• Re: More on wifi range - Pi PICO W Oil level sensor

    From c186282@c186282@nnada.net to comp.os.linux.misc on Wed Dec 17 00:20:43 2025
    From Newsgroup: comp.os.linux.misc

    On 12/16/25 05:43, The Natural Philosopher wrote:
    On 16/12/2025 06:50, c186282 wrote:
    On 12/15/25 12:49, The Natural Philosopher wrote:
    On 15/12/2025 06:55, c186282 wrote:
    In his app, does he even NEED "real time" ?
       "Close Enough" may be Good Enough perhaps.

    No,m I get a report every two hours

       So seven milliseconds is CRITICAL obviously  :-)

    LOL


    I did design my own PCBs

       I've done a number from scratch ... photo-masks,
       evil chemicals, the whole bit.

    It was cheaper to send the files to china than buy the kit..

    https://dirtypcbs.com

    Back in about 10 days at silly prices.

    Must have been a bit later ... no China-Fab
    when I was doing it - or, at least, you had
    to order 100 or 1000 units.

    Besides, Do It Yourself meant you could do
    prototypes, find the issues, tweak - THEN
    make the final boards. Anything using high
    frequencies can't be prototyped using mere
    experimenters plug boards, you HAVE to see
    how the positions of the traces, shields,
    affect thing on a real board. Parasitic
    capacitance can be a serious issue, unexpected
    induction/signal-jumping also.

       SOME use XYZ mini-routers to literally machine-out
       copper from boards. Ugly, but DOES work for many
       applications.

    Ah. Mr Usagi.

    Hey, worked ...

    Some use 3D printers to deposit low melting point solder as well

    That's new ... SHOULD be good for a number of
    things however.

    IF 3-D printing can produce <= millimeter traces
    on a fiberglass board these days in one pass then
    that's the proverbial cat's meow for 'home'
    board design.

    We use what we have to hand. I use cCorel draw. There is an online
    facility that turns that into the right format for PCB manufaturers.

    That'll work. There are also some more dedicated
    PCB design apps - some free, some very NOT free.
    Never had great luck/love for the free ones you
    can get for Linux alas ...

    I have a nano timer that reboots the board every 2 hours and then the
    pi shuts itself off again.

       I have a few PIs that just use crontab to
       reboot once/twice/thrice a day. Working on
       cleaning up the app that clutters/slows
       them down. Found some of the problems.

    Yes. That reminds me, I need to put the watchdog in some of my daemons too But in the case of the oil sensor clienbt this is physical 'shut off all power and stay down for two hours' Draws about 50nA in that period


       Now STARTING a Pi from scratch a few times
       a day, THEN you need an external timer device.

       Microcontrollers, most have low-power libraries
       that 99.99% shut 'em down until there's an interrupt,
       perhaps from a time chip. PIs however - they're
       natural power hogs ... NOT good for battery/solar
       operation.

    Looked at all that. Not good enough. I was going to build a CMOS timer
    then I stumbled across the TP1550 chip.

    liitle 'breakout' boards were dirt cheap


       I used a lot of Ard 2560s for solar field devices.
       They have low-power libs and reduce to seriously
       low current draw. DID slice the traces out of
       the cheapo time chip on the board and used
       "ChronoDots" instead. Interrupt, and it all
       comes back to life.

       Oh, where an oSillyScope is a good thing ... had
       an odd issue where the board would reset for no
       obvious reason. Turned out I'd attached the time
       chip to a current-limited line ALSO shared by an
       external device. The device didn't use much power
       EXCEPT for maybe a sub-millisecond when it powered
       up. Probably caps charging. FOUND that deadly
       voltage dip with a scope, little else would have.
       Had kept looking for a software problem, but ...

    I used my sillyscope to see why the *!%$£ things was oscillating under power.

    turned out to need serious capacitance on the output to stop the wrong
    sort of feedback triggering it



    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From c186282@c186282@nnada.net to comp.os.linux.misc on Wed Dec 17 01:01:36 2025
    From Newsgroup: comp.os.linux.misc

    On 12/16/25 05:43, The Natural Philosopher wrote:
    On 16/12/2025 06:50, c186282 wrote:
    On 12/15/25 12:49, The Natural Philosopher wrote:
    On 15/12/2025 06:55, c186282 wrote:
    In his app, does he even NEED "real time" ?
       "Close Enough" may be Good Enough perhaps.

    No,m I get a report every two hours

       So seven milliseconds is CRITICAL obviously  :-)

    LOL


    I did design my own PCBs

       I've done a number from scratch ... photo-masks,
       evil chemicals, the whole bit.

    It was cheaper to send the files to china than buy the kit..

    https://dirtypcbs.com

    Back in about 10 days at silly prices.

       SOME use XYZ mini-routers to literally machine-out
       copper from boards. Ugly, but DOES work for many
       applications.

    Ah. Mr Usagi.

    Some use 3D printers to deposit low melting point solder as well

    We use what we have to hand. I use cCorel draw. There is an online
    facility that turns that into the right format for PCB manufaturers.

    I have a nano timer that reboots the board every 2 hours and then the
    pi shuts itself off again.

       I have a few PIs that just use crontab to
       reboot once/twice/thrice a day. Working on
       cleaning up the app that clutters/slows
       them down. Found some of the problems.

    Yes. That reminds me, I need to put the watchdog in some of my daemons too But in the case of the oil sensor clienbt this is physical 'shut off all power and stay down for two hours' Draws about 50nA in that period


    PIs *are* a problem. As said, power hogs.

    Many microcontroller-based units let you shut
    down most every bit of hardware on the board
    and put the processor into a super-efficient
    idle mode awaiting an interrupt. Not so with
    the PIs, they're closer to "traditional
    computers".

    DO have some watchdog routines for a few of my PI
    programs, running as little daemons. The easiest
    approach has been to have the pgm touch a file
    every x-seconds/minutes. If the watchdog finds
    a too-old date then do something drastic. These
    watchdogs can literally be half a dozen lines
    of Python, depending on the IQ needed. Stupid,
    simple, EFFECTIVE.



       Now STARTING a Pi from scratch a few times
       a day, THEN you need an external timer device.

       Microcontrollers, most have low-power libraries
       that 99.99% shut 'em down until there's an interrupt,
       perhaps from a time chip. PIs however - they're
       natural power hogs ... NOT good for battery/solar
       operation.

    Looked at all that. Not good enough. I was going to build a CMOS timer
    then I stumbled across the TP1550 chip.

    liitle 'breakout' boards were dirt cheap


    Lots of 'time'-type chips, and they'll be a LOT more
    accurate than any home-built circuit.

    The DS-3231 chips are just great, very well temperature
    compensated, accurate as hell long-term. Some say the
    newer MAX-31328 time chips are even a bit better.

    I did lots of 'field devices/dataloggers' for awhile
    and accurate date stamps were very important. Also
    the chips DO have 'alarm' functions you can set just
    before yer board shuts down ... 'wake me up in 30
    minutes' or similar. A naked wake-me-up chip with
    just a few extra transistors could indeed be used
    to wake up a PI or even something bigger. Look
    into 'solid-state relays', hell, you could turn
    on a whole house/factory :-)


       I used a lot of Ard 2560s for solar field devices.
       They have low-power libs and reduce to seriously
       low current draw. DID slice the traces out of
       the cheapo time chip on the board and used
       "ChronoDots" instead. Interrupt, and it all
       comes back to life.

       Oh, where an oSillyScope is a good thing ... had
       an odd issue where the board would reset for no
       obvious reason. Turned out I'd attached the time
       chip to a current-limited line ALSO shared by an
       external device. The device didn't use much power
       EXCEPT for maybe a sub-millisecond when it powered
       up. Probably caps charging. FOUND that deadly
       voltage dip with a scope, little else would have.
       Had kept looking for a software problem, but ...

    I used my sillyscope to see why the *!%$£ things was oscillating under power.

    turned out to need serious capacitance on the output to stop the wrong
    sort of feedback triggering it

    Capacitance will getcha - or SAVE ya !!!

    Oh, dealing with sensor inputs in electrically
    noisy environs ... if possible try optoisolators.
    The LEDs require CURRENT, not just voltage spikes,
    to light up. Made some automotive/ag devices,
    some had 'pony' motors that gave off noise like
    mad. Optos were the salvation. They don't work
    with high frequencies, but for devices that mostly
    send 'pulses', low-baud serial and such ...

    Higher freqs ... unintended HF/RF coupling between
    components/stages creeps in.

    Nothing is as easy as it first seems :-)

    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From c186282@c186282@nnada.net to comp.os.linux.misc on Wed Dec 17 01:20:14 2025
    From Newsgroup: comp.os.linux.misc

    On 12/16/25 17:31, Carlos E.R. wrote:
    On 2025-12-15 07:55, c186282 wrote:
    On 12/14/25 13:23, Fritz Wuehler wrote:
    The Natural Philosopher <...@invalid.invalid> [TNP]:
    Depending on how long the batteries last I may increase the
    frequency of operation.  But it can never be 'real time'

    Have you considered using a couple of old 12V car/motorcycle batteries
    (one at a time) instead?

       Those are BIG (and heavy and need special recharge).
       At least they don't detonate like Li-Ion batts.

       Had a weather station that used up batteries too fast.
       Wanted to convert it to solar charge with a small lead
       battery - 6v. Well, first you needed a 9-12v solar
       panel, easy enough, but then a converter to get that
       into the range a dedicated lead/acid charger could
       cope with AND the charger could NOT cause the circuit
       to see any more than 6v regardless of sun levels.
       Turned into a MESS.

       Solution ... 9v little solar, a resistor bridge to
       drop to about 6.5v and a 10w zener to suck up any
       over-voltage across the device power leads. The batt
       just got a small harmless 'trickle charge' which was
       enough to deal. Less tech = MORE in this case.

    Doesn't anybody sell, ready to use, solar panel plus battery and
    controller, ready to use?

    Umm ... haven't SEEN any - but that doesn't mean they
    don't exist.

    The prob is 'application' - HOW would you use the thing ?
    This could complicate any commercial offerings.

    There ARE some odd things like 'emergency radios'
    which combine a solar panel and battery charger
    and voltage regulator. They're mostly not very
    expensive. Crack 'em open .....

    AND you can play tunes :-)

    Oh, 'solar-rechargers/power-banks' for phones and
    such, many are reported to be very UNSAFE, prone
    to fires. Little/no regulation or safety features.
    Sorry, $19.95 won't buy you everything ....

    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From c186282@c186282@nnada.net to comp.os.linux.misc on Wed Dec 17 01:38:07 2025
    From Newsgroup: comp.os.linux.misc

    On 12/16/25 19:35, Bobbie Sellers wrote:


    On 12/16/25 14:31, Carlos E.R. wrote:
    On 2025-12-15 07:55, c186282 wrote:
    On 12/14/25 13:23, Fritz Wuehler wrote:
    The Natural Philosopher <...@invalid.invalid> [TNP]:
    Depending on how long the batteries last I may increase the
    frequency of operation.  But it can never be 'real time'

    Have you considered using a couple of old 12V car/motorcycle batteries >>>> (one at a time) instead?

       Those are BIG (and heavy and need special recharge).
       At least they don't detonate like Li-Ion batts.

       Had a weather station that used up batteries too fast.
       Wanted to convert it to solar charge with a small lead
       battery - 6v. Well, first you needed a 9-12v solar
       panel, easy enough, but then a converter to get that
       into the range a dedicated lead/acid charger could
       cope with AND the charger could NOT cause the circuit
       to see any more than 6v regardless of sun levels.
       Turned into a MESS.

       Solution ... 9v little solar, a resistor bridge to
       drop to about 6.5v and a 10w zener to suck up any
       over-voltage across the device power leads. The batt
       just got a small harmless 'trickle charge' which was
       enough to deal. Less tech = MORE in this case.

    Doesn't anybody sell, ready to use, solar panel plus battery and
    controller, ready to use?

        Via DuckDuckGo using your terms above.

    <https://www.amazon.com/Renogy-200-Watt-Monocrystalline-Controller/dp/ B07RFQVB9M>

    That's a panel with integrated 12v charge regulator.
    You still have to buy the batteries. Likely some
    restrictions on those.

    Not a bad thing however - if you want to pay $227 USD.

    Now *I* didn't need 200 watts, more like TWO watts.
    That's still mostly "do it yourself" territory.

    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From c186282@c186282@nnada.net to comp.os.linux.misc on Wed Dec 17 01:55:25 2025
    From Newsgroup: comp.os.linux.misc

    On 12/16/25 20:05, The Natural Philosopher wrote:
    On 16/12/2025 20:40, rbowman wrote:
    On Tue, 16 Dec 2025 10:43:05 +0000, The Natural Philosopher wrote:

    On 16/12/2025 06:50, c186282 wrote:
    On 12/15/25 12:49, The Natural Philosopher wrote:
    On 15/12/2025 06:55, c186282 wrote:
    In his app, does he even NEED "real time" ?
        "Close Enough" may be Good Enough perhaps.

    No,m I get a report every two hours

        So seven milliseconds is CRITICAL obviously

    LOL


    I did design my own PCBs

        I've done a number from scratch ... photo-masks,
        evil chemicals, the whole bit.

    It was cheaper to send the files to china than buy the kit..

    I wish I had that option back in the '70s. Doing the art work, creating
    the silk screens, etching, and drilling was a pain in the ass.
    Fortunately
    I hired an older woman who knew more about the etching part than I did
    and
    a lesbian who was happy to stand in front of a Bridgeport punching holes.
    I got funny looks from the management since it was an all male shop
    except
    for the front office but they weren't DEI hires. They were the best
    candidates for the positions.

    Back in the 1970s I worked for companies that could pay the rates for commercial prototypes.

    But I built it all on pegboard first.

    Mostly good ... but if it's higher-frequency circuits
    you can't properly proto on such a board. You need a
    REAL board to see if there are HF/RF cross-over and
    capacitance problems and such. Sometimes just slight
    repositioning of the traces can have large effects.

    Oh, wifi, had a remote PI on my network. Performance
    had become horrible since I got a new router+wifi.
    Moved it exactly FOUR INCHES plus a slight rotation
    the other day and the data rate went up 5X :-)

    Black magic !

    Hmm ... may re-enable 2.4ghz on the router. The
    lower-freq signal seems to be kinder and gentler
    relative to device positioning. Yea, 2.4 is
    'slower' ... but if you can barely get a 5ghz
    signal then that's even MORE slow !

    PIs ... found an article, THINK it was about P4s,
    with instructions on how to put a REAL antenna
    on the things. Turns out they can have a VERY
    long range (maybe illegal in some places) with
    a spliced-in proper antenna.

    Now USA, with the recent demolition of over-fat
    federal agencies, NOBODY to look for your signal !
    MAYbe if you're 'commercial', but Joe Average will
    have to be ignored.

    DISadvantage of really long range - you become
    noticed/attractive to possibly malicious neighbors.
    They can let something like 'hydra' run for a
    week and find yer PW by brute force if they want.

    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From rbowman@bowman@montana.com to comp.os.linux.misc on Wed Dec 17 07:58:25 2025
    From Newsgroup: comp.os.linux.misc

    On Wed, 17 Dec 2025 01:20:14 -0500, c186282 wrote:

    Oh, 'solar-rechargers/power-banks' for phones and such, many are
    reported to be very UNSAFE, prone to fires. Little/no regulation or
    safety features.
    Sorry, $19.95 won't buy you everything ....

    They're probably improved but a friend had a solar device to recharge
    NiCad AAs. It worked, if you had 4 days to spare in southern Arizona.

    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From c186282@c186282@nnada.net to comp.os.linux.misc on Wed Dec 17 03:24:13 2025
    From Newsgroup: comp.os.linux.misc

    On 12/17/25 02:58, rbowman wrote:
    On Wed, 17 Dec 2025 01:20:14 -0500, c186282 wrote:

    Oh, 'solar-rechargers/power-banks' for phones and such, many are
    reported to be very UNSAFE, prone to fires. Little/no regulation or
    safety features.
    Sorry, $19.95 won't buy you everything ....

    They're probably improved but a friend had a solar device to recharge
    NiCad AAs. It worked, if you had 4 days to spare in southern Arizona.

    The slow charging was probably a GOOD THING, relatively
    speaking ... would not cook/explode the batteries.

    NiMH batteries are even worse than NiCad ... you literally
    have to monitor how HOT they get. When the temperature
    suddenly goes up, they're fully charged.

    LiPo ... slower IS better, 95% IS better than trying
    for 100%. The damned things LOVE to explode. I've
    seen tech stories that like HALF the LiPo batts in
    the USA are factory REJECTS.

    HAD one explode in my office once ... it hadn't been
    recharged for months. Barely BRUSHED it with my arm
    and it went off. Big crimson flames, massive noxious
    smoke, small burn on my arm. I just *really* don't
    trust the things. Have some big LiPo batts for large
    drills and such ... I charge 'em and then KEEP THEM
    IN THE OVEN (which I never use for cooking). If they
    explode, perfect heat-proof place.

    Hmmm ... when I retired, a fellow made a 'character'
    reference - upon seeing me just calmly watching the
    battery spurt huge flames. Hey, nothing to DO about
    it after all ... ! :-)

    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From The Natural Philosopher@tnp@invalid.invalid to comp.os.linux.misc on Wed Dec 17 10:31:24 2025
    From Newsgroup: comp.os.linux.misc

    On 17/12/2025 06:55, c186282 wrote:
    On 12/16/25 20:05, The Natural Philosopher wrote:
    On 16/12/2025 20:40, rbowman wrote:
    On Tue, 16 Dec 2025 10:43:05 +0000, The Natural Philosopher wrote:

    On 16/12/2025 06:50, c186282 wrote:
    On 12/15/25 12:49, The Natural Philosopher wrote:
    On 15/12/2025 06:55, c186282 wrote:
    In his app, does he even NEED "real time" ?
        "Close Enough" may be Good Enough perhaps.

    No,m I get a report every two hours

        So seven milliseconds is CRITICAL obviously

    LOL


    I did design my own PCBs

        I've done a number from scratch ... photo-masks,
        evil chemicals, the whole bit.

    It was cheaper to send the files to china than buy the kit..

    I wish I had that option back in the '70s. Doing the art work, creating
    the silk screens, etching, and drilling was a pain in the ass.
    Fortunately
    I hired an older woman who knew more about the etching part than I
    did and
    a lesbian who was happy to stand in front of a Bridgeport punching
    holes.
    I got funny looks from the management since it was an all male shop
    except
    for the front office but they weren't DEI hires. They were the best
    candidates for the positions.

    Back in the 1970s I worked for companies that could pay the rates for
    commercial prototypes.

    But I built it all on pegboard first.

      Mostly good ... but if it's higher-frequency circuits
      you can't properly proto on such a board. You need a
      REAL board to see if there are HF/RF cross-over and
      capacitance problems and such. Sometimes just slight
      repositioning of the traces can have large effects.

      Oh, wifi, had a remote PI on my network. Performance
      had become horrible since I got a new router+wifi.
      Moved it exactly FOUR INCHES plus a slight rotation
      the other day and the data rate went up 5X  :-)

      Black magic !

    Yes. The RF board got to revision 13...
    ...

      DISadvantage of really long range - you become
      noticed/attractive to possibly malicious neighbors.
      They can let something like 'hydra' run for a
      week and find yer PW by brute force if they want.


    Nearest neighbour is outside wifi range. 250 metres or more.
    --
    For in reason, all government without the consent of the governed is the
    very definition of slavery.

    Jonathan Swift


    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From c186282@c186282@nnada.net to comp.os.linux.misc on Wed Dec 17 06:25:15 2025
    From Newsgroup: comp.os.linux.misc

    On 12/17/25 05:31, The Natural Philosopher wrote:
    On 17/12/2025 06:55, c186282 wrote:
    On 12/16/25 20:05, The Natural Philosopher wrote:
    On 16/12/2025 20:40, rbowman wrote:
    On Tue, 16 Dec 2025 10:43:05 +0000, The Natural Philosopher wrote:

    On 16/12/2025 06:50, c186282 wrote:
    On 12/15/25 12:49, The Natural Philosopher wrote:
    On 15/12/2025 06:55, c186282 wrote:
    In his app, does he even NEED "real time" ?
        "Close Enough" may be Good Enough perhaps.

    No,m I get a report every two hours

        So seven milliseconds is CRITICAL obviously

    LOL


    I did design my own PCBs

        I've done a number from scratch ... photo-masks,
        evil chemicals, the whole bit.

    It was cheaper to send the files to china than buy the kit..

    I wish I had that option back in the '70s. Doing the art work, creating >>>> the silk screens, etching, and drilling was a pain in the ass.
    Fortunately
    I hired an older woman who knew more about the etching part than I
    did and
    a lesbian who was happy to stand in front of a Bridgeport punching
    holes.
    I got funny looks from the management since it was an all male shop
    except
    for the front office but they weren't DEI hires. They were the best
    candidates for the positions.

    Back in the 1970s I worked for companies that could pay the rates for
    commercial prototypes.

    But I built it all on pegboard first.

       Mostly good ... but if it's higher-frequency circuits
       you can't properly proto on such a board. You need a
       REAL board to see if there are HF/RF cross-over and
       capacitance problems and such. Sometimes just slight
       repositioning of the traces can have large effects.

       Oh, wifi, had a remote PI on my network. Performance
       had become horrible since I got a new router+wifi.
       Moved it exactly FOUR INCHES plus a slight rotation
       the other day and the data rate went up 5X  :-)

       Black magic !

    Yes. The RF board got to revision 13...
    ...


    Well, this was a Pi4 ....

    Anyway, NO other revisions ... just MOVED the thing
    a TINY distance and got VASTLY better performance.
    Just shows how iffy 5ghz wifi can be.


       DISadvantage of really long range - you become
       noticed/attractive to possibly malicious neighbors.
       They can let something like 'hydra' run for a
       week and find yer PW by brute force if they want.


    Nearest neighbour is outside wifi range. 250 metres or more.

    Lucky !

    I have more nearby.

    At least ONE might be an asshole. I think they
    injected a load of NAZI bullshit into my old
    printer - only stopped because all the paper
    was used up. Getting into a printer directly
    like that is not dead newbie skills.

    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From Carlos E.R.@robin_listas@es.invalid to comp.os.linux.misc on Wed Dec 17 20:39:58 2025
    From Newsgroup: comp.os.linux.misc

    On 2025-12-16 23:31, Carlos E.R. wrote:
    On 2025-12-15 07:55, c186282 wrote:
    On 12/14/25 13:23, Fritz Wuehler wrote:
    The Natural Philosopher <...@invalid.invalid> [TNP]:
    Depending on how long the batteries last I may increase the
    frequency of operation.  But it can never be 'real time'

    Have you considered using a couple of old 12V car/motorcycle batteries
    (one at a time) instead?

       Those are BIG (and heavy and need special recharge).
       At least they don't detonate like Li-Ion batts.

       Had a weather station that used up batteries too fast.
       Wanted to convert it to solar charge with a small lead
       battery - 6v. Well, first you needed a 9-12v solar
       panel, easy enough, but then a converter to get that
       into the range a dedicated lead/acid charger could
       cope with AND the charger could NOT cause the circuit
       to see any more than 6v regardless of sun levels.
       Turned into a MESS.

       Solution ... 9v little solar, a resistor bridge to
       drop to about 6.5v and a 10w zener to suck up any
       over-voltage across the device power leads. The batt
       just got a small harmless 'trickle charge' which was
       enough to deal. Less tech = MORE in this case.

    Doesn't anybody sell, ready to use, solar panel plus battery and
    controller, ready to use?

    I was thinking of something like in this advert:

    <https://photos.app.goo.gl/HmmTvmem7vVhDJzX7>
    --
    Cheers, Carlos.
    ES🇪🇸, EU🇪🇺;

    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From Carlos E.R.@robin_listas@es.invalid to comp.os.linux.misc on Thu Dec 18 19:00:50 2025
    From Newsgroup: comp.os.linux.misc

    On 2025-12-17 12:25, c186282 wrote:
    On 12/17/25 05:31, The Natural Philosopher wrote:
    On 17/12/2025 06:55, c186282 wrote:


    Nearest neighbour is outside wifi range. 250 metres or more.

      Lucky !

      I have more nearby.

      At least ONE might be an asshole. I think they
      injected a load of NAZI bullshit into my old
      printer - only stopped because all the paper
      was used up. Getting into a printer directly
      like that is not dead newbie skills.


    Maybe there is a virus that does that.
    --
    Cheers, Carlos.
    ES🇪🇸, EU🇪🇺;
    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From c186282@c186282@nnada.net to comp.os.linux.misc on Fri Dec 19 00:55:10 2025
    From Newsgroup: comp.os.linux.misc

    On 12/18/25 13:00, Carlos E.R. wrote:
    On 2025-12-17 12:25, c186282 wrote:
    On 12/17/25 05:31, The Natural Philosopher wrote:
    On 17/12/2025 06:55, c186282 wrote:


    Nearest neighbour is outside wifi range. 250 metres or more.

       Lucky !

       I have more nearby.

       At least ONE might be an asshole. I think they
       injected a load of NAZI bullshit into my old
       printer - only stopped because all the paper
       was used up. Getting into a printer directly
       like that is not dead newbie skills.


    Maybe there is a virus that does that.

    Nah ... it was some "helpful" connectivity protocol
    HOPEFULLY from yesteryear. The printer would ID and
    let anybody anywhere dump stuff into it directly
    via wifi.

    Was never sure I managed to turn that off entirely,
    so I had to keep the printer turned off instead.

    I do SUSPECT who the perp was ... fortunately he
    grew up and switched over to loud hot-rods instead :-)

    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From Carlos E.R.@robin_listas@es.invalid to comp.os.linux.misc on Fri Dec 19 15:16:57 2025
    From Newsgroup: comp.os.linux.misc

    On 2025-12-19 06:55, c186282 wrote:
    On 12/18/25 13:00, Carlos E.R. wrote:
    On 2025-12-17 12:25, c186282 wrote:
    On 12/17/25 05:31, The Natural Philosopher wrote:
    On 17/12/2025 06:55, c186282 wrote:


    Nearest neighbour is outside wifi range. 250 metres or more.

       Lucky !

       I have more nearby.

       At least ONE might be an asshole. I think they
       injected a load of NAZI bullshit into my old
       printer - only stopped because all the paper
       was used up. Getting into a printer directly
       like that is not dead newbie skills.


    Maybe there is a virus that does that.

      Nah ... it was some "helpful" connectivity protocol
      HOPEFULLY from yesteryear. The printer would ID and
      let anybody anywhere dump stuff into it directly
      via wifi.

      Was never sure I managed to turn that off entirely,
      so I had to keep the printer turned off instead.

      I do SUSPECT who the perp was ... fortunately he
      grew up and switched over to loud hot-rods instead :-)

    The router should block attempts to connect to internal printers from
    outside.
    --
    Cheers, Carlos.
    ES🇪🇸, EU🇪🇺;
    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From c186282@c186282@nnada.net to comp.os.linux.misc on Sat Dec 20 03:26:46 2025
    From Newsgroup: comp.os.linux.misc

    On 12/19/25 09:16, Carlos E.R. wrote:
    On 2025-12-19 06:55, c186282 wrote:
    On 12/18/25 13:00, Carlos E.R. wrote:
    On 2025-12-17 12:25, c186282 wrote:
    On 12/17/25 05:31, The Natural Philosopher wrote:
    On 17/12/2025 06:55, c186282 wrote:


    Nearest neighbour is outside wifi range. 250 metres or more.

       Lucky !

       I have more nearby.

       At least ONE might be an asshole. I think they
       injected a load of NAZI bullshit into my old
       printer - only stopped because all the paper
       was used up. Getting into a printer directly
       like that is not dead newbie skills.


    Maybe there is a virus that does that.

       Nah ... it was some "helpful" connectivity protocol
       HOPEFULLY from yesteryear. The printer would ID and
       let anybody anywhere dump stuff into it directly
       via wifi.

       Was never sure I managed to turn that off entirely,
       so I had to keep the printer turned off instead.

       I do SUSPECT who the perp was ... fortunately he
       grew up and switched over to loud hot-rods instead :-)

    The router should block attempts to connect to internal printers from outside.


    Oh NO NO NO ! THIS protocol was made for "convenience"
    and works (hopefully WORKED) entirely within the printer.
    The IDEA was to make it hyper-simple for people in an
    office to connect direct to the printer, even by wifi.
    Didn't matter if the router/main-PC was on or not.

    If I can find the manual I'll report the exact protocol.

    The printer is now on the JUNK pile ... leaks blue
    toner to the max. NOT worth trying to fix. Got
    a new color-laser all-purpose unit - do NOT
    enable anything 'net' however. Use thumb-drives
    to xfer stuff.

    Hmmm ... gotta look up local JUNK services to take
    it, and a bunch of other old stuff, AWAY. My
    damned house has 3+ generations of STUFF inherited
    inside it, the junk piles are getting large ....
    have a VIC-20 buried under there somewhere, Sinclair
    ZX-81, Apple II, bits of a DEC-based H-11, even
    one of those Radio-Shack "original laptops" meant
    for news reporters (with actual Bill Gates code
    inside it). Never mind all the crap from Weird Aunts.
    Wow ........

    No, not gonna bother with e-Bay - don't like it.
    Money paradigm is a bit iffy for my tastes plus
    I'd actually have to pack/box stuff and take it
    to UPS or FedEx or whatever .........

    Can't even LIFT the H-11 dual 8" disk unit
    anymore ......

    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From Carlos E.R.@robin_listas@es.invalid to comp.os.linux.misc on Sat Dec 20 13:00:29 2025
    From Newsgroup: comp.os.linux.misc

    On 2025-12-20 09:26, c186282 wrote:
    On 12/19/25 09:16, Carlos E.R. wrote:
    On 2025-12-19 06:55, c186282 wrote:
    On 12/18/25 13:00, Carlos E.R. wrote:
    On 2025-12-17 12:25, c186282 wrote:
    On 12/17/25 05:31, The Natural Philosopher wrote:
    On 17/12/2025 06:55, c186282 wrote:


    Nearest neighbour is outside wifi range. 250 metres or more.

       Lucky !

       I have more nearby.

       At least ONE might be an asshole. I think they
       injected a load of NAZI bullshit into my old
       printer - only stopped because all the paper
       was used up. Getting into a printer directly
       like that is not dead newbie skills.


    Maybe there is a virus that does that.

       Nah ... it was some "helpful" connectivity protocol
       HOPEFULLY from yesteryear. The printer would ID and
       let anybody anywhere dump stuff into it directly
       via wifi.

       Was never sure I managed to turn that off entirely,
       so I had to keep the printer turned off instead.

       I do SUSPECT who the perp was ... fortunately he
       grew up and switched over to loud hot-rods instead :-)

    The router should block attempts to connect to internal printers from
    outside.


      Oh NO NO NO ! THIS protocol was made for "convenience"
      and works (hopefully WORKED) entirely within the printer.
      The IDEA was to make it hyper-simple for people in an
      office to connect direct to the printer, even by wifi.
      Didn't matter if the router/main-PC was on or not.

    Ah, I think I recall reading about this. Something "direct"?



      If I can find the manual I'll report the exact protocol.

      The printer is now on the JUNK pile ... leaks blue
      toner to the max. NOT worth trying to fix. Got
      a new color-laser all-purpose unit - do NOT
      enable anything 'net' however. Use thumb-drives
      to xfer stuff.

      Hmmm ... gotta look up local JUNK services to take
      it, and a bunch of other old stuff, AWAY. My
      damned house has 3+ generations of STUFF inherited
      inside it, the junk piles are getting large ....
      have a VIC-20 buried under there somewhere, Sinclair
      ZX-81, Apple II, bits of a DEC-based H-11, even
      one of those Radio-Shack "original laptops" meant
      for news reporters (with actual Bill Gates code
      inside it). Never mind all the crap from Weird Aunts.
      Wow ........

    Sounds familiar.


      No, not gonna bother with e-Bay - don't like it.
      Money paradigm is a bit iffy for my tastes plus
      I'd actually have to pack/box stuff and take it
      to UPS or FedEx or whatever .........

      Can't even LIFT the H-11 dual 8" disk unit
      anymore ......

    --
    Cheers, Carlos.
    ES🇪🇸, EU🇪🇺;
    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From mm0fmf@none@invalid.com to comp.sys.raspberry-pi,comp.os.linux.misc on Wed Dec 24 07:58:11 2025
    From Newsgroup: comp.os.linux.misc

    On 11/12/2025 21:18, Carlos E.R. wrote:
    Home made with a box of Pringles. just google for "pringles wifi antenna".

    Also Google cutoff frequency and see that the Pringle tube is too small
    in diameter to be effective at 2.4GHz.

    Of course, designs on the internet do not have to follow the laws of
    physics! :-)

    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From The Natural Philosopher@tnp@invalid.invalid to comp.sys.raspberry-pi,comp.os.linux.misc on Wed Dec 24 12:16:27 2025
    From Newsgroup: comp.os.linux.misc

    On 24/12/2025 07:58, mm0fmf wrote:
    On 11/12/2025 21:18, Carlos E.R. wrote:
    Home made with a box of Pringles. just google for "pringles wifi
    antenna".

    Also Google cutoff frequency and see that the Pringle tube is too small
    in diameter to be effective at 2.4GHz.

    Assuming that is a relevant issue.

    Shouting down a pipe whose diameter is way less than the wavlength of
    voice frequencies, still works....

    Of course, designs on the internet do not have to follow the laws of physics! :-)


    ..especially for people who don't fully understand them...
    --
    "It was a lot more fun being 20 in the 70's that it is being 70 in the 20's" Joew Walsh

    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From John R Walliker@jrwalliker@gmail.com to comp.sys.raspberry-pi,comp.os.linux.misc on Wed Dec 24 14:04:03 2025
    From Newsgroup: comp.os.linux.misc

    On 24/12/2025 12:16, The Natural Philosopher wrote:
    On 24/12/2025 07:58, mm0fmf wrote:
    On 11/12/2025 21:18, Carlos E.R. wrote:
    Home made with a box of Pringles. just google for "pringles wifi
    antenna".

    Also Google cutoff frequency and see that the Pringle tube is too
    small in diameter to be effective at 2.4GHz.

    Assuming that is a relevant issue.

    Shouting down a pipe whose diameter is way less than the wavlength of
    voice frequencies, still works....

    Of course, designs on the internet do not have to follow the laws of
    physics! :-)


    ..especially for people who don't fully understand them...

    Indeed. And I'm sure you are perfectly well aware of the difference
    between longitudinal sound waves propagating down a narrow pipe and
    transverse electromagnetic waves in a waveguide.
    If a Pringles can were highly conductive it would have a cutoff
    frequency of close to 2.4GHz so the attenuation would be very high.
    However, a very thin layer of aluminium on the inside of a cardboard
    tube will be so resistive that it will not make a lot of difference.
    For many purposes a well made half-wave dipole or quarter-wave
    monopole gives excellent results which are far better than anything
    that can be achieved with small pcb antennas.

    A quarter wave monopole made from relatively thick wire or rod can
    be an excellent match to 50 ohm coax so long as the ground plane
    is at least a few wavelengths across.

    A half-wave dipole combined with a coaxial balun can also be a very
    good match but has a slightly narrower bandwidth due to the
    frequency dependency of the coax balun. The choice of which one to
    use depends mostly on how the antenna is to be mounted.

    An almost omnidirectional antenna with very low losses can be
    more effective than a lossy directional one.

    John

    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From The Natural Philosopher@tnp@invalid.invalid to comp.sys.raspberry-pi,comp.os.linux.misc on Wed Dec 24 14:23:45 2025
    From Newsgroup: comp.os.linux.misc

    On 24/12/2025 14:04, John R Walliker wrote:
    On 24/12/2025 12:16, The Natural Philosopher wrote:
    On 24/12/2025 07:58, mm0fmf wrote:
    On 11/12/2025 21:18, Carlos E.R. wrote:
    Home made with a box of Pringles. just google for "pringles wifi
    antenna".

    Also Google cutoff frequency and see that the Pringle tube is too
    small in diameter to be effective at 2.4GHz.

    Assuming that is a relevant issue.

    Shouting down a pipe whose diameter is way less than the wavlength of
    voice frequencies, still works....

    Of course, designs on the internet do not have to follow the laws of
    physics! :-)


    ..especially for people who don't fully understand them...

    Indeed.  And I'm sure you are perfectly well aware of the difference
    between longitudinal sound waves propagating down a narrow pipe and transverse electromagnetic waves in a waveguide.

    An antenna is not a waveguide.



    If a Pringles can were highly conductive it would have a cutoff
    frequency of close to 2.4GHz so the attenuation would be very high.
    However, a very thin layer of aluminium on the inside of a cardboard
    tube will be so resistive that it will not make a lot of difference.
    A statement which clearly contradicts the well known skin effect of
    conductirs at high frequencies.

    For many purposes a well made half-wave dipole or quarter-wave
    monopole gives excellent results which are far better than anything
    that can be achieved with small pcb antennas.

    Sure. Most routers come with wavelength sized wobbly penises that give
    you a few dB.

    A quarter wave monopole made from relatively thick wire or rod can
    be an excellent match to 50 ohm coax so long as the ground plane
    is at least a few wavelengths across.

    A half-wave dipole combined with a coaxial balun can also be a very
    good match but has a slightly narrower bandwidth due to the
    frequency dependency of the coax balun.  The choice of which one to
    use depends mostly on how the antenna is to be mounted.

    An almost omnidirectional antenna with very low losses can be
    more effective than a lossy directional one.


    John


    Nevertheless I have seem that sort of design work.
    I worked around radar antennae briefly in the 1960s.

    What I learned was that theory is too simplified to actually be able to
    design a real antenna: All our designs were field tested and adjusted.

    I am not advocating Pringle cans. I wouldnt use one myself. But I am
    not so quick to rubbish them as you are.

    RF propagation is tricky, and real world objects of no apparent value
    often have enormous effects.
    --
    The lifetime of any political organisation is about three years before
    its been subverted by the people it tried to warn you about.

    Anon.

    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From Kerr-Mudd, John@admin@127.0.0.1 to comp.sys.raspberry-pi,comp.os.linux.misc on Wed Dec 24 17:00:18 2025
    From Newsgroup: comp.os.linux.misc

    On Wed, 24 Dec 2025 14:23:45 +0000
    The Natural Philosopher <tnp@invalid.invalid> wrote:

    []
    What I learned was that theory is too simplified to actually be able to design a real antenna: All our designs were field tested and adjusted.

    I am not advocating Pringle cans. I wouldnt use one myself. But I am
    not so quick to rubbish them as you are.

    RF propagation is tricky, and real world objects of no apparent value
    often have enormous effects.


    Prsumably you're saying Mythbusters-style "not proven"?
    --
    Bah, and indeed Humbug.
    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From The Natural Philosopher@tnp@invalid.invalid to comp.sys.raspberry-pi,comp.os.linux.misc on Wed Dec 24 20:07:30 2025
    From Newsgroup: comp.os.linux.misc

    On 24/12/2025 17:00, Kerr-Mudd, John wrote:
    On Wed, 24 Dec 2025 14:23:45 +0000
    The Natural Philosopher <tnp@invalid.invalid> wrote:

    []
    What I learned was that theory is too simplified to actually be able to
    design a real antenna: All our designs were field tested and adjusted.

    I am not advocating Pringle cans. I wouldnt use one myself. But I am
    not so quick to rubbish them as you are.

    RF propagation is tricky, and real world objects of no apparent value
    often have enormous effects.


    Prsumably you're saying Mythbusters-style "not proven"?


    I am saying that a blanket denial 'because the theory says no' is not
    good enough for me, personally.

    To make a waveguide, which is analysable, is quite tricky. To throw
    something in place that 'does something' and clearly is *not* a
    waveguide, and is essentially unanalysable, is another matter.

    With Gigahertz, as with Heffalumps, you never know...

    --
    If I had all the money I've spent on drink...
    ..I'd spend it on drink.

    Sir Henry (at Rawlinson's End)

    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From Lars Poulsen@lars@beagle-ears.com to comp.sys.raspberry-pi,comp.os.linux.misc,alt.unix.geeks on Wed Dec 24 22:53:21 2025
    From Newsgroup: comp.os.linux.misc

    [Folloup-to alt.unix.geeks]

    On 11/12/2025 21:18, Carlos E.R. wrote:
    Home made with a box of Pringles. just google for "pringles wifi antenna".

    On 2025-12-24, mm0fmf <none@invalid.com> wrote:
    Also Google cutoff frequency and see that the Pringle tube is too small
    in diameter to be effective at 2.4GHz.

    Of course, designs on the internet do not have to follow the laws of physics! :-)

    As a wireless applications engineer, I have always shaken my head
    at the pringles can antenna, which if well made can have a gain up
    to at most 8 dBi or so. For less than $50, you can buy a professinally
    made flat patch antenna with 8 dBi gain at L-com.com. Apparently, you
    can find used ones on eBay for under $30, but why take the risk?
    If you need circular polarization, it will be closer to $75.
    --
    Lars Poulsen - an old geek in Santa Barbara, California
    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From Lars Poulsen@lars@beagle-ears.com to comp.sys.raspberry-pi,comp.os.linux.misc,alt.unix.geeks on Wed Dec 24 23:01:30 2025
    From Newsgroup: comp.os.linux.misc

    [See also my previous followup - also shifting to alt.

    On 2025-12-24, John R Walliker <jrwalliker@gmail.com> wrote:
    On 24/12/2025 12:16, The Natural Philosopher wrote:
    On 24/12/2025 07:58, mm0fmf wrote:
    On 11/12/2025 21:18, Carlos E.R. wrote:
    Home made with a box of Pringles. just google for "pringles wifi
    antenna".

    Also Google cutoff frequency and see that the Pringle tube is too
    small in diameter to be effective at 2.4GHz.

    Assuming that is a relevant issue.

    Shouting down a pipe whose diameter is way less than the wavlength of
    voice frequencies, still works....

    Of course, designs on the internet do not have to follow the laws of
    physics! :-)


    ..especially for people who don't fully understand them...

    Indeed. And I'm sure you are perfectly well aware of the difference
    between longitudinal sound waves propagating down a narrow pipe and transverse electromagnetic waves in a waveguide.
    If a Pringles can were highly conductive it would have a cutoff
    frequency of close to 2.4GHz so the attenuation would be very high.
    However, a very thin layer of aluminium on the inside of a cardboard
    tube will be so resistive that it will not make a lot of difference.
    For many purposes a well made half-wave dipole or quarter-wave
    monopole gives excellent results which are far better than anything
    that can be achieved with small pcb antennas.

    A quarter wave monopole made from relatively thick wire or rod can
    be an excellent match to 50 ohm coax so long as the ground plane
    is at least a few wavelengths across.

    A half-wave dipole combined with a coaxial balun can also be a very
    good match but has a slightly narrower bandwidth due to the
    frequency dependency of the coax balun. The choice of which one to
    use depends mostly on how the antenna is to be mounted.

    An almost omnidirectional antenna with very low losses can be
    more effective than a lossy directional one.

    John

    --
    Lars Poulsen - an old geek in Santa Barbara, California
    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From mm0fmf@none@invalid.com to comp.sys.raspberry-pi,comp.os.linux.misc on Wed Dec 24 23:17:41 2025
    From Newsgroup: comp.os.linux.misc

    On 24/12/2025 20:07, The Natural Philosopher wrote:

    With Gigahertz, as with Heffalumps, you never know...

    Some of us do know.
    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From c186282@c186282@nnada.net to comp.sys.raspberry-pi,comp.os.linux.misc on Wed Dec 24 21:16:01 2025
    From Newsgroup: comp.os.linux.misc

    On 12/24/25 15:07, The Natural Philosopher wrote:
    On 24/12/2025 17:00, Kerr-Mudd, John wrote:
    On Wed, 24 Dec 2025 14:23:45 +0000
    The Natural Philosopher <tnp@invalid.invalid> wrote:

    []
    What I learned was that theory is too simplified to actually be able to
    design a real antenna: All our designs were field tested and adjusted.

    I am not advocating Pringle cans.  I wouldnt use one myself. But I am
    not so quick to rubbish them as you are.

    RF propagation is tricky, and real world objects of no apparent value
    often have enormous effects.


    Prsumably you're saying Mythbusters-style "not proven"?


    I am saying that a blanket denial 'because the theory says no' is not
    good enough for me, personally.

    To make a waveguide, which is analysable, is quite tricky. To throw something in place that 'does something' and clearly is *not* a
    waveguide, and is essentially unanalysable, is another matter.

    With Gigahertz, as with Heffalumps, you never know...

    Yep, once you get into gHz things get really weird.

    "Solutions" here can work for NO GOOD REASON, pure
    chance, funky reflections.

    Moved a unit literally four inches the other day and
    the data rate went up nearly 5X.

    Yes, a cheapo foil-lined tube CAN be a a waveguide,
    but the math has to work out.

    Me, I can't use waveguides because I need an OMNI
    signal the way things are spaced out in my home.
    I use one, sometimes two, wifi repeaters instead.
    Fairly cheap, work well.

    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From rbowman@bowman@montana.com to comp.sys.raspberry-pi,comp.os.linux.misc on Thu Dec 25 02:29:33 2025
    From Newsgroup: comp.os.linux.misc

    On Wed, 24 Dec 2025 12:16:27 +0000, The Natural Philosopher wrote:

    On 24/12/2025 07:58, mm0fmf wrote:
    On 11/12/2025 21:18, Carlos E.R. wrote:
    Home made with a box of Pringles. just google for "pringles wifi
    antenna".

    Also Google cutoff frequency and see that the Pringle tube is too small
    in diameter to be effective at 2.4GHz.

    Assuming that is a relevant issue.

    Shouting down a pipe whose diameter is way less than the wavlength of
    voice frequencies, still works....

    True, but the benefits of impedance matching with a megaphone were
    discovered long ago.
    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From The Natural Philosopher@tnp@invalid.invalid to comp.sys.raspberry-pi,comp.os.linux.misc on Thu Dec 25 03:23:15 2025
    From Newsgroup: comp.os.linux.misc

    On 24/12/2025 23:17, mm0fmf wrote:
    On 24/12/2025 20:07, The Natural Philosopher wrote:

    With Gigahertz, as with Heffalumps, you never know...

    Some of us do know.

    Some of us have worked with RF.
    --
    “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

    --- Synchronet 3.21a-Linux NewsLink 1.2
  • From rbowman@bowman@montana.com to comp.sys.raspberry-pi,comp.os.linux.misc on Thu Dec 25 07:32:08 2025
    From Newsgroup: comp.os.linux.misc

    On Thu, 25 Dec 2025 03:23:15 +0000, The Natural Philosopher wrote:

    On 24/12/2025 23:17, mm0fmf wrote:
    On 24/12/2025 20:07, The Natural Philosopher wrote:

    With Gigahertz, as with Heffalumps, you never know...

    Some of us do know.

    Some of us have worked with RF.

    That was my first real job -- 15 kw of it at around 100 MHz. Why is that fluorescent tube lighting with nothing attached to it?
    --- Synchronet 3.21a-Linux NewsLink 1.2