From allison.122@osu.edu Wed Aug 3 08:13:09 2011 Date: Wed, 3 Aug 2011 14:12:09 -0400 From: Patrick Allison To: ara-firm@ara.icecube.wisc.edu Subject: [Ara-firm] Board/firmware status Hi all: Just a quick update on the board status: We currently have 8 DDA rev C's produced and working: we can actually produce 3 more with the IRS2s we currently have. The boards are already half-populated so they don't take very long. The testing procedure for the DDAs is a bit rudimentary: we check 1) all daughterboard connections manually 2) the board powers on and all I2C devices (except for the EEPROM, which isn't a high priority) are functional 3) TSA produces a TSAOUT (actually TSA_CLOSE, I think, but from the firmware those two are linked) 4) TSTOUT is toggling 5) Vped1/Vped2 are present 6) VrefP is (2.5V - VrefN) 7) When a readout occurs, the RAMP output (RampMon) contains a valid ramp 8) Impedance from C[A,B,C,D]1 to Vped1 is ~45 ohms 9) S11 VSWR on channels 1, 2, 3, 4 look normal (under ~1.5 from 150 MHz to 650 MHz) 10) Readout of the IRS2 contains a valid clock-like signal on channel 7 I'd love to do a more functional test for the IRS2 but the software isn't there yet. The NTU software that they've been testing might be a help for this, but I also don't know what kind of tests you would want to do, or how to do them. Note that for the board that was sent to UH, step 9 (S11 on all 4 channels) wasn't done, so that's why the open on the filter on channel 1 wasn't caught. Manufacturability of the DDAs is a bit of a pain because of the poor footprint on the IRS. Unfortunately I'm really at a loss for a better footprint, since the ones I've seen around vary by a lot - and I think the 'standard' IPC footprint would need much tighter manufacturing tolerances in the future (something more like the ATRI board production used, but possibly even tighter). Still that might be the way we have to go. Right now the main problems are that the pads are too narrow, so the IRS doesn't self-align, and shorts form very easily under the legs. As for the TDAs, we currently have 8 total produced and working (including the 1 at UH). Manufacturability of these guys is *also* a pain because (as Gary - and myself - suspected, but we didn't really have a choice) the components are very crowded. I've got a few ideas there for a next revision to move a few components around to generate space, but that again is for the future. As for the firmware status: The big change was moving to a completely unified architecture between the DDA_EVAL and the ATRI, using the "atri_core" module. The atri_core module is now in SVN all by itself, in ara/firmware/atri_core/trunk The DDA_EVAL and ATRI firmware now simply instantiate a few support modules for the physical interface, and then instantiate an atri_core module (parametrizing the number of daughterboards). This was done because the previous attempt at the ATRI firmware's USB interface wasn't working: in this case the firmware between the DDA_EVAL and ATRI are identical, so I can have a "known working" solution and then debug the ATRI firmware. The DDA_EVAL firmware's top module is now "DDA_EVAL_ATRI", and the ATRI firmware's top module is now "ATRI_revB". More info in a bit. Patrick _______________________________________________ Ara-firm mailing list Ara-firm@ara.icecube.wisc.edu http://ara.icecube.wisc.edu/mailman/listinfo/ara-firm