Development of a payload for the characterization of commercial microcontrollers to radiations

The main scope of this thesis is to prepare the base for the use and the radiation charac-
terization of the new Texas Instruments’ FRAM micro-controllers within the Modular
Architecture for Satellites (AraMIS) developed by the Politecnico di Torino. These kind
of micro-controllers seem to be very appealing for space applications based on Commer-
cial O The Shelf (COTS) components because of their intrinsically radiation hardened
structure and their low power consumption compared with standard FLASH based one.
The idea of using ferroelectric materials to store digital can be dated back to 1952, but
it was practically implemented only starting from the 80s because the needed advanced
technology to develop them wasn’t available before. FRAM based micro-controllers are
instead available on the market since about one year and an half. The ferroelectric RAM
memory, known as FeRAMF or FRAM, is conceptually similar to the DRAM cell, but
there is an important dierence that lies in the dielectric of the storage capacitor: while
DRAM cells use a layer of standard linear material, the dielectric of a FeRAM cell is
made of ferroelectric material, usually lead (Pb) Zirconate Titanate (PZT).
Using a ferroelectric dielectric leads to a dierent behavior of the cell compared
with a DRAM one, leading to many advantages especially for what concern the overall
power consumption in read/write cycles. Furthermore, the material exhibits two stable
polarization conditions and it’s possible to switch between them by means of an electric
eld with opposite polarity. Since the polarization will be kept after the applied eld
is removed, it is possible to link the polarization state to a logic state and so these
materials can be used to build a non volatile memory device. No periodic refresh is so
necessary to keep the information, like in a DRAM memory.
The reading process is destructive: it is not possible to read the content of a cell
without actually clearing it, because of the way the information is stored in the device.
iii
To know which of the possible polarization states the dielectric holds, the only way is to
write a new value to the cell with the bit-line pre-charged but in high impedance state
and depending on the previous polarization, this process will or won’t produce a voltage
pulse out of the bit-line. Read and write cycles require basically the same operations
and can both be completed in times in the order of tens of nanoseconds and without
using high voltage charge pump like in FLASH memories.
The three main design parameters of the electronic systems of small satellites are:
 power consumption;
 physical dimensions;
 radiation environment behavior.
The electric power in the satellite comes from solar panels, which are necessarily of
small dimensions because of the mechanical structure, leading to few Watt of average
power to cover all the needed functions. It is so necessary to make the best use of any
mW of available power. Furthermore launch costs are directly proportional to the mass
of the system, so it is absolutely necessary to reduce as much as possible dimensions
and mass of the electronic system. We said that FeRAM memories are RAM devices,
meaning that read and write procedures do not dier signicantly and random write is
possible without the need of a previous erase of a cell, but they are also non volatile,
we can so for sure state that this leads to save power. In fact in DRAM devices most of
the power is used by the refresh procedure otherwise the stored informations are lost.
Furthermore the refresh process leads to a decreasing in the overall speed performances.
At the moment there are no big FRAM memory available on the market, but in any
case, memory requirements of small satellites are normally compatible with the size of
available FeRAM, except for imaging payloads if local storage of a certain number of
images is mandatory.
Because the FRAM cell stores the state as a PZT lm polarization, an alpha hit
have a very small possibility to cause a change in the polarization. FRAM terrestrial
Soft Error Rate (SER) is not even measurable. This “radiation resistant” characteristic
of FRAM makes it attractive for use in several medical applications and space one.
Arturo Guadalupi iv
Keeping in mind the concepts exposed above, this work is focused on the development
of a payload tile for the AraMIS structure called 1B521 Radiation Characterization
Payload whose aim is to introduce the usage of new FRAM micro-controllers within
the AraMIS nano-satellite structure and characterize them for low cost space applica-
tions in therms of radiations. In particular it is requested to characterize the use of
a FRAM micro-controller (MSP430FR6989) in therms of Total Ionizing Dose (TID),
Single Event Eect (SEE), like Single Event Upset (SEU) and Single Event Latch-up
(SEL), power eciency and reliability in general. No scientic data coming from real
space experiments or terrestrial simulations (using for example particles accelerators)
are available at the moment. It is furthermore requested to show the eciency of the
AraMIS’ developed software hardening library in order to have a direct comparison be-
tween a standard compiled code and an hardened one.
The AraMIS radiation-hardening technique, is based on the use of appropriate C++
classes from the hardened data (Hdata) package developed in house, which can be used
in a common C++ program instead of standard data type. For instance, a short can be
substituted by the so-called TripleShort, which automatically and transparently stores
three copies of the same value and votes or recovers data whenever required. A normal
C++ program can so still be compiled by modifying only the data type denitions.
This makes possible to reuse software algorithms and procedures which have already
been validated and tested without any specic eort apart from redening data types
drastically reducing the development time.
This thesis has to be considered an user guide manual about the developed payload
tile, and a base for future developments on FRAM microcontrollers within the AraMIS
nano-satellite structure. The rst part of the developed work in-fact makes possible
to introduce and start using any kind of FRAM micro-controllers that belongs to the
family MSP430FRxxxx without an heavy eort. All the hardware-dependent choice
that have been made are explained and the software commented in order to be easily
understandable and useful for feature developments.
Arturo Guadalupi v
Here a little overview about the structure of the thesis.
Chapter 1 gives an introduction about the space radiation environments, its interaction
with the electronics and the used shielding techniques.
Chapter 2 gives an overview about the FRAM technology and some concept about
their pro and cons about their use in space applications.
Chapter 3 and Chapter 4 give an overview about the UML approach in the AraMIS
structure and how it is organized.
Chapter 5 explain the design of the developed PCB and what hardware has been chosen
in order to give support to the developed software.
Chapter 6 shows the software structures behind the designed tile, how to it commu-
nicates with the OBC and the type of tests that are executed.
Chapter 7 gives an overview about the tests that has been made to validate the work
and the reached results. What can be done in the feature to improve the what have
been done is also mentioned.