[Date Prev][Date Next][Thread Prev][Thread Next] - [Date Index][Thread Index][Author Index]

IDEFIX thanks



Hello John,

thank you for the geetings, but I verily was not alone and IDEFIX is the
result of a real teamwork as you might dream of every day where each one
took its place seamlessly.
You may pay them tribute by looking at the website and keeping on
sending tlm reports.

The pass of last midnight was a real pleasure for eyes and ears as the
sky was clear and elevation almost at zenith and a lot of people were
able to see IDEFIX passing overhead and hear the signals that were very
strong. The visual magnitude was around 0 and even in the lights of the
center of Paris it was as plain as a bright star.
I heard it with a 4 el handheld yagi from my balcony while visually
tracking it.

Last news:
The internal temperature is around 32 deg celsius, decreasing now. Last
monday, it was around -20 deg celsius after having been in total
obscurity for more than 24 hours not even seeing the earth reflection as
indicated by the telemetry. The third stage slowly revolves around
itself and spins too. Both effect combine to affect illumination. In
fact, we expected even worse conditions and the temperature range is
within the limits.
The batteries are in good health.

Starting from wednesday-thursday night, the telemetry channels of CU1
are all zero while the processor still keeps on transmitting frames and
timing the payload. There are three possible causes in the order of
highest to lowest probability :
- thermal crackdown in the pcb affecting the ADC SPI bus or power
supply.
- thermal crackdown in the ADC chip
- chip destruction by high energy particule.
- uP port destruction.

When the payload cools down and the pcb shrinks, we'll have a chance to
recover the telemetry if the first hypothesis is the right one.

As for the translation of the terms:
OPTRO :each payload is fitted with two light sensors which are basically
a BPW21 photodiode behind a  2mm thick PTFE windows giving a cosine
response of incident light flux. The photodiode is wired as a photocell
loaded by a 5k resistor. Under full sun at AM0, the max voltage is
around 150 mV depending on temperature. One the two light sensor is
fitted with a LM335Z temperature sensor to allow for correction (TEMP
OPTRO channel).
On the pictures, the light sensors are the two brass bolt like
extrusions ontop the payload housing (Z+ side). One is besides the X+
side (coax connectors), the other at X- side (arming plug connector).
The description of the light sensors is given at http://www.satedu.net

TEMP BOITIER : case temperature taken at the bottom of the X+ side of
the payload. You can see the exact place in the pictures at
http://www.idefix-france.net. LM335Z sensor.

T ISD : a temp sensor glued on the ISD25120 voice recorder chip. One of
the voice message is a 1800 Hz tone recorded at 20 deg celsius. As the
internal free running clock of the ISD is temperature dependent, the
measure of the received tone frequency gives a very rough estimate of
the internal temperature of the payload. < 1800 Hz => < 20 deg, > 1800
Hz => >20 deg .

MOY ORBITE : on CU2, three channels are monitored for mean, max and min
values.
OPTRO1
OPTRO2
TEMP OPTRO
There are two mean values : a 10 minutes one, which integrates the
values of the last ten minutes and an 90 minutes one. The last one is
reset every 90 minutes while the first one is a sliding mean.
That was devised to get some tlm values around the orbit as it was
impossible to have amateur stations continuously monitoring the sat.
In the decoded tlm channels, you read :
OPTRO1 mean value, 10 and 90 minutes (orbit)
OPTRO1 min value since mean reset and time stamp
OPTRO1 max value since mean reset and time stamp
 and the same applies to OPTRO2 and OPTRO TEMP.

So we need only a full super frame ( 5 frames, each with subframes)
every 90 minutes.
The instantaneous light sensor value (E subframe on CU2, A and B
subframe on CU1) is transmitted during each frame i.e. every 12 seconds
on CU1 and every 22 seconds on CU2 to adequately sample the spin
rotation.

The place of the other temp sensors is visible in the pictures on the
web site.

The voltage of the Li Thyonile batteries is measured after the schottky
coupling diodes (400 mV drop).
On CU1, it is measured during standby, on CU2, it is measured during
transmit.
On CU2, the RF power indication is taken from a diode rectifier located
near the PA output coil and the spacing has been adjusted to get roughly
600 mV at 1.25W ouptut. The indication is temperature dependent and the
TX temp sensor is located just besides the PA which is a BLT50.

The processor is a 80c32X2 on the two payloads.
On CU1, it comes with a 32k ROM and its internal RAM. On CU2, 32kRAM are
added to store the values for the mean, max, min calculations.
On both CU, the processor generates directly the 1200 Hz bpsk Manchester
modulated subcarrier which  feeds the FM TX BF input. It also drives the
SPI bus (1 MAX186 on CU1, 2 MAX186 on CU2), the ISD25120 voice recorder
and TX on/off control. The clock frequency is 6.144 MHz.
The processor also generates an internal time stamp which slightly
drifts with time, that's the reason why the telemetry decoding software
adds the PC date to appreciate the time stamp drift quite exactly. The
drift is regular and has almost a fixed value slightly depending on
temperature. The time stamp is generated with the internal timer.
The processor is placed in sleep mode when not in use to reduce the
current drain.
The processor program architecture is a basically a state machine whose
all states are defined and looped allowing for the processor to reset
and recover even in the case the stack is corrupted by a SEU.

Two switching mode power supplies are used, one for 5V processor and
sensors, one for 7.5V Tx generation. The 5V TX for vco/buffer and synth
(a few mA) is derived from the 7.5V via a linear regulator. The on/off
control of the switcher is used for TX control.

Mechanically, the PCB are located on an H like structure 'floating' in
the case along with two batteries.
The 10 other ones are fitted on the bottom of the case with thermal
insulation.
The cases have been mirror polished while the internal faces have been
left rough. That was the compromise we choosed to cope with the totally
unknown illumination or dark periods along with internal dissipation. It
seems to be right altough we have to carefully study it with a post
analysis of the telemetry.

There are 12 Li Thyonile, 13 A.h, 3.6 V batteries (LSH20 SAFT, space
qualified model) arranged in 4 series 3 parallel giving 14.4 V no load.
A better arrangement would have been to get higher voltage and lower
current as the capacity is inversely related to current draw. The
switchers allowed for that but there was a technical issue that we
hadn't the time to care for leading to that suboptimal stacking.

Hope it gives you some insight of IDEFIX and answers your questions.

Best 73.
Ghislain F1HDD/ON1RG


Hi Guys,

Thanks to Amsat-F and F1HDD
Having a lot of fun with IDEFIX
and the WDECPSK  telemetry decode software.

Works fine with my soundcard.
Plenty of valid telemetry frames collected.

Could someone post a translation on this bb for
some of the main terms seen in the telem. please.

Optro
Moy Orbite
Temp Boiter
T ISD
Moy IO

Thanks

and 73 all

John G7HIA


----
Via the amsat-bb mailing list at AMSAT.ORG courtesy of AMSAT-NA.
To unsubscribe, send "unsubscribe amsat-bb" to Majordomo@amsat.org



AMSAT Top AMSAT Home