SG April 12th, 2004 

Dear all,

for commissioning of the Run2b CTT with the new Boston DFEAs we are 
already planning to have a new DFEA crate running in 'spectator mode' 
(using the LVDS splitter boards).

Pushing this idea one step further, we propose to install a few extra CTOCs
and another CTTT. These boards are spares for the current (and future) system
and are needed anyway.

We could then setup a complete parallel chain all the way to crate
x13 and the Trigger Framework, and fully test a few sectors, an octant or even 
a quadrant of the new DFEAs without disturbing physics datataking in any
significant way.  

We have space for the extra CTOC/CTTT boards in PW03-2, 
and we have spare inputs on the three CTMs in PW03-0.
 We would 'borrow' one of the VRBs in x13 (e.g. from the CTQD),
and we could even borrow one of our own framework cables with
16 And/Or terms. (Availability of a framework cable depends on where 
we are with CTT and FPS trigger term usage at that point (coming winter). 
Feeding all the way into the framework is not strictly necessary, 
but it might facilitate some of the trigger studies.)

The advantage of this approach (over just using the 'slow control'
output buffer readout through the optical link from the new DFEAs) is
that one can use all the existing tools to study trigger performance.
(A new data chunk for the extra CTOCs would need to be defined,
and the tools would need to be modified to read these data.)

This way we can not only debug the physical layer, but iterate on equations 
and firmware until the new system has the desired performance, 
and at that point replace the DFEA crates. 

Please distribute as you see fit, and let me know what you think. 

cheers,

 Stefan