Earth Notes: On Website Technicals (2022-01)

Updated 2022-10-23 12:25 GMT.
By Damon Hart-Davis.
Tech updates: sitemap timestamp changes, power system tweaks, RE Utils V1.1.12, Christmas dip, desktop page experience, not indexed.
tools
Still winding down for Christmas (2025?), but while waiting I might as well tweak sitemaps, battery-bank power management, RE Utils to produce extra info...

2022-01-30: GSC Crawled - currently not indexed

Am I the only one to use the "Index > Sitemaps > sitemap.xml > Excluded > Crawled - currently not indexed" list as a clue for pages that could do with a bit of TLC and tweaking?

2022-01-18: GSC Desktop Page Experience

A Desktop Page Experience report just became available in Google Search Console and in the page experience summary I see the headline:

96% URLs with a good page experience on mobile
100% good URLs on desktop

2022-01-17: Christmas Dip

Screenshot 20220117 GSC Performance impressions dip on Christmas Day
Christmas Day traditions #8734: impressions dip as seen in GSC.

2022-01-10: RE Utils V1.1.12

Working up to integrating the heat battery, I need to make the current grid intensity easily available, in the file system or over HTTP. The latter if I run a second RPi just to talk to the heat battery, possibly at lower power than a long RS422!

The V1.1.12 reutils release publishes live raw plain-text gCO2e/kWh value eg suitable to poll over HTTP, eg to decide when to fill a thermal store from the grid for hybrid DHW provision.

2022-01-09: Power System Tweaks

I need to reduce mid-winter load a bit (go back to just one RPi for example) or add a bit more generation (~250Wp of upward-facing / near-horizontal panel?). In the interim I have tightened up the power management code to be a little more protective, sooner, after an extended period of weather too dull to make the battery FULL.

% /usr/local/bin/powermng -nv
Verbose mode ($Id: powermng.cpp 43291 2022-01-08 16:23:37Z dhd $).
ISEXPPI=1: Expander Pi is in use.
LOAD POWER (mW): 2439
SS-MPPT-15L Vb_f slow battery voltage (mV): 12012
SS-MPPT-15L Adc_il_f load current (mA): 203
Input flag DUMPING is unset.
SS-MPPT-15L Adc_ic_f battery charge current (mA): 218
External flag file /var/log/SunnyBeam/LOW.flag is present.
NET battery charge current (mA): 15
SS-MPPT-15L T_batt (C): 6, temperature compensation 210mV.
Estimated LA battery impedance: 113mohm.
Estimated LA battery sag/(-rise) from internal impedance, and nominal internal
voltage, effective w/ temp comp: 1mV, 12011mV, 11801mV.
Estimated actual battery sag from load (mV): 0mV.
LA/B1 dynamic threshold adjustment for load current 0mV.
B1 and wiring dump load sag allowances 146mV, 194mV.
Battery voltage 1 local (LA) 11854mV.
Supply drop B1 to BV 158mV; implied supply impedance 778mohm.
Adc_vl_f load voltage (mV): 11985
Supply drop B1 to controller load terminals 27mV; implied controller impedance
133mohm.
Supply drop controller to BV 131mV; implied wiring impedance 645mohm.
Input flag EXTERNAL_BATTERY_NOTVHIGH is set.
Long time since VHIGH (or system not been up long enough to be sure).
Input flag EXTERNAL_BATTERY_NOTFULL is set.
Long time since FULL (or system not been up long enough to be sure).
Battery voltage 1 basic compensation 210mV.
Battery voltage 1 'very low' threshold 12410mV (raw 12200mV).
Battery voltage 1 'low' threshold 12660mV (raw 12450mV).
Battery voltage 1 dump margin threshold 12810mV (B1 delta 150mV, sag for dump
load 146mV).
Battery voltage 1 'not high' threshold 13450mV (raw 13450mV).
Battery voltage 1 'very high' raw threshold 13600mV.
Input flag FORECAST_PV_GEN_GOOD is set.
Battery extra USoC reserve for poor forecast / VHIGH and FULL not recent 15%.
Battery 1 effective/usable SoC 0%, dump thresholds min 65% higher 82%
(uncompensated USoC 0%); usable to dump = 0.
SS-MPPT-15L Power_out A1P (mW) 2628.
SS-MPPT-15L Adc_va_f array voltage (mV): 23783
LASTDATA: 2022/01/09T09:08:55Z AL 218 B1 12012 B2 -1 P 2439 BV 11854 ST VL - L
A1P 2628 B1T 6 UC 0 A1V 23783

2022-01-06: Sitemap Timestamp Tweaks

I was basing the timestamps in sitemap.xml and sitemap.atom on the timestamp of the .*.html source files. (Usually: if there was an insert for the file then its timestamp could win.) This was to avoid bothering search engines (SEs) with non-content changes, eg when one of the build scripts was updated, or there were some site-wide style changes.

But this was starting to get complicated.

These have now been changed to use the build time of the .html HTML files. Note that the pre-compressed .htmlgz and .htmlbr files still may have later timestamps than the sitemap entries.

Also: AMP has finally gone from the GSC 'Enhancements' list in 'Overview'!

~503 words.