[NSRCA-discussion] Futaba and pattern ( A plus for Futaba)

[Bob Richards]

Jim,
   
  IF the stick centered roughly in the middle of one of the "steps", such that slight pressure either way would not force the stick to either of the adjacent steps, then, yes, the 256 would mask the poor stick. If, however, the stick centered roughly on the border between two steps, then the 256 resolution would highlight the problem since the minimum step size of a 256 system is four times the amount of a 1024 system. This is the problem I was having. Some days it seemed as if "center" was in the middle of a step, other days it was bouncing between steps. I started recalibrating the sticks every time I went to the field, but then gave up on using the radio.
   
  I still say that using 256 resolution doesn't really "mask" anything and/or make it seem better. If 1024 does, as you say, emphasize problems, it is because you have that problem whether you are 256 or 1024 or even analog. I will never buy the argument that going to lower resolution is a solution to anything, like poor stick centering, servos drifting, pushrods expanding or contracting with temp, sloppy linkages, etc etc. It is simply one more thing in the long list of things that can contribute to the control surface not being where it should be, IHMO.
   
  I like the idea of programming a deadband for the stick. I will have to try that when I get a chance. But, for you to be able to trim the neutral in .5 microsecond steps, you can't with a 256 system. :-)
   
  In a well trimmed pattern plane, there was no comparison between 256 and 1024. You have to remember, I really wanted the MP8K to work for pattern. I ordered that radio for that very purpose. I was very disappointed when I finally came to the conclusion that it would not be optimum. 
   
  My MP8K was FM. In fact, when I sold the radio, I kept the FM module and rx so I could eventually use them with my old Silver Seven radio (which I still have).
   
  Bob R.
  

James Oddino <joddino at socal.rr.com> wrote:
  Bob,  

  I would think that 256 resolution would mask poor stick centering.  Think about 3 step resolution. Right, neutral left.  When you select neutral it is always the same unless you move the stick to the extreme where you get right or left.  The problem is you can't trim neutral unless you do it with another channel like folks did back in the reed days.  Move the servo with another servo.  

  You had many factors affecting trim in your 8 bit system starting with drift in the servos and an AM link that introduced large changes in pulse width to the servos as signal strength changed.  You could adjust linkages everyday and not keep up with everything that was fighting you.
      So if low resolution masks poor stick centering then more resolution must emphasize it.  This has always bothered me.  I learned long ago that the servos can resolve pulse widths that would appear to be in the servo deadband.  In the early dual rate schemes we had to adjust the neutral if we changed the dual rate setting.  I tried using a servo to do this.  Adjust the pot until the servo didn't move as I switched the dual rate switch back and forth. Go fly.  Every time I threw the rate switch the trim would change.  Went home looked at it on a scope and sure enough the pulse width was changing.  Adjusted it so it didn't and the trim didn't change with the position of the rate switch when I flew.
  

  This week I got a Futaba 14MZ.  I checked the stick centering with a scope.  Not perfect.  Plus or minus a few bits.  Now a bit in a 2048 is only about a half of a microsecond change in pulse width to the servo so most folks don't care.  However I found that with the line or spline curve programming of the control sicks I can put in a plus and minus .5% deadband in the stick.  Now the centering is perfect no matter how I return the stick to neutral and I can trim it in .5 microsecond increments.  
  

  I love this high tech stuff.
  

  Jim O
  

  
   



-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.nsrca.org/pipermail/nsrca-discussion/attachments/20071224/d9ceb43a/attachment.html