HP Forums
48SX and 82215A 128KB RAM card odd behavior - Printable Version

+- HP Forums (https://www.hpmuseum.org/forum)
+-- Forum: HP Calculators (and very old HP Computers) (/forum-3.html)
+--- Forum: General Forum (/forum-4.html)
+--- Thread: 48SX and 82215A 128KB RAM card odd behavior (/thread-953.html)



48SX and 82215A 128KB RAM card odd behavior - cruff - 03-22-2014 03:47 AM

I have an 82215A 128KB RAM card that recently started causing my primary 48SX to behave oddly. The symptom is that the stack contents disappear across a power cycle, but the calculator is not complaining about invalid card data at power on (as it does when you remove or replace the memory retention battery). Simple stack manipulation and arithmetic work correctly when the card is installed. The other contents of memory seem to be retained, such as variables, but I haven't filled memory completely either to see if something else might be disappearing. This happens regardless of if the card is merged or not. Physically removing the card results in normal behavior. If I recall the memory layout correctly, variables are kept starting in the on board 32KB of memory and the stack starting in the expansion cards when they are merged.

My first thought was a battery issue. The card's battery was at 2.9V, and a new battery (3.2V) does not change this behavior. The calculator's batteries are at 4.5V. Do I remember correctly that while plugged in, the calculator will supply power to the RAM card when the calculator is off?

Oddly, when I test this same card in my spare 48SX, everything behaves normally in my limited testing. My primary 48SX also behaves itself when a TDS 256KB RAM card is installed (but of course the 48SX can only see half of that one).


RE: 48SX and 82215A 128KB RAM card odd behavior - Dave Britten - 03-22-2014 09:47 PM

(03-22-2014 03:47 AM)cruff Wrote:  My first thought was a battery issue. The card's battery was at 2.9V, and a new battery (3.2V) does not change this behavior. The calculator's batteries are at 4.5V. Do I remember correctly that while plugged in, the calculator will supply power to the RAM card when the calculator is off?

The 48 will only power the card while it is switched on. This is why you should leave the calculator ON when replacing the battery in an expansion card, so the calculator can keep the SRAM alive.

When swapping cards, the 48 will warm start, resulting in a clear stack. Perhaps your 48 thinks the card is being removed whenever it powers off. What happens if you FREE the card, and toss a couple objects into port 0? Do they persist after a normal power cycle?


RE: 48SX and 82215A 128KB RAM card odd behavior - cruff - 03-23-2014 01:06 AM

(03-22-2014 09:47 PM)Dave Britten Wrote:  When swapping cards, the 48 will warm start, resulting in a clear stack. Perhaps your 48 thinks the card is being removed whenever it powers off.

That appears to be the case, the primary calculator's WSLOG shows entries with both of the following:

3 Run through address 0.
B Plug-in card pulled or "bounced" (port status changed).

and even one of these:

C Hardware Reset (ESD or the hole under the "random" foot).

The spare calculator's WSLOG only shows the "B" entry when the card is inserted.

Quote:What happens if you FREE the card, and toss a couple objects into port 0? Do they persist after a normal power cycle?

Objects stored in ports 0 and port 1, which represents the RAM card, do persist across the power cycle, and in port 1 when swapped between the calculators or when moved to slot 2.

I hope my primary 48SX is not getting flaky, as it is in great shape physically.