[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Paparazzi-devel] apogee board 2.0 - feature questions

From: Prof. Dr.-Ing. Heinrich Warmers
Subject: Re: [Paparazzi-devel] apogee board 2.0 - feature questions
Date: Sat, 17 Oct 2015 10:22:14 +0200
User-agent: Mozilla/5.0 (Windows NT 5.0; rv:12.0) Gecko/20120428 Thunderbird/12.0.1

Dear Jan,
in the data sheet a current of 1.4 µA only for the 32Khz oscillator is given. To use a special chip on a small separate pcb connected via i2c is a better way.

Is there a chance to use a vibration resistance accelerator sensor chip like the MXR9500 as option? The use of this chip gives a better flight behaviour mentioned by the KroozSD- and Mikrocopter- autopilots.
In the first  mikrocopter-autopilot only the z- axis is used by this sensor.

Is the coupling from the power circuit a mechanical or a electric problem?


Am 14.10.2015 20:23, schrieb Jan Čapek:

let me comment on this: Gautier was pointing out to eliminate the backup
capacitor since the RTC could get synchronised from the GPS or from the
ground station if no GPS is available. We have tested the gold cap
setup on Apogee 1.0 and it doesn't seem to hold the RTC system backup
for a very long time. It looks like the power is gone within 0.5-1hour
after the system has been disconnected from the main power source. I
would be interested in other people experience. Unfortunately, the STM
datasheets don't seem to specify the expected current that is being
drawn by the battery backup subsystem. We have tested the same goldcaps
separately outside of the board and they retain their voltage for days.

Removing the goldcap would give us some extra space to move
the IMU further away from the switching power supply - thanks Gautier
for this valuable input.

Back to your question - the port has fully integrated RTC - external
32kHz oscillator and the system enables/uses the interal RTC subsystem.
And it will do so even if the goldcap will get removed. Just the source
for synchronizing the clock will either be GPS or the ground station.


  On Wed, 14 Oct 2015 17:26:34 +0200
"Prof. Dr.-Ing. Heinrich Warmers"<address@hidden>  wrote:

is the port for the additional clock  32 kHz free ?
If this is true you can use the inside RTC of the STM processor by
adding a clock resonator, a gold cap a resistor and a diode.


Am 12.10.2015 22:23, schrieb Gautier Hattenberger:

As a side note, I would suggest to remove the RTC backup capacitor
to save some space on the board. The clock is set using the GPS
time and the backup is used to set the time and date of log files.
So having a backup is only needed if you fly without GPS not too
long after having a GPS fix in a previous flight. Worse case, the
date of your log file is wrong.
Even this could be fixed with some messages from the ground station
to have an approximate time.


Le 12/10/2015 17:59, Jan Čapek a écrit :

thank you for the feedback, please, see my further comments below
On Mon, 12 Oct 2015 16:44:04 +0200
Michel GORRAZ<address@hidden>  wrote:

Hello Jan

- IMU will be updated with MPU9250. This will eliminate need for
    external magnetometer

Apogee v1.00 is MPU9150 compliant (it includes a _crappy_ mag) but
using a MPU9250 is a good idea.
If possible, put the 9250 as far as possible from the switching
power supply.
Good point here, too, not sure if there is space on the board for
this to be moved. Let's see what we can do here.
- SD card sold is discontinued, and will be replaced by the one
    currently in production by molex

New Molex ref is 503182-1853 (exact replacement of 503182-0853)
Yup, this one will be used instead.
- mounting holes - we find the 1.8mm holes pretty difficult for
    mounting the board. The holes will be changed to 2.5mm

v1.00 holes are 2mm, if you really want to increase the diameter,
use 3mm, easier to find than 2.5mm
You are right, however 3mm holes already seem to leave too little
material on the edge of the PCB. We will consider that, too.
- servo connectors - we would like to turn the 3 pin connectors
180 degrees, so that pin 1 corresponds with servos signal. This
will make the board consistent with the standard molex
1.25->futaba/JR harness cables


For some historical reasons, GND is always on pin 1 of all
connectors, so doing this you will be hobbyking compliant but not
pprz anymore (hello magic smoke)
You better buy a tool to swap the hobbyking's harness molex pins :
I am aware of this, swapping is not a problem. However, a regular
user may not notice that reverse layout. Magic smoke won't happen
in either case since the power+ is always in the middle, in the
worst case you would just ground the servo signal. We will
consider your objection.

- power connector - we find the power connection to the board a
bit difficult without a proper connector. The 4mm connectors (the
distance between the 2 pads) are rather too big for such a tiny
board. We will place a 4 pin 1.25mm molex blade header - pins 1,2
as Vdd+, and pins 3,4 as GND. This will still leave the option to
solder wires directly to the board.T

Don't forget the current of the entire aircraft (servos, modem,
GPS...,) will go through your tiny 1.25mm Molex (0.8A max with
32AWG cable)...
That is why we would use 2+2pins. Anyway, I will consider if there
are other options for the connector.
Good luck for routing ;-)
Thank you for all the valuable comments,


Paparazzi-devel mailing list

Paparazzi-devel mailing list

Paparazzi-devel mailing list

Paparazzi-devel mailing list

reply via email to

[Prev in Thread] Current Thread [Next in Thread]