[Arm-netbook] [EOMA68-A20] SATA testing

[joem]

________________________________________
From: arm-netbook [arm-netbook-bounces at lists.phcomp.co.uk] on behalf of luke.leighton [luke.leighton at gmail.com]
Sent: 04 September 2013 23:15
To: Linux on small ARM machines
Subject: Re: [Arm-netbook] [EOMA68-A20] SATA testing

On Wed, Sep 4, 2013 at 8:14 PM, joem <joem at martindale-electric.co.uk> wrote:
>
> um... anyone got any ideas as to why the voltage levels on Tx+ and Tx-
> are 0.5V?  is that normal?
>
> --
> Yes its a differential interface that works on current instead of using voltage signaling.

 yyyyeeuuurgh.  ok.

Seriously its a lot better. Its not common to use it because it drinks power.
USB is differential!!  RS232 is similar and can work 1km wires because of it. 

> If you want to see the signals, you could cut the track, put a serial 0.1R resistor
> attach the scope probes across the resistor, and turn up the scope sensitivity.
> You might see the packets flying past but probably not the signal itself as that
> is running into GHz range. Higher values of resistor may work, but it all depends...

 ok.  current signalling.  via capacitors.  wow.  so what the heck is going on?

Capacitors behave like short circuits at those frequencies, (but will block any DC build up).
The signal alternates between +/- and -/+, so by cutting track and putting
a serial resistor will allow the current to be measured.

 hmmm... let me check the track lengths, make sure they're precisely balanced...

you can balance the tracks by cutting both tracks and inserting the 0.1R resistor
in each track and connect both resistors with scope probes to further balance it.
(A bit excessive me thinks for the kind of results you can extract from it.)