LORAN

[connieguy]

In the course of assisting Jorge with his weather gauge I have come across this extremely detailed site on the Loran system, which I don't think has been posted before. If it has Tom can delete this thread. There are detailed histories of (I assume) all the stations and if you zoom into the coordinates in Google Earth you can sometimes see remains of the buildings. I assume that it is not possible to recreate the Loran system in FS9 or that there would be little point in doing so?

www.loran-history.info/default.htm

[Tom/CalClassic]

Hi,

We have talked about LORAN and this site before, but I think it bears repeating.

This 1955 LORAN manual discusses the theory of LORAN.

www.loran-history.info/chains/Loran-A/Operational_Instructions/A-OI-GLLKA-JUN-1956-CG-222-Amend-4.pdf

Basically, there are two radio stations, called Master and Slave. The Master station sends out a radio pulse and when the Slave station receives it, it waits a certain amount of time and then sends a pulse of its own. These two pulses are received by the plane or ship, and from the time difference between the two pulses (typically measured on an oscilloscope screen) the navigator can determine a line of position, similar to when using celestial navigation. If you can receive a second pair of stations, another line of position can be determined and where those two lines intersect is your position.

A picture of this theory is on page 18-4.

The problem is that these lines of position are hyperbolic (i.e. curved), depending on your angle from the two stations. This is displayed in the figure on page 18-7. Therefore, to be able to determine these lines of position you would need a LORAN chart that plots these lines on a map of the area. These charts are somewhat hard to find.

Here are a few examples of such charts (do a web search for Loran Chart for more):
North Atlantic:
i.redd.it/5gwonjnihcq21.jpg
davidrumsey.georeferencer.com/maps/84736626934/
North Carolina:
www.greasechart.com/images/AC002_lrg.jpg
Scotland:
www.radar-room.co.uk/html/loran.html
San Diego:
www.amazon.com/NGA-Chart-18766-Santos-LORAN-C/dp/B00JV5IWJQ

I guess it would be possible to set up such a system in FS, if you had these LORAN charts, but you would also need the frequencies of the various stations, and some way to read the distance between pulses. But it would be a massive undertaking, and quite difficult programming all the curves, which would probably have to be done one by one (or using math beyond my capabilities). That could take years just for the North Atlantic?

Because of this, most of us who wish to simulate using LORAN just peek at the GPS about once every 20 minutes or so. The typical accuracy is 1% of the distance, so if you are 500 NM from the station (range was around 750 NM in the daytime and up to 1400 NM at night), your typical accuracy would be about 5 NM. That's pretty close to our little GPS gauge accuracy.

Hope this helps,

[blockwood]

The more modern Loran-C receivers gave you a readout much like a GPS without the need for special charts. Work has begun to use eLORAN as a backup for the GPS that provides the same thing. For flightsimming, the GPS is fine to simulate LORAN. As Tom has said already, you would need specialized charts and a whole lot of programming to simulate the LORAN that was available in the 1940s, 1950s and 1960s. I was an Air Force Navigator and learned LORAN-C in training and it was a pain to master. LORAN-A, which is what you see in the classic period, was even more difficult.

[connieguy]

Thank you gentlemen. That is rather what I expected because I knew that Loran was very similar to Gee.

[Jorge]

All,

I have a few minutes before sacking out due to an early day tomorrow (0330 EST), but just wanted to say that the current "LORAN" function that Finn made into the gauge is just that - a "direct-to" function that plots a line in the gps to the station coordinates that are placed on the right hand side of the screen. The concept of the LORAN gauge can be done by hitting the "map" button on the right of the "bar" that has the station information. If you need another LOP, you change station to something in the area and hit "map" again and get another "direct-to" line. If you use Google Earth you can plot the first, get a "direct-to" the second and plot that one in Google Earth, etc. You can do as many as you may need.

Ken, I sent you a response just now.

Once the "main" HFDF gauge is complete I wouldn't mind trying to make one that would basically be the "direct-to" type by itself to simulate LORAN - or at least try to, anyway. The code that Finn put together means that you can do the direct-to function to any place that you can think of on the globe. All you have to do is make the "database" set in the gauge labeled SYS1, SYS2, etc. This is because the limit to data in any one section is 50k in bytes, so the "<Update></Update>" section that drives the information is limited to 50k, not counting comments. As for frequencies, you can put anything. The only reason we're using the NDB type frequencies is because that is what is normally associated with things pre-VOR, but it can be anything when you take into account the information that is being used in these gauges. You could, for example, make an entry that like this based on what is currently in use:

51.9929361 (&gt; L:Sta_Lat_Adak, number)
-176.6126278 (&gt; L:Sta_Lon_Adak, number)
1L2 (&gt; L:Pair1_Adak, string)
380 (&gt; L:Pair1_Rng_Adak, number)

and then have something like:

(L:Pair1_set, string) (L:Pair1_Adak, string) ==
if{ 380 (L:Distance, number) (L:Pair1_Rng_Adak, number) &lt;=
  if{ plot/go "direct-to" Adak
     }
   }

else{ nothing happens }

Not sure if this would work since I'm still learning, but I'm basing it on some of the code I can think of off the top of my head that Finn used in the HFDF gauge. Again, not a LORAN per-se, but it should do the job (maybe). If the info in the Pair1 that is set (i.e. the frequency windown in the HFDF) matches the info for the station then check the distance to the station (using an fs9gps: function that exists in the HFDF gauge to give you bearing and distance which is too in depth for me to remember right now, but it's there) compared to the range the LORAN pair has. If within this range (i.e. less than or equal to) then plot the direct-to.

I'll have a look at the code in a few days when I'm back home, but it doesn't seem too off from what the HFDF gauge is doing now. Is it LORAN? No, but we can imitate the system's LOB capability with the GPS fairly well as Tom and the other poster said. With the two station system you could come up with a code that says, "once you plot direct to this station go and plot direct to this other one" which - I'm guessing - would simulate the two station concept of taking a bearing for the aircraft? Maybe?

Well, it's late. Off to bed. More to come (hopefully) in a few days!

Regards,

Jorge
Miami, FL

PS: If I'm spewing exaust please let me know. You guys that know more about xml can probably correct anything that I'm doing wrong here, please!  Also, let me know if the ampersand "l" "t" and ";" worked in the message above.  Thanks!

[Tom/CalClassic]

Hi,

I think it would be workable, AFAIK.

Thanks,

[Defender]

Ken, Jorge and Tom,

Many thanks for all you're doing on long range navigation methods. Here's some useful charts you can download, either for LORAN details or just flight planning. They're very high resolution so you can zoom right in.

North Atlantic historicalcharts.noaa.gov/image=3071-7-1955

North Pacific historicalcharts.noaa.gov/image=3094-12-1956

Caribbean historicalcharts.noaa.gov/image=3073-8-1956

Bill

[connieguy]

Many thanks Bill. Although I know that site I had missed the very interesting Caribbean chart; the Loran one I had seen but dismissed as of no practical use for classic flight simulation in FS9. Thanks also to Tom for earlier posting the link to the most informative Loran manual.

        This post describes a flight using Jorge's version for FS9 of a gauge originally made for FSX/P3D and available on the A2A forum. It is principally intended to overcome the FS9 feature which limits the maximum range of NDBs to about 150nm no matter what range may be specified for them. The new gauge gives access to ocean stations, to a number of HFDF beacons and to 4 Loran stations, although these act purely as radio beacons rather than representing Loran as it really was. Jorge is now involved in further exciting work which may or may not come to fruition, but of that it is for him to speak when and if the time comes.

         The flight takes place in a Jahn/Connie Team MATS C-121C on 17th September 1956 and is from one of my favourite airfields, Burtonwood in the UK, to Lajes (LPLA) in the Azores, a distance of about 1300 nm. I have chosen it because I did the same flight not long ago using Tom's Ocean Station Kilo at the appropriate point. Departure is around 16:30, which means that the sun sets during the flight and in the later stages the stars are (with any luck) available for navigation. Prior to that the flight plan takes me down the Irish Sea, guided by NDBs at Strumble on the Welsh coast and Tuskar Rock and Old Head of Kinsale on the coast of Eire. As is my usual practice, I am using a FSGRW dynamic weather file starting on this occasion at 16:12 on 17th September, 2019.

Ready to go. Burtonwood scenery by Glen Broome, modified by myself.




Take off was at 16:32. We climb to FL140 over the Welsh mountains.



Cruising at FL140 at a constant power of 1700 BHP with throttle, rpm and mixture settings long familiar from frequent use of this aircraft. The rmi needles point to Strumble to port and Tuskar Rock to starboard.

         Two or three minutes later I have opened the new gauge (via an invisible click spot over the left hand manifold pressure gauge), switched the unit on and tuned it to the HFDF beacon at Foynes, near Shannon. Tuning is done by placing the mouse cursor over the station list at the bottom of the gauge and scrolling up or down until the desired station appears. This is quick, and when the station has been found clicking on the three digit number above automatically produces the required frequency. Then the compass heading on the gauge is adjusted to the compass heading of the aircraft, and the loop aerial is turned until signal strength is at maximum. Foynes is about 140nm away but the signal is good and the needle on the compass points to it. With a loop aerial it is necessary to be careful that one has not tuned to the reciprocal position, but providing that you have a good idea where the target is and have not been drinking before take-off this should not be a problem.

      The Welsh coast on the port beam. I should not be as close to it as this, which is a warning that I am slightly off course already. If I had followed my usual practice of flying out on the reciprocal of a beacon (the Burtonwood Range would have done) this would not have happened.



         Almost two hours later and the sun is about to set, but I have picked up Ocean Station King (later Kilo). I pause the sim and go into Plan G to measure the distance - it is 161nm. However, on an earlier test flight it was about 190nm away. I can explain this discrepancy. In this gauge the Ocean Stations only transmit for the portions of the hour shown in red on the UTC gauge clock. If one just misses the contact and then flies for at least another ten minutes a Super Constellation moving at 250KTAS will travel a fair way and the station will eventually be picked up rather later and at a shorter distance than it would have been had events fallen out otherwise. Note also that Jorge says that the range is determined partly by the altitude at which the aircraft is flying. As it is the compass needle is to port but not by as much as it should be, so our heading needs adjusting to starboard. I am happy to leave it at that but a real navigator would have done more, because by now the moon is visible. Thus a sextant shot could have been plotted on Google Earth accompanied by a line of position derived from the magnetic bearing to the ocean station, once that bearing had been adjusted from magnetic to true.

        Later still. Ocean Station King is now almost on the port beam and much closer, as the signal strength shows.



       Attention now turns to the Azores HFDF which is on an island about 75nm west of the position of Lajes. Not being an Ocean Station this transmits continously and I pick it up at 20:01 UTC, again going to Plan G for the distance. It is 581nm away. The compass needle is in about the right place given that distance.

        The crucial part of the flight is now over. As we get nearer to Lajes the Lajes Range (my own creation) is picked up on the rmi and I descend for a good landing on Runway 33. The night is clear and the heavens are full of stars. Unfortunately, in a sense, they are not necessary. The landing:

       Taxiing in at Lajes. Scenery by John Stinstrom.

        Jorge's gauge works flawlessly and brings to FS9 a facility that some have long desired. Congratulations to him and comments from others welcome.

[Jorge]

All,

Thanks for all the help and work on the HFDF gauge. After six months or so it's been nice to see that all came to fruition. I had the idea of "retrofitting" when working on Platinum Wings for FSX and thinking that it would be nice to have the gauge in FS9 for Golden Wings. I then looked at the default Vega and realized someone else had stated that in order to get the ADF gauge to work in FSX you needed to change some of the lines in xml. I wondered if the same type of "logic" worked both ways - change some of the lines in xml and you can get a gauge to work in FS9. Platinum Wings went on hold (possibly for a while since I'm thinking of going back to FS9 from FSX for the thing) so I could figure out how to do that. Six months later ...

In the process I've also wondered if LORAN may be possible in FS9. Ken and Tom have been great support on this, so I figured to give it a shot. According to Bob at FSDeveloper (the guy that wrote the great GPS book) the short answer is "yes" it is. The big caveat is how to write the "miles" of code required to achieve all the math to accomplish hyperbolic equations. Apparently xml can handle the math. Unfortunately, I can't. Not at my stage of programming, so a better answer would be: "not yet". I can, however, try to accomplish the end result of the system. LORAN-A gave navigators a chance to find their latitude and longitude based on the hyperbolic lines of radio waves that were transmitted from several stations that were "paired" so that the pulses were timed a certain way. By "matching" the pulses, you could eventually reference a book/table/map that would tell you how to interpret the information and give you the assumed position.

My system will not be that complicated. Thanks to Tom and Ken, I've been able to get a heap of input and it looks like there may be something down the line. No timeline promises, but definitely a possibility we can have something.

With that being said, I've dropped version "1" and have started on version "2" of an attempt to create the LORAN-A system on or about 1955. So far I have the data in an excel spreadsheet for the U.S. East Coast "Chain" (the -A didn't use chains, per say, but that's the name on the history website) that includes Nantucket, Cape Hatteras, Hobe Sound, Folly Beach, and Sandy Hook. The graphics are done for the AN/APN-4 receiver, the first practical unit for airborne use. Many of the post-war airlines would have used these as surplus units on their DC's and Lockheeds. Many British aircraft would use this type as well since it was basically an American version of the British GEE receiver. I this works well (and I can get pictures of the British one) I'll try and make background images for the gauge to turn it into the British one if there is interest (again, after this one is actually working). The images are based on the one found online at the National Air and Space Museum website and that is on display in Washington, D.C. at the museum. Here is a tentative "work in progress" of the "navigator" station:



No real coding yet for this, but by doing copy and paste I can probably get the knobs to work and the switches to flip. The real "bear" will be to get the thing to interpret the information from the "system" gauge (the one on the main panel that is invisible) since I ran into some hiccups with version "1" already. Per Tom's advise, I'm going to make it as "simple" as possible with regards to the information, but the user will have to go through some steps in order to get the final answer of where the airplane is.

The plan is to have the person turn the unit on. You have a switch for 20V and 115V up top. I'm hoping to get the thing to be like the CalClassic aircraft that let you use a type of GPU so you can "use" the unit on the ground on 115V, but MUST turn the thing to 20V when the engines are on (i.e. internal power). When you first "tune" the station pair by setting the Channel, PRR, and Station, the gauge automatically calculates and store two sets of aircraft fixes based on the Azimuth1 and Length function of the FS GPS. These are then compared for distance and the one with the longest distance is the one that is used as the "final" set of coordinates. You then go through several steps that simulate the sine wave on the scope which you have to line up using consecutive sweep settings. Each sweep setting is more precise until you get to the last one which activates a mouse area which will enable you to get to a "chart" image. On the chart image there will be a "PLOT" "button" that you then click which will place the assumed lat/lon on the clipboard hanging from the cork board, keeping in mind that this position is the one determined at the beginning of the process. Also, the "map" function of the FS map will draw a line that also gives a distance, track, etc., on the GPS.  Not sure how to get around this for now since it's what is done with the "map" button on the HFDF gauge.  The time it takes to "line up" the sine waves and then get to the "plotting" will be the "error factor" in your position since the lat/lon displayed is where the aircraft was when you first tuned the stations. If you need to start over, then tune another pair and re-tune the one you want to reset the lat/lon. Right now the "map" will be a static bmp file until I can figure out the whole overlay thing to put an "X" on the static background. Considering most folks use Google Earth or Plan-G, I'll probably just leave it as a static bmp file from the different maps already provided on this post.

Of course, you could always just hit the GPS button, but that wouldn't be sportsmanlike.

Hope everyone is healthy and well, especially with all the stuff going around these days. Stay safe and hope to have some updates soon.

All the best,

Jorge
Miami, FL

[awralls]

Jorge,

This is potentially the answer to a prayer. Any chance it will work in FSX?

Andy

[Tom/CalClassic]

It should?

[Erik]

Gentlemen, a big thank you for all great things going on and visible here through yet another fascinating topic. Apart from some tests, real life has kept me from simflying for a year already, I recently discovered. However, staying tuned to the CC forum keeps the feeling alive, so to say. All your efforts are much appreciated!

Erik

[herkpilot]

The man behind LORAN as well as airborne RADAR was a retired Wall Street tycoon named Alfred Lee Loomis. The original name was LRN for Loomis Radio Navigation. Almost totally unknown today, he was the man who moved many of WW II s most important scientific developments from ideas to reality using both his (and others) funding and many social connections throughout the establishment. He was obsessively private and stayed as far behind the scenes as possible. The story of his research lab in the 1930s is told in the 2002 book “Tuxedo Park” by Jennet Conant.

[mrcapitalism]

Jorge,

That looks amazing! I had some thoughts on how you could resolve the navigation solution. It seems a lot of these navigation techniques require some way to chart the solution and the intended flight path/aircraft location. The Sextant gauge has a small chart which displays the various LOP's.

Users such as connieguy of course does an excellent job of using sextants to navigate propliners using vintage techniques. From watching the experience on Youtube (I haven't taken the plunge yet, although I did buy and read a copy of the FAA's Flight Navigator Handbook), Google Earth seems an excellent navigation tool for plotting LOP's.

I wonder, perhaps the LORAN system can take advantage of this solution? It might be a little inconvenient for users to be forced into a charting program, but it could add an extra layer of immersion for the experience! My suggestion is to either upload real LORAN charts, or else we could recreate a Google Earth layer which shows the location of the lanes. Instead of you having to code a map, and resolve GPS coordinates, your gauge would do the basics that only the real unit performed, i.e., you tune a station, and use it to locate the current "lane." There, the user uses either real .pdf maps or a GE layer and locates their lane, drawing the appropriate LOP.

[Jorge]

Thanks for all the support. I've been in touch with Ken and Tom to see what input they could have and I must say they have been invaluable so far - even if it's just testing.

I've run into some snags, however. It seems my knowledge of xml is far less than I thought, despite over six months of looking over the stuff. I've managed to get the thing to move dials and flip switches, but that's about it for now. I can't even get some of the knobs to move properly such as the Gain and the Amplitude Balance. I know it's in the coding somewhere. I was hoping that using the other gauge as a basis it would be possible to have something along the lines of this, but we'll see. I've looked at the code that was used in the RR4 gauges for the Radio Range System and was more lost than when I started. So far I've done more head banging than a fan at a Metallica/Anthrax concert, but still am no farther along than I was a week or so ago.

We'll see what Tom has to say, but for now here's a shot of the thing in the VC view of the JBK Comet 2 in RAF colors (I know, only the inside here). The areas of the knobs and switches have tootips that give you info, but - again - nothing that means anything right now since I can't get any information to parse.  The notepad will give you number calculations as well as allow you to put numbers in such as the three microsecond readings you are supposed to get from the scope (that's the plan eventually, anyway).  The red book will be a hotspot for one reference book and the blue one for a gauge-specific thing I'm going to put together with station info, codes, etc.  Also a pop-up from within.  Simple stuff like that I can do so far, just need to work on the graphics for them.



I like the idea of the Google Earth and would be a nice way to go. The real challenge would be to get the actual lines to plot on the Google Earth layout. It would be almost as much of a challenge as the gauge itself!

I'll hopefully have some movement on this soon. The process should be:

1. Turn on unit.
2. Dial in your channel, station, and PRR. This is the three (3) character code such as 1H4 for the Nantucket/Cape Hatteras pair.
3. Flip back and forth the Filter switch to go between information from either of the two (2) stations. This wasn't the real function of the switch, but it serves a purpose here since that seems to be the best way to keep the two stations identified in the pair (i.e. Station A switch down and Station B switch up).
4. Once "tuned" you then ensure you are on sweep number 1.
5. You will see two sine waves on the scope, one above and one below.  They will have a pedestal and a spike somewhere on the screen.  You move the "spike" on each trace over the left part of the "pedestal" image of the sine wave to be displayed.
6. You switch to sweep speed 2 and start placing the sine waves one over the other lined up.
7. Sweep three is the same, but at a different scale.
8. Sweep 4 is the same, but now the waves are on one trace across the screen. The goal is to make them appear as one. You use the amp balance to change the height of the wave and the gain to change the width of the wave until they are on top of each other. In reality, you'll only be changing one wave here since the math would be too much for me to handle to get this done.
9. Sweep 5 is the first of the calibration sweeps. They are 5, 6, and 7. Each is a different scale. Sweep 5 is 50 microseconds, 500 for sweep 6, and 5000 across for sweep 7. You take readings (best guess) across the sweeps and add each of the three guesses. Lets say the first is 25 across, the second is about 200, and the last one is about 2000 across. You enter the three numbers on the notepad and get 2225 microseconds.
10. Using the notepad again, you see that 2225 microseconds is about 360 nm (it will convert the info for you once you put the numbers in from your readings).
11. Check where this may be on the Google Earth overlay (I like the idea of this, by the way) and note the LOP.

12. Repeat for another pair of stations.

13. Check microseconds and distance again in notepad.

14. Access to the lat/lon numbers as well as something you can use in Google Earth "should" be available by this point.

Okay, that's the theory. We'll see how far the lack of coding experience on my part goes. I'll keep working on the graphics while I wait to hear back from those that are more experienced than me on this.

All the best to you all and stay healthy,

Jorge
Miami, FL