dgriffith

@dgriffith@aussie.zone

I’m a technical kinda guy, doing technical kinda stuff.

This profile is from a federated server and may be incomplete. Browse more on the original instance.

dgriffith,

For some they want that feeling of control.

Get posts while they’re new, push them into the negative so they’re less visible, look at me doing the Lord’s work protecting this sub!

dgriffith, (edited )

Except for using the pen, IR-cameras, booting from USB…

Reminds me of android ROMs about a decade ago.

“My new L33tM@st3r ROM has just been released! Now with kernel tweaks for buttery-smooth performance and major improvement to stock battery life! Comes with it’s own tuning app so you can adjust it the way YOU want!

(Not presently working: bluetooth/wifi/camera/NFC/dialler/headphones but everything else is awesome!!)”

dgriffith,

Cat 3 is a thing and is basically unshielded twisted pair. You can abuse it quite a bit from its voice grade days to cram a few hundred megabits of VDSL over it if it’s only from your house to the curb.

dgriffith,

Accessing that from my phone shows the actual site content for about 3 seconds and then an endless array of “popup and notification blocked messages” as well as the usual combination of “oh noes your PC has three dangerous viruses click here to resolve” type redirects.

Is this just a cunning way to weed out the normies or can I expect the same kind of thing from their allegedly-excellent app? Because they’re not leaving a good impression right now.

dgriffith,

Chrome mobile put up a pretty good fight. I have been meaning to put Firefox + ublock on my phone, maybe I’ll stop being slack and finally do it today.

dgriffith,

As the answer in the link explains, it’s adjustment of your scaling factor to the nearest whole pixel, plus a loss of precision rounding to/from single/double floating point values.

dgriffith, (edited )

A lot of the software components under the hood in Linux are replaceable.

So you have a bunch of different CPU and disk IO schedulers to suit different workloads, the networking stack and memory management can be tweaked to hell and back, etc etc.

Meanwhile Windows Server 2022 has… ?

dgriffith,

Battery chemistry produces fixed voltages depending on what you use. It depends on where the active components sit on the electronegativity table.

The typical ones are:

Zinc-carbon and alkaline - 1.5 volts per cell.

Lead acid - 2 volts

Nickel Cadmium - 1.2 volts

Nickel Metal Hydride - 1.4 ish.

All the Lithium ion combos - 3.4 to 3.7 volts.

dgriffith,

The voltage range depends a lot on cell construction, temperature, load or charge rate, and chemical mix.

For example “lead acid” batteries with lead and sulphuric acid have a cell chemistry voltage of 2.05 volts but their nominal range is 1.8 to 2.4 volts per cell. Translating that to a 6 cell “12 volt” car battery gives you a range of 10.8 to 14.8 volts.

dgriffith,

I remember helping a teacher at school who had installed a CopyIIPc card on one of our computers. They used it to make everyday copies of the master disks of the copy protected educational software we used in our room full of Sperry IBM compatible PCs.

The card went in between the floppy controller and the drive and could do a pretty good job at duplicating all the physical copy protection tricks of the time.

They copied a lot of stuff, not for pirating reasons but simply because they were literally 5 1/4" floppy disks back then and school kids were not kind to them. Either it was simply jamming them into the drives, or touching the exposed disc surface, or chucking them around the room, those disks didn’t last long.

dgriffith,

You’re asking for help in entirely the wrong way, here I’ll fix it for you:

“Linux is shit. I have an Asus Transformer T100TA and I thought I’d try MX Linux and fluxbox. I poked around for a few minutes and couldn’t get any sort of battery display to work on it, when it’s fully functional by default in windows. If this is the best that Linux can do in 2023, forget it, I’m sticking with windows.”

There. That oughta do it!

dgriffith,

But look how fast we can make those little fuckers go!

It’s just like slot car racing, round and round, but… you know… faster. And yeah, it’s more expensive than a regular slot car track, I guess. But still, those particles will beat any slot car you care to pick! So there’s that. Welllll not those fancy slot cars with them high performance motors, I mean, that’s a completely different ballgame there, we can’t compete with that.

But still, those particles whizzing around, it’s gonna be pretty cool. I reckon we should do it.

So anyway, thank you for reading my financial proposal for the SuperLHC.

dgriffith, (edited )

Mmm I have a general dislike of systemd because it doesn’t adhere to the “do one thing and do it well” approach of traditional Unix systems.

It’s a big old opaque blob of software components that work nicely together but don’t play well with others, basically.

Edit: but it solved a particular set of problems in serverspace and it’s bled over to the consumer Linux side of things and generally I’m ok with it if it simplifies things for people. I just don’t want a monoculture to spring up and take root across all of Linux as monocultures aren’t great for innovation or security.

dgriffith,

Eh? You’ve got your thoughts backwards on the pins.

The inside of a USB C socket is simply a central tang. The inside of a USB C plug has the flex pins that grip said tang and provide the electrical connection. The USB plug, on the cable, is designed to wear out before the socket and it’s easily replaced.

The issue however is that some manufacturers skimp on the mechanical mounting of the socket on their device and flexing of the connector + socket damages its connection to the board.

Funnily enough, lightning connectors are designed the other way around with the components that wear out inside the expensive device. How strange…But that’s coming from a company that doesn’t even want to make a robust covering for the wires in their cables.

dgriffith, (edited )

At my latitude, never below 8 or 9 degrees C in winter. Maximum temperature is about 35 or 36 degrees C in summer.

Older systems here were simple thermosyphon designs on the roof with the holding tank closely coupled directly above the collectors and they were quite effective.

My system has a 330 litre mains pressure hot water tank, with the usual cold and hot fittings bottom and top. There is a third inlet about 1/3rd of the way up the tank, and a little bit above that is a heater element and a controlling thermostat. The lower third of the tank is circulated out via the cold inlet and up through the collector by the pump, which can do about 300 litres an hour. Hot water returns from the collector at the 1/3rd location, and rises to the top of the tank via stratification. The copper pipes to and from the collectors are 1/2" and are insulated, but their surface area to volume ratio is quite large and heat is easily lost over the 15 or so metres of pipework.

On sunny days I can see the inlet temperature at the bottom of the tank slowly rise from approx 20-25 degrees C to 30-35 degrees C as the bottom third circulates, the hotter water rises to the top to give the small temperature rise I can see at the top of the tank and the warm water layer slowly lowers towards the bottom of the tank.

Essentially I want a higher temperature returned to the tank, which I suspect can rather paradoxically be done by increasing the circulation rate when running the pump, peaking the collector temperature higher with the pump off then dumping that back to the tank rapidly, rather than the current controllers method of pulsing the pump and slowly circulating it to maintain a moderate collector temperature and losing the heat in the lines.

I have the service manual for the controller and it mentions frost protection, where it will circulate warmer water back to the collectors in low temperatures, but this would be insufficient for your colder climate.

At some stage I would also like to control the heater element, it runs on an off peak circuit which is switched by the electricity company when it suits them, so it is cheaper to run. It also means that it regularly “tops up” the top 2/3rds of the tank temperature, which negates the solar contribution a fair bit. I would like to be able to disable that top up if the system can sense adequate heating from the collectors, but that requires switching control to a 3kW element, and something I will leave alone for now.

dgriffith,

Yes the traces are pretty light for those kinds of currents, and I did have some concerns on the track spacing around the SCR and the screw terminals. The fatter tracks use a spacing for 240v that I looked up online.

Humidity here isn’t “tropical” as such, the board will probably get a coat of lacquer anyway.

The circulating pump is rated at 240v/20 watts (so not much current when it’s running steady state), and the SCR I’m using is good for a few amps of continuous current draw with a fairly high surge current. The SCR has minimal heatsinking so it will go first in extended high current situations, there’s going to be a panel mount fuse before all of this that I’ll probably set to 2 amps or so.

The zero-crossing SCR controller I’m using “should” prevent switching on the pump at the peaks of the mains cycle so max current should just be the initial stall current from the motor before it gets up to speed.

The pump is an induction motor so the only concern I’ve got is false triggering of the SCR and being unable to turn the pump off once triggered. I used an example snubber circuit from the data sheet of the zero crossing controller and of course they said “your milage may vary”.

dgriffith,

It’s generic foam pipe insulation, approximately 3/4" thick.Originally there was also a 3-4 meter section of pipework from the collector output to inside the roof that was completely uninsulated, I’ve gotten that sorted.

I bought the place a couple of years ago and after replacing the seized circulating pump early last year I was pretty disappointed with the system’s performance.

At that point I put onewire sensors on the tank and lines on the ground level and measured it for a few months and didn’t see much heating at all from the collector.

I couldn’t get to the collectors for a long time, not having a three storey ladder, but I got suspicious after sunny + windy days resulted in very little heating. Finally got up there about three months ago to discover that bare section of pipe and insulated that.

Collector temperature I haven’t measured using the original analog sensor yet, I just let the original controller do its thing to get a baseline. But currently in a “warm” Australian spring the maximum outlet temperature at ground level is a brief peak of about 65 degrees C at around 1pm. Going off the circulation rate of the pump it’s probably 40 litres of water at that temp before it starts dropping back down in the afternoon.

Insulating that bare section of pipe improved performance but it’s still not fantastic, so now it’s time to drive the pump with my own algorithm and see if I can improve things.

dgriffith,

The human brain is hardwired for an “Us vs. Them” viewpoint. As Bush once said, “You’re either with us, or you’re against us.”

Advertisers, politicians, and those that seek to influence, they know this. They don’t project any shades of grey or balance into their arguments. They attempt to “match groups” with you, and then point out that other group, over there, aren’t they terrible, we should do something about that.

So you get videos going, “Look at those shoplifters, doing wrong things! We’re glad that they’re getting their just desserts, aren’t we?”

They don’t show anything that might evoke some empathy in their target audience. You don’t want people to identify with your enemy, you want them to identify with you.

dgriffith,

Soft-wired , perhaps.

The brain is a lazy thing. It falls into patterns easily, it takes mental shortcuts whereever it can. All that comes from our evolutionary history, where energy (literal energy for thinking) was limited.

Your subconscious mind handles all the everyday trivia to maintain your existence in a low energy way, and your mind is very quick to shunt things down that low energy path. That path is all our “instincts” (fight or flight, nurturing instincts, forming into groups etc etc ) and is based upon nomadic living in the savannah plains of Africa three million years ago.

If someone presents something in a way that can easily fall into one of our evolutionary shortcuts then your subconscious will run with it without you even realising.

dgriffith, (edited )

Hmm I’m not sure of the pin drive currents on the Pico, but can you power the sensors off a pin? At least then you can programmatically power cycle them if you need to.

The Pico also has a watchdog, you could set it up to give it a reboot if things don’t respond in time. It doesn’t solve the issues of course but at least it gets it back to a workable state. And if the watchdog fails, or it works but there’s still no USB serial, then that would point towards power instabilities or somesuch.

dgriffith, (edited )

Perhaps slightly adjust your logic a little and see what it does.

Read from the sensors first, then enable and connect to wifi, send the data, then disconnect. That would reduce the maximum power draw as only one function is active at once.

Small edit: I have a MagTag ESP32 board with circuitpython that can’t read onewire devices while the wifi is active. Whether that’s because of supply instabilities when wifi is transmitting, or interrupt conflicts, or just plain poor programming in the onewire drivers or the wifi drivers, I don’t know. But reading the devices first and then connecting to wifi and sending the data afterwards works.

  • All
  • Subscribed
  • Moderated
  • Favorites
  • •
  • localhost
  • All magazines
    •
    Loading…
    Loading the web debug toolbar…
    Attempt #