Make New Entry, Make Followup Entry
User name R. Michaels
Log entry time 14:44:42 on April27,2009
Entry number 269138
keyword=L-HRS Retiming check and adjustmentsFor the 1st time, I checked RT on L-HRS.
If L1A is too late it makes a forced gate. The headroom is measured
by triggering on L1A and measuring how much time the ADC shifts
when RT is removed. One runs with psN = 1 and other ps infinite.
One problem is that T4 is later than T3 by 110 nsec. Normally we
want this much smaller. I decided not to change it (it might mess
up something at Bigbite) and instead increased the delay of L1A
at L-HRS by ~50 nsec (put a cable).
Before adding delay
T3: L1A earlier by 200 nsec, has 30 nsec headroom
T4: L1A earlier by 90, has 142 headroom
After
T3: L1A earlier by 156, has 88 headroom
T4: L1A earlier by 52, has 196 headroom
After adding delay I ran the DAQ with two conditions to check
VDC T0. I show only one plane. The others were similar.
Run 1570 ps3 = 1, ps4 = 1, and normal T4 definition. Fig 1.
Run 1575 ps3 = 2, ps4 = 3, and T4 comes from 1st output MLU, hence
should be T3 and we get a good mix. Fig 2. This looks good !
Why fig 1 shows a shift: Cosmics are dirty, the track angle distribution
is different for T3 and T4. We may want to look at this with beam
too.
Last thing I did was look at ADC gates on scope for purely T3.
The gate is 250 nsec wide. The scint were coming about 42 nsec
into this gate which is a bit tight. They might chop when you
account for delays through back of fastbus. So I took out 16 nsec
on RT to make this 58 nsec. This changes the above numbers too;
it makes T4 L1A earlier by only 36 nsec.
Note: L-HRS T0 will change !! by 16 nsec.
FIGURE 1
FIGURE 2
