PSFM Update

Robert's web site Posted on by

I thought it might be time for an update on what’s going on with the PSFM Frequency Monitor. In the past few weeks I tweaked the trigger settings so it’s less sensitive. This should help eliminate many of the small fluctuations that show up in the trigger table. The PSFM software has two event detection triggers – the “delta” trigger and the rate of change trigger. The new settings are shown below.

Delta trigger settings as of early May 2020
Rate of Change trigger settings as of early May 2020

As you can see each trigger has a built in version of the other trigger to help validate that the trigger is valid.

I’ve also been working on the uploader program. Up until a few days ago I had a simple program that converted the trigger csv file to html for the web site. This depended on the Windows Task Scheduler. At 58 minutes past the hour one task ran to generate the html file and then at 59 minutes past the hour another scheduled task would run to upload the new html file to the web site. I have now developed a new program that watches the trigger file and when it sees a change it will generate a html file and upload it to the web site within a few seconds of the event occurring.

The new (May 2020) PSFM Uploader

My next project will be automatic uploads of the frequency plots and their associated csv data. There’s no timeline for when that’ll happen. It could be next week or next year. We’ll see.

2020-05-27 Frequency Event

Robert's web site Posted on by

There were several very unusual frequency events on the Eastern Interconnection this afternoon. They all happened within 15 minutes. My PSFM triggered several times. I graphed the data in DPLOT.

Huge frequency drop!

On the graph above you can see a huge drop in frequency beginning around 18:47 UTC (2:47 PM EDT). A large drop such as this is usually caused by a generator tripping off-line suddenly, however, a drop of 83 mHz (0.083 Hz) is pretty rare on the Eastern Interconnection. To get an estimation of the magnitude there is a value called beta with units of MW/mHz. Normally beta is around 28 or so mHz. On lightly loaded days it can be lower. But if we assume it’s around 25 today then that means a total loss of around 2000MW of generation.

Following the generator trip event the frequency got to around 59.927 Hz, which is to be expected for such a large generation loss. However in the following minutes the frequency shot up to 60.067 Hz, which is very rare. The system frequency rarely goes above 60.04, so this is an unusual occurrence. It’s probably due to the light load conditions today, but I don’t know for sure.

Because there were three rare events within 15 minutes today I thought it was worth sharing.

IRIG and Arduino

Robert's web site Posted on by

I just added a page that explains the IRIG-B decoder project I am working on.  It involves an Arduino and some Windows programming.  Pretty basic stuff, but it has proven to be extremely useful.  This is still a work in progress.  The page is here:

https://www.rmonet.com//index.php/irig-b-monitoring-with-an-arduino/

IRIG connections to Arduino board

Ice Box Rolls

Robert's web site Posted on by

This is the recipe my grandmother used to make dinner rolls.  It’s a bit of work, but definitely worth the effort.  This makes about 3 dozen rolls.

Ice Box Rolls

Ingredients

  • 7 Cups flour
  • 2 packages yeast
  • 1 egg – beaten
  • 2 cups very warm water
  • 1/3 cup melted crisco or butter
  • 1 teaspoon salt
  • 2/3 cup sugar

Directions

  1. Mix warm water, yeast, sugar and salt. Add egg and melted butter (or crisco) and mix well.
  2. Add the mixture to the flour and knead until dough is made.  Add water or flour as needed to make the dough not too dry or sticky.
  3. Let rise in a warm place (around 80 degrees) until it doubles.  Usually about an hour to hour and a half.
  4. Punch down and make rolls a bit smaller than ping pong balls and place into greased pan.
    Let rise again for about an hour.
  5. Bake at 350 until brown, about 15 – 20 minutes.

Robert’s Chili recipe

Robert's web site Posted on by

I always wanted to make chili from scratch and several years ago I just started throwing things into a crock pot to see what would come out.  The first few batches were pretty bland.  It was edible, but not good chili at all.  So I looked for ways to spice it up, and after many batches I’ve pretty much settled on what I think is a pretty good recipe.  If you cook much at all, then you know that in this case the amounts are not terribly critical, so you can adjust as necessary to your own tastes.

I use a 3.5 quart crock pot to cook the chili.

Robert’s Chili recipe

1 lb ground beef
1 lb Jimmy Dean Sage sausage
1/2 bag red kidney beans (16 oz bag, so I guess that’s 8 oz of beans)
1 can (14-16 oz) diced tomatoes
1 6 oz can tomato paste
1 24 oz jar Pace Salsa or Picante sauce
1 6 oz can pineapple or apple juice
1 tsp Chili powder
1/4 – 1/2 tsp mustard powder
1/2 tsp cream of tartar
1 or 2 tsp Texas Pete
1 tbsp Heinz 57
1 or 2 tbsp Ketchup
1 tsp yellow mustard
1 or 2 tbsp brown sugar
1 or 2 tbsp molasses
1 tsp Worcestershire sauce

  1. Brown the beef and sausage in a pan, drain, and put it in the crock pot.
  2. Add diced tomatoes, Picante sauce, tomato paste, and juice.
  3. Add the beans dry, they will cook in the chili.  (this makes better chili beans than pre-cooking them)
  4. Add all the seasonings.
  5. Using the picante jar, add about 1 to 1 1/2 jars of water.  This is for the beans to cook.  I usually go by looks here, but it’s always a bit more than one jar.  Just enough water to make it soupy, the beans will soak it up.
  6. Stir the whole thing well.
  7. Set the crock pot on high and cook for 6 – 8 hours, stirring occasionally.  It usually takes about 6 hours for the beans to fully cook.
  8. After it cooks on high, I set the crock pot to low and let it cook for several more hours.  After an additional 2-4 hours the chili reaches a really good consistency.

 

Photo file naming

Robert's web site Posted on by

I imagine everybody thinks about this – I’ve got thousands of pictures on my computer, and they are taken with a variety of devices.  Each device has its own way to name files and most of the time the file names are relatively meaningless.

If there was a way to name all the files in a consistent fashion, then it would be easier to search and find the files you are looking for.

My proposal is this:

YYYYMMDD,HHMMSS,{Camera model},{original filename},{optional description}.[ext]

This makes for a longer name, but the files are sorted correctly, regardless of the device that took the picture.  This naming convention is very similar to the IEEE C37.232 Comnames convention.

So the obvious question now is how to accomplish this.  I use Total Commander with the EXIF plugin.  There’s also a program called Stamp that will accomplish this.  Stamp hasn’t been updated in years, but it still works.

Update:  Here are a couple more programs that will rename the photo files for you:

  • Advanced Renamer – The best and easiest to use of the bunch that I have tried.
  • Bulk Rename Utility – A comprehensive rename utility.  It will rename anything.
  • NameEXIF – A rename utility specifically for digital camera files.

First post

Robert's web site Posted on by

This is the latest installation of WordPress.  Not sure what I’ll put here, maybe a link to the PSFM project.  This year’s Super Bowl generated about the same number of frequency triggers on the Eastern Interconnection as usual.  I’ll put some plots up here later.

PSFM

The current hardware design