A mechanical sensor can detect apogee by detecting the near-zero g
condition there. In a nominal rocket flight, you start at +1 g on the pad,
+multiple gs during burn, then a rapid switch to multiple negative gs at
burnout ("the train wreck"), and then those negative gs bleed down as you
approach apogee and velocity (and drag) drop to near zero. Dan Pollino's
mechanical apogee sensor (described in his books on PVC rockets) used an
ingenious (IMHO) method to do this.
A tilt sensor is a harder problem and I'm glad to see a commercial
magnetometer-based sensor is now available again. I'll be buying a few to
play with.
Seems like it should be theoretically easy to have a gimbaled vertical gyro
in your rocket, "cage" it before launch to prevent drift, then release the
caging at liftoff and monitor its position relative to the rocket body (the
gyro should maintain its original orientation in inertial space). I haven't
tried to build such a thing, and was trying to see if there are any
MEMS-type gyros but they all seem to be rate gyros--they give you an
angular rate output instead of an angle measurement. Any ideas?
Can we borrow anything from quadcopters? Do they have full IMUs or
something simpler to maintain level? Or is it gravity-based and not
applicable to rockets?
On Wed, Jul 22, 2015 at 10:31 AM, Redacted sender JMKrell@xxxxxxx for DMARC
<dmarc-noreply@xxxxxxxxxxxxx> wrote:
The caged ball tilt sensor does have some limitations that we have seen
in the past.
1. High roll rates during motor burn.
2. Vertical flights with backslider stall.
3. <60 degree arc trajectories.
These conditions reduce the reliability of the sensor. Multiple balls have
been used to improve reliability, but this increases the burn out roll rate
issue.
Secondary cage and release systems have been suggested to
improve reliability. I have no data on these actually being tested.
Increased mechanical complexity should reduce the overall reliability. I've
not seen a solution to the high roll rate issue with a mechanical sensor.
A single integrated accelerometer also has limitations. The best solution
is an IMU for apogee detection, but combinations of accelerometer/barometer
or accelerometer/gyro have a higher reliability than any single sensor. I
use an accelerometer/barometer combination.
Krell
In a message dated 7/22/2015 2:18:18 A.M. Pacific Daylight Time,
johndom@xxxxxxxxx writes:
R. Dickinson wrote on 220715
Rick_the fact is, the metal ball switch concept as an apogee detector
appeared interesting when I saw it flight tested & deploying a chute on
YouTube. It is not like the mercury switch which has its contacts in the
top of a tube and like you say fires at burnout when the metal blob goes up
and hits them. It is a little more complicated. Mercury switches have been
used for second stage ignition, not for apogee detection in rudimentary
rocketry.
The ball contacts are not at the end of the tube cage like in the mercury
switch but at the end of V-shaped ball collector gaps perpendicular to the
tube, typically 4 gaps. In which the ball is supposed to roll over at the
moment the rocket tilts or tumbles at apogee. It is a tilt detector.
I intend to verify the tilt option and also test some modern complicated
IC SMD prints which may have drawbacks as well I suppose. I'll let you know.
JD
-----Original Message-----
From: arocket-bounce@xxxxxxxxxxxxx [mailto:arocket-bounce@xxxxxxxxxxxxx]
On Behalf Of Rick Dickinson
Sent: woensdag 22 juli 2015 4:45
To: arocket@xxxxxxxxxxxxx
Subject: [AR] Re: magnetic apogee sensor
John,
Do NOT use any sort of purely mechanical tilt sensor. They WILL NOT and
CAN NOT work, due to basic physics.
The rocket is in free fall, modified by the forces acting upon it. First
force is the thrust of the motor. Second force is drag, which is in the
opposite direction of the thrust. Gravity is already taken into account by
it being in free fall. At burnout, the thrust drops off, and is exceeded by
the force of drag, causing anything not strapped down to float forward.
Your mechanical contraption can sense burnout, but not apogee.
If the rocket flies a ballistic trajectory, and the fins do their job,
keeping the rocket facing in the direction of travel, there is no sideways
force acting on the ball (or at least none that isn't swamped my minor wind
buffeting during flight), so you can't use sideways motion of the ball (or
pendulum, or whatever) to detect apogee.
An accelerometer-based deployment device uses numerical integration of
acceleration values to determine velocity, and goes when calculated
velocity drops to zero.
Any mechanical pendulum or ball and cage or similar device can only detect
instantaneous acceleration. It can't integrate over time to determine when
velocity has dropped to zero.
It will fire at burnout, not at apogee.
Cheers,
Rick Dickinson
On July 18, 2015 10:11:02 AM PDT, John Dom <johndom@xxxxxxxxx> wrote:
So far the only cheap real time apogee actuators I came across are
ingenious metal ball or micro ball tilt sensors. No µprocessor fuss
required. If they are reliable I still have to find out next month or
so.
jd
--
Sent from my Android device with K-9 Mail. Please excuse my brevity.
