HP49G repairs---what's the situation?


I now have a 49G with a largely nonfunctional keyboard. I haven't taken the time to identify all of the nonfunctioning keys but I know that Enter, +, -, /, *, Delete, and Clear are all completely dead. I can still turn the calculator on and off. Not very useful as it is.

Any help would be appreciated.



In warranty, HP will replace it. Out of warranty, they will replace it for a fee. I think it is about $90, don't know for sure.

If you're into DYI, try pressing down hard above the F5 key while trying the dead column. If it works with pressure applied, you'll need to open the unit to fix a lost contact between the logic board and keyboard.

Post again if that works and we can take it to the next level.


Randy is on target! Applying pressure above the F5 key does indeed restore function to the dead keys!

The next question is how to open the case without damaging it? It seems to be just a friction fit at the bottom, but there must be some kind of hidden clips further up---how does one deal with those?



Using a 5/32"/#22/4mm drill bit, drill the tops off the six heat stakes in the battery compartment. Do not use a power drill, but simply twist the bit in your fingers. You'll find the blue plastic drills quite easily by hand. You only have to go deep enough so that the rivet releases, you'll end up with a blue ring when you've drilled far enough.

Now find yourself a clean, dry wooden popsicle stick. Seems silly as everything handy is made of metal but the wood will not mar the plastic case. From the bottom, slide the stick up the channel that the cover fits in about a half-inch. Use the stick as a crowbar and separate the case halves. Repeat on the other side. They will pop open with enough force.

Once both halves are free at the bottom, pulling away from the case, free the positive battery clip from it's retainer.

Hold the case bottom in one hand, the top in the other and pull the halves apart. Don't seperate by more than a quarter of an inch or so or you'll break the case.

Now, push the bottom half up toward the top of the front half. The halves should separate and the top should slide up and free itself. You should not have to pry or do anything other than push up and pull apart at the same time. Once it pops apart, you may trash the battery clip, but it does not matter, it is easily bent back into its original position.

The next step is to separate the logic board from the keyboard. Above and below the three large chips at the top you'll see six twist tabs, three above, three below. Study them carefully. Normally, you will have to twist the bottom tabs clockwise, the top counter or anti-clockwise to release. In some machines the direction stated is not correct and you will need to establish the proper direction to free the logic board. You do not want to twist them in the wrong direction as doing so will surely fracture them and you will be cursing the dyslectic person that assembled the unit. Once the tabs are centered, you can lift the logic board up from the bottom and free it from the six cinch points.

It's now time to inspect the keyboard connector, especially on the far left side as the last connection to the left is the right most row that you are having trouble with. Bend the white plastic of the keyboard connector upward and see if the foam pad underneath is centered in the case. Many times it is not and it causes the loss in connection. The solvent UN-DO (aka heptane) available in hardware stores and Wal-Mart will cause the sticky pad to release so that you can remove and re-position the pad in the proper place. Many times you'll only need to bend the keyboard connector upward to reestablish proper connection. Clean the keyboard and logic board connections with a cotton swab and isopropyl alcohol. Reassemble, twisting the tabs in the reverse of the direction as required to remove. You'll get about four to six cycles on these before they fracture so be very carefull as to not to waste close/open cycle for no good reason. It is best to start with the left and rightmost bottom tabs followed by the center bottom then the left/ right upper and finally the center upper. This is the best order to center the logic board to insure the proper connectiion to the very fine pitch LCD connections.

Once back together, simply snap the black bottom back into place. Install batteries and test. If all goes well, your 49G will be back to normal.

Edited: 13 Sept 2003, 8:25 p.m.



Could you find space on your site for this excellent discription of how to take the HP49G appart (and how to put it together again)?

I realise that the HP49G is far too new for you, but it IS already out of production! ;-)



(Please, if you haven't already done so . . . )


Hello Mark,
Check this site HP recommended him to me, he is ausome. His name is Randy and he fixed my HP 41CV. Give him a shout.


Since we are discussing the opening of the HP-49G, is there a way to determine, once open, what exactly is wrong with the COM port? Or possibly even fix it?


The first thing wrong with it is that is it not true RS-232. Nothing like the 48 I/F at all, it looks like the RX and TX lines pass through a 150 ohm resistor with a transorb to ground. The other end of the resistors go directly to the CPU, same pins as the 48. There are no level shifting circuits like the 48. It looks like there are three other active pins that go to the CPU in the same manner.

I haven't put a scope to one, but I suspect the TX swing to be 0 to 5V rather than the standard +/-12. I know a lot of 232 receivers will handle that, but will it work in all cases?


Note that early (ID93... and lower) 49Gs had a "serial port hardware
bug" that caused a weakened and distorted signal. This would usually,
but not always, still work with most RS232 ports.

See http://www.hpcalc.org/search.php?query=serial+port+bug for
more information.



The old bipolar 1488/1489 chip (whichever is the receiver) switched at about +1V, like a TTL chip. I've wondered whether today's CMOS receivers could be depended on to have a similar threshold. The RS-232 EIA spec says from -3V to +3V is undefined. There is a later spec (RS-485? 422?) specifying 0-5V voltage swings (the original Mac serial ports were based on this) but it also specified differential signals, although the chips could be used single-ended.

Possibly Related Threads…
Thread Author Replies Views Last Post
  HP 41CV Repairs William Mabotja 3 1,602 11-20-2013, 04:45 PM
Last Post: Mike Powell
  wp34s, another shutdown situation Andrew Nikitin 10 2,989 06-19-2013, 01:52 AM
Last Post: Paul Dale
  HP 34 and HP 41 Repairs Needed matthewx 4 1,794 07-16-2012, 09:12 PM
Last Post: matthewx
  HP41 Card Reader repairs Steve Hunt 3 1,474 01-02-2012, 02:52 PM
Last Post: Randy
  HP 41 post repairs, yeah been done before but this is simple and it works.....;-) Geoff Quickfall 7 2,057 04-22-2009, 02:52 AM
Last Post: Nigel Bamber
  Update on my 9114 repairs in 2000 Steve (Australia) 0 815 02-08-2009, 09:41 AM
Last Post: Steve (Australia)
  HP 3x Series Internal repairs Frank Barberis 5 1,682 09-10-2008, 09:49 AM
Last Post: peter a walker (papwalker)
  HP-41 Repairs SteveH 0 750 04-02-2008, 06:41 PM
Last Post: SteveH
  OT: Card Reader Repairs for TI-59 Kostas Kritsilas 2 1,253 08-27-2007, 08:25 PM
Last Post: Kostas Kritsilas
  HP33C repairs Marco 8 2,630 10-01-2006, 09:04 PM
Last Post: Vieira, Luiz C. (Brazil)

Forum Jump: