From p.j.coleman-smith@dl.ac.uk Mon May 12 11:32:05 2008
Date: Tue, 22 Apr 2008 13:59:34 +0100
From: "Coleman-Smith, PJ (Patrick)" <p.j.coleman-smith@dl.ac.uk>
To: Thomas Davinson <td@ph.ed.ac.uk>
Cc: "Lazarus, IH (Ian)" <i.h.lazarus@dl.ac.uk>,
     "Pucknell, VFE (Vic)" <v.f.e.pucknell@dl.ac.uk>
Subject: RE: AIDA clock and power cable

 Hi Tom,
Thanks for the response.

The HDMI cable is to be used as the local distribution. So the
connection to the main experimental areas BUTIS box would be using
whatever cable/fibre etc that is finally decided. The AIDA system would
be able to operate standalone for prototype and development and then
take the input from BUTIS when desired.
So the cables will be quite short.

I think ( currently ) it is a good way to combine a number of
distributions in one cable :- clocks/control/power .

So for the earthing etc the distribution will be on the same earth as
the detector.

I had assumed the detector and electronics would be on a common ground
plate. So another box to distribute the power/clocks etc wouldn't be a
problem.

The Gbit ethernet will be routed via a number of network routing boxes
and is not intended to change to one or more fibres until that is done.
These boxes will also be local to the electronics.

I would be happy for anybody to use the design as a common experiment
local distribution point. However there doesn't seem to be much movement
on this as yet.

I am seeing two reps from Analog Devices this afternoon. Any comments
about the ADC tests ?

Hope this sheds more light.

Regards
Patrick


> -----Original Message-----
> From: Thomas Davinson [mailto:td@ph.ed.ac.uk] 
> Sent: 22 April 2008 12:11
> To: Coleman-Smith, PJ (Patrick)
> Cc: Lazarus, IH (Ian)
> Subject: Re: AIDA clock and power cable
> 
> 
> Hello Patrick
> 
> Apologies for the delay in replying to your proposal to use 
> HDMI to distribute clocks in AIDA.
> 
> Some questions.
> 
> Is this proposal to be regarded as the de facto standard for 
> distributing
> (BUTIS?) clock data to DESPEC/HISPEC detector systems (AIDA, 
> n ToF, 4pi neutron, TAS etc.)?
> 
> Implicitly, the HDMI cable lengths will be long and 
> distributed to a variety of detector systems. I note that 
> ground and shield (also ground?) signal paths are defined so 
> I can foresee ground loop problems (cf. AIDA FEE gigabit data 
> path via fibre) ... ???
> 
> Cheers
> 
> Tom
> 
> --- 8< 
> --------------------------------------------------------------
> -----------
> Dr. Thomas Davinson
> School of Physics,				Email: 
> t.davinson@ed.ac.uk
> The University of Edinburgh,			Tel: +44 131 650 5250
> James Clerk Maxwell Building,			Fax: +44 131 650 7003
> The King's Buildings,
> Edinburgh. EH9 3JZ. UK.
> --- 8< 
> --------------------------------------------------------------
> -----------
> 
> On Thu, 27 Mar 2008, Coleman-Smith, PJ (Patrick) wrote:
> 
> > Hi Tom and Ian,
> > I am proposing to use the HDMI standard cable and connector 
> parts for 
> > the clock and power distribution in the AIDA FEE.
> > I have written a proposal which includes some background 
> info on HDMI.
> > Can you comment please.
> >
> > If this is OK then the FEE card will need only two wires , Gbit 
> > ethernet and HDMI.
> >
> > Thanks
> > Patrick
> >
>