2023 Mustang Mach-E Select AWD (Standard Range, 226M rated)
The most convenient car I have owned.
–No gas station stops.
–No oil changes.
–Other then a tire rotation, no maintenance at all.
I charge at home and its ready to go every day.
Battery: 70 KWh
City: 237.7 Miles
Highway: 206.5 Miles
Combined Range: 224
Most trips are within the range of the car so charging on the road is not common. Most on the road charging has been for the experience and not need. When I actually have needed to, I did get ‘range anxiety’ which isn’t surprising for me.
When I charge on the road its usually pretty easy. Plug in, get coffee, unplug, go.
My most recent frustrating experience. I booked a hotel with a charger so I could just charge and come back home. They had two chargers, one worked, the other was in use. I fought with the broken one for a long time only to give up and go find a rapid charger. I spent about an hour total getting the car ready for the return trip.
The longer range NCM batteries have been meeting or exceeding expectations so I can’t say they are bad choices. For me, a worrier, I like the LFP battery and its longer expected life, more stable chemistry. I look forward to advances in battery tech and longer range in my next vehicle but this one has enough for me.
I commute to an office on average 3 times per week. A typical commute day is about 90 miles. Never have I owned a car that could do this without the need to visit a gas station regularly. After charging at home evenings I can do this every day, no gas station stops required.
Guess-O-Meter regularly showed ranges around 250 miles. Driving under most circumstances I could get this range. Hot days didn’t seem to effect range much, if at all.
Guess-O-Meter regularly showing ranges from 200 to 210. I suspect I would get this range if I were to drive the entire 200 miles at once but with warming up and shorter trips I suspect 200 would be pushing it.
Not much difference between the starting GOM reading between cool and cold. I have been setting to go times so the car does the battery and cabin warming thing. I depart home with about 190 to 210 on the meter. I suspect if I drive continuously I would get that. Driving short trips with the car cooling and re-heating would likely get closer to 160-180 based on the GOM through the day. Since I rarely put more then 90 miles a day on the car I’m not sure where I would have ended up exactly.
We just returned from a trip to Florida. The car was parked at the Portland Maine Jetport for several days.
This screen shot was taken the day we dropped it off, after it had set at the airport around 0C for 12 hours.
This was taken on pick up day after sitting for a week outside. The pick up day average temperature had been about -8C.
This shows the temperatures the car was exposed to for the week.
I used the Fordpass app to remote start the car and it was warm and ready by the time I got from the gate, to the car. No issues.
I really like my car. :)
]]>
I purchased my 2023 Ford Mustang Mach-E in early March 2024 as a leftover, standard range, AWD. It’s been an amazing adventure going all electric, one I am really enjoying.
EV Economics
Environmental Impact
Energy Usage
Range
Charging on the road
Battery longevity
Battery Fires
But my use case is X
The grid will collapse
I have really enjoyed owning and driving this car. I am not likely to own an internal combustion engine vehicle as my primary transportation any time soon, perhaps never again. It seems clear to me electric is the way to go unless you have a use case that it just doesn’t fit.
You should do your own research. My information is from the internet or my observations and can be wrong. If you find something that is just plain wrong I appreciate you letting me know. I will update this blog post.
Driving an EV can reduce your expenses on fuel and maintenance, leading to savings over time. However, purchasing decisions should not be based solely on financial savings. The cost of EVs in the US is now comparable to that of gas-powered cars. For example, my Mach-E Mustang had a sticker price of $42K, which is similar to an equivalent gas-powered car.
A gallon of gas in an efficient vehicle can get you 40 miles. The equivalent amount of energy in an EV can get you about 90 miles. I have owned all my previous vehicles beyond 100K miles. 100K would be about 2500 Gallons of gas or about 82,500 KWH of energy in a gas car. In my EV I will get about 3 miles per KWH so I will use 33,333 KWH of electricity. That is about 60% less energy used in my EV.
The mining of elements for making batteries for EV has an impact but considering an EV will use less than half the energy of a gas car through its lifetime I can’t believe its worse than a gas car. We can’t ignore that obtaining oil and burning it has its costs also.
My EV has a LFP battery. Lower energy density than other technologies but easier to get the elements to make it. Easier to recycle. Longer life span.
Then we could talk about where the energy comes from. There are many solar and wind farms out there now producing an increasing amount of our energy needs. I have some solar panels myself but only 3.2KW array. Its estimated to produce 4,250 KWH per year which if used just for driving would be greater than 12K miles.
If you park your EV in a small garage and turn it on what happens? Pretty much nothing at all. Lets imagine doing the same with a gas powered car. The garage will warm up from the waste heat. The garage will likely get smokey. You may also die.
Once you look into just how much energy your electric car doesn’t use its amazing. Let me give you some points I think really highlight this.
My commute is 44 miles, 88 round trip. A gallon of gas has 33KWH of energy. For the fun of it lets say my last car, a hybrid, got 44 miles per gallon. This means I can drive to the office using 1 gallon of gas. Now lets put the same energy in my electric car, 33KWH. My car is doing about 3.5 miles per KWH for summer driving. That means with the same energy it took to get my gas powered car to the office, 44 miles, I can drive my electric car 115 miles.
A few of my co-workers will often eat lunch in their cars. Its summer time and has been pretty hot so they sit with the AC running. The average 4 CYL car uses about 0.35 gallons per hour when idling, closer to 0.5 with the AC on. Gas has 33KWH per gallon so assuming they sit in the car 30 minutes they will use about 0.25 Gallons of gas. That works out to 8.25 KWH of energy used. So they sit in a hot car while the ~25% efficient engine creates mostly heat, trying to stay cool. The engine block, exhaust system, radiator, etc all radiating waste heat, so they can stay cool.
Now lets talk about one of the ‘bad’ things about an EV. The range drops a lot in the winter time. A large part of the reason for this drop is that there isn’t much waste heat. A gas powered car is about 70% waste heat so if you need some to keep the cabin warm, you’ve got it. Its difficult for me to see this as a negative. I have so much less wasted energy as heat, that if I want heat, I have to make it.
Not sure its easy to ignore just how much more energy efficient an EV is compared to even the best, hybrid, gas powered cars.
My standard range vehicle has an EPA rating of 226 Miles. This summer I’m finding I’m getting 250+. On the coldest days of the winter I suspect I’ll be getting somewhere between 150 and 180. I’ll report back more about that later after I have it through a winter. I’ll use the number 200 as an average moving forward.
How many days a year do you drive more than 200 miles? 200 miles is a long drive. For every day you don’t drive more than 200 miles you only need to charge at home. Every day you get in your vehicle it will be ready to travel another 200 miles. No trip to the gas station required and your energy cost will be significantly less. Since I purchased my car I have charged on the road because of need twice. I have had it nearly 4 months and drove over 7K miles.
When my wife fills her gas powered car her range is closer to 500 miles so this on the surface seems like a big negative for EVs. The funny thing is she will likely spend more time at gas stations than I will at charging stations because every morning I wake up I can drive up to another 200 miles without needing to stop for gas. Its only rare occasions I need to stop and charge on the road.
When I purchased my car I told myself and people around me that I just didn’t plan to use it for long trips. My wife drives a beautiful SUV and the plan was I would just swap vehicles with her for the day. One person I told, the owner of a Chevy Bolt EV told me that wasn’t going to happen. He said charging on the road wasn’t a big deal and I would just take my electric car. He was correct.
My Mach-E, and just about any EV are pretty fun to drive. They are quiet, smooth, and accelerate faster than most cars. The technology packages are impressive, and fun. So there is a strong desire to just drive my own car. Turns out that is pretty easy to do.
At first you’re nervous about charging but it goes away. When I first started driving my car I looked for any possible DC rapid chargers and gave them a try. Every trip I took, even close to home, I would pretend I needed to charge. I learned a lot. For example there is one charging location with first generation DC rapid chargers and they are not reliable, often not working. I have found others I categorize as awesome. I found DC rapid chargers in very convenient places such as shopping centers, super markets and highway rest stops.
Of the two times I actually needed to charge on the road, I chose to only charge for 10 minutes. This is because both trips were just to the high side of my range and a 10 minute charge gets me about 50 more miles. I just didn’t need to charge any more.
Don’t dismiss standard L2 chargers. They are everywhere. Getting 30 miles of range per hour of charging is pretty slow but lets say your visiting a small town to spend some time shopping and getting lunch. You will likely spend a couple of hours there. That’s 60 more miles you can drive.
One negative point of note about DC fast charging on the road. You will pay almost three times the price for electricity you will pay when charging at home. Home charging saves money when compared to having a gas powered car. Charging on the road will cost you more. There are many Level2 chargers that are completely free, as long as you have the time to wait.
Any building with electricity is a potential place you can charge. Most cars come with a charger that can work on 120V or 240V. 120V charging plugs are everywhere but you will only gain about 3 Miles for every hour your plugged in. 240V plugs will get you closer to 30 miles per hour of charging but are harder to find. You will find suitable plugs at campgrounds and commonly used for an electric cloths dryer.
Many think a degrading battery is a major concern for EV owners. It really isn’t. First the manufacturers take away this concern by offering very generous warranties on the EV battery. Not to mention an LFP (Lithium Iron Phosphate) battery is rated for anywhere from 3000 to 10,000 cycles. A cycle is a full charge and fully discharged. In my car that translates to somewhere between 600K and 2M miles before my battery hits a point where it only has 80% capacity.
Lets say you have the longer range battery, the NCM type. You should worry if you expect to put more than 250K miles on your car, very uncommon.
National Fire Protection Association (NFPA) and other studies indicate that gasoline vehicles experience significantly more fires compared to EVs. Think about it. They are frequently hot and carrying around a highly flammable liquid. It only makes sense. We are at a time when any EV catching fire is national news, not so much for a gasoline powered car. They have been catching on fire for years.
Many fire departments have expressed concerns about fighting EV fires, and rightfully so. They are trained how to fight gasoline fires. The training and experience with EVs is a work in progress.
Recently a Tesla crashed driving at a high rate of speed in Naples Maine, USA. It hit so hard it scattered individual battery cells over a wide area, the driver was killed instantly. I watched the fire chief being interviewed. There were small pieces of battery burning all over the place. My take away was that a big problem for them was debris collection. How do we handle the scattered pieces of this battery? Where do we put them to insure more fires don’t start? These are real problems but I don’t believe they are unsolvable.
Not all batteries are the same. Chemistries and manufacturer quality are important. There have been products with design flaws such as the Galaxy Note 7, that sparked a lot of fires.
Lithium batteries are now everywhere. I’m sitting in a room that has 4 I can think of quickly. There are 2 laptops, a cell phone, and a portable battery for cell phone charging. They have a great safety record.
For my EV and battery power tools I keep them in my garage. A garage that has thicker sheet rock that was mandated many years ago to protect me from my gas powered car, that is significantly more likely to catch fire than my battery operated things.
There will be use cases where a gas powered car is a better choice. From what I can tell it takes a lot of energy to plow roads, tow trailers, move cargo long distances, etc. Energy that just isn’t available in an EV with current technology. There are some remote locations, or remote locations with really cold climates. etc..
I think many things can be made to work, if you want them to. For example you own a remote camp you frequently drive to. Take a look, are there charging locations in route? Do you have electricity at your remote camp and can charge while there for your return trip?
Are you on on the road sales person and drive more than 200 miles in a day frequently? Take a look around. Do some of your customers have EV charging? Perhaps a lunch stop charging station. Most cars can charge in less than 30 minutes to 80% capacity, that is pretty quick.
Do you drive 1500 miles to Florida every fall to return in the spring? That can be done in an EV. In my car I would have to charge just under 6 hours total at DC fast chargers. This 6 hours is if I did NOT plan to charge at any hotel I may stay at. So every 2 hours of driving I need to sit at a charger for 30 minutes. Keep in mind this is in my ‘standard range’ 226 EPA rated vehicle. Many EVs can do better. Like I said, it can work if you want it to. If you don’t then you will see this as unacceptable. I hope to do this trip, this year, just to see how it goes. My figures were generated by a trip from Portland, Maine to Orlando, FL using https://abetterrouteplanner.com/ as my trip tool.
The usual narrative is something like that if everyone swapped over to EVs the grid couldn’t handle it. I guess the premise being everyone switched over at the same time. I agree, if the completely impossible were to happen and we all were driving EVs suddenly the grid would collapse. What is possible, and currently happening, is we are swapping over slowly and in a manner that enables our power supply and delivery companies to keep up.
There may eventually be policies put in place to encourage off peak charging. Newer technologies that allow for the fleet of EVs to give power back to the grid during peak demands, etc. Time will tell how this all works out but the collapsing grid is not a likely outcome.
]]>Docked Mode
Processing Power
Cell Coverage in the USA
Battery Life
State of the Software
Camera
Furi Labs FLX1
Furi Labs on Mastodon
Furi Labs on Telegram
The FLX1 has much better specifications then my Librem5 making it faster, longer battery life, better pictures, etc. The Librem 5 is still the winner for me because of the FLX1 having a modem that doesn’t work well in the USA. The software is still a work in progress.
If they ever offer a model with a modem that works well in the USA I’ll likely purchase one.
Here you can see the FLX1, on the left, is physically larger then the Librem5.
I didn’t do any testing but was told there is no support for an external display. This is OK for me as I would want it for a phone mostly anyhow but it should be noted if you are thinking of making a purchase. Docked mode is something the other Linux phones I have messed with can do. (pinephone, Librem5)
A fun test I always like to do on a processor is to calculate PI to 4000 places and see how long that takes. It doesn’t tell a full story but it is informative. A summary would be that the FLX1 is way faster.
Here on my Librem 5. I pick one of the four cores. I don’t think I need to as they are all the same.
purism@pureos:~$ taskset -c 3 bash -c "time echo 'scale=4000;a(1)*4' | bc -l > /dev/null"
real 0m36.805s
user 0m36.793s
sys 0m0.005s
#Now on the FLX1. First on a fast core, then a slow one.
root@FuriPhoneFLX1:~# taskset -c 6 bash -c "time echo 'scale=4000;a(1)*4' | bc -l > /dev/null"
real 0m16.428s
user 0m16.420s
sys 0m0.000s
root@FuriPhoneFLX1:~# taskset -c 0 bash -c "time echo 'scale=4000;a(1)*4' | bc -l > /dev/null"
real 0m41.158s
user 0m41.142s
sys 0m0.009s
Its crap. Let me explain. You can follow this link to see how the modem in this device fairs in the USA.
Compatibility of the FLX1 modem in the USA
I have T-Mobile and work in Portland, Maine, USA. I can get nearly zero coverage with the exception of the phone saying ‘2.75G’ which I’m told is 2G on steroids.
Even with this 2.75G coverage its spotty in Portland. Driving home and the situation was very bad. I would freqently have no coverage at all. At my house nothing. This is really bad considering both my wife with AT&T and myself with T-Moble have excellent signal strengths. You may notice the Librem5 in the picture above is 4 out of 5 bars.
A little digging shows that there is no 2G or 5G service in my area.
Cellmapper.net
Since there is only 4G and not on a band that works with the FLX1, I am out of luck. I have ordered a verizon SIM for testing but my expectations are pretty low. Hopefully I can get 5G in Portland but suspect I will still get nothing at home.
Furi Labs is looking into what it will take to get coverage in the USA. Likely a USA specific model, less likely a modem firmware change.
Nothing but good news to report here. It seems to last for a couple days without issues. Perhaps partly because I don’t keep a SIM card in the device, not sure. When the screen blanks it uses something called ??batman?? to turn off cpu cores. It keeps just core 7, one of the fast cores running.
It doesn’t go into sleep so you continue to get notifications from all applications. This is awesome. To get more then about 5 hours from the Librem5 I have to let it sleep. When the L5 is sleeping only a phone call or text will wake it up. I tend to not let it sleep and just keep in plugged in whenever I can. The FLX1 is a clear winner for battery life.
This phone is very new and its software is in a rapid state of development. Many issues mentioned here may already be fixed.
Firefox was crashing but I was given the fix for it. Seems to have to do with accelerated graphics. I believe the command I was given disabled OpenGL but this phone is powerful enough it still worked quite well.
When I had ‘2.75G’ service I was able to make calls and receive simple SMS messages. MMS did not work. I started looking into that but was told by helpful Furi Labs people on Telegram chat to just wait, I would have working MMS very soon.
Speaking of audio devices. Since the phone modem was such an issue I proceeded to try and setup a SIP account. With the applications I tried I could hear the remote end fine but could not be heard. When I asked Furi they said they are aware of audio issues that should be sorted out with the next release.
I was able to connect my Nextcloud account at murena.io but.. Everything synced in OK but the calendar changes made on the FLX1 didn’t reflect back. While trying to test output audio I was trying to set calendar reminders, which I could not.
I could not make gnome-clocks alarms or timers make any sound. I have since been told I needed to ‘sudo apt install libgtk-4-media-gstreamer’ and sound would have worked.
I transferred some Linux Magazines to the device. After installing evince, so I had an application to view them, I was very impressed by the graphical performance.
I didn’t do a lot of testing here but what I did do was all good. It works and seems to work quite well. Great pictures, fast, and even video. I didn’t get audio with the video but am told that is being taken care of. The one quickly noticeable bad thing is if you keep screen rotation on, putting the camera on its side is confusing as it also rotates the image. I just turned off screen rotation when using the camera.
]]>
Fun project here. Old hollowed out TRS80 case with new components sporting a RISC-V board.
Usability notes: It works! Pretty modern, easy to type on, current apps, low power consumption. What’s not to like? Biggest usability issue is the 1024x768 12” monitor I used. Pretty low resolution for some apps. Keyboard is a Master Keys Pro by Cooler Master.
The top shelf contains three TNCs and radios that are part of a ax.25 packet system. TNCs: MFJ TNC-X, Timewave PK-96, PacComm Tiny-2 MK-2. Radios: Alinco DR-135 MKII, DR-235 MKII, Tait TM8115.
Then we have some 440 duplexers, a Yeasu System Fusion DR-2X repeater, the WOPR (computer running all this stuff), another Yeasu DR-2X, switched receptacles, Pakedge P2 Ethernet controlled receptacles, shelf holding a Alinco DM-330MV power supply, more 440 duplexers, Icom ID-RP2C D-Star repeater Controller, Icom ID-RP4000V UHF digital voice repeater.
To the right of the rack sits two sets of VHF duplexers.
The WOPR computer, a 19” 2U computer in the radio rack that runs this stuff has an ASROCK ITX-3455 main board. This is a low power setup, using less then 25W. Moving around voice packets and being a local NAS just doesn’t need much power. The WOPR runs OpenSuse Leap and hosts 4 virtual machines. Lubuntu for the packet node, CentOS for the D-Star gateway, pi-hole for local add blocking, and a Windows 7 VM for the Yeasu fusion stuff. WTH Yeasu? No Linux support? Really? (Not a Yeasu fan)
The heart of the solar electric system is the Outback Power Systems inverter on the right under the wall mount 19” equipment rack. Its a 3700W continuous, 7200W for 15s utility interactive inverter. To its left is a DC breaker box and one of the two Mate MX charge controllers. The other charge controller is just to the right of the window. Hard to miss are 8 blue Fullriver DC335-6 batteries in the battery rack. Not in the picture is the 2.2KW solar array.
The system chooses solar power first, utility power second, and batteries as a last resort. This is because the grid is a very efficient system for storing energy. A lot more efficient then batteries. On top of that if the batteries were cycled regularly they would need to be replaced more often.
This system was installed in 2005. I have had no issues of mention with the exception of having to replace the batteries twice so far.
Powered directly off the inverter are fridge, furnace, radio gear, garage doors, computer room receptacles, and general lighting circuits. When the power fails these circuits are automatically powered by the batteries with a transition time so fast that you won’t notice anything. Other circuits loose power but can be manually switched over. Originally all circuits transitioned automatically but this is problematic because if the power fails when you are not home you may kill your batteries running things like electric heaters, window AC units, dehumidifiers, etc.
We can run, if conservative, for ? 24 hours off batteries. If we get a lot of sun we can run necessary equipment for long periods of time. Recharging from a generator can be done if needed.
In the 19” equipment rack on the wall has a POE switch, cable modem, pfsense router with 2X LANs. One for the radio network, one for local stuff. The WOPR (mentioned above) has an add on PCIE 4 port LAN card. This allows it to participate in the radio LAN, local LAN, dedicated port for the D-Star controller, and one port bridged into the radio LAN for the Ethernet switched receptacles.
]]>Permissions for the usbtmc device
Hello world
5 Second Delay Reading usbtmc Device
Get a Screenshot
Getting and Plotting Data with gnuplot
Misc Commands I Sent to my Tektronix Scope
I have been experimenting with this scope on and off and have lost track of many references but here are a few:
Wikipedia about SCPI
Awesome c program
The device when connected, in my case, created /dev/usbtmc1. Not sure why its 1 and not 0. One issue is a ‘normal user’ doesn’t have permissions to do anything with it. Two options here. One is to just ‘sudo chmod 666 /dev/usbtmc1’. The next is more permanent and recommended.
Create a file ‘/dev/udev/rules.d/86-tmc.rules’ with the following in it.
KERNEL=="usbtmc*", MODE="0666" Then run: ‘udevadm control –reload-rules && udevadm trigger’
First a simple test. If you ask the device to ID does it?
adamd@pangolin:~/dev/TekTronix$ echo "*IDN?" > /dev/usbtmc1
adamd@pangolin:~/dev/TekTronix$ cat /dev/usbtmc1
TEKTRONIX,TDS 2012C,C015224,CF:91.1CT FV:v24.26
cat: /dev/usbtmc1: Connection timed outOne point of note here is the ‘Connection timed out’ after about 5 seconds.
I spent more time on this then I will admit. From some reading of others posts the kernel driver usbtmc doesn’t set the EOF bit so it just times out. This sucks! It means all my simple techniques can still work but will have a 5 second delay on read.
I tried many things and the only one that worked for me was this. The code came from https://github.com/dpenkler/linux-usbtmc/blob/master/ttmc.c and I just reduced and slightly modified to work for what I wanted to do. It works without the 5 second hang time. It does NOT work for commands that may return more then 1024 bytes. For those commands I still use echo and cat at the command line.
//This was cut from here: https://github.com/dpenkler/linux-usbtmc/blob/master/ttmc.c
//Trying to make the min necessary to ask an instrument questions from the cli
//It will read up to 1024 bytes, no more. For pulling large data sets like when using
//CURVE? I continue to use echo and cat from the command line and deal with the delay.
#include <sys/ioctl.h>
#include <unistd.h>
#include <fcntl.h>
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include <linux/usb/tmc.h>
int fd;
/* Send string to scope */
void sscope(char *msg) {
write(fd,msg,strlen(msg));
}
/* Read string from scope */
int rscope(char *buf, int max_len) {
int len = read(fd,buf,max_len);
if (len < 0) {
perror("failed to read from scope");
ioctl(fd,USBTMC_IOCTL_CLEAR);
return 0;
}
buf[len] = 0; /* zero terminate */
return len;
}
int main (int argc, char *argv[]) {
#define MAX_BL 20480
unsigned char buf[MAX_BL];
// Check that we were provided with a single command line argument
if (argc != 2)
{
printf("Need a singlecommand to send, perhaps start with '*IDN?'\n");
return 0;
}
/* Open file */
if (0 > (fd = open("/dev/usbtmc1",O_RDWR))) {
perror("failed to open device");
exit(1);
}
/* Send device clear */
ioctl(fd,USBTMC_IOCTL_CLEAR);
/* set timeout lower */
unsigned int timeout = 1000;
ioctl(fd,USBTMC_IOCTL_SET_TIMEOUT,&timeout);
/* Send identity query */
sscope(argv[1]);
/* If the command ended with a =? we expect a response */
if (argv[1][strlen(argv[1]) - 1] == '?') {
rscope(buf,MAX_BL);
printf("%s",buf);
}
}I compile it with ‘gcc -o send_this thecode.c’
With it I can do things like this:
#!/bin/bash
./send_this 'MEASU:IMMED:TYPE CUR'
while true
do
./send_this 'MEASU:IMMED:VALUE?'
doneThis has been around a while so I don’t have my reference(s).
#!/bin/bash
dev=/dev/usbtmc1
#script gets TDS2012C screen capture-
if [ "$1" == "" ]
then
echo "Missing file name as command line parameter"
exit
fi
out_file="$1" #output file comes from firs arg of command line
size="0"
totaltries="0"
printf ":HARDCopy:PORT USB\n" > $dev
printf ":HARDCopy:FORMat JPEG\n" > $dev
printf ":HARDCopy STAR START\n" > $dev
printf ":HARDCOPY START\n" > $dev
#I keep trying until my picture is available
#giving up after a bit.
while (( size < 20 && totaltries < 6 ))
do
cat $dev > $out_file 2> /dev/null
size=`stat --printf="%s" $out_file`
totaltries=$((totaltries+1))
echo "Have tried $totaltries times, timeout at 6"
doneUse like this: ‘./thescript.sh bla.jpg’
I use this to get the raw data points. There is a lot of informatin missing such as formating the data, what data we’re taling about, etc. That information is in my TDS 2012C programmers manual. There are lots of options and you should be looking in your specific devices manuals.
# DATA INIT defaults the parameters for what CURV you are talking about.
echo "DATA INIT" > /dev/usbtmc1
# I want it ASCII encoded
echo "DAT:ENC ASCII" > /dev/usbtmc1
# Request the data points
echo "CURV?" > /dev/usbtmc1
# Read data points back dumping the timeout error mentioned above.
cat /dev/usbtmc1 2> /dev/null > thedataNow I have a file with 2500 values all in one big line seperated by commas. It looks like this: 32,32,32,-21,-30,-31,-31,… For plotting it I use gnuplot. I use the following configuration file curv.gnuplot
#curv.gnuplot
set datafile separator comma
row = 0
set yrange [-100:100]
set terminal x11 size 800,600
plot filename matrix using 1:0 every :::row::row with linesI get the plot by running this: gnuplot -p -e “filename=’result’” curv.gnuplot
It looks like this:
Just some random commands I experimented with. I didn’t notice any case sensitivity when sending them. These commands were found in the Tektronix Programmer manual (077-0444-03) for my TDS 2012C osilliscope
*IDN?
ch1:scale?
DATA:SOURCE?
DATA:SOURCE REFA
MEASU:IMMED:TYPE CUR
MEASU:IMMED:VALUE?
HARDCopy:PORT USB
HARDCopy:FORMat JPEG
HARDCopy STAR START
HARDCOPY START
DATA INIT
DAT:ENC ASCII
CURV?
The output from the script looks like this. I was doing an iperf3 test when I grabbed this screenshot.
I tend to over complicate things. Perhaps there is a simple and better way to do this. I see a port utilization value in the switches MIB file but it seems to be slow and keeps incrementing. After the time I have put into this script I’m thinking I should have looked more into what that value is and what I could have done with it. It mostly works and I learned a lot so I guess it turned out well.
#!/bin/bash
# Experimental script to monitor my Ubiquitu 24 + 2 port EdgeSwitch
# If it seems like I may not know what I am doing its because I don't.
# Seemed to also work on my home pfsense box that has 4 ports.
# Known issues:
# --Full port utilization seems to show as closer to 125%. Will update if I figure out why.
# --Could improve by reading port link speeds via snmp
# --I could not figure out how to make the {1..26} to be variables. HELP!
#
# Corrections / improvements appreciated adamd at sdf.org
#
#10,000,000 = 10**7 10Mb
#100,000,000 = 10**8 100Mb
#1,000,000,000 = 10**9 1G
#10,000,000,000 = 10**10 10G
port_speed=$((10**9))
upd_rate=5 #seconds
ip="192.168.0.2"
main () {
while true;
do
InOct=($(snmpget -v1 -Oqv -c public $ip IF-MIB::ifInOctets.{1..26}))
OutOct=($(snmpget -v1 -Oqv -c public $ip IF-MIB::ifOutOctets.{1..26}))
InPkts=($(snmpget -v1 -Oqv -c public $ip IF-MIB::ifInUcastPkts.{1..26}))
OutPkts=($(snmpget -v1 -Oqv -c public $ip IF-MIB::ifOutUcastPkts.{1..26}))
x="0"
clear
echo "Port Utilization ***************************"
for p in ${OldInOct[@]}; do
#TotInOct=$((${InOct[$x]} - ${OldInOct[$x]}))
#TotOutOct=$((${OutOct[$x]} - ${OldOutOct[$x]}))
#TotInPkts=$((${InPkts[$x]} - ${OldInPkts[$x]}))
#TotOutPkts=$((${OutPkts[$x]} - ${OldOutPkts[$x]}))
get_difference "${InOct[$x]}" "${OldInOct[$x]}"; TotInOct=$itemp
get_difference "${OutOct[$x]}" "${OldOutOct[$x]}"; TotOutOct=$itemp
get_difference "${InPkts[$x]}" "${OldInPkts[$x]}"; TotInPkts=$itemp
get_difference "${OutPkts[$x]}" "${OldOutPkts[$x]}"; TotOutPkts=$itemp
((y=x+1)) #Arrays are zero based, ports start at 1
#The Incoming utilization
i=$(echo "((($TotInPkts * (96+64)) + ($TotInOct * 8)) / ($port_speed * $upd_rate)) * 100" | bc -l)
#The outgoing utilization
j=$(echo "((($TotOutPkts * (96+64)) + ($TotOutOct * 8)) / ($port_speed * $upd_rate)) * 100" | bc -l)
#Colors the upload and or download number if it sees above requested percent
echo "P:$y"
if (( $(echo "$i > 1" | bc -l) )); then
printf "\e[31m%2.4f\e[0m, " $i
else
printf "%2.4f, " $i
fi
if (( $(echo "$j > 1" | bc -l) )); then
printf "\e[31m%2.4f\e[0m\n" $j
else
printf "%2.4f\n" $j
fi
#printf "%2.4f, %2.4f\n" $i $j
((x=x+1))
done
OldInOct=(${InOct[*]})
OldOutOct=(${OutOct[*]})
OldInPkts=(${InPkts[*]})
OldOutPkts=(${OutPkts[*]})
sleep $upd_rate
done
}
#Created to deal with packet and octet counter rollovers
get_difference () {
if (( "$1" >= "$2" )); then
itemp=$(($1 - $2))
else
#we must have rolled over the register
#How far was it from the old value to the rollover point?
#4294967296 = 2^32
itemp=$((4294967295-$2))
#add that to the new value
itemp=$(($itemp+$1))
fi
}
main "$@"; exitIf you are willing to install / use bleeding edge software you can have automatic controls on the Librem 5. If you want to give it a shot I just followed the build instructions here:
Link
I’m sure soon, it will be in the current millipixels ap and there will be no need to build manually. I would check before doing a manual build / install.
I am very pleased with the performance and it beats the heck out of doing all the controls manually. As you can see there is still some work to be done. In the challenging light situations the images are too dark IMHO. Manual adjustments were an option, I just didn’t use them here.
I should also note I was also able to take videos although it was rotated 90 degrees. I used ffmpeg to fix.
The one video I took:
The pictures:
Not my every day phone yet but I’m making real progress. My previous issue with 4G calling is resolved. I managed to get encrypted messaging using the matrix protocol working using the Nheko client. Battery life remains an issue but is manageable. There is still work to be done with energy savings, camera, suspend, etc.
Its important to remember that this is running a desktop computer operating system. Software is being developed, or modified, to be usable on a Linux based phone. While it will run many desktop applications they often don’t fit on the screen, have poor touch interfaces, or just use to much power.
When I received the phone, calling didn’t work. I thought calling was completely broken but after traveling a bit I found it did work using 2G in some locations. After some reading I assumed VoLTE or 4G calling just didn’t work yet. That assumption was incorrect. I finally, today, created a support ticket with Purism and received a quick reply with instructions to update the modem. It worked. Now I can make 4G calls.
If you have a similar issue I suggest you email support@puri.sm, as I did.
Subject line should be something like: [Purism_12345678] VoLTE issues (replace the number with your order number)
The email should include what version your modem is currently running and a description of the issue. I found my firmware by installing bm818-tools and running ‘BM818 VoLTE’. This tool is designed to enable VoLTE but also reported the firmware. Funny thing is now the firmware is updated this tool is no longer reporting it. But it did originally work and here is my screen.
I wanted to share the instructions and firmware files but am afraid it could lead to me assisting people in bricking their devices. The instructions and firmware files are likely dependent on what modem and region you are in. So I’ll just stick with contact support.
Short version: Matrix.org account, Nheko client, custom notification script because I didn’t get audio notifications when on the lock screen, NO automatic suspend. I may just be overcomplicating this. If you find that to be the case, please report what setting or option I overlooked.
Long version: I understand Chatty, the default texting ap, can handle the matrix protocol but I don’t think its ready for the masses yet. I had lots of trouble including crashes and not able to read encrypted messages. I tried several other matrix clients and found Nheko works, fits the screen, and seems like a nice program.
In Nheko notifications work fine as long as:
–Your device isn’t suspended. Only texts and calls wake up the device, not messages through Nheko.
–The incoming message isn’t coming in to a currently open chat window. This kinda makes sense. It sees the chat as active, having focus, it doesn’t need to notify you. This is very easy to work around. Hit the arrow to return to the all chats view.
–You are not on the lock screen. If you are, for some reason, I get a visual indication but no audio notification. The LED turns blue but I hear no sound. I don’t know why. I found a work around but it wasn’t a simple thing. I really hope its just me over complicating something there was a setting for.
The script I used:
#!/bin/bash
linesInFile="0"
#close dbus-monitor if its open. Close any running copies of this script if open.
killall dbus-monitor
killall myscript.sh
#start a dbus monitor process and output to a file
dbus-monitor --profile "sender='im.nheko.Nheko',interface='org.freedesktop.Notifications'" > /tmp/dbusmon.txt &
while true; do
linesInFileNow=$(wc -l /tmp/dbusmon.txt | cut -f1 -d ' ')
if [ "$linesInFileNow" -gt "$linesInFile" ]; then
linesInFile=$linesInFileNow;
aplay -q /usr/share/sounds/purple/receive.wav
fi
sleep 1
doneThis script needs to be started after any reboot to work.
I start it with ‘./myscript.sh &’ so it runs in the background. When I close the terminal window it gives a warning but I ignore it and my notifications still work.
Not sure it was needed but I enabled an option in Nheko about exposing room information to dbus.
I need to do more testing but I feel pretty safe saying you can manage about 6 hours when not allowing the phone to suspend. You can let it suspend and you will likely get a lot more but you will only received texts and phone calls. Other chat applications will only function when the phone isn’t suspended. Plug in when you can and have a portable supply as needed for longer outings without power available.
I purchased one of these:
Link
Keep in mind if you have a resource hogging application, its using more power. Perhaps close it. I was keeping a browser tab to Microsoft online open for work and recently discovered it was using a lot of power. I used htop at the command line and did some testing to confirm this.
Using suspend only allows you to receive phone calls and text messages. Other applications stop working while suspended. Also when resuming from suspend I find my cell modem isn’t always working as apparent by there being no cell signal strength or 4G indicators. I fix this by cycling the modem hardware kill switch on and off. For now I am not letting the phone suspend.
These, in my opinion, are awesome. I can save power, reset hardware devices, be more private all with the flick of a switch. I tend to turn off WiFi when traveling. Camera and mic when at home and not using them. One time removing the phone from a pouch, I turned off something by accident. I noticed right away and turned it back on. Other then the single accidental switch toggle, I have had no issues.
Don’t start using this application. It is addictive and bad for productivity. :)
What is going on with Purism? My guess is they have good intentions and ambitious goals and are struggling to keep up. It was pretty frustrating to wait almost 4 years for my Librem 5 phone and easy to feel angry. Its also easy for me to want them to succeed in their goals. I am tired of being the product and want other options. Purism is trying to provide other options.
Now I have my device I am working with them. Here are thoughts on my experiences.
Email support has been great. Pretty sure every Email I have sent, both while waiting for my phone, and since, has received a reply within 24 hours. The replies have been direct and helpful. The only negative has been the replies to ‘when will I get my phone?’ were not accurate. Not sure how much of that was unforeseen circumstances.
I have interacted with them via their matrix channel, again very responsive, direct, helpful.
I have attempted to use the Librem.one services. So far I have used chat(Matrix), social(Mastodon), and email.
–Email works fine, no issues.
–Social works but appears to have no local timeline. I suspect they were getting a lot of hate posts from people tired of waiting for their Librem 5 phones and had to do away with the local timeline. People can be mean, and really mean when waiting 4 years for their Librem 5.
–Chat (Matrix) is outdated and won’t even work with many clients. I have seen this discussed very recently. Seems the DB migration hangs for days with no completion in site. They discuss hiring a professional DBA to fix this for them. I hope they do as I would like to use this service. I am using matrix.org but think it too big. I would prefer to use Purism’s server or another small server.
I plan to continue to support them through their products and services and hope for the best. I think they have lofty goals, good contributions to open source, and I hope they succeed. I’ll just hold off on any pre-orders they offer. :)
]]>