NRAO Home  >  Green Bank  |  Wiki Topic:    GB > Projects > KaBandReceiver > KaRxMappings

Changes | Index | Contents | Search | Statistics | Jump to Topic:

Ka Band Inputs and Outputs

Galen & Brian worked all this out with hot and cold loads and a polarized 33 GHz sweeper signal Thursday 20oct.

Fixed points for the mapping, as you stand with the electronics box between you and the dewar, were:

• feed A (=feed 1) is on your left
• the radiometer labels coming out of the dewar (1s, 1d, 2s, 2d)
• direct detectors (CCB) connections being
  • 1-8 from right to left on the side of the plate near to you
  • 9-16 from right to left on the side away from you
• spectral line connections labelled by magic marker

J Number Physical Interpretation Tabulation

Here is the PreDecKaMappings

Ka Receiver Mappings

NOTE: when reassembled the Ka receiver outputs Feed 1 L polarization and Feed 2 R polarization were interchanged at the output (1L <-> 2R). This affects ports J1-J8 and J17,J18 ("radiometer 2"). This is something that can happen anytime modifications to the Ka band hardware are made and is a fairly subtle effect that's difficult to fix in the hardware, so I believe we should be prepared to swap feeds at the output whenever hardware modifications are made.

• The first 16 J numbers are the CCB direct RF detector lines. They will be connected to the CCB as J1 <-> J1, J2 <-> J2, etc.
• Note that the soothing pattern of the continuum frequency channels inverts between J8 and J9
• The last four ports are the spectral line outputs and will be connected as
  • L1 --> mmconverter 1 --> IF Router J6
  • R1 --> mmconverter 2 --> IF Router J22
  • L2 --> mmconverter 3 --> IF Router J39
  • R2 --> mmconverter 4 --> IF Router J53

Feeds on the Sky

• Feed 1 is what Galen and I called "Feed A" in our re-tracing project Thurs AM.
• If you stand on top of the receiver room over the axle of the turret and look out over the dish, Feed 1 (A) will be on your left.
• Roger has determined that the feed on your left (ie to the left of the boresight) also ultimately looks out to the left as well-- ie, has a negative azimuth offset. similarly if the feed is above the boresight-- between the boresight and you-- it looks up ie has a positive elevation offset.
• Joe Brandt confirms that if you put a postive cross-elevation offset into the antenna database, that positive number is added to the calculated boresight azimuth to obtain the azimuth where the telescope is pointed. Therefore I say that the offsets should be
  • Feed 1 = pointing in a slightly negative azimuth direction = a positive offset to track with this feed
  • Feed 2 = pointing in a slightly positive azimuth direction = a negative offset to track with this feed
• There is a GBT convention that Feed 1 should be the one that (presumably for an equatorial source) sees a source first in a drift scan. According to the above reasoning, by this convention Feed 1 should always have a positive cross-elevation offset. (NB this naming convention is different from the real problem of knowing how things are associated with each other)
• The antenna manager uses the receiver offsets that you can see on this web page (click on model 4c): http://leo.gb.nrao.edu/antenna_db/displaydb.php
• As of 26oct06, Ka has a positive xel offset for Feed 1. This is as we think it should be. Note also that Ka has a positive elevation offset... is this as we expect given the location of the focus relative to the feeds? (The positive elevation offset is certainly correct and the answer to that question, is a check on the zemax calculation)
• Not all receivers follow the beam 1 drift scan convention (in particular K band doesn't).

Changes to the Block Diagram

• Radiometer 2 is the top radiometer. J18 and J17 are attached to this one.
• Radiometer 1 is the bottom radiometer. J19 and J20 are attached to this one. (these are from the labels on the lines coming out of the dewar)
• One problem we have as that we enshrined radiometer 1 being the top radiometer, and radiometer 2 being the bottom radiometer, in the block diagram that was used to assign individual (bias, phase switch, cal) controls in the manager . Therefore the manager controls don't match the labels coming out of the dewar. I guess we eat that and live with inverting the names (so radiometer 2 is controlled by d1bias, s1bias, d1phaseswitch, s1phaseswitch, etc).
• In radiometer 1 swap sum and diff, in both hybrids.
• In radiometer 2 swap J17 and J18 (so J18 is on top)
• then label continuum channels appropriately...
• Feed 2 is the top feed.

• We also need to carefully check which bias points, cal track signals, cal on/off signals affect which outputs We already have one inconsistency and we will have to sort out later where this is (wiring, block diagram, manager MCB adress, ...)

I think this gives the table of mappings at the top but please don't take my word for it!

-- BrianMason - 20 Oct 2005