[NSRCA-discussion] Futaba R6014FS Receivers

[Bob Richards]

Ed,
 
Good information.
 
The 10k pot establishes the output impedance for these tests (high), so the voltage drop of the signal when loaded by the servo will give some indication of the input impedance of the servos. (The DS821, for instance, obviously has a much lower input impedance than the other servos.) This could be a significant factor when loading a channel with multiple servos.
 
What could not be deduced from these tests is the output impedance of the receiver. How much drive do our receivers have?
 
In any IMAC type plane, I will have some type of powerbox arrangement (my favorite right now is Smartfly). No worry about signal levels at all.
 
Bob R.


--- On Sun, 11/23/08, Ed Alt <ed_alt at hotmail.com> wrote:

From: Ed Alt <ed_alt at hotmail.com>
Subject: Re: [NSRCA-discussion] Futaba R6014FS Receivers
To: "General pattern discussion" <nsrca-discussion at lists.nsrca.org>
Date: Sunday, November 23, 2008, 6:22 PM



OK, I set up an attenuator on the signal to get results on a handful of servos.  Jim O. already did this, but I think it was with just one servo?  I don't have his post anymore.  
 
This test was done with a 10K Bournes precision pot, ground to one lead, signal out to servo the wiper, signal in from the receiver to the other end.  Aileron channel on the R921 used for signal attenuation tests, rudder used to add a separate DS8611 servo load to induce crosstalk.  Added lead length of attenuator network about 1 foot of extension wiring. Signal level measured about midway in the 1 foot of added wiring.  This would be rouighly equivilent to what's happening halfway along an aileron extension in a Pattern bird.  Also, I tried a couple of old analog Futaba servos I had.
 
Results:
 
DS8411A
Loads the test network by about 0.05V drop
Below 1.2V, totally uncontrolled response or no response.
1.2V - Very jumpy, not really controllable, even when the 8611 was silent or disconnected.  Significant crosstalk from 8611 when active.  The lightest touch on the 8611 to try to move it would set off the 8411 on the test channel.  You would crash.
1.3V - Jittery, but controllable.  Noticeable, but improved crosstalk from 8611.  You might make it in, but your shorts may need laundering.
1.4V - Solid.
 
DS9411A

Loads the test network by about 0.05V drop
0.1V better than the 8411, otherwise a carbon copy of it's performance, just offset by +0.1V.  The small difference is likely explained by component tolerances and less noise injected on the power rails because the 9411 is a lower torque servo.
 
DS8611


Loads the test network by about 0.05V drop
0.1V worse than the 8411. It's a much higher torque servo and injects about 0.2V of noise on the ground lead every time the motor moves.  The 8411 injects a little less noise, something like 0.5V less, but I can't measure that accurately.
 
DS821


Loads the test network by about 0.2V drop0.85V - No response
0.9V - wild response (see 8411 notes above about crashing at 1.2V)
0.95 - Just barely controllable, a lot of crosstalk.  You would still crash, but it might take a little longer.
1.0V - A little jumpy, some crosstalk.  
 
NES-4131
Loads the test network by about 0.1V drop
1.3V - Very wild response, frequently drives to the stops
1.35V - Slightly jittery, some crosstalk
1.4V - Fairly solid, milder crosstalk
1.5V - Solid, very mild crosstalk
 
Futaba S3101
Requires 1.5V to work solid
 
Futaba S9602
Works well at 1.1V
 
I also tried adding a 2nd DS series servo to the test channel (added a 9411 to an 8411) while the 8411 was set to work at 1.3V (controllable, but jittery).  I deliberately did not adjust the signal level to compensate for the additional load and observed that it became very wild for both servos.  This is to be expected, since the signal level was already a what I would consider the minimum controllable level. The 9411 added load knocked it down roughly another 0.5V, but also added more noise to the equation.
 
So to summarize, when you get out at longer lead lengths, the effects of signal degration to to crosstalk, capacitive loading etc becomes worse than in my test.  In a pattern plane, not so bad, but in a big IMAC thing, watch out.  Double up on servos on an extension, worse again.  Add higher torque servos, worse again. These are intended as useful guidelines and not absolutes.  I still think the margins are uncomfortably low with only a 2.7V receiver output, but maybe Futaba or JR has a different view.
 
 
Ed
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nsrca.org/pipermail/nsrca-discussion/attachments/20081123/deca6b6a/attachment.html>