User name R. Michaels
Log entry time 10:17:16 on February 7,2000
Entry number 33022
keyword=BPM resolution, ADC problemsPlots below show some things about the BPM measurements
using the Struck 7510 ADC ("burst mode" ADC, new for this expt).
I checked several runs, they were similar. Some comments:
1. The oval correlation between BPM and raster current shifts
around (one sometimes sees two ovals in one plot). 2) If I cut
on the raster to require less than 10 micron motion, I get the
BPM distribution in fig 1 bottom two plots. Intrinsic resolution
is about 100 microns (sigma). 3) There is a problem with the
Struck ADC that it sometimes reads a completely spurious value,
see fig 2. There is a 2nd ADC, a LeCroy 1182, which I believe
is very reliable and reads copies of the data. (It is a charge integrating
ADC with 50 nsec gate). One can see the LeCroy is always clean,
while the Struck has about 1% of events that look like junk. But when
not junk, the two are correlated. 4) We are not making optimum use
of the ADC dynamic range (both are 12-bit). At the moment the bit
resolution is ~20 microns, which is better than the total resolution. We
could adjust this with about a 1 hour access, but it seems not necessary.
FIGURE 1
FIGURE 2
