Author Archives: David Sidwell

Apollo 12 and the (virtually unrecorded) eclipse

I’m just reading the final chapters of the excellent Apollo book. I had no idea that the crew of Apollo 12 witnessed the Earth eclipsing the Sun!

Mission planners knew that, on the way back, the Earth would block the Sun. But nothing special was planned.

The crew were stunned when they saw it start. They had to watch through a dark filter so as to not go blind due to looking at the Sun. Thanks to Earth having an atmosphere it was amazing; the colours changing as the Sun got more blocked. But…they had no colour photo film left!   

They frantically radioed down for the settings on the B&W camera to shoot in this complex set of lighting conditions, but apparently it was too late, as it was changing fast.

They grabbed the movie camera and shot some footage, but it seems to have been blurry and had exposure problems. 

Here’s a frame from the movie footage. It looks like you can see some of the colours in the atmosphere.

This was the first time any human had seen this. And the last. No other Apollo came back via this trajectory and we haven’t been that far out since.

Converting TV Recordings (PC) to playable

I’ve known for years how to convert TV recordings (“mpeg-2”) to PC files (avi, mkv).  But yesterday I dived back in and learnt a few things:

  1. My tiny Raspberry Pi media player struggles with raw TV recording playback
  2. Windows batch file programming has a strange way of handling “ things in names

The Underlying Problem:

My main PVR (hard disk recorder) in the TV room is great. But it’s developed issues with Channel 10 recordings. Either the tuner or the aerial/cable struggle and the live picture breaks up and the recording stops or gets corrupt. Thus my weekly dose of Have You Been Paying Attention? suffers.

My PC room also has a TV aerial socket, but unused for 5+ years. I hunted down my USB TV stick and plugged it all together. I was still licensed for the excellent DVBviewer software. Amazingly it all worked first time, both live viewing and recording.

I then tried to play back the .ts (mpeg-2) raw recordings on the Raspberry Pi media centre (Kodi). It struggled a bit and the CPU was high.

Prepare, Convert then Automate

The upshot is I spent a while re-learning how to properly prepare then convert the large .ts file to a smaller .mp4 file (with AVC video and AAC audio).  Then finally automate the whole thing.

Prepare: The raw .ts file usually has hiccups as bits go missing and timecodes go out of sync etc. So I run it through the free Java program Project-X to sort these out and re-mux back into a new, ‘clean’ .mpg file

Convert: I then encode this .mpg file into a much smaller mkv file, with the free ffmpeg converter.

Automate: Both have command line options, so I wrote a script (Windows batch file) to automate and hooked it all up to enable ‘drag and drop’ via the magical Drop-It free program.

It kept failing!  Turns out that there was a hiccup with filenames “That_have spaces in_them like this.ts”   – when you pass them as .bat variable to programs like ffmpeg. The upshot, I learnt, was you have to use an obscure batch file command to remove the “ from the filenames (which have to be there to pass the whole filename to the .bat file).  It involves this line to reformat the variable containing the filename:

set fileName=%fileName:”=%

News to me.   But it worked.  All good now. One drag and drop and it’s done.

Navigation – Getting the basics right (Part 2)

This is not directly related to Part 1, but similar concepts : you need to get the underlying basic setup – or measurement system – right.

Just a week ago we were out for a walk by the sea along Pt Gellibrand here in Williamstown. On the other side of the bay, we spotted an obvious point, but were unsure what it was. It was misty, so hard to see details. I took some phone photos and went back the following day to take some with a camera. Again it was misty (double click to zoom in on all photos):


On that first day, my simple compass (app) implied this was about due East of Pt Gellibrand (we were not far from the footy ground). Looking at Google Earth later, I speculated it was Pt Ormond as that was ~ East. But at least one other person thought it was closer to Black Rock, more South-East.

I went back today (a week later, Sunday 26th June) armed with a new toy: a phone app that overlaid the direction (bearing) the camera is facing. Here’s me pointing roughly at the Mystery Point:


Aha, it’s 133 degrees from me. Sort of South-East.  Not sure how I had logged it as ~90 (East) last week. Maybe the compass hadn’t been recalibrated?  Or Carbon-Based Error?

Anyway, off to Google Earth. Which clearly showed the bearing from Pt Gellibrand to Black Rock was … about 144 degrees. What?  Way outside the assumed 1 or 2 degree error margin.

Something was nagging me. Different “Norths”? The upshot is: yes. I found that the (free) version of the App was locked in to use Magnetic North.  Google Earth is locked into using True North (as in; the lines on the printed map).  They two can be different depending upon your position on the Earth (i.e. the local variation in the Earth’s magnetic field). A quick look-up shows in Melbourne this variance is 11.6 degrees.  133 + 11 = 144.  Nice!

Earlier I’d taken another photo, with the big camera, when the sun was out, then labelled it with my (our!) speculation:

Red Bluff photo from Pt G small

Seems pretty right to me. Just need to work out what the two towers are. And don’t say “a book” Smile

Navigation – Getting the basics right (Part 1)

Something that happened today (see Part 2) reminded me when I first got a GPS. This was 2008 and it was a hand-held Garmin which cost a wee bit ($280 USD, inc $50 USD shipping). Arguably, most of the function I used then is now available in a $4.99 phone app.

So the first time I  used it for navigation testing and route logging was in the Dandenongs. I had a paper map with a grid on it. My planned route (pale red, below) was roughly N-E; Edgar Track, Bills Tk, Camelia Tk.

Yet the GPS was telling me I was hundreds of metres away from where I should be.  I could expect it to be 1 or 2 meters, but hundreds?  If it wasn’t for the signs, I would have missed the turn offs.

Later, when I loaded the GPS log into their map program on the PC, I saw the log was ‘correct’ but displaced, in both directions. The image below is a very rough recreation, showing the walk I really did in pale red and the GPS log in red.

GPS datum error (simulated)

My first reaction was “my new GPS from America is faulty!”  But a bit of reading taught me something: the paper map had the same co-ordinate system, but a different datum than the GPS.

A datum is “a system which allows the location of latitudes and longitudes (and heights) to be identified onto the surface of the Earth – i.e. onto the surface of a ’round’ object.” Source.

There are different ways mathematically of doing this, so lots of different datums. I discovered that my map had an old Australian one (AGD66) but the GPS was set up to use the new Australian one; GDA94

The difference is about 200 metres to the north-east. Which was exactly what I saw.  I’ve never been tripped up like this again.  In fact Google Earth uses an international datum virtually identical to GDA94, so that makes things nice and consistent.  As I said earlier, this is sort of related to what happened today, as covered in Part 2.

Where is that photo? – South West Victoria trip

I was asked where some photos I had (in my Apple TV screensaver) were taken.  Most were from the same trip, it turns out. Down the South West of Victoria.

Photo 1


This is Aire River near the Aire River West Campground. More info here.  It’s between Cape Otway and the 12 Apostles. Even more precise it’s between Cape Otway and Johanna Beach.

Photos 2 & 3


Johanna River as it enters the wild sea at Johanna Beach. Also the scary waves at Johanna Beach. The River is a short walk (South east) from the car park at the Johanna Beach camping area.  Between Aire River and 12 Apostles area.

Photo 4


Northern Grampians.  Near the Mt Zero Log Cabins, looking south towards Mt Zero and Mt Stapylton.

Photo 5


The other end of the Grampians. Coming East from Hamilton area towards Dunkeld in the Southern Grampians. One of those is Mt Abrupt, I think.

R + L = J. Then just heat up a bit.

Quick summary of Internet fan theories. No spoilers coz most of this is not in the books :

  • Ned’s Stark had a sister Lyanna. This was all 20+ years before the books and TV show.
  • Lyanna was ‘kidnapped’ by Rhaegar Targaryen, who is Dani’s (late) eldest brother. Not the tosspot who got the golden crown … poured over him.
  • In fact Lyanna eloped with Rhaegar. And had a baby, called Jon. 
  • So, Jon is Dani’s nephew and Ned Stark’s nephew too.
  • Ned took off after them (update: I mean Lyanna & Rhaegar) but she was dying when he caught up. She made him promise to look after her baby, Jon.
  • So Jon Snow is half Stark and half Targaryen, remember that last bit Smile 
  • Melisandre (the Red Witch) – in the TV show etc – had visions of a leader and assumed it was Stannis Baratheon. Nope. It was someone else…
  • Melisandre is on the scene when Jon Snow cops it.
  • Jon Snow, like Dani, should be able to be re-born via fire
  • ==> Melisandre will assist Jon Snow to be re-born via fire in this season

The trailer for this season has hints of a young Ned Stark going into battle. Probably a flashback when he tried to rescue his sister, who was actually engaged to Robert Baratheon!  This started the recent war….

R + L = J  is the meme for Jon Snow’s parents.

Walk Route: Dandenongs – Doongalla and Rankin Loop

About 6.6 km total. Rankin Track quite steep.

We had brunch at The Basin, then drove up The Basin – Olinda Road towards Olinda. 

Car near the power substation at (1).  Walk quite level, then the descent starts at  (2).  Doongalla site – upper picnic ground (3) – great place for lunch.

Suggest you walk down the hill to the lower picnic ground at Doongalla; the Stables area. It’s lovely. Then back up to Camelia track, up Rankin Tk, then right onto Channel 10 tk. Re-trace route back up to car (i.e. Zig Zag tk etc).   Click on image below for full size.  Note: some track names may have changed.  Update: some photos at the end, after map.



Fog quickly lifts at the very start. On – on near – Kyeema Track.

Hikes (Bushwalks) via Train

An interesting question came up yesterday, what sort of good bushwalks can you do, that are accessible by train?  We are assuming day-trips; back the same day. My first thoughts – as I said Smile where The Dandenongs: Belgrave Station (where Puffing Billy departs from).  I’ll update this later when I work out what walks go from there.

Another idea may be a country bus trip or combined train/bus trip. Time for some research.

Telstra, Netflix, Google and the AFL Broadcast Rights

A little time capsule. Let’s re-open in a year to see how wrong I was Smile 

I have zero inside knowledge. Just bringing together some threads from the recent few weeks;

  • Google talking to NRL about streaming via their YouTube service (erroneously reported to be AFL)
  • Telstra reportedly looking at offering Netflix, despite owning half of Foxtel ; The Age today – July 13th 2015

So here’s my speculation as of 8:05am on Monday July 13th 2015:


Carlsruhe-Daylesford ex-Railway Line

I lived briefly in Daylesford in the late 60s. I remember it being cold and at least three other interesting things Smile , but I don’t remember any trains. Years later I heard the town had re-invented itself “after the railway closed”.   Fairly recently I stumbled upon some excellent scans of very old Victorian Railways engineering documents & drawings# These include  ‘steepness’ guides – called Grade Books – like this one (extract  shown):


You can see the whole document online. My reading is that Grade is Gradient and so the numbers like 50 mean “1 in 50” gradient.  The above is also showing Stations (in Bold) and other items (like the Coliban River). 

Brief History of the Line

To keep things logical, I’ll jump ahead in my discoveries and give a summary of the line. I found this out after I’d found the drawings (above and below).

The Wikipedia article on Carlsruhe station gives a good summary of the Line: “The station was opened in 1862, and became a junction in 1880 when the first section of the line to Daylesford was opened. The Daylesford line was closed in 1978 and staff withdrawn. The station was one of 35 closed to passenger traffic on October 4, 1981 as part of the New Deal timetable for country passengers. The signal box was abolished in 1980 and the station closed in 1982 …The surviving station building is now used as a private residence … Carlsruhe was the junction for one of the lines to Daylesford (the other running from North Creswick) although most trains using this line actually originated in Woodend.”

Here’s my OziExplorer/OSM map of the approximate position of the Line, with the Stations just shown as numbers.  Not how far Carlsruhe station is/was from the town. Interesting!  It was/is on the Bendigo line, which is shown in grey.  Note: virtually all the Stations have gone now as has the Line.



My June 2015 Trip

Further digging (links on the same page as the above Grade Book, if I’m being honest) revealed very detailed 1878 drawings of sections of the entire track.   Such as this one, again from 1878:


The original PDF is online.   Some quick observations on these finely detailed documents:

  • The scale is quite fine. The bottom numbers (10, 20) are chains, which are a cricket pitch, that is pretty much 20 metres. So the little division is 20m.
  • The zero point for this scale – as noted in the top left corner – is the “Centre of S.W. platform Carlsruhe station”
  • It has assorted heights, but that’s not my real focus
  • There is a ‘Curve (of) 17 chains rad(ius)’ not long after the Line breaks away from the main one.
  • It has ‘points of interest’ such as fences and roads, which intersect or are near the Line. I should say ‘as it was in 1878’

That last bit was the reason for my trip in June 2015. After looking at Google Earth:



…I wondered if I could find some of the 1878 features?  I’m pretty sure that’s the route of the old railway line (top arrow), even though it’s gone per se.  Note the curve and it crossing a road (lower arrow).

To cut to the chase, the very helpful owner of the house (ex Station) explained:

  • The trees have covered the first bit of the old track and all traces have gone
  • You can’t get to the rest as it’s fenced off
  • It’s apparently still a ‘railway reserve’ or similar, which means the “Government” owns it.
  • There has been talk of building a Rail Trail bike path on the old route, but nothing so far
  • You can see sections of it, further on, particularly a bit where the ‘gorse trees had been burnt away’

After some exploring, local knowledge was again shown to be best kind; all of this was true. Here’s some photos of my trip (more later)

Firstly the old Carlsruhe Railway Station

DavidS_20150621_122312 Google Image link 

And a bit further on, the old Line looks obvious to me. At least the sheep are making use of it!

DavidS_20150621_124305 Available here too

(I’ll put more up later, plus add more text)


# Drawing is the preferred term for such engineering diagrams even today, when they are done by computers.  As you can see, the ones here seem very much hand drawn.