Kang's Thoughts

First Hydro Bill with Solar

I cannot believe that I am actually excited to receive a bill! Our first bill from Alectra Utilities covering the consumption period since our net meter was installed on May 4th has arrived.

Our July Alectra Utilities Bill Extract (click to enlarge)

This bill covered almost two and half months. As you can see from the above extract, we have given back a total of 2,763 kWh and used 1,621 kWh yielding a net of 1,142 kWh, which Alectra credited back to us at $0.115 / kWh.

The surprise is that this credit can be used to offset both the delivery and other added surcharges. It looks like we still benefit from any other regulatory and provincial rebates and they are incremental to our credit.

The net result of all of this is that not only did we not have to pay any electricity from May 4th until July 14th, a period which definitely covered some heavy air conditioner usage, we also generated $61.80 worth of credit on our account.

So, how much did we save? I looked back at our bills for the same period from last year (2021), and found:

From	To	Amount
2021-04-15	2021-05-13	$132.18
2021-05-13	2021-06-11	$196.41
2021-06-11	2021-07-14	$240.17

Amounts From Last Year’s Bills (from 2021)

Although we do not have perfect alignment for comparison since I cannot find a bill in 2021 that matches the May 04th starting period, I think it is clear that we have saved approximately $450.

Of course it is not only about the money, but this bill has confirmed that our solar system has the ability to not only to stay away from grid electricity (after netting everything out) but also to give some back even during the peak summer months. This is also a confirmation that our Toyota Prius Prime (a hybrid plugin vehicle) is now also indirectly funded by solar. We were not certain about this when we first design the system, so I am ecstatic to have this final confirmation!

Live Monitoring of Electricity Consumption in Ontario

Today I came across Gridwatch. This is an interesting web app that provides a live monitoring of Ontario’s current electrical production and consumption.

As the screen capture shows, we have a heat warning today. This situation is currently being shared by UK and Europe, where they are experiencing temperature in excess of 40ºC, clearly much worse than we have it here.

The hype of these high temperatures in the media caused me to wonder what is our current electrical consumption?

Which lead me to do a quick search on Google, and resulted in me finding the Gridwatch web app, and its home page:

Gridwatch Home Screen (click to enlarge)

Aside from how much electricity we are using, the most interesting part is the CO2e number which is currently at 81g/kWh. Since my solar installation, I have generated 6,450 kWh of electricity through the use of my solar panels from April 12 to today. Do some simple math, this means I have prevented 522 kg (~1/2 tonne) of CO2 emissions.

In a previous post, I mentioned that an average person in Canada uses around 14.2 tons of CO2 per year. From this perspective, we are just placing a dent on our overall CO2 footprint. However in this Reuters article, it claims that we have to remove 1 billion tonnes of CO2. If we do some rough math, the average per capita onus is (1 billion tonnes / 7 billion people) = 143 kg. Given our household is 4 people, we have an obligation to remove 572 kg of CO2 to meet the goal of 1 billion tonnes of CO2 worldwide removal.

I say we are pretty close! Okay, we can obviously do more and we will.

AirPods Pro Replacement

I purchased my AirPods Pro as soon as they were launched back in 2019. During the pandemic, I had one earbud replaced due to crackling noises when I moved my head. This replacement was performed under warranty, so Apple provided great service.

About two weeks ago the same thing happened with the other original earbud. I called it in, and was advised to book a Genius Bar appointment, which I did yesterday.

Today I went to the Apple Store, waited for 15 minutes, and they replaced both my earbuds no questions asked. Apparently to Apple’s credit, they found out that not one but both earbuds were defective.

The breakdown of the earbud was fortuitous timing. About a month ago, my wife informed me of this AirPods Pro Service Program. I did not think much of it because both were working. Now I am glad such a program existed.

When I first started to use the brand new AirPods Pro, I found out how bad my previous pair really were. The noise cancellation on these new earbuds worked twice as good as my defective pair.

Once again Apple Services can be dependent upon. I am glad to be sporting brand new AirPods Pro earbuds as a result of their service quality. These new earbuds will have a 90 days warranty.

What more can I say? I will continue to purchase Apple products not only because of their quality but also because of their prompt service and retail presence, a competitive advantage or feature that is often overlooked.

Rogers All Day Outage

Today we experienced the longest residential Internet outage that I have personally experienced since I have been a Rogers customer for more than 20 years.

My firewall records show:

As per the records above, one can observe that initial instability started at around 2:15am and then at around 4:51am it failed for a very long time. There were attempts at around noon and in the evening to resurrect the system. These attempts proved to be unsuccessful. In each of these attempts, there were issues with domain name lookups, and possible routing problems.

Our Koodo LTE on our mobile devices were able to make phone calls, but data rates were very slow. Usability was intermittent at best. To get the latest news, we had to resort back to our emergency radio.

We found out that this radio’s dynamo generator no longer works, and we should probably replace it soon. We tried solar but finally settled for batteries to operate the radio.

It turns out the whole country is impacted with specific incidences of:

Point of Sales (POS) terminal issues with using and clearing debit cards;
Work at home businesses can no longer function;
Issues with ArriveCan applications for those entering the country;
Possible 911 reception issues in certain areas;
This Toronto Star article has more.

I am writing this post at around 10:15pm. At this point the Internet is backup now longer than any of the previous Rogers attempts in fixing the outage. I hope we are up for good.

My speculation is that Rogers’ provisioning database was hacked and got corrupted causing the disablement of Internet access by all of their customers. This is my GUESS. I have no proof. We will see what really happened if Rogers choose to divulge such information in the near future.

There has been precious little information at this point on what caused the outage. My curiosity is piqued to find out what happened, and how Rogers can prevent this in the future.

Another NAS Storage Upgrade

Our home Network Attached Storage (NAS) media server is going below 4 Terabytes of free space. The Seagate IronWolf 12TB hard drives were on sale with Amazon offering them below $300. I figure that I swap out two old 6TB drives with these new 12TB drives resulting in a net increase of a further 6TB of storage.

The last time this was done was around two years ago when I replaced 4TB and 6TB hard drives with 10TB hard drives.

So far the mdadm and LVM storage architecture has proven to be very flexible. I am able to mix drives of different sizes and able to grow our media storage volume over time.

Previously I had to make two swaps, each swap for each drive in the array. Effectively I am changing two 6TB drives for two 12TB drives because they are in a Raid 1 array. I cannot swap both at the same time, because I have to incrementally sync the data from the old drives to the new ones.

This has always been inconvenient because it means opening the physical server twice. However, this time I used my USB 3.0 HDD dock. I inserted one of two 12TB new drives into the dock, and then I temporarily created a three disks Raid 1 array. Once the sync is completed, which took 10+ hours, I remove one 6TB drive from the array configuration and I then physically replace both 6TB drives with both 12TB new drives in the server chassis, and place one old 6TB drive into the dock. The 6TB drive in the dock is the one that is still in the array configuration. I then add the second 12TB drive that is already in the server chassis to the three disk array. Once again, a sync is required to accommodate the second 12TB drive. This also took 10+ hours. Once the second sync is completed, I can finally remove the second 6TB drive in the dock from the array and have the array returned back to a two disk Raid 1 array.

The above description is probably quite confusing, but this technique allowed me to just have a single down time for the server instead of two when swapping hard drives in the server chassis.

There will be an additional downtime when I grow or resize the LVM volume and file system.

After this upgrade I should have the following Raid 1 (fully mirrored) arrays:

An array with 2 x 8TB
An array with 2 x 10TB
An array with 2 x 10TB
An array with 2 x 12TB

The above four arrays are combined into a logical volume using LVM that results in a total volume size of 40TB (fully mirrored) or a little over 36TiB of usable space (increasing from the old 31TiB).

% df -h
Filesystem                        Size  Used Avail Use% Mounted on
udev                              7.7G     0  7.7G   0% /dev
tmpfs                             1.6G  3.4M  1.6G   1% /run
/dev/sdj1                         454G   64G  367G  15% /
tmpfs                             7.7G   37M  7.7G   1% /dev/shm
tmpfs                             5.0M  4.0K  5.0M   1% /run/lock
tmpfs                             7.7G     0  7.7G   0% /sys/fs/cgroup
/dev/mapper/airvideovg2-airvideo   37T   26T  9.1T  74% /mnt/airvideo
tmpfs                             1.6G     0  1.6G   0% /run/user/997
tmpfs                             1.6G     0  1.6G   0% /run/user/1000
/dev/sda1                         5.5T  548G  4.7T  11% /mnt/6tb

As you can see from above, the /mnt/airvideo now has 9.1TiB free!

The NAS motherboard and CPU is now over three years old. I may give it a couple of more years before considering another hardware upgrade.

Thoughts on Inflation

Like others, I have been reading and experiencing about inflation. It seems like everything but GPU graphics cards are going up. Everything from prices at the pump, grocery shopping, and eating out are all going up. In short, these are the commodities that are needed to sustain life.

Normal people do not stock up on extra food and gasoline, not in huge supplies any ways. Most of us typically buy and use them as we need them. Therefore, price increases in these categories of goods are most likely caused by supply constraints and not by excess demand.

When central banks around the world sees high inflation rates, their immediate monetary policy change is to increase interest rates. This increases the cost of borrowing and impacts on cash flow. For example, you tend to use your credit card less and therefore you end up buying less. There will also be less investments and ventures. People will tend to save more.

If the current inflation is a result of supply constraints of key commodities necessary to sustain our current way of life, then a forced demand reduction means a pare back to our quality of life. The increased interest rate will have a dampening effect on items other than food and energy, at least not immediately.

It will eventually reduce demand on the essentials when people are indirectly impacted through layoffs due to lower business activities in other areas. This means people will have to buy less food and consume less energy. A lower quality of life.

Okay. Why raise interest rates then if it is going to do more damage than good? Unfortunately interest rate is the only main lever that central banks have. Interest rates will have to increase to the point where demand of the essentials are align back to their corresponding supply levels. This means to fight this type of inflation, we will necessarily see a recession in the economy and our livelihood.

The alternative is to allow for inflation rate to go uncheck and await for supply to naturally fix itself. I do not think this will work as high inflation effectively act as a brake to economic growth for everyone. The central banks now face a difficult decision. Let inflation create hardship for all or raise interest rates to cause pain for some (less than all), and reduce demand on the essentials. I fear they will have no choice but to choose the latter. As a result of this choice there will be a further widening of our social class gap. The rich will be more prepared to weather through the increase in borrowing costs. They may even seek opportunities to increase their social status. The poor will have to deal with making ends meet with less necessities.

In short, things will get much worst before they get better. When inflation is tamed, and I think this will be some time considering the current Ukrainian war, sanctions, and logistical issues, the poor will be less well off than the rich in the next cycle.

Let me know your thoughts on my thoughts.

Frequency Shift Power Control

If you follow my posts in the Solar category, I mentioned Frequency Shift Power Control. What is this? I will try to answer it in this post and explain why this is important in the context of full house backup and grid connectivity is lost during a power outage.

There are two types of inverters in my house. The Schneider inverters are the hybrid inverters that charge and discharge the battery and manage the battery power in conjunction with grid power. During normal operations when the grid is connected, the Schneider inverter can consume or send electricity to the grid.

The other type of inverter is the SolarEdge inverter, which is a solar inverter. Its job is to convert DC power from the solar panels into AC power which your house needs.

Having both hybrid and solar inverters connected on a common AC bus is known as AC coupling.

First, let me explain how it works when the grid is happy and we have connection. The solar power from the solar inverters will do the following in priority:

Power the house; and send excess to
Charge the batteries; and send excess to
Send back to the grid

If there is not enough power from solar, then power is supplemented by the batteries or the grid depending how you configure the Schneider inverters. I will not get into how Schneider inverters are configured here.

Now when the grid is out (power outage) and the batteries are fully charged, the solar power has no where to go. To regulate solar power production, the hybrid inverters will communicate to the solar inverters via Frequency Shift Power Control. The hybrid inverters will intentionally shift the output power frequency from its normal 60Hz to something like 61Hz. This is done slowly. The solar inverters detects this anomaly and either throttles its power output or stop producing power completely. Once power regulation is completed, the hybrid inverters will shift back to 60Hz.

The solar inverters will come back online after an arbitrary down time. For SolarEdge, I have observed this to be around 5 to 10 minutes. The idea of having this down time is so that batteries can be discharged while powering the house and create more head room for new solar energy. Unfortunately the SolarEdge down time is not long enough, at least not for my house and my battery storage system. Once solar power is restored, you can end up in a cycle if the batteries are still topped up. The hybrid inverters will again shift its frequency to force the solar inverters to reduce or stop producing power. The cycle repeats until the grid comes back on.

Instead of a repeating the cycle every 5 to 10 minutes, it is better to manually turn off the solar inverters and wait for the batteries to draw down until you have enough head room for the batteries to absorb the solar energy again. For example, you may want to draw down the batteries from 100% to 80%, and then turn back on the solar inverters. In this scenario, excessive solar power can then be used to charge the batteries.

Another matter to pay attention to is UPS (Uninterruptible Power Supply) units. Certain models are very finicky to frequency shift and will actually shutdown instead of continuing to supply power. This is very frustrating as it cuts power to the very items that you are trying to protect from outages.

I had UPS units spread through out the house prior to our solar installation. Now I am simply going to replace these with surge protection power bars.

First Hand Supply Chain Impact

Our last power outage on May 21st, caused one of my two SolarEdge inverters to stop generating power. This effectively caused 50% of solar panels to be ineffective. Considering that the system was recently commissioned fully on May 4th, it was a bit frustrating to go through this. At least the other SolarEdge inverter is still generating power from the remaining 50% of my panels.

Upon a remote assessment of the inverter by SolarEdge, they determined on the 25th that the inverter requires replacement. As of this writing, I have not been given a formal response by SolarEdge that explains why the inverter cease generating power. I assume it is a quality issue on their end, and something from the grid outage during the storm fried something in the inverter. For an inverter that was less than 2 months old, this was slightly concerning.

It took SolarEdge from May 25th to June 10th to ship a replacement. Today is June 15th, and the UPS tracking number still shows “Label Created”.

Apparently recent supply chain issues has caused much shortages resulting in the delay.

Kudos to New Dawn who was able to source another inverter and they promptly installed the inverter today. All is good now.

In summary, we had 50% of our solar panels not producing power for 24 days (from May 21st to June 15). SolarEdge to this day still has not met its warranty commitments because its replacement unit is still in “transit”. If it was not for New Dawn being creative, I would still be out 50% of my solar capacity.

If you are using SolarEdge for off grid applications, perhaps it is wise to store an additional unit if you think solar generation is critical to your livelihood.

Safari Cannot Load Certain Pages (iOS 15.5 & macOS 12.4)

I upgraded my iPhone and iPad to iOS 15.5 and my Mac’s to 12.4. I noticed in the past couple of days that certain web pages would not open. I have discovered that the following domains were effectively blocked:

assets.adobedtm.com
fonts.gstatic.com
platform.twitter.com

This causes web sites containing references to the above domains to load very slowly or not at all.

I first took the attitude that this was a great feature and Apple is helping me to remove more SPAM or prevent further identity tracking. However, this morning when I try to access canada.ca, which is our government site also has the same symptom. Clearly Apple’s technique of applying privacy tracking has crossed the Rubicon sort of speak and is now generating enough false positives that is hindering day to day use.

I do have the option of using Chrome on my Mac’s but this is a huge CPU guzzler. I prefer to stay with Safari.

I came across the following thread on the MacRumors forum. The discussion pointed to the Limit IP Address Tracking feature to be the culprit. After turning off the feature on my iOS and macOS devices, I am now able to load the problematic web sites without any issues.

This may be a temporary problem with Apple with their cloud relays, so perhaps it will be automagically fixed in the future.

Below are screen shots from my iPhone and my MacBook that shows where you can turn off this feature.

If you are experiencing similar difficulties, I hope you will find the above useful and you can continue to visit certain web sites without issues.

Update: It looks like you also have to check to make sure the Safari Privacy Settings as well:

Reminiscing PC Tech

We are in the process of shredding our paper records from twenty years ago, stuff like tax returns, invoices, insurance papers that are no longer relevant or needed.

During this paper demolition process, I came across this invoice from 24 years ago!

A Windows 95 PC costing over $4,500 in 1998 money. Wow. I did not realize I had such an appetite to spend that much money back then.

Notice that we were still using dial in modems back then. No high-speed internet yet. Most motherboards did not come with sound and video either. I also have to laugh at the hard drive specification. At 8.4 GB, that would be smaller than most USB flash drive you can now get at Staples.

Fascinating walking down memory lane!