Four batteries

 I now have four LiFePO4 batteries in place and balanced. The batteries store 4.8 kWhr of energy. 

The RPi is now set up and running a 24-hour schedule.

New Batteries: Three and counting

 There are now three LiFePO4 batteries wired in parallel, connected to my system.

   There is a fourth on order.

   Eventually, I want to have six for a total of 7.2kWh of stored electrical energy. 

   The catch is, there is a high probability tariffs wiil be levied against Chinese products by the incoming President. Prices will go up. There's no way to know how high prices will go and if availability remains steady.

   For that reason, I'm considering buying the remaing two before next year (2025). Another, more positive incentive, is a 30 day price guarantee for $147.16 per 12V/100Ah battery.

Currently, the retail is $169.99. I will have to think about this some. The guarantee ends on the 23rd, I believe. I'll have to tighten my belt, if I do.

New batteries

 Ever since the lightning strike of last year, I've been recovering, slowly, but surely.

A few weeks ago, the first of several LiFePO4 batteries arrived. The old lead-acid battreries were relatively short-lived and probably suffered damage when lightning ran in on the steel guy wire.

Use of the type battery has taken some getting used to, but I'm finally settling in and ready for the next two or three.

The lithium-iron-phos[phate battery can be dischrged completely and recharge in as little as three hours. Lead-acid batteries should not be discharged below 50% and are slow to recharge. Think days in the winter time. I had six Duracell Ultra Marine lead-acid batteries.

The old batteries were sold and the money placed toward the purchase of the first LiFePO4 lithium iron phosphate battery. The new batteries cannot start a car, but they last longer, charge quicker, and now are much cheaper, and they can run a lead-acid charger to charge an auto battery. The new battery cost about the same as a lead-acid battery with core charge. The new batteries are 12VDC@100Ahr where the lead-acid batteries were 12VDC@50Ah. I will have the same capacity in three lithium batteries as I had in six lead batteries. Note : I am moving quickly now since the incoming president has stated tariffs of 30% will be placed on imports, especially, Chinese, where the inexpensive batteries come from.

There is a need to understand the new battery type, for their are settings associated with the type that differ from those of the old lead-acid type. I corresponded with three sellers regarding the settings. 

Epever - for the Triron charge controller.

LiTime - for the new LiFePO4 battery

WattCycle - for the two LiFePO4 batteries on their way.

I also consulted several YouTube videos on the best battery brands and the settings.

I am confidant the setup is pretty much optimized. The next step is to re-build the panel array and supporting framework. The ladder support toppled due, I believe, to a high wind one night. The array needs an improved support, one that is stable in high winds. The panels need to be replaced by higher wattage panels. This may, in turn, require modification of the inverter-controller setup.

Solar power station Limitations

Charge controllers:

Epever Triron 4210N / 40A@24VDC

Xantrex C-60 PWM / 60A@24VDC

Inverters:

Xantrex Freedom X 1200 / 1200W

Batteries:

12V/100Ah / 1200Wh

Panels:

5x100W+1x85W / 585W

Upgrades needed for 2kW station:

Approx. 2400W

2400/24v=100A

The source for panels will probably be used panels from 'SanTan Solar' with a branch in Atlanta.

Changes done

The changes mentioned in the last post have been completed. 

In addition, two three-to-one solar cable splitters have been replaced.

The new relay card will arrive mid-November.

I'm looking for software scripts to SSH into the RPi and program the Triron. I found some but the Boost Duration and Equalize Duration registers cannot be read or written to correctly.

UPDATE:

The relay card has arrived.

The LiTime LiFePO4 12V/100Ah battery has replaced the 12V lead-acid battery. More on the lithium battery, later.

Mechanical and Electrical update with possible future software updates

Two basic mechanical changes will have been made:

1. replacing the temporary round aluminum tube with Superstru half channel.

2. lift the array several feet off the ground.

Two electrical changes:

1. replacing the existing, underrated card with a more suitable card.

2. The duracel batteries are being replaced by LiTime 12V/100Ah LiFePO4 batteries. Communications over a couple of weeks with agents of LiTime and Epever plus extensive online searches, has resulted in a usable set of parameters for the Triron-LiTime pairing.

Future software change: The Dygraph software may replaqce the existing hml graph.

 The array framework has long needed some modifications. Some of the originlal work was haphazard and only temporary in intent. Yesterday and today and tommorrow will have been spent making the needed changes. The array is also being raised higher off the ground. There have problems with the lawn grass wilting under the lower end of the array. The new arrangement will leave the bottom of the array several feet off the ground. A five foot section of pipe was attached to the existing pipe using a junction/coupler. The most needed change is the replacement of temporary round aluminum tubing as a base for one of the panels. Superstrut half channel is the replacement and matches the rest of the framework.

The wench used to raise the array onto the pipe, originally, is being used to lift the array several more feet. The wench had to liberally oiled to overcome friction. The pipe clamps on the pipe were also oiled to make the lift easier. One additional benefit of the lift is the full west facing tilt available to catch late afternoon sun.

A new relay card is on the way. The recent card worked when underated. That was good luck. The replacement did not last so long.  The reason turned out to be the ratings of the relays used. The ads stated 12V, 24V which I assumed was the working load voltage. I was wrong. The relays are marked 14V, which is why they failed with a 24V/13A load.

Programming the RPi

 Each time I've had to start programming a Pi from scratch, there've been two ordeals to overcome:

1.) Compile the driver for the charge controller and

2.) Program the Pi for WiFi Access Point operation.

Variables:

1.) The model: 3b+, 4b

2.) The opsys: Jessie, Buster, Stretch, Bullseye, Bookworm

3.) The program code source

All can effect the result. As time passes inevitable incompatibilities arise and have to be reconciled. Thus, I am attempting a period setup. The term 'period' is borrowed from the world of entertainment media, e.g., period clothes, period furniture, etc.. I attempt to match the software with the operating system that I have been successful with. The model may have less impact. The installation software, Raspberry Pi Imager, automates some of the work.

Currently, I have the system up and running except for a csv to image script. I'm looking at alternatives like gnuplot and matlibplot.

I expect a motor control module in a few days. The unit seems perfect, except for the use of relays in place of MOSFETs.

RPi and obsolescence

 Attempts to get my system up and running have failed, so far.

The software has changed with each update, from Jessie to Buster to Stretch to Bullseye to Bookworm. The Exar driver has always been difficult to install, but now the driver will not compile.

I have two options. Wait till someone solves the problem or revert to an earlier version of the opsys, namely, Stretch or Buster.

Inverter/generator disaster.

 A couple of months ago, I was using the generator alone, with some electrical tools and disconnected from the house. The Predator 2000 was running as usual when I heard a loud slapping-banging noise.

I ran as fast as I could to turn the thing off. By the time I reached the generator, the noise had gotten louder and a thrashing noise had started. I turned off the machine and the engine chugged to a slow stop.

The coiled spring-loaded strap in the recoil starter assembly had come loose and damaged the motor and surrounding space. I ordered a new rope starter assembly and a new ignitor. I installed the parts and decided to adjust the valves while I was at it.

The unit did not start. The compression has improved but there's no spark getting to the cylinder as far as I can tell. There is one other part that may solve the problem: the ignition coil.

Now, Harbor Freight calls an 'ignitor', what most would call the 'ignition coil'. And calls an 'ignition coil' what most might call a 'pick-up coil' or 'sensor coil' mounted deep inside the generator. Their 'ignitor' is mounted outside and on the motor housing and plugs onto the spark plug electrode, as in a car.

My guess is the machine will run if I replace that part they refewr to as the 'ingnition coil.'

I've been in no hurry since, as quiet as the generator is, I enjoy the silence and zero gasoliine costs I save by not running the thing daily. But, I will evetually need to get the unit running for Winter.

Lightning strike!

 Several months ago, a lightning storm occurred. During that storm, a bolt struck my solar array framework. The array was not adequately grounded. A steel cable connects the array to the house, outside, and is used to support the electrical wiring that runs from the array to the control center inside the house.

Subsequently, when the bolt struck the frame, the cable acted as an electrical conductor. The lightning jumped from the outside metal wall fastener to the system iniside. There was a flash of light from the storm's bolt and a smaller flash inside the house along with a loud pop sound.

The lightning blew out the charge controller and damaged the inverter. The batteries, two, did not last long after that.

The charge controller was replaced by a better version from the same company, used but cheaper. 

The inverter was a stroke of luck purchase on eBay. The unit has an automatic sitch inside that detects an existing line voltage. This makes transistioning from generator to invereter/batteries easy and automatic. Prior to that, I had to manually connect and disconnect power cables.

I added a bolt and washers to connect the cable to the array frame. This creates a direct path to ground for lighning.