ADF suggestions, tweaks, and ideas
-
If this wasn't abundantly clear by the nature of my aircraft, you're talking to one of the other sickos who enjoys old school navigation! Although I would love to implement some of the more advanced features that you list here (the lighting one is a particular challenge that I will be thinking about for days...), I do fear that we are but a sparing few who would truly enjoy it
I am interested, however, in point #2. I see in my code that it is not a 45° angle but a very specific 36°. Before I release the Baron and Bonanza Professional, would you be able to find me documentation that specifies the horizontal orientation of the inactive needles? Do you also know if this is the parked position when there is no power to the unit? I would love to get this correct, if possible, because the 36° angle has always felt wrong to me even if I may have known it to be correct from some documentation deep dive long ago.
Ah ha. Found it on the KI-229
https://www.scribd.com/document/687669843/Installation-Manual-KI229-RMI
-
Oh and several sources have stated that the requirement even for a brand new RMI installation is +/- 5 degrees, both for NDB and VOR tracking. Heading tracking is +/-2
-
Oh and while we're on it and I've got permission to be extra picky, the needles move too quickly overall. Thinking back to my avionics days, the number of 25 degrees of rotation per second sticks out. I'll do some more digging on that too. That would, of course, vary from unit to unit and there was apparently a capacitor mod that could be done to some of the Bendix/King units to speed them but but I never saw that in action.
-
Ah ha. Found it on the KI-229
https://www.scribd.com/document/687669843/Installation-Manual-KI229-RMI
@CdnCptMoustache Wonderful!! Both pointing to the right? I would have assumed one right and one left. So it shall be!
That's some pretty impressive RPN for someone who claims not to be an expert in RPN. The challenge is that I'm already doing a lot of signal attenuation calculations in my javascript based on position relative to the station and its emission pattern, not just the signal strength variable, which I have found to be unreliable in MSFS. In theory, my code should do the exact same thing that yours does, minus the banking effect, and only with random noise, instead of pseudo-random sinusoid addition. The banking effect should be very easy to add. I will see about that sometime soon!
-
@CdnCptMoustache Wonderful!! Both pointing to the right? I would have assumed one right and one left. So it shall be!
That's some pretty impressive RPN for someone who claims not to be an expert in RPN. The challenge is that I'm already doing a lot of signal attenuation calculations in my javascript based on position relative to the station and its emission pattern, not just the signal strength variable, which I have found to be unreliable in MSFS. In theory, my code should do the exact same thing that yours does, minus the banking effect, and only with random noise, instead of pseudo-random sinusoid addition. The banking effect should be very easy to add. I will see about that sometime soon!
@Black-Square Yes, both pointing to the right
That "impressive" RPN was many hours and many times walking away from it. There may have been some whiskey involved too. That was also a year or more ago so I've forgotten how to do it for the most part. However, my plan from the start was to add the bank error in as an after-thought once the main signal is determined.
As a few basics, and to get your head around it, turning errors have a few gotchas. Thinking of it like compass dip helps. At a Relative Bearing (RB) of 270 or 90, you'll get basically no bank error at all. You'll get it the worst at 0 or 180 RB and high over (ie close to) a station. Taking it to the extreme, if you're almost over the station at a RB of 0 and bank 90 degrees to the left, the needle will show 270. Ditto with 180.
The additional fun comes when you realize that pitch angles will behave the same, just the most at 90 or 270 RB. It's conceivable you could be very high over a station with a RB of 0, pitch up enough and watch the needle go to 180 early. But I'm less concerned about that. Generally you're not using large amounts of pitch (as compared to roll) when navigating via NDBs or VORs on the RMI.
I'm sure somewhere someone has made an equation that will figure out the needle indication based on relative bearing, relative altitude, bank angle, pitch angle, etc but I haven't gone looking
-
Since you were on such a roll later, could you take a look for anything regarding the parking positions for the Collins RMI-30? That's the only other RMI that I've modeled, and I've already made the changes to the Bendix/King units in my other aircraft. Sorry to ask for help on this one. I'm trying to fit this in before the Baron and Bonanza are released, but I'm a little busy this week, as you can imagine.
-
Since you were on such a roll later, could you take a look for anything regarding the parking positions for the Collins RMI-30? That's the only other RMI that I've modeled, and I've already made the changes to the Bendix/King units in my other aircraft. Sorry to ask for help on this one. I'm trying to fit this in before the Baron and Bonanza are released, but I'm a little busy this week, as you can imagine.
@Black-Square Super easy. Barely an inconvenience
https://www.scribd.com/document/862377894/RMI-30-MM
Looks like the needles respond very quickly in the RMI-30, upwards of 180 degrees per second
-
As an added tidbit and if it helps with your ADF functionality, an ADF set to ANT mode will mildly boost audio receive while parking the needle at 90 degrees, even with a good signal. It won't point in that mode. Because of that, ANT mode can be briefly used to ensure you are actually getting a good signal and that the unit itself hasn't died or is drifting towards something else
-
Thank you so much for finding that! I'm ashamed at how easy that seems, but when you're trying to keep track of some 20,000 lines of code per aircraft and everything else, it can be very easy to let something like this hold over from years ago. I've been meaning to implement the voice receive mode, since that's something I'm very familiar with myself. I'll add that to the list to do at the same time as the banking error. As you can imagine, the real time sink is not the coding, it's the testing. That's the fun of turning your hobby into a business, unfortunately. Thanks so much! You have been one of the most helpful and knowledgeable contributors of ideas that I've had in a while!
-
My pleasure! If you need any testing done, please let me know. My years of knowledge have to pay off somewhere. Oh and you can expect this level of detail with the Turbo Commander tips too. It's my curse
