4·
2 years agoPeople around here usually have trailers behind their normal cars for that. Works fine.
- 1 Post
- 68 Comments
Joined 3 years ago
Cake day: April 26th, 2023
You are not logged in. If you use a Fediverse account that is able to follow users, you can follow this user.
The severity of the problem depends on external factors:
- do you live in a dry and salt-free environment? No problem.
- did the maker of your car implement disc wiping as a precaution? No problem.
- did the maker of your car implement intermittent brake blending during recuperation? No problem.
- is your car less driven in wet and salty conditions and then always store in a dry, warm garage? No problem.
- does your car have drumsin the rear and near-undersized discs in the front? Less of a problem.
I drive an electric vehicle and the unfortunate reality is, that brakes that are large enough for an emergency stop get underutilized during everyday, regenerative braking. So they rust.
My solution is: Every time I get back to our village in wet weather, I accelerate hard on the last long straight and brake hard to almost a standstill once. No more brake trouble, all four rotors squeaky clean, but any measuring device would write me up as the biggest idiot on the roads out there.
That is too general of a statement. I have three EVs in my family, none of them do any temp condition of the battery just by being plugged in. However, EVCC turns off the wallbox when they reach 75% SoC and there is no appointment that day in our shared calendar. Sitting at high SoCs kills batteries, especially in warm climates.
12·2 years agoEh - YT music (premium) offers 256kB-ish Opus now. That’s on par with 320kBps MP3.
YT Music Premium delivers 256kBit/s Opus now, Deezer delivers FLAC.
Yep. And in my case, the backup battery is connected to another DC input on the inverter and the inverter pretty much manages everything. As I understand the documentation, there is no other way to use solar AND a battery at the same time as a power source for islanding. Switching over manually with a short disruption in-between is always possible of course, as is charging an AC coupled battery from an islanding solar inverter.
An engineer dabbling in such things explained to me, that it is hard enough to regulate a small island network frequency and voltage-wise from a single point. Reacting to whatever another source (something like another solar inverter out in the garden with a few panels of its own, e.g.) in the same island grid does could easily lead to potentially destructive oscillations in the regulation circuit. Large grids have “mass” - literally, because large generators and electric motors are spinning at whatever speed they are spinning in whatever phase they are in. So small disturbances from regulating too quickly or a little wrong just disappear into that. The same doesn’t go for a small island grid, so at Fronius they have decided to put 52Hz on the grid which by standard prevents other sources from syncing. Electric utilities do the same when they have to power small villages from diesel generators temporarily - 52Hz and the house mounted solar generators don’t sync.
That sentence does not make any sense whatsoever. SATA and SAS are hot pluggable as well.
Not necessarily. While running parallel to the grid or needs to sync to it of course, but when running in island mode it can do whatever it feels like - if it supports that. My Fronius runs at 52Hz e.g. to keep other generators in the island from starting up.
During the gold rush, shovel manufacturers had a steady income.
51·2 years agoYeah, yeah, but, numbers please, not the PR talk.
What’s the efficiency of the process then? Say, I generate 1kWh of electricity somewhere in the middle of the day on a sunny field, how much of that kWh arrives at the wheels of a car at the end of this described H2 workflow?
Dunno. I’ve got 3 phase, 400V, 100A service which results in 68kW useable. However, because one of sub junction boxes which, unfortunately the wall boxes are connected to is wired internally with 10mm², I’ve enabled peer to peer load management across my wallboxes for now so they never pull more than 28A per phase. Most cars here support three phase charging so that’s still fine even for two cars. I’ll get to rewiring that, for now it works.
The go-e wallboxes I have support a central controller which in turn can measure current on all three phases into the home, e.g. to use a solar system to its maximum, but also to limit absolute load on the house connection. They just use three hall sensors for power measurement as far as I know, so installation is relatively unintrusive.
I went without that and solved the whole solar optimization using EVCC and regarding absolute load I’m just yoloing it, but then again, I do have a neat safety margin.
100A*240V is 24kW. I don’t know what else you’re doing, but a 7.2kW car on board charger is well within that.
I’ve got some vservers and some block storage at hosters in a different country and I terminate a VPN tunnel there from yet another country. Storage is LUKS encrypted and VMs are configured to halt hard automatically in case of certain triggers that hint towards hypervisor side fuckery.
It’s not bullet proof, but it’s fire and forget, basically seed boxes where I’ve split knowledge further.
I used this as an exercise in provider agnostic deployment and I’m mainly hosting archive.org torrents that can use some seeding there.
I haven’t found the time to research an answer for you, sorry. The way I’d go is: create a veth of your physical uplink and stuff it into its own namespace with dhcp client and wg userspace tools. Do not configure the original interface in your initial namespace. Use the approach wg-netns uses to spawn the tunnel interface in initial network ns. Done.
There’s readily available docker containers for it but I wanted to build it by hand. Well, more or less, Extremely hacky but it works, so fine for me.
I started out with cheating and used this wrapper around wg-quick that gives us a persistent network namespace with the tunnel interface in it:
https://github.com/dadevel/wg-netns
cat /etc/systemd/system/wg-qbittorrent.service [Unit] Description=WireGuard Network Namespace for qBittorrent Wants=network-online.target nss-lookup.target After=network-online.target nss-lookup.target [Service] Type=oneshot Environment=WG_ENDPOINT_RESOLUTION_RETRIES=infinity Environment=WG_VERBOSE=1 ExecStart=/opt/wg-netns/bin/wg-netns up /etc/wireguard/wgconfig.yaml ExecStop=/opt/wg-netns/bin/wg-netns down /etc/wireguard/wgconfig.yaml RemainAfterExit=yes WorkingDirectory=%E/wireguard ConfigurationDirectory=wireguard ConfigurationDirectoryMode=0700 CapabilityBoundingSet=CAP_NET_ADMIN CAP_SYS_ADMIN LimitNOFILE=4096 LimitNPROC=512 LockPersonality=true MemoryDenyWriteExecute=true NoNewPrivileges=true ProtectClock=true ProtectHostname=true RemoveIPC=true RestrictAddressFamilies=AF_INET AF_INET6 AF_NETLINK RestrictNamespaces=mnt net RestrictRealtime=true RestrictSUIDSGID=true SystemCallArchitectures=native [Install] WantedBy=multi-user.targetThen I built a static binary of qbittorrent using this really neat docker image: https://github.com/userdocs/qbittorrent-nox-static
…and stuffed the result into a systemd service that runs it in the namespace wg-netns provides:
cat /etc/systemd/system/qbittorrent-nox.service [Unit] Description=qBittorrent-nox service Wants=network-online.target wg-qbittorrent.service After=local-fs.target network-online.target nss-lookup.target wg-qbittorrent.service [Service] Type=simple PrivateTmp=false #User=qbittorrent ExecStart=/usr/sbin/ip netns exec ns-qbittorrent sudo -u qbittorrent /opt/qbittorrent/qbittorrent-nox TimeoutStopSec=1800 RestartSec=15 RestartMaxDelaySec=600 RestartSteps=10 Restart=always [Install] WantedBy=multi-user.targetTo get the webui out of that I stuck two instances of socat together at the stdout and from there it depends on whatever you want to use as a reverse proxy on the host - or you bind to a network interface if you trust the network:
cat /etc/systemd/system/qbittorrent-webui.service [Unit] Description=qBittorrent-nox webui forwarding into its namespace Wants=network-online.target wg-qbittorrent.service After=local-fs.target network-online.target nss-lookup.target wg-qbittorrent.service [Service] Type=simple PrivateTmp=false ExecStart=/opt/qbittorrent/forward-webinterface.sh TimeoutStopSec=1800 Restart=always RestartSec=10 [Install] WantedBy=multi-user.targetcat /opt/qbittorrent/forward-webinterface.sh #!/bin/sh set -eu exec socat tcp6-listen:"8080",reuseaddr,fork,range=[::1]/128 "exec:ip netns exec ns-qbittorrent socat stdio 'tcp-connect:127.0.0.1:8080',nofork"Works, is reboot safe, stopped caring about beauty at that point.
When I design something, critical applications get their own network namespace with only the VPN interface inside anyway. So, yeah.
Yep. I dabble in recruiting related stuff at conferences and expos for our company occasionally and I usually meet one or two young people that get the “get in contact!” remark on the protocol sheet. They’re out there, they’re just rare.