Sunday, May 31, 2020

A Ham Radio Series 1 - Sporadic E Propagation

The reactions to my “Got a Question for You” post surprised me.  Mostly by the quantity; I don’t have a way to check but it’s in the top couple of comment generating posts I’ve ever put up.  Some of specific questions were a bit of a surprise, too.

One of the subjects that came up has a time sensitive aspect to it, so I’m going to take a look at  propagation; in particular E skip, or sporadic E.  It’s going on right now, as I write and we’re at or close to the peak time of the year for sporadic E; at least here in the northern hemisphere. 

As usual, let me start at the beginning for those who aren’t familiar with it.  Both references are to propagation in the E layer of the ionosphere.  The ionosphere is called that because the air density is so low that individual atoms are far from each other, far enough that energy from the sun, mostly UV radiation, can cause electrons to be stripped from some atoms, giving rise to both negatively charged free electrons and positively charged atoms that can stay separate for long times.  The density of the air is so thin that it takes a long time for collisions between these ions to neutralize them.

The ionosphere is characterized by layers, conventionally called the D, E and F layers – in order of height.  If solar energy is high enough, the F layer can split into two layers called (imaginatively enough) F1 and F2.

Because the ionization is driven by solar energy, the general rule is these layers appear in the daytime, and the highest levels of ionization are in the area where the sun is most intense; so midday and directly under the sun.  While the D layer getting denser during the day shuts down lower frequencies that tend to be open all night (say 160 through 30m), the higher frequency bands’ signals (20 to 10m) travel through the D layer to the F, so they open during the day and are shut down at night.  This isn’t like a switch, it’s a more gradual transition.  For example, it seems that 30 is intermediate between 20 and 40; most days it seems to open earlier than 40m in the afternoon, and close later in the morning than 40m.  Back around the Spring Equinox, I’d hear Australia on 30m – close to my sunset and their sunrise.

It’s probably generally known that solar activity - sunspots and higher Solar Flux Index – increases the density of the ionosphere, making the F layer propagation last longer into the night, sometimes completely overnight.  Bands that are only useful for local communications during sunspot minima can be open to somewhere in the world 24 hours a day at the peaks of the solar cycle.

E layer propagation is left out of that overview because it’s harder to fit in.  Sporadic E has been known for a long time, but that word “sporadic” in the name is an indication of the toughness of predicting when it will occur.  We know there’s no correlation between solar activity and Sporadic E; in fact, the last few years have made it seem that it could be inversely proportional to solar activity (my observation – I haven’t seen anyone else say this).  We know that the mechanism is fleeting clouds of sufficient density in the E layer, and that it tends to occur more at certain times of years than others (now in the northern hemisphere).  We know that it tends to form later in the day and linger into the evening; that is, you'd be more likely to find it at 3PM than 3AM (local time).  Finally, we tend to observe it on higher HF and lower VHF bands; 28 and 50 MHz are peak bands.  It happens on occasion on the 144 MHz band.  Hams have long noticed that sporadic E clouds seem to form above the tops of thunderstorms - long before mechanisms that could extend the charge into the ionosphere, such as red sprites or dark lightning were discovered.

3PM?  This is a plot of reported sporadic E contacts at 10:14 AM EDT, Saturday, on 6m.  Each arc is a reported contact between the two stations at the ends of the arc.  Most of those arcs are in places where it's 9:14 AM.  It’s impossible to read all the call signs in this screen capture; there’s simply too many.

The plot is from a website I visit regularly called DX Maps.  This link should bring up this display for 6m over the US as of the moment you go there - I think those reports stay there on the order of a half hour.

The area in the east central US is blanketed by reported contacts.  There are many arcs starting in the Caribbean, (Puerto Rico, Guadeloupe, St. Lucia, Trinidad) going into the states.  As I write this, 11 hours later, the map looked much the same.  The density of the red traces (reported as Sporadic E, although always hard to know) is down, but the distribution is similar.  A couple of days ago, the concentration of red traces was over the western states: Utah, Wyoming, Montana, Eastern Oregon and Washington and virtually nothing on the east side of the country.  In other words, the opposite of this.  Another day, the east coast was full of paths reported into Europe, still another was full of paths reported from the west coast into Alaska and then Japan.  26 hours later, that previous plot looked like this:  

You can see the propagation to Europe marked by dark red arcs is dense into the NW US, with some into Mexico.

They call it Sporadic E because it’s not dependable or particularly predictable; you kind of know something is likely to happen, but don’t know what that “something” is going to be. It varies all day every time.  We know late spring into early summer is better than mid summer or mid winter.  I’m not sure that doesn’t vary with where you are.  I’ve heard from a guy between my location and the Space Center that around the Winter Solstice in December, he has worked 6m into New Zealand.  Note that’s around the equivalent of our first day of summer in NZ, in terms of heat and constancy of sunshine.

Because of that unpredictability, it’s fair to ask what practical use Sporadic E is.  I know of no systems that rely on it.  For hams, it’s mostly a way to rack up novel contacts that we don’t or can’t get any other way.  The band I’ve been on, 6 meters, is just below the bottom of the old VHF TV Channel 2.  A few days ago, I had a contact with a guy in Spain on 6m.  That’s similar to (while different in important ways from) watching Channel 2 TV from Spain and I get a kick out of the novelty.   

I find the study of propagation one of the most interesting aspects of radio. 

Saturday, May 30, 2020

Not Much More That I Can Say

This screen capture kinda sums it up for me.  Except that the actual mission for the crew had barely started.  The screen capture is from moments after the radio callout that the second stage had just shut down and the Crew Dragon capsule was in the desired orbit.  Then the video changed scenes to include the first stage booster had been successfully recovered.

On the left, you see the booster, now a used rocket that's experienced one flight, standing in the circle marked on the deck of OCISLY.  On right, you see the view we've never seen before: Commander Doug Hurley (l) and pilot Bob Behnken (r) at their control screens.  Instead of rows of big toggle switches, almost everything is on a touchscreen.  

The entire flight went like it had been done a hundred times before.  I think this was the 85th Falcon 9 launch. 

Friday, May 29, 2020

SpaceX Loses Starship SN4

“Loses” in the sense that it blew up so badly it's tough to figure out where all the pieces went.

SN4 has survived more than any other Starship prototype; it has been on the pad since the end of April, had its first static firing on May 5th, and then another couple of static firings. After that, they replaced the engine with a newer Raptor, and static fired the new engine.  Yeah, they had a bit of a fire and had to wait a day to inspect and clear the area, but the damage was minor.  Yesterday, they added 55,000 lbs of ballast to it and static fired the engine again; then this happened today after yet another static firing.  A commenter to that Ars article said this was the fifth static firing of the second engine. 
On Friday, SpaceX prepared its latest iteration of the Starship prototype vehicle, known as Serial No. 4, or SN4, for a static fire test in Texas. The Raptor engine appeared to fire nominally for a couple of seconds at 1:47pm local time and then shut down as planned.

However, about one minute after engine shutdown there was some kind of uncontrolled gaseous leak, and one minute later the vehicle exploded almost instantaneously—a truly rapid Rapid Unscheduled Disassembly.
RUD - for Rapid Unscheduled Disassembly - is a term that Elon Musk and his engineers seem to use regularly.  Personally, I hadn't heard it before, but I don't know they made it up.

I'm not going to speculate on what happened, but the vehicle is lost and the test stand appears badly damaged.  There will be a few days work and SN5, which has been ready for weeks, will be transported to the test area. 

Too bad.  They got their FAA approval to conduct suborbital test flights yesterday after the successful static firing.  It looked like SpaceX would lay low, let Crew Dragon get to the ISS and then try a hop test of SN4.  Maybe even Monday, June 1.  That doesn't seem possible today.

The remains of the test stand and much of the hulk of SN4.  A small flame is visible - orange, just left of center, a quarter of the way up from the bottom of the scene.

Thursday, May 28, 2020

A Little More on RF Power Safety and Ham Radio

The previous post on this topic touched off a lot of discussion and I can see it's something that many people are concerned about.

I thought I'd look at a couple of examples and go through how its calculated. The ARRL has a sample worksheet to fill out that can be seen and downloaded here (pdf).  They also have a book they give away to help with this (also pdf format) – with details in chapter 5.  A word of warning: that book is copyright 1998 and based on the original FCC documents.  All of the distance limits are higher than the referenced “FCC Guidelines” from the previous post and now in effect.

Filling in every blank on the form seems a bit like overkill, but I’m sure there are questions in there that the majority of people just haven't dealt with so I thought I’d try to help with those. 

I’m going to go through an example based on a modest station to show how you’d calculate the things they ask for.  For this demonstration, I’m going to assume a 100 W transceiver and a kind of typical low power/portable stations, with a wire antenna in a nearby tree.  Let's assume they're using 30 feet of RG-58 coax to get to the feed point.

A quick look at Table 1 in the previous piece shows that nothing I can radiate matters until I get to 12m (24.9 MHz) because no matter how much power I put out, I lose some on the way to the antenna and every band from 15m (21 MHz) down has an exemption from analysis for powers less than or equal to 100 W.  That power limit for the 12m band is 75 Watts.

That 75 W is power into the antenna and although the radio puts out 100 W, there are losses in the cable, and switches or anything else in the path to the antenna.  Let’s start with loss in my coax.  There are lots of brands of RG-58 coax, so I'll choose Belden 8259 or RG-58C/U.  The table on Belden’s website doesn’t say what the loss is at 24 MHz, but it gives these points:
10 MHz  1.5 dB/100 ft
50 MHz   3.7 dB/100 ft.
For another type ("RG-" number) or brand of coax, you should try to look up the corresponding numbers for what you have.  

I’ll do a linear interpolation (assume attenuation vs frequency is a straight line).  That’s
(3.7-1.5) divided by (50-10)  or 0.055 dB/100ft per MHz.   
There’s 15 MHz from 10 to 25 (the top of 12m is 24.980 MHz – close enough), so we calculate the attenuation by:
1.5 + 15*.055 = 2.33dB /100ft.   A 30 foot run of this coax then has 0.3 * 2.33 or 0.698 dB of loss at 25 MHz.  I can use the same linear interpolation to show it has 2.58 dB/ 100 ft at 29.7 MHz (top of the 10m band). 

The problem with using the loss in dB is that the power is stated in Watts, so we need to convert that loss to a percentage of 100 W.  I’m guessing that’s probably a new thing to people.  Remember a dB is a power ratio, where
10*log (ratio)
is the value in dB and log denotes the base 10 logarithm.  We want to find the ratio that gives a loss: loss means negative dB.  It’s -0.698 dB not +0.698.
(-.698) = 10 log (ratio), so 
(-.698/10) = log (ratio)
10^(-.698/10) = ratio = 0.852 or 85.2%

That’s saying the power at the antenna is 85.2% of the transmitted power.   The worksheet asks for the percentage of power lost, so 100%-85.2% = 14.8% lost, but they only do that to get the first number we solved for, the percentage of power that gets to the antenna.

There’s a trick you can do to be finished right here.  Remember that Table 1 says my power at the antenna has to be below 75 watts.  I could just crank down my power until the power at the antenna is 75W.  That’s picking a power output from the radio that combined with the cable loss puts the power at the antenna at 75 W.
75W/.852 = 88 Watts
Which says if my output is 88 Watts, I’m always at or below that 75W limit, but I’m not going to go there, I’m going to the next step at Table 4a in the FCC report and the second table in that blog post.

Looking at the table for 24 MHz, I can see that assuming I put 100W into the antenna, I have four options for my antenna that lead to four different distances to choose from.  The first column after the band (24.99 MHz, (12m)) shows the antenna gain in dBi.

I chose an antenna that isn’t particularly good; it’s something like a random length wire, a multiband wire, G5RV or something low gain like that.  Is it 0 or 3 dBi?  Let’s be generous and give it 3 dBi (everything has gain compared to a theoretical isotropic antenna, just not much).  You’ll see two numbers: 1.7 or 3.8.  That’s the distance to keep people away from the antenna, in meters.  The difference is the smaller distance, 1.7m  is for people who are aware of the risk of being near or touching the antenna – like you and your family – while the other (uncontrolled) column is for the neighbors.   The number is 12-1/2 feet.  The smaller is 5 feet 7 inches.

That distance can be reduced because heating is different depending on the transmitter duty cycle.  The league says you can use 40% for CW, 20% for SSB with no speech processing, 40% for SSB with heavy speech processing,  50% for FM (full carrier, talk half the time, listen half the time), 50% for most digital modes – (also based on transmit 50%/receive 50%).  The bottom of Table 4a says to multiply those distances by .707 if the duty cycle is 50%, so since the others are less than 50%, we can use that as an upper limit.  Those become 1.2 m and 2.7 m, or 3.9 and 8.9 feet.

As the infomercial guys say, “but wait! There’s more!”   Since those numbers are based on 100W and we know we’re only putting 85 Watts into the antenna, we know the minimum safe distances are even less than that.

Although values can be hard to find, anything you put around the antenna is going to attenuate the signals somewhat.  The antenna distance recommendations in that FCC document are based on mounting antennas in the open.  As a rough example, HF signals will go through most walls pretty well as long as their isn't a wire grid in them.  Shortwave pocket radios with small antennas work acceptably in most houses and lots of people use attic-mounted antennas to avoid property restrictions.    

So what happens if you do a station evaluation and conclude one of those limits is exceeded?  In this case, what if that 8.9 feet just falls beyond your property line and there’s nothing you can do to guarantee that the neighbor’s kids can’t get into that your RF field.  At the bottom of the League’s worksheet for evaluating your station is this section:

Using this method, did your station exceed the FCC RF exposure limits? (Y/N)
Controlled exposure: ___________(Y/N)    Uncontrolled exposure: ____________ (Y/N)
If the station is not in compliance under all circumstances of its expected operation, attach a separate sheet describing any limitations of methods that the station operator will use to ensure compliance if people are present in areas that could be out of compliance.

In this case, you could state you limit power to keep your power below 75W at all times, in compliance with table 1, or you could say you'll only operate your station at times when the neighbor's kids are not potentially too close to your antenna, such as during the evenings or when nobody can be seen outside - if you can see outside from your operating position or with a camera.    

EDIT 052820 2235 EDT:  Left out a link for Table 4a.  Added just above the small portion clipped out of that table.  Roughly just above the middle of the post. 

Wednesday, May 27, 2020

The World Isn't Complying With Me

One of those nights where I bite off a subject and don't get as far as I wanted.  So - attempt at humor:

Everyone who cares realizes the SpaceX launch was scrubbed in the last few minutes due to the weather refusing to clear up.  They'll try again Saturday at 3:22.  Which is just as good from my standpoint; I wouldn't have seen a second of the launch.  I watched the live stream on SpaceX and turned it on when Doug and Bob were arriving at pad 39A in their Teslas - both with a license tag that read ISSBND - ISS Bound. 

There was a brief interview with Chris Cassidy, currently the ISS Commander.  He pointed out something that I somehow failed to think of.  This isn't just the first launch of astronauts from the US since 2011.  It's the first new, manned spacecraft in the US since 1981, when the first Space Shuttle Mission Launched.  Almost 40 years. 

Tuesday, May 26, 2020

The Only Hurdle Left to Demo2 is the Weather

After the years of effort and billions of dollars spread among the handful of contractors, NASA and SpaceX are at the mercy of the weather for tomorrow's 4:33 PM EDT Demo2 mission to the ISS.  To be honest, it's the best possible position to be in.  All the pre-launch reviews and dress rehearsals have been passed, the approvals have been granted and we're in the final 24 hours to launch.  The chances have improved over the last 24 hours so that the current forecast from the weather squadron at Patrick Air Force Base (pdf) is for a 40% chance of weather violating launch criteria.

There is no mention of a launch window for this mission, just the launch time.  If anything comes up to delay the countdown, the launch will be scrubbed.  The next calculated launch time will be Saturday, May 30, at 3:22 p.m. with Sunday at 3 p.m. as a backup to that.
"Everything is looking good. As of right now, we are 'go' for launch," NASA Administrator Jim Bridenstine told reporters Tuesday morning. "Our country has been through a lot. This is a unique moment where all of America can take a moment and look at our country do something stunning again."

Expected at KSC for the launch: President Trump, Vice President Pence, SpaceX CEO Elon Musk, at least half a dozen current and former astronauts, and countless other VIPs. They will be joined by a much smaller press corps and other visitors than had been expected before the outbreak of the coronavirus pandemic.
If all goes according to plan, Behnken and Hurley will make a 19-hour trek to the ISS and attempt a docking at 11:39 a.m. Thursday. A little over two hours later, onboard crew Chris Cassidy, Anatoly Ivanishin, and Ivan Vagner will help open the hatch, ending the first phase of the demo flight.
Something that might not be well known is that SpaceX engineered the suits for this mission as well as the capsule and rocket system.  They'll suit up in the Operations and Checkout (O&C) building; as has been tradition.  The departure from tradition is that instead of the Airstream RV to the launch pad, Bob and Doug will be driven in Tesla Model Xs

Doug Hurley (l) and Bob Behnken with the twin Model X Teslas that will ferry them the couple of miles to pad 39A.

While prepping this post, one of the voices in my head said, “hey, you've got pictures of the last shuttle mission; the last manned launch from the KSC.”

That's Doug Hurley, second from the left. 

Sunday, May 24, 2020

On Memorial Day 2020

Let me join the chorus of folks saying that while you're enjoying your day, be it beach, barbecue, pool or whatever, take a moment to remember or think of and thank those who gave their all in service to us. The ones who don't get to mark the holiday with us.

It's a weird Memorial Day here weather-wise.  The temperature isn't supposed to get over 80 and the National Weather Service is calling for 100% chance of rain, showers, and thunderstorms with 2 to 3 inches of rain possible.  This has been an unusually mild May, with only a day or two hitting 90.  Ordinarily by now, the last week of May, 90 is an everyday thing and the afternoon thunderstorm system has started.

With 100% chance of rain, there will be no barbecue today, and probably not much of anything productive.  Good thing I just put fresh batteries in my UPS for this computer and the house internet/video streaming hub.

I've run this picture more than any other on my blog.  I guess it resonates with me.
In a final act of loyalty, Hawkeye, the dog of slain Navy SEAL U.S. Navy Petty Officer 1st Class Jon T. Tumilson walked up to his fallen master’s casket during the funeral in Rockford, Iowa, and then laid mournfully down beside the body for the rest of the proceedings  [Note: Petty Officer Tumilson was one of the 30 killed in Afghanistan in the shoot down of Extortion 17 which the families blame squarely on the Obama administration - SiG]


Saturday, May 23, 2020

A Question for You

A question for you, dear readers. 

My article on interpreting the FCC's RF radiation limits drew several thanks.  It makes me wonder what else you're interested in reading about in the general topic of radio.  Since I made my living in high-performance, high-reliability radio design and I've been a ham for 44 years, that's naturally what I gravitate toward. 

I was thinking about a poll, but I don't even know enough to ask the right questions.

So, comments, please!


Mississippi Governor Pwned on Graduation Facebook Live

As high schools across the country are holding virtual graduation ceremonies amid the pandemic, graduating seniors are finding alternative ways to leave a lasting impression on their communities. One such way appears to be pranking the governor into congratulating "Harry Azcrac," class of 2020, in a live broadcast.

He appeared to notice the prank, pausing briefly after reading the name, but continued on without losing his composure.  Later, as the reaction across the Interwebz went into overdrive, he retaliated, tweeting “In 10 minutes we’ll be honoring more graduates on Facebook Live. Maybe even Ben—the pride of the Dover family.”

Oh, we could go for days with these, but it would just be between friends, not on a "serious" livestreaming event.

Friday, May 22, 2020

The Milestones To Wednesday's Manned Launch Keeping Going By

Today was a big day in the preparations for Wednesday's return to manned space flight.  First, today's Flight Readiness Review took place behind closed doors and was declared a success with no issues raised.  Second, the flight-ready Falcon 9 and Crew Dragon capsule underwent a booster static fire test for a few seconds on historic pad 39A.   (That's a surprisingly low quality 5:38 long video that I set to start at 2:45 and the static fire is about 3:10)  The launch vehicle was rolled to the pad yesterday.
On Thursday and Friday, senior managers from NASA, SpaceX, and the space agency's international partners held long meetings to review all of the aspects of an upcoming flight of the Falcon 9 rocket and Crew Dragon spacecraft.

These discussions must have gone well, because on Friday afternoon, NASA officials emerged with a clear message: "There are no significant issues," said NASA Associate Administrator Steve Jurczyk, who led the meetings behind closed doors at Kennedy Space Center. "In the end, it was a very clean review. We are ready to launch."
That article, on Ars Technica, reminds us that it was just over a year ago (Saturday April 20) that the test Crew Dragon capsule was destroyed on the pad when the Super Draco thrusters exploded milliseconds before they were supposed to fire.
Since that time, SpaceX has addressed not only the root cause of that problem but has also overseen a complete redesign of the vehicle's parachute system. The company closed out dozens of other significant issues to reach this point.

NASA's manager of the Commercial Crew Program, Kathy Lueders, acknowledged that the process has been a whirlwind. "Last April, I probably wasn't thinking I was going to be flying in a year, but you know what—you can never sell this NASA and SpaceX team short," she said. "They've always accomplished miracles for me. And I'm very, very proud of them."
It's a busy time until launch and there will be several more milestones to be met before then.
Saturday the crew will conduct a "dry" dress rehearsal in which Hurley and Behnken suit up for launch day, and there will be a Launch Readiness Review meeting on Monday. But the biggest concern is probably weather; there are multiple constraints for the Falcon 9 launch and emergency abort scenarios down range. Florida may see some scattered to widespread showers next week, according to medium-range weather models.

Yet NASA and SpaceX are very close. (NASA Associate Administrator) Jurczyk said Friday it "is hard to believe" we are just five days from launching this crewed mission, and we have to agree. We can't wait.

Thursday, May 21, 2020

The Numbers Don't Lie - NASA's Move to Commercial Space Has Saved Money

Ars Technica's Rocket Report this week includes an excerpt from a longer article with the same title as this blog post.
As part of its initial investment of $396 million into SpaceX, NASA got development of the Cargo Dragon, Falcon 9, and a launch site at Cape Canaveral.
A cost of 50 times more
... At the same time, NASA was developing the Ares I rocket to fly crew into low-Earth orbit. Independent estimates placed the cost of Ares I at about $20 billion. President Obama ultimately canceled the Ares I, projected to have a similar lift capacity to the modern Falcon 9 booster, because it was behind schedule and over budget. The agency, in turn, got a bargain.
The beginnings of the commercial space program goes back nearly 15 years to NASA administrator Michael Griffin, appointed by George W. Bush in 2005.  NASA placed a small bet on the nascent commercial space industry when it sought to diversify its fleet for delivering cargo to the International Space Station.  In 2005, NASA had the space shuttle to ferry supplies, of course, but recognized that the aging system was not going to fly forever.  So NASA administrator Griffin, committed $500 million in seed money for the development of new, privately built spacecraft.
Griffin may not have realized what he had unleashed. The first small “Commercial Orbital Transportation Services” contracts awarded to SpaceX and Orbital Sciences have since expanded into other areas of spaceflight while multiplying in value from hundreds of millions of dollars into billions of dollars. NASA now looks to private companies for not just cargo delivery to orbit but, with Crew Dragon, people. NASA also recently sought commercial services for sending supplies to the Moon and even landing humans there. What began as a pebble tossed into a pond has become a wave.

Critics of this commercial approach certainly remain—it has disrupted the business models of traditional aerospace powers like Boeing and Lockheed Martin, which have long profited from lucrative cost-plus contracts. Some at NASA, too, still don’t trust commercial providers, and they’re especially wary of Elon Musk, the brash founder and chief engineer of SpaceX.
Insiders are still fighting over cost-plus contracts.  In the days before the advances in commercial spaceflight, NASA would specify precisely what they wanted and monitor the contractors with intensive audits, visits and oversight.  If a vehicle ran five years late and doubled its original budget, (cough - SLS - cough) NASA was on the hook for cost overruns. This tended to incentivize programs being late and over budget, but eventually the government got what it wanted. And those big contractors absolutely loved it.  To be fair, when the government is asking some company to do something no one on Earth has ever done, cost-plus is a way of reducing the risk.  The reality is the method becomes harder to justify when a company is building the next generation of something that has been done many times before. 

As the infomercials say, "but that's not all!"  Like every other government agency I'm aware of, there's a conveyor belt of managers going back and forth between the agency and the private sector; between the buyers and the companies they buy from.  That makes it in the interest of some NASA people for the contracts to go late and over budget, too, further incentivizing cost-plus contracts. 

Griffin used the analogy of having a house built to describe his vision for commercial space.
“The contractor builds homes for a living, I’m not creating the contractor’s company. He has to have a company before I will consider allowing him to build a home for me. He builds his homes, and if I like them, I can buy a design that he offers. At different stages of completion he gets money from me if he’s building my home, but he doesn’t get all the money until he has furnished all the product.”
An important point about this analogy that may not be obvious is that if the contractor screws up, it's on his nickel.  That would make the contract for commercial space much more like the "Firm, Fixed Price" model the DOD uses.

Ars presents this table of the highlights:

ProgramCargo DragonCommercial crewConstellation
Type Private, fixed-price Private, fixed-price Government, cost-plus
NASA costs $396 million $5 billion $34.5 billion
Deliverable Cargo Dragon Crew Dragon Ares I rocket

Falcon 9 rocket Starliner spacecraft Orion

Florida launch site Booster integration (costs are estimated)

Results Cargo Dragon went on to fly Crew Dragon to fly in May, Ares I canceled, Orion to be

20 ISS deliveriesStarliner next yearused only in deep space.

Don't make the mistake I made; the entry under Commercial Crew for Starliner isn't the SpaceX Starship we frequently talk about.  Starliner is the Boeing private capsule, which (it seems to me) will be lucky to be flying in a year.  That $5 Billion is the total amount NASA has spent among the various contractors in the running.
Put another way, while SpaceX developed a cargo version of its Dragon spacecraft, the Falcon 9 rocket, and built its launch facilities at Cape Canaveral, the Constellation Program was toiling away on Orion, the Ares I rocket, and ground systems. “We were effectively doing what the Constellation Program was doing with about the same amount of money, total, that they were burning in a single month,” said Mike Horkachuck, the NASA engineer originally assigned to SpaceX for the commercial cargo program. “So that kind of puts it into perspective.”

But that’s just for cargo missions. Now compare the costs for crew transportation. All told, NASA will invest nearly $5 billion in SpaceX and Boeing to bring their Crew Dragon and Starliner systems to the launch pad. Phil McAlister, a NASA manager for commercial space, noted that NASA was on track to spend about six times more for the single Ares I-Orion system than what it ultimately paid for two distinct private spacecraft launched on private rockets.
It's hard to overlook that SpaceX has done much more with much less than the old line contractors have.  For cargo, it has flown more missions for less. As part of the crew development program, NASA paid Boeing about 50 percent more than SpaceX. Despite this, SpaceX completed Crew Dragon about a year ahead of Boeing, which, again, is unlikely to fly a crewed mission before next spring at the earliest.

Wednesday, May 20, 2020

The Test Was Successful - If You Ignore The Fire and Shutdown Afterwards

Yesterday, SpaceX successfully had a static fire test of SN4 at Boca Chica, Texas.  This was at the end of several days of delays and lots of construction being added to the Starship.  Before the end of this video, fire is seen at the bottom of the booster that goes on for quite some time.  Water systems on the test stand spray the fire and eventually put it out, but the results weren't quite clear as of earlier today

Now more than 12 hours after Starship SN4 fired up its new Raptor engine, the ~30m (~100 ft) tall, 9m (~30 ft) wide prototype is apparently trapped with one or both of its propellant tanks still partially filled with liquid (or gaseous) methane and/or oxygen. An initial road closure scheduled from noon to 6pm local quickly came and went and SpaceX and Cameron County Texas have since modified the paperwork, extending the closure a full 24 hours. In other words, SpaceX has reason to believe that Starship SN4 may continue to be unsafe (i.e. pressurized) as many as ~30 hours after it technically completed its third static fire test – extremely unusual, to say the least.

There’s only one obvious conclusion to draw. Whether it was something invisible to the public eye or damage related to the off-nominal fire that burned for some 15 minutes after Raptor shut down, SpaceX appears – to some extent – to have lost control of Starship SN4.
I've tuned into the live feed a few times today, and there's no visible work being done.  The road alongside the test stand is closed (it was opened yesterday a couple of hours after the test) and the info panel says the road is scheduled to stay closed until 8AM (CDT) tomorrow.  There have been no workers to the pad since the incident.
Given that safety is almost certainly the priority, chances are that some combination of fairly mild hardware failure and telemetry/control loss has left SpaceX with just enough uncertainty that it can’t risk sending technicians to the launch site to inspect the damage and reestablish control. As a result, the only option left is to quite literally sit and wait until it’s once again safe to approach the rocket. Thankfully, at this point, the risk of the mystery problem actually destroying Starship SN4 is very low. If, as it appears, only its methane tank is affected, leaving some unknown quantity of latent liquid methane trapped inside, it’s possible that waiting will actually solve the problem and safe the rocket.

The fact that Starship hasn’t exploded yet strongly implies either that the amount of propellant trapped is minuscule or that the vast majority of SN4’s propellant management systems (including vents) remain functional. Assuming that’s the case, any remaining cryogenic propellant will eventually boil into gas, increasing the pressure inside Starship’s tanks, while those tanks will continue to vent to prevent an explosion or rupture. Eventually, Starship SN4 will be empty once again and SpaceX will be able to approach the rocket to regain control and begin inspections and repairs.
At some point, they're going to have to do something with SN4.  If they can't get to the pad at all for days or a week that throws a big monkey wrench into their operations.  It sounds like SN5 is either ready now, or not far from it, so it's possible that they'll swap the prototypes, but not if they can't work around SN4.  The next few days were supposed to include a hop to 150m (~500 feet) but that's on hold now.  If they were to scrap 4 and go to 5, they'd need to start the entire testing routine over. 

Honestly, I don't see any problem with SN4 that a .308 couldn't solve, from far enough away.  One through hole, get rid of the residual gas (methane or LOX), and weld a patch over the holes. 

In other space news, as I expected, SpaceX cancelled the Starlink mission which had been scheduled for early Tuesday morning and slipped it off until after next Wednesdays' historic Demo 2 mission.  When the stories started Monday morning about fairing recovery ships Ms Tree and Ms Chief were headed for shelter in North Carolina because of Tropical Storm Arthur, it seemed a foregone conclusion the mission would be rescheduled.  The other two ships on the mission, Go Quest and Of Course I Still Love You, stayed offshore and dealt with the seas from the storm.

Preparations continue for Demo 2 next Wednesday afternoon with the crew arriving at the KSC today for final preparations.   
On Thursday, May 21, NASA and SpaceX will conduct a flight readiness review system to evaluate the Crew Dragon and deem it ready for flight. If the vehicle passes that, then on Friday, a static fire test will occur, followed by a crew dress rehearsal. On Monday, NASA will hold its final launch readiness review. If all goes as planned, Bob and Doug will board their spacecraft approximately three hours before launch.

Astronauts Bob Behnken and Doug Hurley (left), NASA Administrator Jim Bridenstine and an unnamed guy.

Liftoff is scheduled for 1633 EDT on Wednesday the 27th.  It has been a lot of years since I've seen one of these for a manned launch.

Tuesday, May 19, 2020

A Little Radio Safety Techno-Geekery

I think everyone who reads here regularly knows that I've written on the topic of RF safety; probably the most in-depth look was a Radio Sunday post last summer.  You'll probably have noticed that I consider RF (Radio Frequency) energy as visible light only lower in energy, because, well, it is!  Sunlight or electrical light, if it's too bright or too powerful, can damage delicate tissues.  Nobody disputes that RF can cause injury from heating, everyone knows we have physical therapy diathermy machines and microwave ovens (systems which depend on exactly the same phenomenon - dielectric heating); it's the "other stuff" that gets blamed on RF, everything from cataracts to hemorrhoids. 

If you're an amateur radio operator, you're required to attest to the safety of your station on your license renewals, and it's typically not even something that needs to be analyzed.  If you run the typical 100 Watt transmitter and like to hang out on 40 or 80 meters, you don't even need to do the simplest analysis.  

The FCC publishes an excellent (and very readable) document with a dense title: “Evaluating Compliance with FCC Guidelines for Human Exposure to Radiofrequency Electromagnetic Fields” (pdf warning) and subtitled “Supplement B (Edition 97-01) to OET Bulletin 65 (Edition 97-01).”    Some title, isn't it?  To paraphrase Mark Steyn, it's not written like a refrigerator manual for people who really love refrigerator manuals.  It's well written - but it is kind of dense.   

I thought it would be helpful to reproduce some information out of this.  The first is the power limits for a ham station that don't require any analysis at all.  This is Table 1 in the document.

Reading directly across, say you're running that 100W transceiver on 20m.  Regardless of your antenna, you can't generate enough RF fields in a nearby observer to be even potentially hazardous.  You need to look closer if you're running 225 W PEP into the antenna, which is going to be less than your radio's output because of loss in the cable or switches or anything else the signal goes through.  Notice though that if you run that same radio on 10 meters you need to do some analysis because the limit there is 50 Watts.

Reading farther, though, you see if you're running more than 50 W on any VHF band (6m, 2m and 220MHz in the US), you need to look a little farther.  As the frequency goes up through UHF into the microwaves and higher, those powers go up. 

So what if you want to run full amateur power on HF through 6m, 1500 W PEP?  Then you need to look at these tables. 

This takes you through HF and the next one takes you as high as the FCC document goes. 

The tables all use the same basic approach, so let me grab the top of Table 4b - the 6m (50-54 MHz) section.  There are six entries for six different antenna gains.  There's a couple of ways of expressing antenna gain but these all use dBi; that's the ratio of the power in the forward direction with respect to a theoretical radiator that is equal power in all directions (an isotropic radiator - where the "i" comes from).   It takes a pretty phenomenal antenna to get to the last one at 15 dBi. I'll grab the 9 dBi antenna because it's a good antenna and it's an example.

What the numbers across tell you is the safe distance from the antenna for Controlled access (someone who knows it could be dangerous, like you) and Uncontrolled access.  That would be your neighbors, kids in the street and so on.  If you're running 100W, the closest those uncontrolled people should get is 9 meters (about 30') from the front of the antenna.  If you're running 1500 W, those kids shouldn't be within 28.4m or 93 feet. 

It's important to note that those numbers are for 100% duty cycle: full power, all the time.  Some digital modes put out full power for transmissions that might be 30 seconds out of a minute while conversational exchanges will be less.  It is fair to de-rate those numbers and say that during your contacts you listen about half the time and transmit about half, and that means lower average power. 

Let's say you're pointing your antenna in the direction of some neighbor (unavoidable in the suburbs and many places).  If their house is more than 93 feet from your antenna, they won't be exposed to a level of RF that could be dangerous. 

The same basic analysis approach applies from as low as 160m at 1.8-2.0 MHz to as high as the 23cm ham band at 1240-1300 MHz. 

It's easy to miss an important point here.  These are straight line distances, so if your antenna is up on a tower and the neighbor's house is one story, you calculate the hypotenuse of the triangle from your antenna at (for example) 66 feet and the straight line distance to their house.  If the distance from the front of your antenna to their house is 200 feet and the antenna is 66 feet up the distance you care about is 211 feet.  If your antenna points at the window of a multi-story building, you're back to the 200 feet. 

Monday, May 18, 2020

And the Anti-5G Attacks Spread to the US

Back in early April, I ran a story about attacks on 5G cellular infrastructure in the UK, setting fire to infrastructure and attacking workers who were laying so-called 5G cables.  Since the term 5G describes an over-the-air set of protocols, it may be terribly pedantic, but 5G doesn't exist in cables, just in radio signal over the air.  The cables carry data, just like a Godzillion other cables.  The attackers are attacking people for nothing.

I learned today that the Department of Homeland Security is issuing notices to the industry about attacks on cell towers and telecommunications workers by 5G/coronavirus conspiracy theorists. The DHS warned that there have already been "arson and physical attacks against cell towers in several US states," and warns that more may be coming.  They're advising the wireless infrastructure providers to pay more attention to securing their properties.
The preposterous claim that 5G can spread the coronavirus, either by suppressing the immune system or by directly transmitting the virus over radio waves, led to dozens of tower burnings in the UK and mainland Europe. Now, the DHS "is preparing to advise the US telecom industry on steps it can take to prevent attacks on 5G cell towers following a rash of incidents in Western Europe fueled by the false claim that the technology spreads the pathogen causing COVID-19," The Washington Post reported last week.
Did you catch that they talked about "directly transmitting the virus over radio waves?"  That's physically impossible.  If anyone believes that enough to go damage a cellular tower or hurt a worker, they should do themselves a favor and try to prove it.  If they can demonstrate that it's possible to transmit a virus by radio, they can win a Nobel prize in medicine and get themselves mentioned with greatest scientists of all time.  Plus, they'll never have to work another day in their lives. 

A telecom industry official said that carriers in the US "have seen sporadic attacks on their cell towers that were apparently prompted by COVID-19 disinformation" over the past few weeks, and the DHS said that attacks on cellular infrastructure date back to December of last year.
"since December 2019, unidentified actors conducted at least five arson incidents targeting cell towers in Memphis, Tenn., that resulted in more than $100,000 in damages... Additionally, 14 cell towers in western Tennessee, between February and April, were purposely turned off by way of disabling their electrical breakers."
Considering that 5G doesn't exist anywhere, yet; at least not in the form with millimeter wave systems that seem to cause the most fear, these attacks are particularly stupid.  At best, the widely deployed systems are 4-1/2 G.  The attackers aren't even close to their target.
The warning to law enforcement agencies said that an April 22 Facebook post "encouraged individuals associated with anarchist extremist ideology to commit acts of sabotage by attacking buildings and 5G towers around the world… in furtherance of an 'International Day of Sabotage'" and that videos have been posted online "showing people how to damage or destroy cell towers," according to ABC.

"Violent extremists have drawn from misinformation campaigns online that claim wireless infrastructure is deleterious to human health and helps spread COVID-19, resulting in a global effort by like-minded individuals to share operational guidance and justification for conducting attacks against 5G infrastructure, some of which have already prompted arson and physical attacks against cell towers in several US states," the DHS report said. The DHS report also warned of possible attacks against the electric grid.
The conspiracy theories also hold that the 5G radio waves suppress the immune system.  Again: 5G describes a system for using radio frequencies that have been in use widely for decades.  There is no known mechanism by which radio can suppress the immune system and no known evidence that it has ever happened.  The only linkage between 5G and Covid-19 is the coincidence of 5G starting to be talked about when the virus first got out into the population.

A reader sent me a link to this video about 5G that seems to be technically solid, just not covering everything I'd like to see covered.  On the other hand, if it was everything I'd like covered it would easily be a half hour long.

Sunday, May 17, 2020

Weekend Shop Update

Seriously, I sometimes wonder why do I do these on the weekend?  The days when I could only work in the shop on the weekend are long gone, and I'm out there some amount of time every day.  I guess the reasons are twofold: first, not to overwhelm with every little thing I'm up to all the time, and second, it's the weekend and everybody's pace of posts is down, be it Sunday Music or just lower activity.

First item, the problem I mentioned about my Fogbuster not spraying seems to be resolved. I only ran it for a few minutes, but it behaved more normally in that test than any other time last week.  The problem, as a comment from reader Leigh suggested
Yeah, it sounds like the issue is between the needle and the outlet orifice of the spray head.
and it was.  I had noticed what looked like a line around the tip of the needle valve but assumed that would cause a leak - too much coolant not none.  I chucked the needle valve in the lathe and held a file against the taper briefly, then finished it with 240 grit paper, followed by 320, 500, then finally with Simichrome polish.

Next, I finished the counterweight that I included a drawing of last week.

Those are 1/4 and 5/16" dowel pins in the two holes. They're not a press fit; you can lift the counterweight off them without moving them, if you're careful. 

After that I spent time working on the approach to machine the rocker arm that gets controlled by a cam on that crankshaft.   The drawing looks like this, top view on top, face view on the bottom:

This is to be made from steel, 1018 cold rolled.  I've cut a piece of bar to be the rough it will be machined from.  The top view is a little tricky; it looks like it's 1/4" across and it is - until you get to that 1/4" square tab on the right that adds 1/16" to the width, making it 0.313".  There's not much metal in this whole piece to hold onto while you're milling it. 

The other things I'm working on are some Tee nuts for the small (Sherline) mill, my version of a product made by a company that went out of business,  and a way to make an enclosure for a little ham radio project that doesn't involve hollowing out an aluminum bar and splitting 80% of it between the recycling bin and the vacuum cleaner.

Saturday, May 16, 2020

Weekend Looking Pretty Much Like Expected

This morning's launch attempt for the X-37B was scrubbed, currently rescheduled for tomorrow morning at 9:14 local. 

The possible tropical or sub-tropical storm has developed to be Tropical Depression #1, also pretty much as expected.  I'm a little surprised that Tomorrow morning's 3AM scheduled launch for the SpaceX Starlink mission 8 has been rescheduled for Tuesday morning, according to the local fish wrapper.  As that link says, Tuesday is cutting it close, especially with seas in the category of "partly to completely crappy," which the storm brings.  Booster recovery ship OCISLY was on location waiting for the launch as of this morning.

NOAA/National Hurricane Center graphic.  The "S" just north of the 8 PM Sat spot indicates they expect it to be a named storm by the next forecast.  That will be Albert.

A tropical storm before the start of the season isn't unusual. As a rough guess, I'd say it happens every other year; maybe every third year.  The one thing I'm a little surprised at is that it happens while we're having the mildest May I can recall. We're coming out of a 10 day period where a solid week never made over 80 degrees for the high. The forecast for Memorial Day is for a high of 84; granted that's a 10-day forecast which isn't the most accurate, but I recall a lot Mays that had afternoons in the 90s by about the end of the first week, and the every-afternoon-quitting-time thunderstorms starting by then or the next week. As it is, the only day in the Forecast with a high touching 90 might be Monday (day after tomorrow), although the official forecast is 89. 

Friday, May 15, 2020

Some Virus Stories That Have Grabbed My Attention

Over the past month, as the nation struggles with trying to resume some degree of normalcy and apparently is fighting off tyrannical state authorities without actually, so to speak, fighting those state authorities, a handful of stories have caught my eye and got stuck in a folder full of stories.  I thought I'd post about that here, but I could put dozens of pictures in this post and take dozens of screens to tell the story.  After some consideration, I thought I'd offer a link and a brief explanation. 

Bear in mind, I'm an advocate of the idea that the important things in any learning situation are the things that aren't expected.   It's often stated as “the most important phrase in a scientific investigation is rarely, 'Eureka! I've found it!!'; it's usually 'that's funny'.”  As one professor told me years ago, if you're not surprised, you're not learning anything new.

To begin with, there have been several papers that say lockdowns are ineffective.  The first I noted was in an Israeli publication called Mako and linked to by Townhall.
Professor Yitzhak Ben Israel of Tel Aviv University, who also serves on the research and development advisory board for Teva Pharmaceutical Industries, plotted the rates of new coronavirus infections of the U.S., U.K., Sweden, Italy, Israel, Switzerland, France, Germany, and Spain. The numbers told a shocking story: irrespective of whether the country quarantined like Israel, or went about business as usual like Sweden, coronavirus peaked and subsided in the exact same way. In the exact, same, way. His graphs show that all countries experienced seemingly identical coronavirus infection patterns, with the number of infected peaking in the sixth week and rapidly subsiding by the eighth week.

The Wuhan Virus follows its own pattern, he told Mako, an Israeli news agency. It is a fixed pattern that is not dependent on freedom or quarantine. “There is a decline in the number of infections even [in countries] without closures, and it is similar to the countries with closures,” he wrote in his paper.
Another one of those sayings is along the lines of "one paper doesn't mean much; several might".  An Oxford Professor says the UK's Covid epidemic peaked in March and the lockdowns now are doing more harm than good.

We can't talk about this story without Sweden coming up.  Sweden, of course, had a very laissez-faire response to the pandemic, and yet their results are "middle of the pack" among the European Union and neighboring countries.  As in other examples, most of Sweden's infections and deaths were in the highest population density portions of the country - around Stockholm.
Critics point out that Sweden’s per capita COVID-19 death rate is higher than several of its Scandanivan neighbors—Finland, Denmark, and Norway. Proponents point out that Sweden’s death rate is lower than many of its European neighbors—Belgium, France, the United Kingdom, Spain, and Italy—who initiated strict lockdowns. Proponents also point out that Sweden has “flattened the curve,” noting that the nation of 10 million has not seen its hospitals overrun or experienced medical equipment shortages.
“Whether or not they have openly embraced the Swedish approach, many other countries are now trying to emulate aspects of it. Both Denmark and Finland have reopened schools for young children,” wrote Professors Nils Karlson, Charlotta Stern, and Daniel B. Klein. “Germany is allowing small shops to reopen. Italy will soon reopen parks, and France has a plan to allow some nonessential businesses to reopen, including farmers’ markets and small museums, as well as schools and daycare centers.”
A standalone article on The Spectator about how Sweden lowered its 'R naught' number without lockdowns.

A professor of Political Science isn't the kind of guy you expect to have insights about viruses, but professor Wilfred Reilly does numbers and that's what it's about.  He writes "There is No Empirical Evidence for These Lockdowns" on Spiked.  This is quite possibly the money quote:
The question the model set out to ask was whether lockdown states experience fewer Covid-19 cases and deaths than social-distancing states, adjusted for all of the above variables. The answer? No. The impact of state-response strategy on both my cases and deaths measures was utterly insignificant. The ‘p-value’ for the variable representing strategy was 0.94 when it was regressed against the deaths metric, which means there is a 94 per cent chance that any relationship between the different measures and Covid-19 deaths was the result of pure random chance.  [Bold added - SiG]
If you love data visualizations, you might like this place.  Big shocker, I know: the best predictor in this multivariate analysis for fatalities is population density.

An interesting study by Stanford Medical School doctor Jay Bhattacharya was to test antibodies to the virus among Major League Baseball team staff members. This focused on staff; not the players, the other thousand people in the various offices.
The results of the MLB study showed that only 0.7% of the staff had the antibodies indicative of having had COVID-19.

Some staff is located in areas where antibody testing has been conducted for the general population such as New York, Los Angeles, and Santa Clara. Most of these locations showed the MLB staff had a much lower rate of prior infection than the general population. For example, the New York City metro area showed 25% of the population had antibodies. The MLB staff for the Yankees only showed antibodies in 1.64% of the employees. The Mets tested positive at a rate of 2.61%.
Dr. Bhattacharya said this seems to be basically their socioeconomic status and it's a trend he has seen in other studies he has done. Lower income residents had higher exposure rates to COVID-19 while the higher income workers had lower exposure.  Perhaps the lower income residents ride the subways while the higher income workers take private cars or other means?

The implication to Dr. Bhattacharya is that lockdowns might even be counterproductive.
“I think in the back of people’s heads there is this idea that somehow we can eradicate this disease if we just stay locked down. That is not possible. The serologic evidence, even the MLB study, suggest this. It suggests the epidemic is too widespread to eradicate. It spreads via asymptomatic contact. Like people who don’t have very many symptoms, even mild cold symptoms can spread the thing. They aren’t going to show up for testing. They aren’t going to show up at a hospital or a doctor.”
“There is no safe option. If you think that having a lockdown will provide you safety, you are mistaken. Because the problem is this lockdown has had enormous negative effects on the health of people in the United States and around the world.”
"Enormous negative effects on the health of people...?"  Since the lockdowns started, I've been saying we're only measuring one thing - the fatalities - and we're not measuring that accurately.  No one is tracking the actual costs and deaths from the lockdowns.  The National Institutes of Health ran this paper simulating the effects of the lockdown on diabetic patients, for example.  They conclude the longer the lockdowns, the worse the number of patients with diabetic complications.  

Early on in the spread of the disease, it was pretty easy to come to the conclusion that eventually everyone on earth will be exposed to this.  It's a new virus, and new viruses jump from animals to humans pretty regularly; it's simply how the world works.  I don't have strong numbers to base this on, but I believe the nationwide downward trend in new infections will continue downward, this numbers guy says the worst of the epidemic is over, but the virus will be back next fall.  It's my understanding that coronaviruses, as a group, are hard to create vaccines for.  The common cold is one of many such viruses that we're always exposed to, for example, and there's no good vaccine for the common cold, right? 

I'm not saying this is a cold; I'm saying it's in the same family of viruses.  OK, Google? 

So what do you do?  Covid-19, if it's unique at anything, seems to be very good at attacking the metabolically unhealthy.  The reason the death rates are the worst for those over 80 is that the majority of people in that group are metabolically unhealthy.  It's worse than that.  In the American population at large, less than 20% are ranked as metabolically healthy on five indicators that are assessed.  If you have high blood pressure, diabetes, a bad HDL:triglycerides ratio, existing heart disease or lung diseases, it might a good idea to isolate yourself. 

The surprising thing to me was how good blood levels of Vitamin D were as an indicator of surviving Covid-19.  Those patients with a measured blood level in the deficient range (under 20ng/mL) had a 10x worse chance of surviving the virus than those with normal levels of the vitamin (>30ng/mL), correcting for age, sex and other comorbidity factors (right plot here).  When you add that to the widely reported intolerance of the virus for heat and humidity, that might be a factor in the better numbers of Florida, Georgia and southern states as opposed to the Northeast.  Get outside in the sun, a few minutes a day to start if you've been locked indoors too long. 

A screen capture from an hour and a half podcast with those two guys whose images are stuck in the right graph; an American M.D. and an Irish biochemical engineer.  Interesting, but probably too geeky for most people. 

Wednesday, May 13, 2020

The Biggest Risk for the Next Starlink Launch is Weather

June first is the start of the Atlantic hurricane season, and it's rather common for some development to be getting started just about any day now.  This year is like that, with the NHC assigning a greater than 60% chance of development of a subtropical system to develop off the NE Florida coast in the next five days.

The next batch of 60 Starlink Satellites is scheduled for launch in the early morning hours of Sunday, four days away, at 3:53 am EDT (07:53 UTC), May 17th.

There are other interesting aspects to this launch, such as booster B1049 going for the elusive 5th landing that SpaceX has so far been unsuccessful at pulling off.  They say the boosters should be reusable 10 times, but have yet to recover any booster more than four times.  

It’s unusually important that this 7th Starlink launch goes perfectly, as any in-flight anomaly would almost certainly delay Crew Dragon’s crucial NASA mission. Additionally, if Starlink-7 slips more than a day or two, it could easily force SpaceX to push the mission into late May or early June, as Crew Dragon’s first crewed launch will also need a drone ship to recover its brand new Falcon 9 booster.

SpaceX moved their Pacific recovery drone Just Read The Instructions (JRTI) to the Atlantic in support of the increased launch cadence over here.  They've been doing some serious upgrades to JRTI, but don't think it's likely to be ready for these next two launches.  Which leaves them with one drone, Of Course I Still Love You (OCISLY).
Typically, OCISLY has taken around 7-10 days from port departure to arrival to recover Falcon 9 boosters after Starlink missions, most of which is spent being slowly towed by tugboats. In simple terms, assuming no technical or weather-related launch delays, that would give SpaceX just a handful of days to remove booster B1049 and turn OCISLY around to recover Crew Dragon Demo-2 Falcon 9 booster B1058. Unfortunately, to recover Starlink-7 Falcon 9 booster B1049, OCISLY is heading more or less straight for a tropical depression forming in the Atlantic Ocean. High seas in the recovery area are an almost guaranteed launch delay unless SpaceX is willing to expend B1049 (very unlikely).
Part of the whole mission to launch Starlink satellites is to test these "experienced" (not old!) boosters on their own nickel and not on flights that customers have paid for.  I tend to think they really wouldn't want to lose either B1049 or the brand new B1058.

Booster B1049, looking like a proper used spaceship, after her January 9th Starlink mission.  Richard Angle photograph.

Anything I'd put here for what I think is going to happen is just a WAG.  My guess is that they'll watch this weather system for a few days and decide as late as practical if they're going to go for it or delay the Starlink 7 mission until after the critically important Demo2 flight for NASA on the 27th.  OCISLY would have to be almost all the way to its assigned location out in the Atlantic by that time.  I have even less idea if they'd order OCISLY back in to Port Canaveral or not.  Although the article didn't mention it, at least one booster was lost after a successful landing because the seas were too rough and the booster slid around on the deck before it could be secured.  Nobody wants to see that happen. 

Tuesday, May 12, 2020

Have You Had Enough Stimulus? Part Deux

I've been beating the dead horse of creating money out of thin air for the life of this blog, so this is part two only in the sense of a post with a similar name in early April.

I heard today that House Democrats are working on another stimulus bill, this time in the vicinity of $3 Trillion.  It's all very preliminary at this point, but the Financial site The Motley Fool reports on what they're seeing in the early discussions.  Note that this is not an exhaustive list -- the actual bill is over 1,800 pages long and you can only imagine what sorts of handouts for favored groups are included in this.
  • Almost $1 trillion for state and local governments;
  • Another stimulus payment to individuals;
  • Hazard pay for essential workers;
  • $75 billion for coronavirus testing and contact tracing efforts
  • Instead of expiring at the end of July, the $600 weekly boost to unemployment benefits would last through January;
  • $175 billion for rent, mortgage, and utility assistance for affected households;
  • Funds for election safety and to help make voting by mail easier;
  • Money to help the struggling U.S. Postal Service.
Thanks to the first three (four? who knows?) packages, the US Debt Clock shows our total debt has ballooned to over $25.1 Trillion.  The Motley Fool isn't very confident that anything that passes the house will pass the senate and points out some obvious contentions, such as those last two bullet points.  That doesn't mean that something won't pass, though it's just a matter of how much pork gets pushed through.

How big is a trillion dollars?  I thought I'd pass along a couple of visualization exercises that are my favorites.  The first one is that the number of seconds in a year is within 1/2% of pi times 10 million (that was in a freshman physics class nearly 50 years ago and stuck in my mind).  So what?  That fun fact leads to simple visualization; if someone sat at a window and threw out a $1 bill every second of every day for a year, they'd throw out just over $31 million.  Make it an 8 hour day (1/3) of a year, and it's $3/second to throw out $31 million and so on.  To throw out $3 TRILLION at $31 million/year would take 96,774 years - and 10 weeks.  Clearly we'd need to step up our game.  If you can imagine something that threw out $95,066 per second, 24/7/365, or $285,200 every second for an 8 hour working day, that would get you there.

In the first year of this blog, I used some visualizations from and I was pleased to find they're still there.  They present a series of pictures on the national debt (they stopped updating the numbers back in '17 when the total debt was a mere $20 Trillion.  Ah, the innocent days of our youth.)  Here's $1 Trillion in pallets stacked with $100 bills. 

One Trillion Dollars.  
 Each of those cubes you see is a pallet of $100 million in $100 bills.  At the far end you can see an American football field with a Boeing 747-400 parked on it. You can see the White House with both wings to the right.

Just triple that area and you get the $3 trillion they're talking about for this spending bill.  Alone.

Look, I know that there's a lot of suffering out there, I just think this is exactly the wrong thing to do.  When someone has a compound fracture, you don't just give them a pain reliever; you fix the problem.  This is a band-aid.  We're doing Modern Monetary Theory and that's going to end in a worse disaster than we have now.  This could be the thing that destroys the economy and the entire western world.  

Monday, May 11, 2020

Starship Getting Ready to Hop

In the two weeks since I last posted about Starship SN4, SpaceX has continued to make progress, steadily checking off one test after another until now they're one set of tests short of waiting for the formality of approval to make the 150 m (~500 foot) hop flight. 

In the two weeks since then, they've done a "wet dress rehearsal" and two short (~3 second) static test firings of the Raptor engine (first and second).  Then they pulled that Raptor engine off SN4 and did another cryo pressure-proof test that was done to higher pressures than the previous test: 7.5 bar (110 psi) versus the 4.9 bar test from two weeks ago. 

Today we learn that they replaced the engine today, with a second Raptor engine, serial number 20, seen here.  Because that's a new Raptor, there will be a first static test of that engine, currently looking like it will be May 13th, with a test window open from 9am to 9pm CDT (UTC-5).  Backup days for road closures and such are the next two days.

What's that up there?  The world's highest thrust per buck rocket engine, being driven down a beach road by a fork lift?  Must be Boca Chica and SpaceX.  That was yesterday morning.  Taking a look at the live video feed just now it seems like the engine has been mounted. 

Final words to Teslarati, with one modification.
If that static fire and included wet dress rehearsal (WDR) is successful, Starship will technically be cleared for flight. The only obvious missing piece is an attitude control system (ACS). Starhopper, for example, used cold nitrogen gas thrusters quite literally taken off of flight-proven Falcon boosters. It remains to be seen if SpaceX will take the same approach with Starship SN4 or if a different kind of ACS thruster is already installed on the rocket and hidden in plain sight. For now, it looks like we won’t have to wait long at all to find out.
Instead of "If that static fire and ..." is successful, I'll say "When that static fire..." is successful.  After watching them for these last several months, they are nothing if not relentless.  If the tests don't go well, they'll keep working until the tests pass.  Starship SN5 is almost ready to roll to a test stand now.  Construction on SN6 has started as well.

Sunday, May 10, 2020

Weekend Shop Update

Last time, I showed the piston rod and mentioned it wasn't Done done.  I needed to make a couple of small bearings out of bronze and the piston wrist pin.  I had made the wrist pin but grabbed a piece of the wrong size stock - the only piece of the 5/16 drill rod I had - when it was supposed to be 3/16".  I got started on the bearings and came to the realization I didn't have a reamer for the big one.  It's to be reamed to 0.281, or 9/32".  I ordered that last Sunday from an eBay seller and received it Thursday.

Now for the Ta Da! moment for the connecting rod with the bronze bearings in place.

With the wrist pin alongside.

There's another piece I started and made good progress on, but it's not quite ready for its photographic session yet.  This is a beginner's engine and so a lot of parts are acceptable without touches that might be done on a more advanced engine.  One of these is on the crankshaft.  The shaft itself is called out as just a piece of 5/16" Cold Rolled Steel.  I found a piece of drill rod in that diameter at one of the metal dealers and that became the crankshaft.  The way the end of the crankshaft is drawn, this is unbalanced (left side of this picture shows the end).

I've seen various solutions to this; some guys use the square tab for the top and then go into a semicircle for the bottom.  That seemed like it would be trading one imbalance for another, so when I saw the idea on the right here, I thought I'd opt for that.  It's roughed out on the mill but not ready for prime time.

I spent some time on a problem with my Fogbuster cooling system.  It stopped misting the coolant although the air flow is normal.  I've done a couple of things to try to fix it and haven't gotten anywhere.  I cut out the outline of that sector of circle with no liquid, just the air blowing and while the aluminum chips didn't melt onto the cutter, the metal came out looking like I'm not used to seeing.  Shiny but looking ripply.  Almost as if it had melted and then quickly solidified. Some of the aluminum cut away was still warm when I picked the pieces up after cutting.  I don't think I'll be doing much cutting until I get that fixed.  Maybe some of the smaller parts on the smaller machines.