Open question for Howard at FMA

[vanguy]

Howard, given the discussion on whether CP2 regenerates the signals or act only as a pass through. Can you shed some light as to what the anwser is.

[Howard Matos]

Originally Posted by vanguy

Howard, given the discussion on whether CP2 regenerates the signals or act only as a pass through. Can you shed some light as to what the anwser is.

Hi,

"Direct from the Engineer"

O.K. You asked for it. When I posted in the past that CPII computer
"passes through" receiver signals when the remote is OFF, this was a
simplified explanation of how the device works. In actuality, the computer
module reads inputs from the RX, and then in effect mixes these with
compensation derived from the sensor data in the firmware. A new serial PWM
is then generated and fed into a shift register. The outputs of the shift
register feed each of the SVO outputs. Most any RX with uC's and failsafe
capabilities work in much the same manner converting RF to digital words,
and then generating PWM outputs based on whether or not RF is present and if
so, the data that the transmission carries. The following highly technical
information should answer any question you may have about how and why CPII
generates PWM outputs as well as commentary about it's ability to achieve
failure analysis:

Startup:
CPII outputs 1.7mS pulses if nothing is connected to the input. In another
thread, a customer has reported 2mS, but I think possibly his o-scope is not
calibrated properly, or perhaps he was rounding off the pulsewidth for
purposes of illustration. There is a reason for this carry over from the
original CPD4. Some people mistakenly don't connect all the control wires
to the receiver. For example: Suppose the user didn't connect the elevator
wire to the receiver. Or suppose they had an intermittent break in the
elevator wire at the receiver. If they powered up CPII and it centered the
elevator automatically AND the user didn't actually verify that the elevator
is working before takeoff, there would be a certain crash. Making CPII move
the elevator to an extreme position notifies the user that the elevator wire
isn't connected or is broken.

Failure analysis of CPII:
1) An elevator wire is connected at takeoff but breaks in flight. CPII
will do last frame hold. The 1.7mS output only applies on a hard start when
no receiver signal is detected. It takes 5 pulses in a row to detect a
valid receiver signal on the channel.
2) An elevator wire isn't connected at takeoff. CPII moves the servo to
extreme position on a hard start showing the user the elevator isn't
connected.
3) Firmware is corrupt. CPII verifies a firmware checksum.
4) Setup is corrupt. CPII verifies a setup checksum.
5) In flight brown out occurs. CPII restarts in less than 10mS and
issues an error code.
6) Poor sensor connection. CPII verifies a 16 bit CRC from the main and
vertical sensors every 10mS. All network communication has a 16 bit CRC.
7) Receiver voltage below 3.3V. CPII shuts down the sensors before they
are affected. The surfaces go to neutral.
8) Intermittent connection to the receiver. CPII verifies a valid receiver
signal on each channel before it is passed thru to the output. If the PWM
is out of range, CPII will output the last valid frame.
9) Firmware hang. CPII has a watch dog timer running on an independent
oscillator. WDT events will soft start CPII and issue a system error.
10) Possible 3D mode activation. All three 3D (non level) modes must be
armed in the appropriate flight attitude first. Flipping the remote to
inverted will not invert the aircraft from level.

How CPII handles PWM signals:
1) CPII has dedicated hardware timers on each receiver input. This timers
are handled in interrupt at assure that are processed in time. Resolution
is 0.5uS.
2) CPII output PWM to the servos thru a shift register. The shift register
allows a single hardware timer to output all 4 servos with 0.5uS resolution.
3) CPII processes the signals in 16 bit 0.5uS resolution.
4) CPII uses software passwords to assure the correct flight mode is
running.
5) To minimize the PWM delay turn CPII, the output PWM refresh rate is
synchronized with the input PWM from the receiver. This makes the
processing delay 1 frame long (typically 20mS).
6) CPII will probably handle future 4mS frames from faster receivers, but I
have not tested this yet. Firmware updates might be required.
7) The receiver inputs are Schmitt triggered.
8) Once 5 valid servo frames are detected in a row, the input is passed
thru to the output.
9) CPII has a soft start detection circuit. A soft start will not move the
servos.

[Teej]

Thanks for the info for us techies.

Yes, my scope does read a bit below 2.0ms... 1.7ms is very believable. Due to limited ranging options (If you've seen my scope in the other thread, you'll understand...heh...) it's difficult to read that precisely.

[jpe9]

Bonjour, j'ai vraiment de gros problème de réglage, en effet lorsque je suis en mode OFF la queue fait yoyo( haut/bas), et surtout je suis obligé de trimer en début de batterie et de trimer en inverse au fur et a mesure que la batterie se décharge (hélico elec).
Je ne comprends pas les réglages.
Hello, I really have of big problem of regulation, indeed when I am in mode(fashion) OFF the tail makes yoyo (top / bottom), and especially I am obliged to toil in the beginning of battery(drum kit) and to toil there opposite in the fur and has measure that the battery(drum kit) passes off (copter elec).
I do not understand the regulations
JP

[Howard Matos]

Originally Posted by jpe9

Bonjour, j'ai vraiment de gros problème de réglage, en effet lorsque je suis en mode OFF la queue fait yoyo( haut/bas), et surtout je suis obligé de trimer en début de batterie et de trimer en inverse au fur et a mesure que la batterie se décharge (hélico elec).
Je ne comprends pas les réglages.
Hello, I really have of big problem of regulation, indeed when I am in mode(fashion) OFF the tail makes yoyo (top / bottom), and especially I am obliged to toil in the beginning of battery(drum kit) and to toil there opposite in the fur and has measure that the battery(drum kit) passes off (copter elec).
I do not understand the regulations
JP

Salut,
S'il vous plaît essayez ceci, débrancher le capteur principal pour désactiver complètement CPII, même si à distance est ON ou OFF et voir si l'oscillation disparaît.

[jpe9]

Bonjour, merci pour votre post en français, c'est vraiment sympa.
J'ai débranché tous l'ensemble de CPII = hélico parfaitement stable en stationnaire sans oscillations.
J'ai rebranché le CPII et mis sur OFF, auto-trim sur OFF = OK, lorsque je bascule le CPII sur ON = l'hélico part en arrière, je suis obligé de trimer.
Auto-trim sur 6° CPII sur ON = trim, passage sur OFF = retrim

Je pense que ceci est peut du à une mauvaise position du capteur principal, à modifier avec les angles.

Par contre pourquoi je suis obligé de trimer dans un sens et après dans l'autre au fur et à mesure que la batterie se décharge, je n'ai pas eu le temps de réagir et j'ai cassé le train d'atterrissage.
Faut-il traduire en Anglais ??
JP