Rockbox.org home
release
dev builds
extras
themes manual
wiki
device status forums
mailing lists
IRC bugs
patches
dev guide



Rockbox mail archive

Subject: Re: Recorder noise
From: Andrew Jamieson (ajamiesn_at_mira.net)
Date: 2002-02-16


>> * If the noise goes away when the music is stopped, then it's not the amp
>> that's causing it, but the pre-amp.

>There are actually two levels of noise. One louder during playback and one
>quieter that is always present when the unit is turned on.

>> If the noise is of a very low frequency (such as 15hz) then the Archos
may
>> be missing the higher rated caps (uF).

>The noise itself is more or less white noise. The 10-15 Hz I was talking
about
>is the oscillating changes heard in the noise while displaying the VU
meters,
>sounding "dzi-dzi-dzi-dzi" at a rate of approximately 10-15 "dzi" per
second.

>> * The schematic does not show a capacitor to PVDD. This is an auxilary
>> supply to the analog circuitry, and is specified as requiring a 3nF
bypass
>> cap to AVDD1. It's possible that hooking this up would help ....

>In one of the photos on the Yahoo group page I can see pin 63 (PVDD) is
>connected to one side of a tiny component, probably a capacitor, and pin 64
>(AVDD1) is conntected to the other side. Here's the URL (you have to
register
>with yahoo):

Yeah, you're right. The component's definitely a cap. Guess we need to
update the schematic.

>> * The input lines on serial audio input (SIC, SID, SII) are left
floating.

>In the above photo, pin 45 (SID) is connected to something marked "103"
while
>pin 46 and 47 (SII and SIC) are indeed unconnected or routed under the
chip.

Right once again. The part is a 10K resistor, pulled to ground.

It's interesting that the cap on pin 1 looks like a mod - there's no way
it's attached to an actual pad that close to the IC pin (let's face it; it's
soldered directly to the pin). As this is the AGND pin, it strikes me that
this has been done to address the problem of noise post PCB layout. Of
course, if the noise is still there, either it doesn't work, or the noise
was even worse before ......

Jeeezus .... I just noticed that the big electrolytic cap (big black
thingie) is also a mod; soldered directly across a SMT resistor (180k, or
18R maybe ... Looks like its attached to the ISD circuit). If they had to
mod this cap into the circuit to fix the power rail .....
If we knew what part of the circuit this was used for, it may help.

Also, the actual power input pins (10 - 14) appear too far from any bypass
circuitry (the crystal is directly next to them on the component side (from
the MAS point of view), and there does not seem to be any caps on the solder
side - although there could be something under that tape/glue. It's
possibly of interest that the crystal is located right next to the power
input pins - they are, of course, a great source of noise (mostly too high
frequency for audio, but with harmonics that extend into both upper and
lower ranges). The fact that they are located right next to the power pins
on the MAS doesn't help.

Errrr ... I gotta go play stamping ..... I'll keep looking.

A

--
Björn



Page was last modified "Jan 10 2012" The Rockbox Crew
aaa