Jekyll2024-03-18T18:41:10-04:00https://mouse.mousetrap.net/feed.xmlthe mousepadunending dorkinessfrater secessusbackchannel - solar rules of thumb2024-03-18T13:04:35-04:002024-03-18T13:04:35-04:00https://mouse.mousetrap.net/blog/backchannel---solar-rules-of-thumb-[this is a response to [a thoughtful post](https://www.reddit.com/r/vandwellers/comments/1bh9iwt/buying_a_van_soon_are_these_specs_enough_to/kvcggt2/) on reddit.]
> two ratios that I followed in my build
Valid starting points. I want to bring up a couple things that may warrant consideration.
> Solar should be twice the Watts as batteries have **usable** Amp-hours.
The *usable* part could result in undersizing the array, especially problematic with lead chemistries.
* The traditional [ratio rule of thumb](https://rvwiki.mousetrap.net/doku.php?id=electrical:solar:panel-bank_ratio) is 1:1 panel-to-Ah. So 200w for 200Ah.
* basing panel wattage on *usable* Ah would mean 100w for 200Ah of lead (ie, 100Ah usable), or 0.5-to-1.
* 100w of panel would require ~**14 hours of FSE** (full sun equivalent) to fully recharge 100Ah usable (lead), about 2x what is available anywhere in CONUS. ([blog post on this topic](/blog/2020/07/02/the-problems-with-11-solar-charging/))
Undercharging isn't a problem for Li chemistries. But since usable % is greater (~80%) the *usable* criteria trips us up in a different way. Even assuming perfect charging efficiency a 0.8:1 ratio won't cut it in CONUS. Let's use 200Ah again for consistency.
* we need to replace 2,048Wh (160Ah x 12.8v) with 160w of solar to get back to 100%
* assuming 85% solar harvest efficiency this will require over **15 hours of FSE** (2,409Wh÷160W), again ~2x what is available in CONUS even under optimal conditions.
> Batteries should have triple the Ah of daily usage to account for cloudy days.
Reasonable, assuming lithium. There is little downside to running [larger Li banks](https://rvwiki.mousetrap.net/doku.php?id=electrical:12v:drop-in_lifepo4#sizing_the_bank).
With lead banks having more bank than can be charge fully/regularly is a recipe for [battery murder](https://rvwiki.mousetrap.net/doku.php?id=electrical:batterycide).frater secessus[this is a response to a thoughtful post on reddit.]a string of successes2024-03-17T16:12:05-04:002024-03-17T16:12:05-04:00https://mouse.mousetrap.net/blog/a-string-of-successs-even a loser like me can win sometimes
## boondocking
[![view from the road](https://img.mousetrap.net/2024/thumbs/IMG_20240315_132830.jpg)](https://img.mousetrap.net/2024/IMG_20240315_132830.jpg)
[![weekenders](https://img.mousetrap.net/2024/thumbs/IMG_20240315_103355.jpg)](https://img.mousetrap.net/2024/IMG_20240315_103355.jpg)
I'm back to boondocking and it does my heart good.
This spot is small and overused but still better than being in the city. In the first pic the van is center-left. There were probably a half-dozen rigs on hese few anchors when I got here. Over the weekend it got quite full (2nd pic) and noisy. :-\
## sun
The solar harvest has been terrific. Even on overcast or partly cloudy days I've been able to do all my cooking and other duties off solar.
Today has been partly cloudy, I'd say 50% cloud cover. But I hit Vabs by 1300 and made coffee, oatmeal, cornbread, and red beans off excess solar. Must have got some cloud-edge because the controller maxxed at 45A at least once.
## VictronConnect-x86
<a href='https://img.mousetrap.net/2024/victron-linux.jpg'><img src='https://img.mousetrap.net/2024/thumbs/victron-linux.jpg' align='right' /></a>
I'd downloaded the [Victron Connect app for linux](https://community.victronenergy.com/questions/43667/victronconnect-for-linux-download-instructions.html) a couple years back but never got it to work right. It could see the devices but crashed consistently when I tried to select one.
*Something* (linux update? a recent Vic BT update?) changed in the interim because ``VictronConnect-x86_64-v5.90.AppImage`` loads and runs correctly now. Coolio. Newest V6 core dumps immediately.
## scripting
This is one of those cases where you struggle with something, lay it aside, return later on and things just fall into place.
### normalization
I've been wrestling with audio normalization on the Myth pi. Challenges:
* some OTA broadcasts (almost always HD) have very low audio
* the Pi doesn't have much grunt so I want to minimize the workload
I started playing with [ffmpeg's speechnorm filter](https://ffmpeg.org/ffmpeg-all.html#speechnorm) (``-filter:a speechnorm``) and it's starting to come together. Single pass, not too heavy, does a decent job with the levels.
It went so well that I wanted to leverage it on my podcasts, yt vid, and movie DLs that are Processed Elsewhere. Took me a while to figure out that the ffmpeg on the box Elsewhere has an older version of ffmpeg that predates speechnorm. Oh. I d/l'ed a new statically-linked version and it's working well.
Previously I was unpacking all my podcasts to wav so I could compand/normalize and delete silence. Then encoding to ogg. Now with speechnorm and native ogg support it's a one-liner: ``-filter:a silenceremove=1:0:-50dB,speechnorm``.
### subtitles
My previous movie processing script was not preserving embedded subtitles. It would crash when none were present so I stopped trying. This time around I used ``ffprobe`` to check for the presence of subs. If present I define the sub-related command otherwise set it to null:
~~~
if [ "$CHECKSUB" = "subtitle" ]
then SUBS=" -c:s mov_text "
else SUBS=""
fi
~~~
I inject the $SUBS string into the ffmpeg command later on; if it's null it has no effect on the command. Nothing there.
I use *mov_text* since that's mp4's preferred format. I output mp4 because mp4 with x264 plays back with hardware decoding on the pi.
## comments
[mastodon comment thread](https://vmst.io/@fratermus/112113387455087795) for this postfrater secessuseven a loser like me can win sometimestwo week banking nightmare appears to be over2024-03-15T13:49:23-04:002024-03-15T13:49:23-04:00https://mouse.mousetrap.net/blog/two-week-banking-nightmare-is-over-## TL;DR
I deposited a check. The sending bank acted squirrelly[^squirrelly], confusion ruled the land, my bank ([Ally]({{ site.afflinks.ally }})) suspected foul play and locked all my accounts on Feb 29th.
[^squirrelly]: they made one attempt (!) to contact the check sender for verification on a phone number from the 90s instead of what's on their account (!!) then marked it [Refer To Maker](https://www.routingtool.com/refer-to-maker.html) (!!!). A snowballing shitshow ensures.
During the lockout my direct deposits and 2023 tax refund were deposited in the inaccessible accounts, so I couldn't use them for damage control or to access cash. Luckily I still had my [Amazon Prime credit card]({{ site.afflinks.prime }}) for short-term needs. Unluckily the payment date was coming up and I had no accounts to pay it from....
I scrambled to open accounts elsewhere and moved all my direct deposits over. I had Chase pay my Prime CC from the new savings[^funded]. All those emergency maneuvers were completed by yesterday. This morning (Mar 15) the two banks finally talked to each other and the sender; it was resolved within an hour. grrrrrr......
[^funded]: funded by a replaement check which went through with zero drama
## lessons learned
1. if things go awry, **have a new check cut** instead of redepositing (even if CS says it's ok to do so)
1. have accounts in different financial institution so one bank can't do critical damage to your liquidity
1. have some cash on hand[^cash] in case you need to deposit something to open another account
1. have any direct deposits go to separate institutions for the same reason
1. you can't pay a [Chase]({{ site.afflinks.chase }}) CC balance with a Chase savings account
1. you can't Zelle from a Chase savings account although [Zelle itself allows it](https://www.zellepay.com/how-it-works)
1. USAA "disables email verification 1x/year", which may result in being locked out of the website.
There were also some subjective lessons. Ally and USAA customer service is exceptional; Ally fraud department ranges from unhelpful to bizarrely hostile. Chase's in-person lobby experience continues to feel predatory, like walking onto a used car lot.
[^cash]: I had $12 cash, not enough to meet the CC $35 minimum at least.. Normally I'd have a bit more but I'd expected the check to be deposited so I hadn't made a routine withdrawal from my retirement account. I couldn't make a withdrawal *now* because it too would be direct deposited in the locked account. Then I remembered: **I had pesos** in my passport pouch. I had ~700 pesos which was ~$40 after conversion. I used that cash to pay the minimum and paid off the rest a few days later when the new savings balance was available.
## conclusion
*I really like Ally* and I am glad they reopened my account[s]. My Ally savings account yields ~44x (times, not percent) what my new Chase savings yields. My Ally checking yields the same as Chase savings.
I will keep the new checking and savings (different institutions) open just in case.
### silver lining
I didn't realize it at the time but my new checking might be eligible for a $200 incentive. I opened the account on the last day of the offer:
>> A qualifying direct deposit is a single Automated Clearing House, also known as ACH, electronic deposit of $100 or more of your United States military salary pay, retirement pay, **military disability pay**, or annuity pay for a surviving beneficiary from Defense Finance Account Service....
My VA disability benefit is small but should meet those requirements. If so, it feeds the emergency fund. If not it's no harm, no foul. I'll report back in ~June when the payout timeline appears to be satisfied (next direct deposit + 60 days).
## comments
[mastodon comment thread](https://vmst.io/@fratermus@lemmy.sdf.org/112101485413922662) for this post
[lemmy comment thread](https://lemmy.sdf.org/post/13851008) for this postfrater secessusTL;DRboondocking resumed2024-03-12T11:11:05-04:002024-03-12T11:11:05-04:00https://mouse.mousetrap.net/blog/boondocking-resumed-I got my ducks in something resembling a row yesterday and departed El Paso this morning for the Alamogordo area.
## opened another savings account
<a href='https://img.mousetrap.net/2024/IMG_20240311_085233-rotated.jpg'><img src='https://img.mousetrap.net/2024/thumbs/IMG_20240311_085233-rotated.jpg' align='right' /></a>
The replacement check arrived at the USPS and I picked it up. On the way out I got a pic of NASA's [super guppy](https://en.wikipedia.org/wiki/Aero_Spacelines_Super_Guppy) parked in the adjacent airport. I'd seen it flying for the first time a few days earlier and seemed like the most unlikely shape for something in the air that wasn't a blimp.
I went to Bank of America first to see if they could pre-clear the check so [another slow-motion disaster](/blog/2024-03-05-preparing-to-leave-wintering-spot.html#bank-shenanigans) doesn't occur. The clerk (a manager filling in for someone on break, one supposes) said it was impossible. There is, in effect, no way to deposit a check without endangering my accounts. WTF.
Since I have an account at Chase I opened a savings account there and deposited the check. I hope it is drama-free.
## VA meds
I went to the VA to pick up a blood pressure refill before leaving town, as this will save me remailing costs. They'd done it before but this time the clerk said she couldn't fill the whole thing and would have to mail the rest. So not only will I have to pay for remailing but I'll get get double-dinged by the VA for the fill. I did not scream in frustration..
## laundry
Laundry was the only thing that day that went to plan.
## back in the scrub desert
I am happier and Muffin is *much* happier out of town. I'll be here for 14d then move on to Capitan or Albuquerque, depending on weather.frater secessusI got my ducks in something resembling a row yesterday and departed El Paso this morning for the Alamogordo area.MythTV pi DVR project- finding new OTA shows2024-03-09T12:21:13-05:002024-03-09T12:21:13-05:00https://mouse.mousetrap.net/blog/finding-new-OTA-shows-One of the things that's been meaning to address is how to find new programs that show up Over The Air without having to slog through the ``Schedule Recordings | Program Finder`` alphabetical list. vgrepping[^vgrep] is prone to error and fatique. IOW, I want to find when a new show called Awesome Show is found in EIT data for the first time.
[^vgrep]: *visual grep*, aka eyeballing it
To avoid impacting the Pi's microsd (where mysql/mariadb is stored) I dump the known names into a text file on the SSD drive. The basic idea is:
## pull the current showname list
Get the current list of upcoming shows from the mythtv database:
``select DISTINCT title from program``
and write them to a file.
## and compare to the complete list
``comm -13 $LOGDIR/list.archive $LOGDIR/list.latest``
the ``-13`` suppresses the LEFT and BOTH columns, so just shows the new entries in the .latest file.
## and make a new complete list
by catting old and new together and deduping.
## sample output
~~~
# # ###### # # #### # # #### # # ####
## # # # # # # # # # # # #
# # # ##### # # #### ###### # # # # ####
# # # # # ## # # # # # # # ## # #
# ## # ## ## # # # # # # ## ## # #
# # ###### # # #### # # #### # # ####
Campeón sin corona
Dyson Cleaning
Holmes & Watson
Killer USA
Línea de emergencia
On the Red Carpet
Rápidos y furiosos 8
Spiders 3D
Support Your Local Gunfighter
Support Your Local Sheriff!
~~~frater secessusOne of the things that’s been meaning to address is how to find new programs that show up Over The Air without having to slog through the Schedule Recordings | Program Finder alphabetical list. vgrepping1 is prone to error and fatique. IOW, I want to find when a new show called Awesome Show is found in EIT data for the first time. visual grep, aka eyeballing it ↩resuming snowbird circuit, van repairs, Ally locks my accounts2024-03-05T13:06:43-05:002024-03-05T13:06:43-05:00https://mouse.mousetrap.net/blog/preparing-to-leave-wintering-spot-## leaving El Paso
I've been wintering in El Paso both to get my V.A. appointments done and as part of my [reduced snowbirding circuit](/blog/2023-04-24-a-more-modest-snowbirding-loop-.html). Temps are starting to warm up (nearing 80F) so it's time to move on.
Some things have been delaying my departure
### tire puncture
<a href='https://img.mousetrap.net/2024/IMG_20240208_143756.jpg'><img src='https://img.mousetrap.net/2024/thumbs/IMG_20240208_143756.jpg' align='right' /></a>
I picked up nasty piece of debris in my driver-side rear tire; pretty sure it was a carriage bolt. The tire lost ~7lbs of pressure overnight so I used my compressor to top it off while I looked for a tire shop.
Because of the size of the bolt I was worried I might need a new tire. But Bori at [M&M Ecomony Tires](https://maps.app.goo.gl/CxWB3898hKqUhwsv9) thought it would be a simple patch. He showed me the backup plan (a patch with an integral centerpost [like this](https://amzn.to/3TxLTEl) IIRC) but a plain patch workd fine.
Fast, friendly, and $15. 100% satisfied.
### van codes
[note: I'm pulling/clearing codes with [this cheap OBDII bluetooth dongle](https://amzn.to/49CM8U9) and [Torque Pro](https://play.google.com/store/apps/details?id=org.prowl.torque). The [free version](https://play.google.com/store/apps/details?id=org.prowl.torquefree) also worked but I had credits on Google Play and wanted to support the author.]
I've had some emissions-related codes since the time a rodent started dragging debris into my engine compartment:[^unrelated] P0455 gross leak, P0456 very small leak, p0457 check fuel cap. I started replacing the least expensive and most likely candidates over time: [vapor canister purge valve](https://amzn.to/3TmCRcT), [leak pump module](https://amzn.to/3wGaFsN), [MAP sensor](https://amzn.to/3T1bzHP). No change.
[^unrelated]: an unrelated coincidence, it appears
The next in line was the [vapor canister](https://amzn.to/3V8zhV9) but it was hundreds of dollars and I wasn't sure that was the problem. I waited.
### more van codes
<a href='https://img.mousetrap.net/2024/Screenshot_2024-03-04-14-52-35-506_org.prowl.torque.jpg'><img src='https://img.mousetrap.net/2024/thumbs/Screenshot_2024-03-04-14-52-35-506_org.prowl.torque.jpg' align='right' /></a>
Recently the van started throwing P0172 and P0175 errors, which are "too rich" errors on both banks. For those keeping track, the full list of codes is shown to the right.[^spoiler] I could clear them with Torque but they'd come right back.
[^spoiler alert]: spoiler alert: these are all historical codes. I didn't screencap the the active codes at the time.
Drivability was not affected so I started googling. And started looking for a mechanic. Everyone was booked out for at least a week so I Googled Harder. It was starting to look like the vapor canister was implicated in both problems. Since the "too rich" codes and fuel trims were similar on both banks I figured it had to be something common to both rather than something like a leaky injector on one bank. I listened around with a tube trying to find vacuum leaks to no avail. I had 3 out of the 4 common symptoms for clogged vapor canister[^symptoms] so I figured I'd need one whether or not it was related to the codes.
[^symptoms]: emissions codes, slightly reduced MPG, pump shutoff while filling tank
Then the engine started running rough on cold start. I decided to dig back into the harness to dike out [the chewed sections of wire](https://img.mousetrap.net/2024/IMG_20240301_141012.jpg) and splice in new wire. No change, but I felt better about it.
Fine, if I need a canister I might as well source and install it myself. It would cost less and (if I were incredibly lucky) solve the issues/codes. I found one at Autozone for ~$250 after tax, which is at least $100 less than the OEM. The Autozone experience was a beatdown (failure to pick, broken part, exchange at different location) so in total it took almost 6 hours to get the part from them.
I installed the canister, cleared the codes, and started the engine. No new codes, just the historical ones. [cautious optimism loading....]
<a href='https://img.mousetrap.net/2024/Screenshot_2024-03-05-08-29-55-837_org.prowl.torque.jpg'><img src='https://img.mousetrap.net/2024/thumbs/Screenshot_2024-03-05-08-29-55-837_org.prowl.torque.jpg' align='right' /></a>
This morning there were no new codes and most of the historical codes were already falling off. Woot! Since there are no new codes I suspect this means the canister was the last piece of the puzzle. The other parts might have helped but I have no way to know. They didn't cost much, anyway.
### general delivery mail
So from a van standpoint I am free to move about the country, as the commercial said.
But I am waiting on a packet of mail to come in by General Delivery. I think it will be here by Friday the 8th. I hope so because by Sun/Mon temps will start touching 80F.....
The mail is a re-cut check.
#### bank shenanigans
[I moved all this noise to [a separate blog post](/blog/2024-03-15-two-week-banking-nightmare-is-over.html)]frater secessusleaving El Pasoimprovements and observations2024-02-22T19:15:47-05:002024-02-22T19:15:47-05:00https://mouse.mousetrap.net/blog/improvements-and-observations-## storage
My food and clothing storage has never been great. I started out with a snap-together cube arrangement which was excellent for organization but not sturdy enough for vanfolk. It collapsed regularly and there may have been cussing.
I tired of that and bought four large stackable bins, two for food (pantry and backup) and two for clothes (in-season and out-of-season). This worked ok, except:
1. I had to remove the top bins to access the bottom ones
1. there was a lot of wasted vertical space
1. and no room for appliances like crock pot, rice cooker, etc.
I picked up a low coffee table thing (?) to stack the bins on. This reduced wasted vertical space and allowed appliance storage underneath. This worked ok for a while then the stacking and unstacking started to aggravate me. Then the bottom bin lids started collapsing from the weight of the upper bins (my fault).
I've been threatening to buy wire kitchen racks since ~2018.
### wire racks
I measured the space and found something that worked: two racks that are 54" x 36" x 14". So the total space is 72" wide. I was looking shelves deeper than 14" but all the 16" and 18" were to tall too fit under my overhead shelving.
<a href='https://img.mousetrap.net/2024/IMG_20240222_103307.jpg'><img src='https://img.mousetrap.net/2024/thumbs/IMG_20240222_103307.jpg' align='right' /></a>
It took several hours to pull everything out of the driver's side of the van, build the racks in place, strap them down, put everything back on the racks and make paracord or bungee retention. The result is very usable. It did not increase my actual storage area but there are substantial benefits:
1. it made every container directly accessible
1. since each container sits on a shelf the broken lids are now just dustcovers -- I don't have to buy any more containers.
1. the narrower width increased "hallway" width by 6-10", depending on the spot one is standing
1. I haven't started cleaning out each container in earnest yet, but direct access allowed me to shift stuff around and find stuff to jettison. The chaos and jumble was starting to weigh on my mental health.
Total cost was $113 with tax and the Home Depot 10% vet's discount. Money well spent.
### the pillowcase trick
I saw a vid once where a person put their off-seaon clothing in a pillowcase. OMG. I did that yesterday, freeing up an entire large bin. I moved the appliances and bulk staples (flour, cornmeal, rice, etc) into it. The bin, not the pillowcase.
## MythTV
### bulking up
The 1TB SSD was hovering around 40% utilzation (~250GB of YT vids and ~150GB of OTA recordings). I decided to bulk up the OTA records to "bank"
them for times I am not around OTA signals.
I increased the number of episodes to keep on shows I like, and added a few shows I wasn't that excited about. It costs nothing and may pay off it I am way out in the sticks.
### MPV is teh MVP
I really, really (really) wanted to get subtitles elegantly included with my YT vids. It seemed straightforward but I ended up have many forms of breakage when YT did weird things with the subs. Finally got that figured out.
Part of that journey involved testing an external player for ``.mkv`` and ``.mp4``. I landed on [MVP](https://mpv.io/) and HOLY CARP!! is it excellent. I remapped the key commands to mimic MythTV's internal player controls. I would switch OTA playback to it, too, but I'd lose the progress icons.
### video sorting
I have a script that auto-sorts YT vids based on title string matching but it's not very precise. I think it gets about 85% correct. It would take a lot of work to make it 100% and it was time consuming to manually move mismatched vids to their rightful homes.
So I wrote a little script that presents each vid in a sorted subdir for categorization and allows one-key sorting to another subdir. Very fast.
My next microproject will be renaming downloaded .srt subtitles to match the movie file's naming scheme so the players will recognize them. It's starting to take rough shape in my noodle but I haven't opened a text editor on it yet...
## Insolation wars: Yuma vs El Paso
I usually think of Yuma as an example of good insolation in deep winter, and frequently use it as an example.
Last year I was doing [some math](/blog/2023-04-24-on-the-possibility-of-all-electric-cooking.html) and noticed that El Paso, TX actually has higher winter insolation -- 3.34kWh/day in December vs 3.16kWh in Yuma. Forgot to mention it until now.frater secessusstoragedoes higher array voltage make more power in the morning?2024-02-15T19:12:52-05:002024-02-15T19:12:52-05:00https://mouse.mousetrap.net/blog/does-higher-array-voltage-make-more-power-in-the-morning--It is often said that higher array voltage means the solar setup will make power earlier in the morning. Let's test it.
## the data
<a href='https://img.mousetrap.net/2024/Screenshot_2024-02-15-17-03-40-321_com.victronenergy.victronconnect.jpg'><img src='https://img.mousetrap.net/2024/thumbs/Screenshot_2024-02-15-17-03-40-321_com.victronenergy.victronconnect.jpg' align='right' /></a>
I've observed this phenomanon with different panels, panel arrangements, and controllers. Today I finally made a screenshot so we could talk about it here. This particular setup is 750w at 20v nominal (30.3Vmp) on a [Victron 150/45 controller]({{ site.afflinks.victron15045 }}).
This is the timeline:
* 0610 - array voltage (*Vpanel*) starts to come up from 0v. Harvest is unmeasurable.
* 0649 - Vpanel has risen to 14.4v but harvest is still unmeasurable.
* 0712 - Vpanel has risen to 24v and harvest becomes measurable at 5w (0.7% of rated). This would ~14.4v for 12v nominal panels.
* 0720 - the array has reached Vmp and harvest jumps to 25w (5.5% of rated).
* 0753 - the array has been at normal voltage for a half-hour and has just risen to 75w (10% of rated).
## analysis
In this example The 20v array starts making measurable power 23 minutes earlier than a 12v array would.
Actual harvest during those 23 minutes is tiny: less than 1% of the array's rated wattage. With a 200w array it would be a an extra **0.19Wh** harvested each morning.[^difference]
If there were thunderstorms and it was very dark all day it would be a whopping gain of 11.2Wh.[^allday]
[^difference]: 200w x 0.007 x [8/60 minutes] = 0.19Wh
[^allday]: 200 × 8 hours x .007 = 11.2Wh
## conclusion
So, yes, a higher voltage array will make *more* power in low-light conditions. After all, 0.19Wh is *more* than 0.0Wh. In practice the difference is so small it could be a rounding error.
There are valid reasons to run higher-voltage arrays but IMO increased low-light harvest is not one of them.frater secessusIt is often said that higher array voltage means the solar setup will make power earlier in the morning. Let’s test it.tuning the Pi4 MythTV rig2024-02-15T15:28:28-05:002024-02-15T15:28:28-05:00https://mouse.mousetrap.net/blog/tuning-the-Pi4-MythTV-rig-## from 32-bit buster to 64-bit bookworm
## frontend without graphical desktop
Since the only thing the pi4 displays is MythTV I didn't want to lug around the a whole desktop environment or even just a window manager. Under Debian Buster I was able to start just one graphical window from the console:
xinit "mythfrontend" $* -- :0 vt$XDG_VTNR
but this didn't work after the OS upgrade. It took a couple hours over a few days to figure it out. It seems that X under bookworm requires root privs to run ``xinit``. Fine, I'll do it:
sudo xinit /usr/bin/mythfrontend -- :1
## screensaver
The monitor blanking after sitting a while is annoying so I wanted to stop that from happening. Many sources said that:
xset s off
xset s noblank
xset -dpms
... should work. I could do it from the commmand line (ssh'd in from the laptop) but couldn't it to take effect programatically from the usual places on the pi like ``.xinitc``, ``Xsession``, etc. . If I called it before xinit that program complained it couldn't find the display. True enough, there wasn't one running yet. If I moved xinit to a background process with ``&`` and called xset afterwards it *still* complained about no display. What? I knew there *was* a display running that point.
An hour of Google-fu revealed frequent mention of *timing issue* and *sleep*. Ohhhh. xset was running before the X window was fully on its feet. The working invocation is now:
sudo xinit /usr/bin/mythfrontend -- :1 &
sleep 5
export DISPLAY=:1
xset s off
xset s noblank
xset -dpms
## dark theme
The ``Mythcenter-wide`` theme that comes [bundled with MythTV](https://www.mythtv.org/wiki/Category:Themes#Bundled_Themes) has excellent functionality for my use case:
1. shows when shows are recording normally (green dot) or if there are signal losses (yellow dot)
1. shows the progress of the recording ("40% recorded")
1. if the show is partially-watched it will show the position ("15% watched")
[<img src="https://www.mythtv.org/w/images/thumb/d/de/MythCenter-wide.png/450px-MythCenter-wide.png" align='right' />](https://www.mythtv.org/wiki/Category:Themes#Bundled_Themes)
The problem it is blindingly bright blue, a problem at night especially when one is [trying to be stealthy](https://rvwiki.mousetrap.net/doku.php?id=camping:stealth_camping).
I tried all the dark themes but each was missing one or more of the features above. I decided to see how hard it would to modify the functional theme to be darker. Turned out to be much easier than I thought.
1. copy ``/usr/share/mythtv/themes/Mythcenter-wide/`` to ``/usr/share/mythtv/themes/dark-wide/``
1. make a black verion of ``/usr/share/mythtv/themes/ui/background.png`` to replace the blue original.
1. edit ``/usr/share/mythtv/themes/themeinfo.xml`` to change the name so there are not two themes that show up as Mythcenter-wide in the theme chooser.
1. select the new theme in the theme chooser to see if everything looks ok
1. take a screenshot of the main menu to replace ``/usr/share/mythtv/themes/preview.jpg`` so it shows up properly in the theme chooser
<a href='https://img.mousetrap.net/2024/dark-wide.jpg'><img src='https://img.mousetrap.net/2024/thumbs/dark-wide.jpg' align='right' /></a>
The resulting theme isn't pretty but it *is* much darker. I'm calling it [Good Enough](/goodenough.html).
## disk i/o hangs
While performance is quite good, sometimes if I'm hammering the frontend with rapid changes (like holding down the key to rapidly scroll through the recorded programs) is the pi's green LED goes solid for a few seconds and everything hangs for a bit...
I wanted to know what was clogging the SD card I/O so I triggered the problem a few times while watching ``fatrace``:[^fatrace] over ssh:
mariadbd(859): W /var/lib/mysql/mythconverg/program.MYI
mariadbd(859): R /var/lib/mysql/mythconverg/program.MYD
mariadbd(859): W /var/lib/mysql/mythconverg/program.MYD
``mariadb`` is the FOSS fork of mysql after [that project was sold](https://techcrunch.com/2008/01/16/sun-picks-up-mysql-for-1-billion-open-source-is-a-legitimate-business-model/). We can see that both Read and Write processes in the MythTV database are involved.
At this point I don't know if it's a tuning issue with mariadb, an SD card issue, or something else. It happened on two different Class 10 cards so who knows. If I figure it out I'll update y'all about it.
[^fatrace]: this is not a slur against our heavy brothers and sisters. It appears to mean "file access trace"frater secessusfrom 32-bit buster to 64-bit bookworminstalled the replacement shunt2024-02-10T12:00:27-05:002024-02-10T12:00:27-05:00https://mouse.mousetrap.net/blog/installed-the-replacement-shunt-Last post [I mentioned](/blog/2024-01-31-still-alive-in-El-Paso-.html#shunt) that I'd ordered a replacement [battery monitor](https://rvwiki.mousetrap.net/doku.php?id=electrical:12v:battery_monitor).
## the original shunt
[<img src="https://img.mousetrap.net/2024/drokshunt.jpg" align="right">](https://amzn.to/3uBau11)
In 2018 I installed [a 75A Drok shunt](https://amzn.to/3uBau11) ($27, but discontinued). It served me well with the 220Ah FLA bank, 100Ah LFP bank, and [100+50Ah bank](/blog/2023-02-01-leaving-ELP--adding-battery-bank-capacity.html#adding-to-the-battery-bank) until I smoked it with a careless short during some kind of battery bank twiddling.
A 75A shunt might seem like a limitation but I don't have any loads larger than that. My theoretical load is the Instant Pot, ~54A if pulled from battery alone. But I rarely run it without solar doing some/most of the lifting. My biggest practical load, when I am confident in the day's future harvest, is the coffeemaker. It pulls 46.5A and is running in the early morning before solar harvest hasn't even met my background loads.
I knew I liked the Drok's display but didn't give it much thought. Spoiler warning: this will bite me in the butt.
## the shuntless interim
The 100Ah batt has a bluetooth BMS but the [50Ah Chins]({{ site.afflinks.chins }}) does not. I *suspected* they would take/contribute current according to their relative capacity and it was very close. The 100Ah takes a tiny bit more of its share (like 68% instead of 66.6%) but for practical purposes I can look at the BMS and multiply current by 1.5.
But I liked the ability to walk by the display and see what's happening without opening an app.
## the replacement shunt
The Aili is the cheap-shunt darling so I was leaning that way. The one I selected had free shipping to locker and I snagged it.
After installation and a bit of use I realized why I like the original Drok so much: it had all the data I wanted on one screen:
* Amps charging/discharging
* Watts charging/discharging
* volts
* SoC%
* remaining Ah
* time to full (while charging) or empty (while discharging)
<a href='https://img.mousetrap.net/2024/aili.jpg'><img src='https://img.mousetrap.net/2024/thumbs/aili.jpg' align='right' /></a>
The Aili-style monitors don't have the time estimate[^time] or watts. That doesn't bother me, but what *is* available is on different screens and requires interaction. I find that annoying.
[^time]: The time estimate is more interesting than useful for prediction. It assumes a constant charge/discharge rate but solar is highly variable. I suppose if one were idling to charge with a DC-DC it could be useful. Or if one had nearly no solar coming in it could tell you how long until the power goes out...
Even more annoying is I knew that's how they work. I'd seen the pics and vids of them in use. It meets needs, though, and I don't care enough to replace it. If I were doing it again I'd probably pick [this drok](https://amzn.to/4bAODaE). The color screen is not as clean as the old green one but it has all the data on the same view.
{edited to add pic of the offending display}
### observations
* the shunt setup continues to align with [BMS x 1.5)
* my daytime background loads right now are ~6A, including the mattress pad on lowest heat for Muffin's napping comfort.
* cloudy today, but was clear for about 15 minutes just before 10am. 30A net into the bank that early is a nice change from December's low yield.
### review
The Aili-style monitors are fine if you like the display. They are *acceptable* if you don't like the display.
Some reviewers said the wiring was too short but I'd say the included 3ft was sufficient for my needs. In any case there is [a 16ft wire](https://amzn.to/42CltE5) available for $3. I might pick one up in case I decide to remount at eye level some day.
Some reviewers dislike the backlight but the manual shows how to turn it off. I don't mind it.
## recent solar harvest
There were two utterly overcast days this week where I didn't reach Vabs. One day produced 780Wh (ouch!) and the day after 1.64kWh. The 2nd day would have reached Vabs if it wasn't playing catch-up from the abysmal day before.
On clear days I'm starting to see max harvest in the 500s. I've seen 600+ but it was cloud-edge effect.
## supahbowl
I found a spot with good reception to record the superbowl. I don't care much about it but it's a cultural touchstone.
## comments
[mastodon comment thread](https://vmst.io/@fratermus@lemmy.sdf.org/111908657839548461) for this post
[lemmy comment thread](https://lemmy.sdf.org/post/11963594) for this postfrater secessusLast post I mentioned that I’d ordered a replacement battery monitor.