HP-71B + 96KB Port 5 == short battery?



#2

I have two 71Bs now, The first one came with a 96K module plugged into the card reader port. I thought that was pretty cool, but I started noticing difficulty with powering the machine on. If I wasn't hooked up to AC, I tended to get "Memory Lost". Checking the batteries, they were indeed flat. This wasn't a huge deal as I could reload what I wanted from my 9114, but it was annoying. I switched the expansion RAM over to the second machine when it arrived, and discovered that its batteries too were dead, after only 24 hours in the machine.

My question is, generally, how much life do people get out of a set of alkaline AAA cells in the 71B? I may just be leaving the darned things on by mistake, assuming they have an "auto off" feature.


#3

Howard,

The 71B has a very slow resting metabolism, and when off it should maintain memory for many months on a set of alkaline AAA batteries. If left on by mistake, it should turn itself off automatically after ten minutes.

From what you have said, it sounds as if that 96K memory module has a short of some sort. In my experience, a 71B with 4K, 32K, or 64K front port RAM modules, or with a 128K card-reader port module, can still retain memory for many months on a set of batteries.

It is certainly possible for a faulty RAM expansion module to cause rapid battery drain. This has happened to me twice with expansion cards for my HP Omnibook 300/425/430 computers. I elected to throw the expansion cards away rather than lose the long battery-life in those computers. You will no doubt want to do further testing before taking such drastic action with that 96K module.

Good luck in solving the mystery,

Tom


#4

Is 96K a strange number for port 5? You mentioned 128K modules, which would seem to be more sensible. Maybe it's missing 32K, and that's another symptom.


#5

My only experience has been with the 128K module for the card-reader port, but 32K and 64K were also made for that port, so 96K is not unlikely. Someone else on this forum can probably tell us whether a 96K was made.

Tom

#6

Howard,

According to this thread from the archives:

http://www.hpmuseum.org/cgi-sys/cgiwrap/hpmuseum/archv001.cgi?read=2052

a 96K RAM module was made by Handheld Products for the 71B card-reader slot. That is probably what you have. And they were apparently quite rare -- a consolation to a collector, if not to a user!

Cheers, Tom


#7

Thanks for the reference!


#8

Hi Howard.

I have no direct experience with those products but I think you may have a defective (static damaged) 128k memory module. Maybe it's a four chip module (32k per chip) and one ram chip is faulty or electrical connections to it inside the module are are bit funny
(bad). If so then that may explain botht the odd size and the
rapid power drain you get. The fact that it runs indicates to me it's probably quite salavageable.

I used to fix video arcade gae equipment so I've seen some real bizarre logic faults in my time. It sounds like maybe even you have
a case where one or two chip select lines to one memory chip has been lost (open circuit) and it is (1) not "selecting" and (2) leaking power when system is shutdown. CMOS draws power at a high leakage level of up a few milliAmps if the logic control pins are left floating (open circuit)... and there are mains electric fileds around, which get picked up by the high impedance inputs and affect the circuitry (putting nodes all through the chip in a partially on (=leaky) state...).

Just a hunch, hope this helps...

Don W


#9

I wondered about that. 96K seemed like a strange number. I believe the ports on the 71B take a maximum of 128K, without bank switching.

I've just been leaving the machine connected to line power all the time. The extra RAM is too useful for me to risk cracking the module open at the moment. (It has screws, so "crack" is metaphorical, probably.) I'm also deterred by the fact that I'd be clueless as to how to diagnose any fault that wasn't apparant from corrosion or heat damage. If they used standard RAM, I guess I could dig up pinouts for the ancient parts. And I'd presumably have three to 6 good chips to compare to one to two bad ones. But I'm nervous about sticking probes onto the pins of CMOS parts.


#10

Howard.

When you get a round tuit ;-) you could always take photos and
e-mail them to me or post them here. That would help.

The approach I would take is to find out if the chips are
socketed (unlikely) and swap chips. Failing that, if possible
write software which tries to read and write to particular bytes. You won't see much with a 'scope, let alone a multimeter, but if the chips are socketed or other wise can be electrically disconnected (disconnect the "chip select line from the pc board and tie it "off" [usually +5 volts=logic 1]") that takes it out of circuit. A meter is good for checking power supply is good n all memory chips, though (first step I would take).

Assuming four 32kB chips (big assumption), doing this to each chip, one notes what memory capacity the machine reports or which bytes don't exist when peeked and poked ;-). When you find a chip which doing the above on gives no change, you have found your problem.

The other (perhaps better) way to do things is to use an electronic
multimeter in OHMS mode (NOT the continuity test mode!), to check to see if all the common buss lines are intact from chip to chip.


(In ohms mode, the meter generates 0.3 volts across the tips, quite safe. In continuity mode it puts 3 volts across the tips with more current behind it. The logic chips may not tolerate such "abuse".
I am being cautious, but that pays...)

In general do tests with all equipment power off and batteries removed! When working on the logic, touch a GROUND buss first to put yourself at the same potential as the logic to avoid static damage.


For example, one can check the address buss one line at a time,
then the data buss, power supply rails and also the chip enable control lines. Laborious, but effective.

Also, if you had a practised eye you could look for "cold solder joints". I used to fix WYSE WY-60 terminals and they were
notorious for really bad solder joints to the stainless steel
jumper wires that were used in copious numbers to avoid the cost
of manufacturing a double sided PCB. They were a real pain in the ass, so I found it better to do all the joints that looked even slightly suss in one hit. That always worked...

Also, chip sockets sometimes give trouble.

Hope this helps.

DW

#11

I have a 96K RAM in my main HP-71's card reader port, made by CMT. They offered 32K, 64K, 96K, 128K, and 160K in the card reader port. I wished I could have gone for the whole 160K, but memory was expensive at the time, and I set a budget and got what I could afford. (Later I did buy another 64K front-port RAM when I was able though, again CMT.) My card-reader RAM port numbers there are 5.00, 5.01, and 5.02. My batteries need changing maybe every year or two, but I have a friend who only has a 32K (I think) front-port RAM in his 71 and his batteries don't last nearly as long. I wonder if your internal module battery is not just dead, but maybe even shorted. It does seem however like the power-down memory-protection circuit should make the AAA batteries' current unable to get to the lithium battery.

There was a company at that time where you could send your 71 and have them install 256K of RAM internally, so it didn't take up any of the ports.

A couple of years ago I got to wondering if the old lithium batteries might leak, and if I should take them out. I called up CMT to ask if these batteries were capable of leaking, but there was no longer anyone there who even knew what an HP-71 was. What a shame! They took the stance though that if in doubt, you better remove the batteries. With the scores of files I have stored in there, I'd have to take some time to move things around to remove the modules, take the batteries out of them, and plug them back in with their memory lost, ready to re-load. I never did all of that, but I did open up the card-reader-port module without removing it, and found a little battery with leads. If memory serves me right, it was about the size of a shirt button. It showed no signs of leakage. I just put a sleeve over it and put things back together. Again if memory serves me right, I saw no IC packages in there-- just dice with little blobs of epoxy over them, right on the PC board. In any case, you certainly won't find DIPs or sockets if you have the CMT module.

The brushed aluminum piece around the LCD seems to be grounded internally through a very high resistor value for static-dissipation purposes, but it would be good to verify that before counting on being able to avoid damaging CMOS parts when you probe them by keeping a finger in contact with that aluminum. Does anyone else here know?


#12

Yeah, ESD protection scheme for sure Garth.
Being metal, the face would store a charge.

DW

#13

There's no maker's mark on the outside of the module. Just a serial number. There are two screw holes. One has a very tiny torx head for which I don't have a small enough driver. The other looks into the module showing the side of one chip and part of the PCB. That side is held down with electrician's tape. The two halves of the module case don't mate very well, and you can see the PCB, plus chips on both sides.

I suspect that what I have here is a hardware hack that went horribly wrong. 8) ("Horrible" is overstating the case. The RAM works as RAM.)


Possibly Related Threads...
Thread Author Replies Views Last Post
  HP-41 module covers and side port cover Nate Martin 1 169 12-02-2013, 01:51 AM
Last Post: Giancarlo
  HP Prime - Short "learning" modules CR Haeger 1 149 11-27-2013, 02:13 PM
Last Post: Jonathan Cameron
  I have written a short introduction to the HP Prime Michael Carey 7 342 11-18-2013, 08:04 PM
Last Post: Michael Carey
  HP-65 short circuit Ignacio Sánchez 2 199 10-22-2013, 08:27 AM
Last Post: Ignacio Sánchez Reig
  Yet another benchmark port on the wiki: Savage Pier Aiello 35 1,239 09-26-2013, 03:22 AM
Last Post: Pier Aiello
  HP 50G serial port programming Chris Smith 2 200 06-15-2013, 02:47 PM
Last Post: Chris Smith
  OT: RPN port to Casio fx-cg10/20? Matt Kernal 4 253 04-30-2013, 12:16 PM
Last Post: hugh steers
  HP-42S with Electroluminescent screen and simple I/O port Jose Poyan 8 376 03-27-2013, 07:11 PM
Last Post: Jose Poyan
  AME Design Port Extender for the HP41 Dane Spillers 2 175 03-14-2013, 07:58 PM
Last Post: dane spillers
  FRAM71 for HP-71B, short update #3 Hans Brueggemann 15 589 01-20-2013, 10:22 AM
Last Post: Jerry Raia

Forum Jump: