Updated documentation for helicity circuitry for FPP/GDH. R Gilman 10/23/98 v2 ----- Helicity scheme ----- In original FPP runs, we used "toggle" mode, with 30 Hz of pulses, 33.3 ms + alternating with 33.3 ms minus. From the source we got 2 helicity correlated logical pulse signals, + and -, that were complements of each other, and we sent both downstairs. During GDH, we are running in pseudorandom mode. We get pairs of 1 s long pulses. The second of each pair is the complement of the first, but the phase of the first is chosen pseudorandomly. Thus the first pulse of a pair may be the same as the second of the previou pair, and apparent pulse widths vary from 1 to 2 s. We get two signals from MCC, a correlated signal plus a clock signal. We use the correlated signal to generate an h+ signal and the complement h- signal that we send downstairs. NOTE: close examination downstairs shows that the h- signal end overlaps the start of the h+ by ~2.1 us, and the h+ end overlaps the start of the h- by ~100 ns. This has to be checked further. ----- Cables to downstairs, hadron arm ----- The h+ and h- signals go downstairs to the equipment aisle through patch 1H75B03, numbers 17+18. The signals are connected with 4 foot BNC cables to the patch to the hadron arm "nose", through numbers 9+10. At the back of the detector stack, 10 foot BNC cables bring the signals to patch 161B, numbers D5+D6, which takes them upstairs to the top level of the stack. ----- Circuit in hadron arm ----- The signals sent downstairs are complements, and do not blank off the first ~200 us, to give time for the helicity to stabilize and the electrons to reach the hall. This blank-off period is introduced in electronics in the hadron detector stack. For h+ and h- signals, the circuitry is a little different... h+ signal | | LRS 429A linear fan out-----\ | | | | GO output Phillips 794 gate+delay | from TS 200 us delay$ | | 200 us long gate output | | | | | | | | | | | veto input input | LeCroy 465 coincidence module | logical out (~100 ns wide) linear out | | / | | | to e-arm | | Phillips 794 | | gate+delay input input 500 ns delayed LeCroy 465 coincidence module 100 ns wide linear output logical output output to scalers | \ | | \ - unused - | \ LRS 429A gate for linear fan out scalers% | | \_ | | \_ / | \_ / | \_ / | \ ADC for backup ADC LeCroy level shifter& event for event channel 6+, 10- for helicity helicity input to scalers -- tagging* tagging* level shifter output | is fanned out through LeCroy ECL to electron arm delay box to inputs to the MLU/TS scaler and the helicity gated versions of it $ The h- appear to be blanked off for 201 us, while the h+ appears to be blanked for 198 us. This aproximately cancels the h- being 2 us longer when sent downstairs. % Note that the scaler gate is an AND of run and good helicity. * For the event data, the long helicity gates are: h+ gate into channel 15 of the S1 and gas Cerenkov 4413 discriminators h- gate into channel 16 of the S1 and gas Cerenkov 4413 discriminators Experimentally we find that ~50% of events are +, ~50% are minus, a small fraction are neither, and perhaps 1 in 10^6 events are both helicities, a situation which should not happen. & We originally used scaler channels 9 and 10, but this was shifted to 6 and 10, by Mark Jones to check for crosstalk. The electron arm is about the same circuit, with the following exceptions: o There are only three coinc units in the LRS 465, so the helicity.AND.run.AND..NOT.first-200-us-veto is done in a single module. o To generate the helicity-minus-first-200-us signal to go to the electron arm, we use an LRS 222 gate and delay in latch mode. It is started with the 200-us-delay-output signal of the h- Phillips 794, and stopped with the LRS 429A h+ signal fan out. o The short logic output on this channel of 465 is not giving a signal, so we use the long signal from the 429A fan out to be sent to the scalers to be counted. -- Note that this puts run.and.helicity into the scalers! ----- Cables to electron arm ----- To enable independent DAQ for hadron and electron arms, we need to bring the first-200-us-blanked-off signals to the electron arm. The ~1 us delay from hadron to electron arms should not matter. The cabling is as follows: From the top level of the stack, the cables go to the back of the stack through patch panel 161B, numbers D7+D8. Two ~10 foot cables then connect the helicity signals to the "hadron nose" patch panel, cables 1 + 2. Two short BNC cables connect the "hadron nose" channels 1 + 2 in the equipment aisle to "electron nose" channels 3 + 4. In the back of the electron spectrometer, we cable the "electron nose" to patch panel Alfred D 13+16 to take the signals to the upper electronics level. As of 10/22/98, we have seen that we get good complementary, 200-us-blanked-off signals in the electron arm; we have not identified into which two ADCs to send the signals.