GSM-noise "buzz" issue

[Werner Almesberger]

Joerg Reisenweber wrote:
> Werner, we're not short on devices to gain decent results

Excellent :-)

> Argentina customs is a different issue.

It's actually going pretty smoothly lately. Seems that we found the
right procedure.

> Tell me what you like to do exactly and I'll see whether I can set up a test 
> environment :-). Actually got some spare devices.

How about this ?

- Take two units and collect the NMEA data under a set of different
  conditions. Compare signal strength (from the GSV sentences).

  The goal is just to establish how much the two devices differ
  and to have a reference for the one that's getting reworked.
  If they differ a lot, swap the one that's worse and repeat.

- Rework one as follows:

  - remove R1764 and use it to inject 3.0V into RF_3V

    (Maybe check that the trace leading to B7602 doesn't cause
    trouble, e.g., let HF riding on the injected power crawl
    into other elements inside the GPS can. If it does, find a
    solution :-)

  - connect H-TP7602 to a 3V-compatible RS-232 port. You can
    probably just use a debug board for this.

  - make the injected 3.0V the only power source of that unit

- Do a set of measurements with the same two units under roughly
  the same set of conditions.

  - The unmodified unit should perform roughly as before.

  - The modified unit may perform differently, particularly
    relative to the unmodified unit.

Then post the results and see if we have a winner ;-)

- Werner