Class, I have updated the schedule to reflect the completion of your
projects. There are 4 weeks left in the semester at this point. In
order to make it easier to finish I am setting a paced
schedule. Remember that the project requirements are

(1) Development of performance and design specs
(2) Design of the instrument
(3) Test of the instrument on breadboard
(4) Design of PCB
(5) Comprehensive report

At this point (1) and (2) have been mostly completed, but there are
still some issues that remain. I think the magnetometer needs only a
current sensor. The dosimeter still needs to design the cutoff
switch. The EHM appears to still have some parts of the circuit and
some voltages which are not yet defined.

So that you are not left with a large amount of work at the end I am
going to request that you are going to build your final report section
by section during the rest of the semester as you also finish your
instrument and its testing. I will ask for your reports as PDFs posted
on the website. If I have requests for revision I will do my best to
get those back to you as quickly as possible.

I see the report as having the following sections.

(1) Performance specs
(2) Design specs
(3) Instrument design to spec
(4) User documentation
(5) Test plan
(6) Test results
(7) The PCB design


The schedule I am setting is the following:

April 11: Sections 1, 2, and a draft of sections 3 and 4
April 16: Sections 3 and 4 complete, draft of section 5, revisions
April 23: Section 5 complete, draft of section 6 and 7, revisions
April 30: Initial version of final report
May 4: Second version of final report, if needed
May 7: Final report

I also want to see a 1-hour final presentation which covers all the
aspects of your instrument. The presentations will take place April 30
and May 2.

For prototyping and test you can use my lab, but keep in mind there
are other tenants there as well. A couple of you have a key, but we
will need to make sure at least one person per team has a key.

A few words about sections 1 and 2. The performance specs cover things
such as the characteristics of the phenomenon being measured, for
example what should be measured and with what accuracy, whether
quantitative or qualitative. It also covers interfacing requirements,
mass, volume/area, power, etc. Design specs set requirements on the
components that you use. An example is the resolution of the ADC,
perhaps ranges of inputs, and other quantities. In some cases the
performance specs may have been vague to being with but later
solidified. For example I added the ability to switch off power not
long after we started, and added the ability to measure current more
recently.

A few words on section 4. This provides the essential information that
will allow the person picking up your work in the future to quickly
get up to speed. This is meant as a quick reference for your
instrument which does not cover all the details in section 3. A data
sheet. Snippets of code, and explanations for voltage levels.

In the case of the EHM and the Magnetometer the specs were developed
somewhat backwards. For EHM noise measurement you selected amplifiers
and then we decided that their noise floor was acceptable. For the EHM
DC measurements we elected to use the same ADC, but you should check
that this is a reasonable choice given that the solar panel currents
will be used for attitude control. Attitude accuracy of a few degrees
is desired. For the Magnetometer I told you which chip to use and that
set the measurement accuracy. In those cases simply explain that
situation and then proceed with developing the rest of the specs. For
the dosimeter the performance spec for the dose measurements were
vaguely "want to be able to detect passage through south-Atlantic
anomaly." What dose resolution does that correspond to?

I would like to set a few general requirements on your report: table
of content, explanatory captions on figures/tables, references to all
external content you use. I prefer you err on the side of including
more material, such as figure and tables, to make the report as
self-contained as possible. I also generally prefer single-spaced
text.