|
|
Repairs:
(H): | Front panel lock drilled out due to lack of key. Exterior panels, power supply regulator modules, boards removed for dusting. | |
(H): | Powered up with most boards in. Front panel operational. Problem with some memory locations. The first 64 words on each half-page (512-word) boundary are not accessible. (Appears they may be readable but not writable.) | |
(H): | Front panel lock cylinder cover installed. | |
(H): | HALT, A14 and M15 lamps are burned out. Parity Halt lamp moved to A14. | |
(H): | Broken lower lever on INHIBIT board replaced. | |
(S): | Simple test programs (incrementing loop, walking bit display) manually loaded and successfully executed. | |
(H): | With X/Y driver board reverse-engineered, memory problem traced to bad transistor in driver BR2b-. BC junction reads at almost double 'resistance' than equivalent junctions of other drivers. With the pins disconnected, both junctions read open. | |
(H): | X/Y driver board transistor for driver BR2b- replaced with 2N2219. 8K memory now fully functional. | |
(S): | Assembler and disassembler written. | |
(S): | Simulator written. | |
(H): | Top and bottom panels cleaned, bottom and back structure wiped down. | |
(H): | Original mil-spec power connector (Amphenol MS3102 A? 22-9P) removed, replaced with wired-in power cable. | |
(S): | Time-Base Generator successfully tested with busy-wait program. | |
(S): | HST interface successfully tested with busy-wait input and output programs. | |
(H): | Side and back panels cleaned. | |
(S): | PFR loader routine successfully executed. Initial problem with data overrun when data sent to the HST/loader has only one stop bit (even though HST interface is strapped for one stop bit). Runs successfully when data is sent with two stop bits. | |
(S): | HST interface successfully tested with interrupt-based I/O programs and combined interrupt-based Time-Base Generator / HST interface program. | |
(S): | First version of monitor (memory commands) successfully tested. | |
(S): | 7970A Tape drive exercised via monitor commands. | |
(S): | Data (3 files) recovered from the 'Mag Tape System' tape (the one tape that came with the drive). The first couple of records of the first file had parity errors. | |
(S): | Version 2 of monitor installed. | |
(S): | HST properties examined to determine why it doesn't receive properly when sent characters with a single stop bit. Problem traced to busy flag remaining high through end of first stop bit and into following start bit. | |
(S): | HST jumpered and tested at 9600 and 19,200 BPS. Successful. | |
(S): | Problems with tape I/O, most commands work but any commands involving data transfer fail to terminate (command channel flag is not being set). Problem suggests an absence of read clock from the drive to the controller. | |
(S): | Tape I/O problem seems to have gone away, perhaps as a result of rearrangement of equipment and cabling. | |
(S): | Instruction rate tests performed, establishing basic memory cycle time and instruction period for single-memory-reference instructions at 1.6µS, the same as HP documentation specifies for the HP2116B. Two tests were implemented by counting loops of 8 instructions and 1000 instructions over 1 and/or 10 second intervals. | |
(H): | Overflow lamp burned out. Indirect lamp moved to Overflow. | |
(S): | New version of NSOM using DMA for mag tape tested. Problem with count of words transferred. Traced to lack of upper two bits of DMA length register, resulting in subtraction calculations being incorrect. | |
(H): | Heatsink modules removed, compressed nuts re-tapped so screws don't bind. Screw hole for A301 drilled out for better fit. | |
(H): | 16V zener (2184) on A301 blown while measuring voltages (shorted from +12 to case/base of 2N1304. Replaced with MZ1000-19 (17V, measures at 16.8V). Shutdown tested by opening new zener, still functions, so transistor and IC input are OK. | |
(H): | Power supply diagram / reverse engineering completed. | |
(E): |
Powered up. Blinken program runs after a memory correction.
- MonitorV2 shows boot message but displays 'interrupt at 11' repeatedly after being started at 000100. - BBL has numerous memory errors. Re-toggled. - MonitorV2 reloaded via BBL, runs. | |
(E): | Powered up. Blinken (017100) and Monitor (002000) run. | |
(H): | New CM345 bulbs inserted in HALT switch, M15, INDIR, INTERRUPT & 2 SPAREs. |
2116C Unit Log
| 7970A Unit Log
| 2748A Unit Log
CTµL | 2116 | I/O Interfaces | Programming Ref | Software | 2116C Refurb HP 21xx Series |
bhilpert Jun 2004 |