1
\$\begingroup\$

I have recently acquired an HP 6632A power supply and was inspired by Roberto Barrios' project of replacing the screen with an OLED display to do the same. However, it is evident that the 6632A is much different as it doesn't have a front panel board and separate serial connection. Instead, the LCD driver is controlled directly by the CPU through a latch.

schematic Pg.101 of the service manual

Looking up DFI, ISA, SYNC, INA, etc yielded nothing in combination with LCDs.

Picture of the LCD board: lcd

I probed up the 16-pin connection according on the schematic, though I could be missing something.

Signals when "VSET 0" is displayed: logic signals

These signals are sent at ~2.5Hz when idling: idle Here are the rest of my captures

I'm not sure where to go from here. Probe 6 looks like clock (hooked up to pin 12 on the left), yet it stops when data comes in. Probes 4 and 5 appear to be exactly the same, just shifted 5us. I'm not able to get any sort of serial decoding working through Saleae Logic, obviously. I probed the rest of the pins to make sure I wasn't missing any signals (or negative voltage for the LCD). I've tried to compare the enabled segments' binary representation with the signals, but I couldn't make any correlations.

I would appreciate if someone could point me in the right direction. Thanks!

\$\endgroup\$
  • \$\begingroup\$ Are you sure you have the pins labeled correctly? To me it looks like it is a serial data interface of some sorts, and very little info about HP LCD part. Probes 4&5 look like bi-phase clock, so I would presume they were Phi1&Phi2 (O1 and O2). Rising edge of first clock is used to update data on ISA and INA pins, and on falling edge of first clock it is latched in. Of course rising or falling edge of second clock could be used as data is stable on both edges of it. ISA data is sent when SYNC high, otherwise INA data. They might be register/data pairs what parts of the display to update. \$\endgroup\$ – Justme Jul 7 at 9:21
  • \$\begingroup\$ Data is either character codes what to display, or direct data which segments to light up. It may not be difficult to reverse engineer. Just figure out the difference when maybe one of the segments change, either a digit or any single segment. \$\endgroup\$ – Justme Jul 7 at 9:23
  • \$\begingroup\$ @Justme They probably aren't, that makes sense. I figured O1 and O2 should look similar, so something was wrong... I thought it might be Manchester, but I didn't know about bi-phase. This gives me a lot to go on, thanks! \$\endgroup\$ – sealj553 Jul 7 at 14:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.