Weird HP-48G problem


Hi all,

I have a HP-48G that has a problem with the "ON" key. It doens't work 'alone' (on/off) but WORKS on the auto-tests. I cannot switch it on or off (only by the "OFF" command) and only turns it on by the soft-reset (ON+C).

Anyone have seen this problem?




Seems that the key itself works, but isn't recognized by the calc during 'normal' operation. Have you installed any libraries that could interfere with the ON key? Have you tried memory clear?

Cheers, Victor


I just acquired this calculator (used, described with this problem and a missing line on the display) and put new batteries. No libraries, no software loaded, just plain "Memory Clear" condition.




It is not the on-C that is making it work it is the pressure at the top of the key pad that is making it work. Press on the bezel above the C key, and I'll bet the on button will work at that point. I have not seen the inside of the 48 but my guess is that is where the Keyboard connects to the main board. (correct me if I am wrong please)


I disassembled it and cleaned the keyboard contacts, and now all keys work just fine. I also cleaned all LCD contacts but there is a horizontal line missing (#32, if I counted right). There isn't any corrosion or internal leakage. The calculator works fine, pass all self-tests.

I had this kind of LCD problem when I tryed to replace the old LCD from my HP-42S with one (identical) from a 'deceased' HP-17B. The "new" display has a missing column. Tryed to clean it in all ways I could imagine... I just put back the old one (some black spots on the left) and the column was there... I looked on the display carefully and can't see any imperfection. Bad luck...

This HP-48G was cosmetically perfect, noone tryed to opened it (before me) and there wasn't any battery leakage.

Any tip on how to fix this LCD problem?

Thank you all!


My experience with Pioneers is that any foreign matter (even dust) on the rubbery contact strips that carry the PCB signals onto the LCD edges will result in one or more missing segments.

Those strips have been called "zebra strips", as the early ones used to be black & white striped -- black conductor stripes with white insulators in between. The Pioneer ones tend to be uniformly gray, but I assume there are mutually isolated conductive layers or channels molded in which connect the PCB traces to the almost invisible contacts on the LCD.

If you haven't already, make sure the rubber strips are absolutely clean and free of lint, dust & whatever, while assembling the PCB/LCD/front case. Also, flipping one of those strips end-for-end, or swapping them top & bottom, will isolate a bad contact strip from a PCB or LCD problem.


On the 42S I tryed clean, swap, exchange with the other strip pair from 17B, no success. With the original LCD, it works, so I discard the strips/PCB contacts as a guilty. I think that is some bad contact/track on the LCD, but it is invisible.

On the 48G, I cleaned also, but I haven't another display/strips to exchange.

I already fixed several other display problems with just cleaning, but this two eluded me...

Best regards!



I would think that swapping the upper and lower strips (I'm assuming there are two) in the 48G should help narrow the list of possible problems.

If, after swapping, the same pixel row is not displaying, that would implicate the PCB or LCD. If, however, a different row or column or annunciator is affected, then a strip would likely be to blame.


I don't swapped the strips, I'll try later. But I don't like to open/close it often, the metal clips that hold the PCB with the case/LCD are too thin and can be broken (as on my 42S...)

I'll try to check the tracks on the PCB also. I don't know if the display of the 48G is "partitioned" with 2 halves controlled by different pins (the 42S is 131x16, but the LCD controller is 66x32, 2 partitions plus symbols). On the 42S I'm shure that is the LCD, on the 48S I'm thinking of the PCB or controller chip, as the missing line is a horizontal one of the entire extent of the display (if the display is partitioned, this is covered by 2 lines).

Best regards,



re: twisting the metal tabs too many times . . .

You might consider building a "jig" -- a fixture that holds the pieces together and applies enough pressure for testing, thereby avoiding metal fatigue.

Randy Sloyer, I think, uses one for work on Pioneers -- I assume something similar could be built for 48G's. It needs to stay clear of the display (so you can see what's going on), press the PCB without shorting traces together (insulate its points of contact), and apply even pressure throughout the PCB/LCD/keyboard contact area. Not easy, but not impossible. (Easy for ME to say!)

Anyway, it sounds as if such a device has helped Randy -- I suppose it might help you, if you do a lot of work on 48G's. (Just a suggestion . . . )


It is really a good idea! I'm fixing calcs that I buy cheap (non-working), fix them for my collection or to sell later. Most of them is Voyagers and Pioneers, this 48G is the first one. Is amazing how resistant those cals are!

I do repairs on my spare time and don't have a dedicated space (yet) to do it. I'm planning to build a more sofisticated "equipment" to fix the calcs, a ultrasonic washer, this "jig" to press the displays (adjustable to more than one calculator serie?) and get some conductive paint, among other things...

Best regards,


Possibly Related Threads…
Thread Author Replies Views Last Post
  hp50g screen weird line Sok-khieng Chum Hun 2 1,581 09-10-2013, 08:11 AM
Last Post: Sok-khieng Chum Hun
  Riemann's Zeta Function update (HP-28S, HP-48G/GX/G+, HP-49G/G+/50g) Gerson W. Barbosa 0 1,059 06-30-2013, 01:01 AM
Last Post: Gerson W. Barbosa
  weird statistics bug in wp34s Andrew Nikitin 5 2,083 06-20-2013, 01:54 PM
Last Post: Namir
  OT: DIFFERENTIAL EQUATIONS using the HP-48G/GX Ivan Rancati 2 1,371 06-04-2013, 04:11 AM
Last Post: Frank Boehm (Germany)
  Able to set HP 48G axis tick marks at integers? Randal B 1 1,054 01-24-2013, 12:40 AM
Last Post: Chris Dreher
  HP 48G -- entering more than one equation into the PLOT application Randal B 1 1,063 01-24-2013, 12:25 AM
Last Post: Chris Dreher
  HP 48G series -> 20 years NACHO 0 923 12-13-2012, 09:40 AM
Last Post: Nacho
  Weird behavior from lithium AAs Dave Britten 4 1,569 11-14-2012, 06:22 PM
Last Post: Lars Bergström
  ROOT bug? HP 48S/48G Eddie W. Shore 8 3,810 07-13-2012, 07:05 PM
Last Post: Eddie W. Shore
  RPL prog for Fibonacci on HP 48G needs minor modification. help. wildpig 68 16,231 07-09-2012, 09:38 AM
Last Post: Gilles Carpentier

Forum Jump: