| 1 | | == Some Thoughts == |
| 2 | | |
| 3 | | 32 channels of 1GSPS is probably too expensive. There is a 12-bit part from TI (ADC12D500) |
| 4 | | which is a two-channel 500MSPS part which can interleave to get 1GSPS. They're about $280 each. |
| 5 | | |
| 6 | | || '''Bits''' || '''MSPS''' || '''Cost/ch''' || '''PN/Notes''' || |
| 7 | | || 12 || 1000|| $280|| TI ADC12D500 (two 500MSPS channels interleaved) || |
| 8 | | || 12 || 500|| $168|| Intersil KAD5512P || |
| 9 | | || 12 || 250|| $44|| ADS4229 (two 250MSPS) || |
| 10 | | || 12 || 125|| $21|| LTC 2145 (2 ch) || |
| 11 | | || 12 || 105|| $23|| ** ADS6424 (4 ch, 12 bit Serial LVDS) || |
| 12 | | || 12 || 105|| $22|| AD9633 (4 ch, 12 bit serial LVDS) || |
| 13 | | || 12 || 105|| $24|| ADC12DS105 (2 ch, 12 bit Serial LVDS) || |
| 14 | | |
| 15 | | Observations about signals: |
| 16 | | |
| 17 | | * "Normal" case is three short pulses (few uS long at most) at 0, 50us and around 1ms |
| 18 | | * Stray discharges are typically quite short (few 100ns), amplitude similar to normal pulses (<1V) |
| 19 | | * Sparks are characterized by large amplitude and significant width. There can be a train of spark pulses spread out over substantial fractions of the 1ms pulse period. |
| 20 | | * Beam timing has ~ 10ms between pulses where the should be no signals |
| 21 | | |
| 22 | | One possible proposal: |
| 23 | | |
| 24 | | * Digitize 32 channels at 125 MSPS (single gain scale), full scale +/-5 or +/-10V |
| 25 | | * Artix XC7A35T has 1.8Mb block RAM. These are around $35 each. Propose to use one per Quad pulser (total of 4 in system). This could store ~ 150us of waveform per channel on all channels, seems fine. |
| 26 | | * Transferring entire 1.8Mb in 10ms elsewhere requires a 180Mbps link (only) |
| 27 | | * There are 4 MGT transceivers on the !PicoZed 7015 or 7030. |
| 28 | | * Store zero-suppressed waveforms with length limit, when limit exceeded record just width or number of zero-crossings or other info to characterize the pulse. Possibly the absolute-value envelope with low resolution but calculated in the FPGA |
| 29 | | * Readout via MGT point-to-point to !PicoZed, thence over Ethernet |
| 30 | | * Worst case bandwidth: 1.8Mb * 4 pulsers * 12Hz = 86Mbps (10 MBytes/sec or 2.7 MHz 32-bit words) |
