Entry: 1 2 3 4 5 6 7 8 9 10 11 12 13 14
We safely arrived at South Pole just before midnight.
ChristinWe took it easy and spent most of the day working with the SWAMP web pages. We also attended the town meeting in the Dome and had a AMANDA at South Pole meeting with Steffen and Robert.
ChristinAfter brunch we walked to MAPO to get an overview of the current situation. We wrote down all prompt and delay channel labels for the crates we will be working with and found the corresponding OMs in the SWAMP hardware database. Finally we made plans for tomorrow. We will try to measure gain with AMANDA running and take waveforms.
ChristinToday we shut down all high voltage and measured the gain for the prompt and delay output for channels 01-04 in rack 1, crate 2, slot 03. Then we ramped up the high voltage again and repeated the measurement and got the same results whereby we concluded that the gain can be measured with the high voltage turned on. Therefor we started a new run around around twelve o'clock and started measuring gain for all channels. The input signal was a 1MHz sine wave with about 30mV peak-to-peak. These measurements were finished by the end of the day. We also started taking waveforms. We set both oscilloscope inputs to 1MOhm and saved two pictures for each channel. One with 500ns/div and one with 5us/div. The pictures contain both the prompt and delay output. Waveforms were taken for all channels except slots 03-06 in rack 5, crate 4.
ChristinAll remaning waveforms have been saved. A power supply from the old B14/19 crate that was replaced last year, was taken and and mounted to be used in the new crates. At around three in the afternoon we stopped the run and turned off the high voltage. All HV and OM cables for crate 4 in rack 5 and crate 2 in rack 2 have been disconnected and the labels written down. We denote the OM cables: string_number_Q_quad_number-OM_in_string/OM_in_string, for example 04Q2-23/24. All lemo cables have been disconnected. We have started taking delay chip from the old SWAMP cards and putting them in the new SWAMP cards.
ChristinWe moved the delay chips from the old crates to the new crates except crate 3, rack 5. 14 OM connectors were changed, and channels 1-4 were tested with the new crate and connector in crate 4, rack 5. Several updates to the scripts and data in the web pages were made, including adding all waveforms before the exchange.
ThomasConnector exchange continued, and all channels in rack 2, crate 2, slot 01 were tested and turned out to be working fine. The noise was around 20 mV as before the SWAMP exchange.
ChristinAll connectors except string 10, three in string 8 (08Q18, 08Q17, 08Q16) and the single connector in string 13 have been replaced. The remaining crate has been mounted and the HV, OMs and prompt cables have been reconnected. There are now four slots in each crate (making a total of 12 SWAMP motherboards) and the two spare motherboards have been marked 'SPARE' and mounted in the bottom crate in rack 9. One spare motherboard lacks delay chips, and the other spare only has delaychips in channels 01-16.
ChristinConnectors in string 10 changed. This finish the connector change of strings 5-10. Front panel delayed lemo cables of strings 5-10 connected. The rest of the backplane OM anf HV cables connected for strings 5-10. The new SWAMPs are therefore installed and ready to be tested.
CarlosBrought back the high voltage. Measured gain for crate 3 in rack 5 and crate 2 in rack 2. Took waveforms for crate 3 in rack 5. Everything seems ok. 11 previosly unconnected channels brought back to an exciting life. Another glorious day in MAPO.
ThomasGain settings finished in rack 5, crate 4. All waveforms taken. Yet another glorious morning in MAPO without Gomez
CarlosAll gains have been set to 100 respectively 40 for prompt and delay. All waveforms have been saved. The previously disconnected OMs in string 10 are working fine. There seems to be a number of channels that can be brought back to life by simply setting a high voltage value. OM 226 was fine before the exchange and is now dead. The high voltage cables for the three string four quads (OMs 81-86) were connected to the power supply. Every second OM was dead, and changing the polarity of OM 85 brought it back to life but at the same time M1 stopped working. We have started looking at changing connectors for strings 11-13 (rack 1, crate 1). Quads 18,17 and 16 in string 8 are still not connected. Some 16 seconds after turning on either of either of the LeCroy 1458 power supplies a 8 MHz noise appears. There's also a 1.2 MHz noise that could not be identified.
ChristinWe try to revive channels 81 to 86 (rack 2, crate 2, slot7, ch 15 to 20). We set HV (1600 V) to them (HV crate M1, channels 129 t0 134. HV cables D01 to D06). Channels 82, 84 and 86 are alive. The others do not show pulses. We switch the polarity on the quad cable of channel 85. Now it works. The HV crate M1 shows problems. It hangs for long periods. We do not try to continue with changing polarity of channels81 and 83. We lower HV and switch the unit off. We start checking the cabling on the backside of rack 1 crate 1 before disconecting the OM quads to exchange the connectors. We find a change from the setup we left last summer: Quad 1.01.2 is connected together with 15.2 in slot 7, channels 13 and 14. Quad 1.01.1 is connected together with 1.14.1 in slot 7,chanels 15 and 16. Cables 15.1 and 1.14.2 are unconnected (hanging loose). We go for dinner.
CarlosChanged connectors on strings 11-13. It seems the low-numbered quads for string 13 have been connected to the pickup-boxes since last summer's visit by swamp team. Quad 1/channel 13 was split and the two channels connected together with quads 14 and 15. One original channel was disconnected in each of quads 14 and 15. We could find no indications of why this was done in the logs. The OMs of string 11 appear not to be numbered sequentially as e.g. 307 and 309 are in the same quad. In string 12 the even-odd assignment within the quads varies. Sometimes the odd one is the first, sometimes the even. We connected the differently coloured wires in the same way for all quads and assumed that the temporary extensions which we removed all did the same thing. We also noted that the HV channels for e.g OM 309 are different in the HV database (RED_M_03) and in the swamp DB (RED_M_02). The one connected is RED_M_02. It appears HVs for OMs 309 and 308 are swapped in the HV database.
klas