Dead reckoning / celestial nav - 1200nm of open Pacific

[stearmandriver]

Hello all,


Since you seem to like classic airplanes here, I thought you might enjoy hearing about a recent flight I completed in the A2A Constellation. I did it the really old fashioned way: dead reckoning and sextant shots only, using the sextant gauge originally developed by Dave Bitzer and Mark Beaumont, and ported to FSX by Kris Ogonowski (Kronzky). I've enjoyed the recent discussions on the A2A board about traditional navigation methods, they inspired me to come back and do a little more of this stuff. So who knows, maybe I can inspire someone else to try it out .

(Due to SOH requirements about 4 images per post, I'll complete the story via several replies to this original post).


So I departed Johnston Atoll in the Pacific at 1251z, (0151 local) on June 01, 2016 (date chosen for relatively quiet weather enroute, just a few areas of storms).


Flight Plan said:
1147 nm
05:03 enroute
Cruising at FL200 at 245 KTAS
3500 gallons fuel on board
(I originally planned this flight to fly in the outstanding Jahn/Visser C-47 v3.14, but then decided to fly it in the Connie instead. That's why the checkpoints are labeled out through "6 Hour fix"; the C-47 is slower. I just re-planned it at Connie speeds and that made every hourly fix happen once every 40 minutes instead lol).

Departure time was chosen using a common technique of the era: plan to fly the majority of the flight at night, so you have multiple celestial bodies to take shots from (resulting in multiple lines of position and therefor a fix); but plan to reach your destination an hour or two after sunrise, so you can intercept and follow a line of position off the sun all the way to your destination. So the plan was to fly a landfall procedure (a turn off course to know for certain which side of destination you're on) to intercept the 158 degree LOP from the sun and follow that into PLCH. Dead reckoning headings / groundspeeds calculated using ASN's wind forecasts, which are pretty true to real life in that they give you an estimate, but what you actually encounter is still a little different so you have to adjust.




Departure was uneventful; smooth climb out of Johnston Atoll, attempting to track 140 true.

Taking the first sextant shot off of Nunki, for a speed line, at the 1st checkpoint:

[CONTINUED]

 

[stearmandriver]

And the results: planned groundspeed is right on. Continue. (Green lines on the maps are Lines of Position calculated from sextant shots)

Took 2 shots at the next checkpoint: off Fomalhau for a speed LOP, off Antares for a course LOP. You can see the results above, plotted after the "2 hr fix". Groundspeed faster than planned, drifting left of course. But not by much, and a bubble sextant isn't that precise, so I'm not going to make any corrections yet. We'll grab a 3 star fix at the next checkpoint and have a better idea then.
Taking the Antares shot:

Heading for the 3rd checkpoint, treated to a nice moonrise:

[CONTINUED]

 

[stearmandriver]

3rd Fix, shots off Fomalhau, Antares, Enif. Remember, bubble sextants - not the most precise navigation device (and this guage simulates that perfectly). And yet, we can reasonable conclude we're somewhere within the triangle formed by the 3 crossing lines of position. Yup, fast and left of course. Off-course correction calculated and applied at this point; turned three degrees right. Groundspeed averages out to 254kts, faster than our planned 242kts.

Now the shots for the fourth fix were a challenge as I ran into an area of weather. This guage is smart; it won't let you take a shot in the clouds. When you try, you see this:

So just like in reality, weather makes celestial nav a challenge. But we're at FL200; we aren't going to be in the clouds forever, and we have math on our side. We can make this work. So I just keep taking shots until I get some clear ones. Our checkpoint was planned for an arrival time of 1541z; I ended up with successful shots off Fomalhau at 1544z, Enif at 1546z, and Antares at 1551z (with periods of clouds in between). Now, deriving lines of position that create a fix from this data means we have to mathematically "advance" the LOPs based on the time elapsed between each one. I actually moved them "backwards" in time in this case, moving them all back to 1541z. Based on our last estimate of 254kts groundspeed, a quick time/speed/distance calculation tells us we're doing 4nm per minute. So, each line adjusted backwards the appropriate amount. Result (with a little more slop which is to be expected considering we had to interpolate, but we should still be within that triangle): still fast and left of course. A couple more degrees right applied for correction.

[CONTINUED]

 

[stearmandriver]

Approaching the 5th checkpoint, weather has improved, and... dawn!

5th checkpoint (the last we'll get star shots for) indicates still fast by the same amount, and trending back on course. After passing the 5th checkpoint, we get sunrise in earnest (several shots, just 'cause she's pretty in the morning!)

[CONTINUED]

 

[stearmandriver]

Checkpoint 6, no stars were available and the sun was a bit low, but I grabbed a shot off of it for a course line, and it showed we were just about back on course. That was validated just after passing checkpoint 6 in the best way possible - visually. After over 4 hours in flight, and over 900nm traveled, our course line passes a piece of land near enough to maybe see it. That would be Fanning Island, a small atoll off to the right a little ways past checkpoint 6:

[CONTINUED]

 

[stearmandriver]

And... eureka!

Home stretch, now. Using time calculated from my most recently calculated groundspeed (which the appearance of Fanning has roughly confirmed), I calculate that I should make the turn onto my landfall procedure - a 090 true course - at 1721z. I do so, and set my sextant up for the most important shot of all - the one that tells me I'm on the 158 degree LOP that passes through our destination of Christmas Island at 1730z (give or take a few). You can see on the map above what we're doing - at "Turn for Landfall", we turn left onto a 090 true track and start taking sextant shots on the sun to intercept the red line on the map. That's our LOP into Christmas Island.

(One thing I like about this one - with a real sextant, when taking a sun shot you put a green filter on for safety. This guage shows that by turning the sun into a little green ball.)


All right... at 1729z, within ONE MINUTE of my original planning, I get a sextant shot that shows we're dead on the 158 degree line of position off the sun. We know that at this time, that LOP passes through our destination, which should only be about 50nm away. Since we're on our line we want to follow to destination, we turn right to pick up the 158 true track (my dead reckoning, corrected for variation and winds, put this at a magnetic heading of 147 degrees). Annnnnd:


LAND HO!

[CONTINUED]

 

[stearmandriver]

Nice gentle approach and a smooth landing. My only regret is that I forgot to activate the nice freeware scenery I found for Christmas Island, so the final shots are default FSX. Oh well... there's always the departure ;-).

So this was a great time. I sometimes feel silly for putting in the time I sometimes do with flight sim... but you know, it's stuff like this that makes me feel foolish for that. I don't think there's any reason to feel silly for gaining a greater respect for and understanding of our history by attempting to experience it to the best of our abilities. And I'm continually blown away by just how GOOD all this is! The whole FSX world, the physics that mean dead reckoning actually WORKS, and of course the extras like the sextant guage and the great planes. This... is pretty good stuff!


If anyone made it all the way through that novel, thanks for reading ;-).

 

[BendyFlyer]

Nothing silly about this Stearmandriver - Great trip using all the older navigation tools in a great aeroplane. And that is a real achievement, well done and thanks for sharing.

This is where I always hoped and knew Flight Simulation would go, community of sincere and interested aviation folk who appreciate the sheer magic that can be achieved in terms of realism, fidelity and plain fun. Folks with eye and appreciation for technical details and aviation history not to mention a little artistry. I cannot think of any other way to recreate great and not so great flights and aviation now and in the past. Every day I am impressed and amazed at how good it is and getting better constantly.

 

[paiken]

Thanks for sharing your adventure. I certainly don't have the patience to do a 4+ hour flight in real time, let alone with dead reckoning and sextant navigation. Well done!

 

[ryanbatc]

Wow that is hardcore... well done!

 

[txnetcop]

Congratulations and excellent navigation! I didn't even know you could use a sextant with FSX and have it be reasonably accurate!
Ted

 

[lownslo]

If anyone made it all the way through that novel, thanks for reading ;-).

Not only made it through but read it a second time! Thanks so much for this lesson. Where can one find that sextant?

Thanks,

Greg

 

[stearmandriver]

Thanks for sharing your adventure. I certainly don't have the patience to do a 4+ hour flight in real time, let alone with dead reckoning and sextant navigation. Well done!

Oh my no, not real time. Time acceleration between all the checkpoints LOL. Although, with the time you put into flight planning first, it probably did take about that long in total, spread across a few nights.

 

[PRB]

That's good stuff SD! I need to learn how to use that sextant!

 

[Willy]

Great job on navigation! This is the kind of stuff that makes FS for me. (Although I'd have ended up in Reykavik)

 

[stearmandriver]

Not only made it through but read it a second time! Thanks so much for this lesson. Where can one find that sextant?

Thanks,

Greg

Right here:

http://www.kronzky.info/fs/sextant/

Couple tutorial linked on that page, and the documentation that comes with the gauge itself (in the form of a .chm help file) is pretty good too.

I've toyed worth the idea of making a youtube tutorial about all this stuff, because there were a few gaps I had to figure out and full in for myself. Maybe an example flight showing how I plot LOPs on Google Earth etc... but I'm not sure there'd be much interest, and I'm not sure I'm THAT motivated!

 

[Flyboy208]

I am very impressed !!! Mike

 

[PhantomTweak]

As has be said, truly impressive feat!
A friend recently did a similar flight with the same Nav method, including that sextant which you used, in a DC-3. He stayed over land, though
His thread is in the Multiplayer forum over on Flightsim for any that may want to see another amazing man's accomplishment.
My father was an old-time pilot, started in AT-6's in the Navy as a Flight Instructor. He tried to teach me Celestial Nav, day and night both, but that was long and long ago. I was very young as well, and head other things (girls mainly) on my mind, not Navigation.
Thanks for a very enjoyable post(s)
Pat☺

 

[roger-wilco-66]

Thanks for sharing this and document it so well! I love FSX / P3D for these capabilities.
I've got to try this on my Connie or C-47 flights as well.


Cheers,
Mark

 

[hairyspin]

Thanks for sharing, it's properly interesting when someone picks up the old methods and shows they still work. There's something beautiful too about the star names which satnav lacks, like the centuries of seagoing tradition which was based on experience, not electronics. Much appreciated!