HP-67 not behaving?

[markhen]

Hi - I’ve had my “best” 67 in a drawer for a while and I decided to use it today after months (a year?). This machine was professionally restored about 10 years ago and is in mint condition. I popped in a battery and plugged it into an USB charge cable from TAS some years ago (that I have used with this 67 many times). It powers up but is not working properly. I hope it’s not dreaded ACT-itis (since my power source is 5V?). Here are the main symptoms (100% repeatable):

- display is permanently in SCI mode with DSP = 0 (only 1 digit of mantissa visible, decimal point, and 2 digit exponent)
- any numeric digit entry gives a display of “-x000000000. 00” (x = 0..9)
- stack lift ops (ex. ENTER) appear to correctly enter single digit into stack (without the minus or the extra 0s)
- arithmetic operators appear to work (single and dual operand) although display remains in SCI DSP = 0 mode
- stack can be rolled and remembers all 4 entries
- primary registers STO and RCL correctly
- PROGRAM mode shows “000” and nothing changes it (entries, SST etc)

Ideas????

Thanks

Mark

[teenix]

I did some experiments with my 67 emulator and if I fail the RAM at addresses 48 to 63 and only read 0's from this memory range, then I can recreate your problem.

It seems like the RAM in this address range has failed. I'm not sure which ROM IC this RAM belongs to as I haven't had a chance to read the 67 ROMs and run RAM tests for real yet.

Sorry :-(

Tony

EDIT:

It may be something simple like a dry joint on the DATA pin for this particular IC. The DATA pin transfers RAM data in and out of the device, it is pin #5. As mentioned, I don't know which IC has the problem, but you might consider checking for corrosion on the ROM IC's. If you are handy with a soldering iron, I would try re-soldering pin #5 on all the ROM ICs.

ID's for these 8 pin IC's are probably: 1818-0231, 1818-0232, 1818-0550 and 1818-0551.

Try not to touch the component connections on the board, the IC's can be damaged by static from your body.

[markhen]

Tony - thank you so much for recreating! It does indeed seem like part of the RAM has failed (for example where the display format is stored, and program steps?). I am experienced with the iron and static prevention, I’ll try not to mess up the rear label and crack this baby open to investigate! Due to travel this probably won’t be until later this week. I will report back!

Cheers

Mark

[AndiGer]

(07-27-2022 06:03 AM)markhen Wrote:  This machine was professionally restored about 10 years ago ...

(07-27-2022 03:30 PM)markhen Wrote:  I’ll try not to mess up the rear label ...

If wise maybe the restorer left out the screws underneath the label?!?

[markhen]

Good point!! I will remove the other screws first - thanks for idea! The restorer did an amazing job of keeping the label smooth (far better than my attempts on other machines).

[Duane Hess]

Since there are so few (as far as I know) people who restore 67/97s who have large numbers of serviced units in their "resume", who restored yours?

After 10 years is this person in business?

Everyone knows Randy Sloyer, there was also a guy in Placerville CA, but both have stopping servicing due to health years ago.

thanks!

[markhen]

Duane - checked my records and my 67 (and 2 65s) were serviced excellently by Richard Anthony in Placerville, CA in early 2010 (>12 years ago now!!). I don't know if he's still servicing?

Regards -

Mark

[markhen]

I eventually worked around to opening the calc - my findings:

- AndiGer was spot on - the restorer did leave out the screws under the label! Thanks AndiGer!!

- The PCB is in excellent shape (no corrosion etc.). I visually inspected (with magnification) all the solder joints and they are excellent. Even so, I did carefully re-solder all the ROM/RAM pin 5s per Tony's suggestion, but sadly no change in symptoms :-(

I'm surprised to see this failure in a clean working unit that was stored in a "cool, dry" drawer, but I guess age takes its toll on everything, eh?

I have some potential donor 67s that I will raid to repair this unit - it's cosmetically perfect and the card reader should still be working (since it was well restored already).

Thanks to all!

[teenix]

The faulty ROM will be either 1818-0231 or 1818-0232. I'm not sure which as both of these ROM's I have on the bench have failed RAMs.

I might be able to get a working one to read, if so I will post the serial number.

1818-0228 (67 and 97), has RAM $00 - $0F
1818-0226 (67 and 97), has RAM $10 to $1F
1818-0231 or 1818-0232 (67), have RAM at $20 to $2F, and $30 to $3F but not sure which one.
You need to replace the one with RAM $30 to $3F.

I am making progress on ROM replacements and you never know, I might be able to offer one in the not too distant future. I know my ROM decoder works, but the ones with RAM included are more complex.

cheers

Tony

[Dreato]

Sadly, applying 5 volts to the classic series works MOST of the time, right up until it does some damage. HP regulated the applied voltage in their early chargers to 4.5 volts for a reason. Later chargers dropped the regulated voltage further to 4.1.

[markhen]

Dreato - your point is well taken... A quick look at the schematics confirms what you say. I will stop using my USB-Classic charging cable (and/or modify it to only supply 4.1V).