MIDAS problem with DCRC2(33), RevC

Had been running for weeks with DCRC1(55) and DCRC2(33), Rev C.1 and Rev C respectively. Both were operating correctly.

Changed the setup to test DCRC4(X) which is Rev C.1. All three DCRCs were enabled and a timing link established between them.

Problem then occurs with DCRC2. Setting Driver Gain to x1 using MIDAS causes pulses to disappear. By checking the registers found that the corresponding gain value on the DCRC is not being set correctly by the MIDAS settings:

The above behavior makes it seem like MIDAS is treating DCRC2 as if it were a Rev C.1 board.

Problem was fixed by turning off MIDAS programs, unpowering DCRCX, reconfiguring to only enable DCRC1 and DCRC2, then restarting MIDAS. Then behavior goes back to normal.

Had been running for a few days with DCRC1(55), DCRC2(33), and DCRC4(X). No problems noticed with DCRC2. We hadn't changed its driver gains, which were kept at a standard x2.

Problem occurred when I tried to perform IbIs study on a DCRC2 channel. Reducing the driver gain to x1 caused pulses to disappear and the channel to become unresponsive to changes in QET bias, Front-end Gain, Polarity choice, SQUID bias, although it did respond to changed in Driver Gain and Driver Offset. I had forgotten about the problem from April 4 so did not make the connection that the Driver Gain settings were probably being applied wrong.

Problem was fixed by turning off MIDAS programs, power-cycling DCRC2, then restarting the MIDAS programs. Note that in this case the programs were restarted with all three DCRCs enabled. After doing this I was able to set the DCRC2 driver gains to x1 and back to x2 and it responded normally.

One day after fixing the problem, with the setup still the same - DCRC1(55), DCRC2(33), and DCRC4(X) being used simultaneously - the problem recurs.

Problem was noticed when trying to perform IbIs on DCRC2 channel: setting Driver Gain to 1x caused pulses to stop and the channel to become unresponsive. This time I made the connection to the April 4 problem and found that the behavior was the same: changing the Driver Gain via MIDAS sent the wrong commands to the DCRC.

Tried stopping then restarting the MIDAS programs. This did not fix the problem.

Fixed the problem by stopping MIDAS programs, then changing the ODB configuration to disable DCRC4(X), unplugging the timing link cable from DCRC4, designating DCRC1 as timing master (previously it was DCRC4), then restarting the programs. DCRC2 worked normally after this.

While still in this 2-DCRC configuration I set all the DCRC2 Driver Gains to 2x. Then I reverted to the 3-DCRC configuration. Checking the Telnet session verified that all the DCRC2 Driver Gains were still at 2x.