HP Forums
HP-67 with some odd good and bad things - 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: HP-67 with some odd good and bad things (/thread-996.html)

HP-67 with some odd good and bad things - [kby] - 03-29-2014 07:21 AM

I recently picked up a '67 fairly inexpensively as it was sold as for parts/not working, the user stating he just got a flickering display. I originally got it as I have another mostly rebuilt -67 that is missing a few keys (they should be around somewhere—lost them when the -67 had some other parts used to fix my true love, which is a -65).
Anyway, I pop the battery in and the calculator acts fine upon turning on. Trying to read a card started the motor but didn't pull the card; figured it was just the usual gummy wheel issue. So I took the back off. There was some white residue mostly in the battery compartment. I presumed it was KOH from a bad battery, but it won't all go away with vinegar, so I'm not sure. Isopropanol actually removes black from the plastic, but when it dries it looks greyish. There was also some tell-tale green (copper) corrosion on some of the connectors, and a significant amount of rust, which looked to be mostly from the screws (the part of the label that covers the screw holes had two solid brown powder prints stuck to the adhesive; there was also a light rust stain on he keyboard. That mostly cleaned up except the plastic.

Now is where it gets to the first interesting part. Upon taking apart the card reader. the wheel wasn’t suffering from the usual gummy wheel problem. And unlike other wheels I've seen it wasn't white or natural latex yellow, but basically black. It had some of the white dust which wiped off, and I treated it with some rubber renue, although not much black came off. The wheel is soft and pliable and (decided to transplant to a -65) works just like new.

I am fairly sure from other signs that this calculator has never been opened up—label was basically pristine, plus, as I mentioned, the rust prints on the underside of the label. Has anyone seen anything like a wheel not made from the polyurethane?

This calculator was made in Singapore and the SN molded into the case is 1808S02243, matching the Singapore label.

Although I have successfully used both the square and regular o-rings, they don't seem to work as well as the original for me; usually the start slipping on the inside unless glued (cyanoacrylate) and they don't stack quite long enough (so if not glued one tends to wander away from the other). I have tried the silicone tubing but the stuff I get ends up being a little too small, stretches, and is hard to cut exactly (too long and it binds; too short and similar issue as with the o-rings, and it seems to be soft and degrade on the surface fairly quicly.

Now for the problem part. I was in the process of adjusting the card reader, and at the beginning of this, it appeared everything was OK with memory and stuff; just not reliable with the cards. Somewhere along the way I found that it couldn't be programmed in the sense that key presses would advance the step number, but all registered as R/S (the default fill). The same keys would work fine in run mode. I thought I saw someone here ask about bad RAM (or I guess RAM/ROM) chips in a -67, but I couldn't find it so if anyone can point me to that I'd be grateful. The issue is on the CPU board; switching to a dfifferent (good) CPU board fixes the problem.

RE: HP-67 with some odd good and bad things - HP67 - 03-30-2014 07:23 AM

If you don't get other suggestions, you can almost certainly get help from hpcc.org. There is at least one member who can fix anything.

RE: HP-67 with some odd good and bad things - Paul Berger (Canada) - 03-30-2014 04:38 PM

I had earlier reported a ROM issue with a 67 in the old forum, in my case, as near as I could determine it performed functions ok but it would not display anything correctly, I did isolate it down to a single ROM, but I was only able to do that by desoldering the chips and swapping them. I have a 97, which is very similar hardware, that I have all the chips in sockets, that I used for that purpose. Since the 97 is similar the service guide for the 97, available on the museum DVD, can be helpful.

Your description of the damage sounds like liquid damage, many beverages are slightly corrosive and can get into almost every corner. The same 67 that I had the ROM issue with showed signs of liquid damage as well. Like your screws where rusty, there was also some corrosion damage on the main board and the aluminum name plate showed some corrosion damage. You may find that there is even some residue on the board depending on how far the liquid seeped in.

One of the first thing I would check is to make sure that Vss (+6.25) and Vgg (-12.0) are correct they don't have to be right on but should be within 10% of nominal value, I have found that faulty capacitors can cause the voltage to be low, and in one case on a 97 the capacitors measured the correct value but just had high serial resistance (ESR). Ground is pin 5 on the RAM/ROM chips Vss pin 1, and Vgg is pin 3.

Unfortunately the RAM is spread over all 4 of the RAM/ROM chips, the 8 pin chips on the board this is why it usually ends up with swapping chips to find the bad one, I am not aware of any documentation that states what portion of the memory is on what chip. Now you say that functions entered from the keyboard work fine, but they would not involve any of the RAM as the stack registers are on the ACT chip. What happens if you try to save a value in one of the storage registers? The 26 registers would take up all the RAM in about 1.5 of the RAM/ROM chips the remainder of the RAM would be the program storage memory. Did you go all the way through the program memory? If the RAM portion of one chip was defective I would not expect you to get an incorrect value stored throughout the whole of program memory. If you do get an incorrect value store throughout the whole of the program memory I would be more inclined to think it is a ROM issue or perhaps even an obscure ACT issue, same thing with the storage registers since they would span more than one chip. Is the card reader working? if so it might be of some value to compare what happens when you read in a program from a card as opposed to entering it form the keyboard. You could do the same thing for the data registers.

RE: HP-67 with some odd good and bad things - [kby] - 03-30-2014 08:56 PM

(03-30-2014 04:38 PM)Paul Berger (Canada) Wrote:  I had earlier reported a ROM issue with a 67 in the old forum,

Thanks for your response. I tried looking for your post but couldn't find it in the old forum or the old forum archives (at least archives 17–21; it looks like the earliest you appear in is archive 19). It sounds like what I might have seen posted and I'm just mis-remembering it. I found one response for a -67 card reader by you, but doesn't seem to mention a bad ROM or what it sounds like you wrote above. [Update—I think I found what you were talking about—could have been the one I saw as I was probably looking a -97 things at the time)

Turns out that part of the issue seems to have "fixed itself" somehow while I was finishing up the reader. I switched to the other CPU board I have and it was fine, although it still took some fiddling to get everything working with that, then, when I switched it back (see below for why), everything was fine, at least for that problem. I hadn't actually changed any individual components at this point.

So…everything was working ok, until I discovered that the CPU board that was switched in had the buggy ROMs (inverse trig functions bugs). So I switched back in the one that had previously appeared to have the bad RAM, and at this point keying in programs and stepping through memory seemed fine. And it didn't have the buggy ROMs, either…but it turns out it had a bad card reader controller…got the WP and HDS transition, but didn't ever turn on WE. So I ended up swapping the card reader controller chip (ugh) and things seem copacetic again for now. I know some people would find it interesting to have the buggy ROMs, and I might, too, if i had another machine complete, but for now, I wanted a "debugged" one.

I'd agree there was some liquid damage, although it seemed to be more of sitting and getting flooded from the bottom if so and then maybe turned over. The stains on the keyboard board were on the back (battery) side whereas the keyboard and the plastic overlay under the keys looked untouched.

I'm still interested to know if anyone's seen the black pinch rollers.

And the hpcc forum no longer exists—they refer one to t his forum!

Thanks again for all of the responses so far!

RE: HP-67 with some odd good and bad things - HP67 - 03-31-2014 06:08 AM

I'm sorry, I didn't mean that hpcc had a forum. Only that if you join you'll have access to their doc which includes teardowns and schematics for a lot of devices, and you can get in touch with the members via email.