src/rf.md

RF Signal from AVF
==================
\author HY

We can make use of the rf signal from AVF cycloton
for the particle identification (PI) of the beams.
The frequency is usually 10--20 MHz, 
which corresponds to the time interval of 50--100 ns.

Cable connections
-----------------

The rf signal is provided in the following way:

1. (**Operators room, next to the waiting room**)<br>
Near the wall on the north side, there are 19-inch racks
with NIM bins and a cable panel.
The signal is: <br>
  + provided at the connector "AVF" at the bottom of a rack,
  + go through a discriminator, in a NIM bin above,
  + then go to the cable panel on the right side.
2. (**Nishina experimental building BF2**)<br>
The corresponding cable panel is located at BF2 of the
Nichina experimental building, near the elevator.
A cable should be connected from this panel to the J1 room. 
3. (**J1 room**)<br>
The signal should be found at the panel in the leftmost rack. 
The signal is:<br> 
  + Discriminated,
  + Downscaled by 1/2,
  + Splitted into two signals (RF1 and RF2),
  + RF2 is delayed by a few 10 ns from RF1,
and reaches at the TDC.<br>
TDC input are as follows:
  - TDC start....trigger (PPAC, for example)
  - TDC stop....RF1, RF2

How it is useful for PI?
------------------------

The TDC  value can be used as a TOF between the F0 target
and triggering detector (e.g. PPAC), but in the reverse 
direction (higher value means faster).

This can be explained as follows:

 - The primary beam is extracted according to RF.
 (The time width of the beam is typically 2-3 ns).

 - Before reaching at F0, there is no time differece 
between beam particles.

 - Thus, RF1 (or RF2) can be regarded as the timing at F0,
but with a fixed offset (T0 = RF interval x *n*):<br>
 T_F0 = RF+T0

 - The TDC is started by the detector trigger, and stopped by RF.<br>
 TDC = RF-T_det.

 - Therefore, the TOF is, <br>
 TOF = T_det-T_F0
     = -TDC-T0,<br>
which means the TDC value represents inverted TOF
with a fixed offset.

Orbit lengths:
- F0-F2 7.81 m
- F0-F3 13.97 m


Why downscaled and splitted? 
----------------------------

The discriminated RF signal must have a finite pulse width.
If a trigger is made within the width, 
the TDC immediately stops, and the value will be 0.
In that case, we cannot have real timing information.

To avoid this "dead time", we downscale the rf trigger.
Downscaling by 1/2, we can record the timing information 
for 1 complete cycle + 1 cycle with dead time.
We still have dead time, however, by 
recording the 2 RF signals, one of them
delayed by about a half cycle, 
we can cover the dead time of each other.

Time calibration
----------------

Due to the downscaling, 
we may have two spots representing the same ion in a plot.
The channel number difference corresponding 
to the interval of RF, which is precisely known by the operators, 
allows as to perform a calibration.