APRS Turn Slope Units

There’s a fundamental error in my writeup about setting the APRS Smart Beaconing parameters for the bike trackers: I blundered the units of Turn Slope.

Rich Painter recently explained how that works:

I ran across your blog on Smart Beaconing and saw something that needed correction.

You state the Turn Slope is in units Degrees / MPH

This is incorrect. Although the term Turn Slope is not a real slope (such as rise/run classically) that is what the originators used albeit incorrectly. They do however correctly attribute the units to MPH * Degrees (a product and hence not really a slope).

In their formula they calculate a turn threshold as:
turn_threshold = min_turn_angle + turn_slope / speed

Looking at the units we see:
= Degrees + (MPH * Degrees) / MPH

which yields
= Degrees + Degrees

Which makes sense. It is too bad that the originators used the wrong term of Turn Slope which confuses most people. A better term would have been Turn Product.

In looking back over that post, I have no idea where or how I got the wrong units, other than by the plain reading of the “variable name”.

As he explained in a followup note:

As for units… I was introduced to making unit balance way back in 1967-1968 science class in HS by a really fine science teacher. It has served me all my life and I’m thankful for that training.

I have ever since told that teacher so!

A while back, our Larval Engineer rammed an engineering physics class head-on and sent me a meme image, observing that I’d trained her well: if the units don’t work out, then you’re doing it wrong.

Yes, yes, I do care about the units:

Give a shit about the units

Give a shit about the units

  1. #1 by Ken Davidson on 2015-01-19 - 09:07

    I am constantly teaching my kids to do unit math when they can’t figure out what to multiply or divide by what. It’s just plain magic.

    • #2 by Ed on 2015-01-19 - 09:21


      The HP48 & HP50 calculators have units and combine them throughout a calculation, but the entry method is so clunky and the display so cramped that there’s no real point. I suppose the TI calculators work much the same way.

      Just writing down the numbers with their units pretty much tells you what to do next and lets you know when you can stop turning the crank. Why are engineers the only people who know this? [sigh]

  2. #3 by david on 2015-01-19 - 23:12

    Do you have an opinion of the hp50 / 49? Some Horrible Person stole my flight bag last year, including my much-beloved hp48g, and I need to replace it with something

    • #4 by Ed on 2015-01-20 - 09:19

      I got that HP50 mostly because it’s the only real RPN calculator available, not because it’s such a wonderful thing. A comment on my post about the screen protector sheet suggests there won’t be any more HP calculators, so this will be my last:

      • #5 by Red County Pete on 2015-01-20 - 22:03

        HP still has a calculator website, but several seconds research doesn’t tell much about the future of the business. I’d guess it will be sold as HP splits again. Hey, if Moog is willing to recreate its modular synthesizers (for a non-synthetic fortune), somebody might keep HP calculators alive. Maybe.

        My 32mumble lost its LCD, and I tried replacing it with a TI. Too many decades of RPN, and I just can’t handle algebraic in calculators. On the gripping hand, my 41CX and the 15C are still going, though the CX uses “N” cells–not my favorite. The 1973 vintage HP45 died a few years back. Loved the monolithic LEDs.

        I lusted after the HP97 desk calculator at our department, but never got one for myself.