i did consider that, but then looked again

Energy density is OK, but what sort of engine will run it?

Now they were talking about fuel cells, and fuel
cells=>electricity=>electric motors=> ducted fans or propellors is not
massively efficient at higher powers and its all a bit heavy.

And that s as I see it the issue with ammonia. Its got the energy
density and its easy to synthesise and it doesn't need scarce carbon
dioxide to make it and it can use atmospheric oxygen to create power so
that doesn't need to be carried by the plane.

BUT if you run it in a fuel cell its gonna be heavy and inefficient, and
if you throw it in to burn in a gas turbine, well the chances of NOx
production are fairly high.

Its already a concern in jet engines, and with nitrogen in the fuel
itself you are going to have to smack a lot more nitrogen through the
engine for a given power level

There are no easy answers or we would already be using them.

As I said, for ships - large ones - the obvious answer is small lifetime
fuelled sealed nuclear reactors.

For aircraft? Ive not been able to construct a realistic scenario for >
1 hour duration other than hydrocarbon fuel. Just possibly lithium air
batteries.

As far as Radio 4 goes, well I grew up with the home service but I don't
listen any more.

The world they seem to live in is like a distant roseate childhood
dream. It has no relation to the reality of my life in the real world of
technology.


>
>


--
“I know that most men, including those at ease with problems of the
greatest complexity, can seldom accept even the simplest and most
obvious truth if it be such as would oblige them to admit the falsity of
conclusions which they have delighted in explaining to colleagues, which
they have proudly taught to others, and which they have woven, thread by
thread, into the fabric of their lives.”

― Leo Tolstoy

[Giangi72]

Il 22/12/2021 13:42, Axel Berger ha scritto:
> Theo wrote:
>> Like I said, the OP's calendar lives on Google Calendar.
>
> Perhaps I misunderstood, or at least understood differently. For you a
> Google calendar is a calendar maintained for you by Google in their
> cloud. For me it's the calendar app in my phone and provided by Google.

Sorry, probably my fault, I mean the google calendar on the cloud.
In fact while waiting for the ordered ESP32 to arrive I already created
a web service reading from my google calendar (+ public holiday
calendar) and returning a json with some informations for each event:
summary, starting and ending date.
Thanks everyone for the help and suggestions
Giangi