Inductor on HP 25C board

[Harald]

I have acquired a 25C that I would like to repair if possible.
So far I have cleaned it and found an inductor falling appart. It is the component on the top right in the picture (above the right ram chip).
Does anybody know what it does and what value it should be? Mine is non-conductive



Cheers,
Harald

Edit:
OK, I answered the question myself. It is the oscillator and according to Jacques Laport it should be 150µH.

[Harald]

The ACT and anode driver needed replacing. But now I get the screen filled with zeros every time I press Enter or a function key.
Am I right in assuming this means that RAM is not working? Shame, would have been nice to get this 25C back to life. Now it appears to be a candidate for Bernhards new ACT.

Edit: The issue has somewhat changed. I now have a non-programmable 25C. I have checked a couple of functions and they work, but whatever I enter in program mode, all I get is "13 00".
Is program memory located in one of the chips and the other one is for all the rest?

[Geoff Quickfall]

That is a classic dead ACT display. Therefore it is hard to tell if the RAM or ROM chip are dead.

Bernhardt ACT would solve all three problems as it has self contained RAM and the ability to switch between the onboard ROM or use the calculator ROM. You can run Bernhards ACT in enhanced mode or original mode for authenticity.

I have an HP 25 that exhibited all the 0's after a full service. Replaced the ACT with a PANAMATIK flashed to HP 25E Ir (print able enhanced mode) only to discover the out gassing nicads had destroyed the ROM also. No problem, switched on power up to the new ACT ROM and never looked back.

So, yes, the 00000000000's display is definitely an PANANMATIK ACT repairable event.

The video is redone and show the very same HP 25 running the version 1.02 67E Ir in conjunction with the infrared printer.

*****Edited due to original post edit. See Berhands response on the RAM chips, that's why there are two in the 25 on the board.****

Cheers, Geoff

[PANAMATIK]

(09-21-2015 03:48 PM)Harald Wrote:  The ACT and anode driver needed replacing. But now I get the screen filled with zeros every time I press Enter or a function key.
Am I right in assuming this means that RAM is not working? Shame, would have been nice to get this 25C back to life. Now it appears to be a candidate for Bernhards new ACT.

Edit: The issue has somewhat changed. I now have a non-programmable 25C. I have checked a couple of functions and they work, but whatever I enter in program mode, all I get is "13 00".
Is program memory located in one of the chips and the other one is for all the rest?

Program memory is located together with LastX Register in one RAM chip, the 8 registers 0-7 are located in the other RAM chip. If you get always 0.00 by recalling a register and GTO 00 in PRGM mode then the respective RAM chip is defective. Theoretically there could also the DATA line of the ACT defective, which reads and writes the RAM data.

The stack registers are not located in RAM chips, they are ACT registers. All calculations are valid. Thus you have a non-programmable HP-25 without registers, perhaps an HP-21 minus one -> HP-20

Bernhard

[Harald]

All the registers are working and retain their contents even when the calculator is turned off.
LastX in deed always returns zero. So I guess it is that RAM chip that is bad. Which one is it? Unfortunately the IC markings aren't legible any more.

[PANAMATIK]

(09-21-2015 08:50 PM)Harald Wrote:  All the registers are working and retain their contents even when the calculator is turned off.
LastX in deed always returns zero. So I guess it is that RAM chip that is bad. Which one is it? Unfortunately the IC markings aren't legible any more.

I don't have knowledge about which RAM represents registers and program. You have to remove one of it to find out.

Each of the two CMOS RAMs in HP-25C contains 8 registers, which is 56 byte. The PMOS RAM in HP-25 contains 16 registers in a single chip. I had a defective HP-25 RAM which consequently lost both memories, program and registers.

Bernhard

[jebem]

(09-21-2015 08:50 PM)Harald Wrote:  All the registers are working and retain their contents even when the calculator is turned off.
LastX in deed always returns zero. So I guess it is that RAM chip that is bad. Which one is it? Unfortunately the IC markings aren't legible any more.

Hi, Harald,

Looking to the posted picture, we can see a lot of battery acid corrosion around the IC's. A few things could be done before concluding it is a defective IC.

1.- Wash the PCB with warm water and dishwashing soap using a soft brush; try to go under the IC's on the component side by using a thin medium soft brush; then let it dry for a while before proceeding with step 2.

2.-Wash the PCB with abundant isopropilic alcohol using a medium soft bush around the IC pins and under the IC's; then let it dry at room temperature for a few hours.

3.- Probably one or more of those IC pins are not making good electrical contact with the PCB holes anymore. This next step is more critical to do as it requires good soldering skills. We need to undo those soldering joints and apply good fresh solder.

As a side note, professionally we start from step 3 to remove the suspected IC's, then proceed with the cleaning, and at last place back the IC's (same ones if there are no stock available).
But in this case, it is much easier to do it as suggested.

Good luck with your new machine, and please keep us posted here about your findings.

[Harald]

Thanks to everyone for their comments and advice!

So far I have cleaned the PCB (the rough way by immersing it into 25% vinegar, the picture was already after inital cleaning, looked much worse to start with). I have replaced the broken oscillator inductor, then the ACT and the cathode driver wich both were not working properly. I replaced the ROM, but I think that was not necessary (but it gave me a chance to clean the PCB under the ROM) and finaly desoldered the RAMs and found that if I swap them (I installed sockets) the registers stop working and program memory partly works. At first I do get the correct key codes, but when I step through the program all I get is key code 40 (rather than 13 00) with the broken memory chip or without one at all. I wonder what that means. Are both memories not working properly? Or is there something else wrong?

Maybe someone has a memory chip he or she no longer needs, maybe because Bernhards ACT was installed, and would consider parting with it (for a resonable financial compensation)?

I know I could get this calculator back to life by simply installing Bernhards ACT. I already did that with one of my dead calculators, but in this case i'd prefere to get it working with the original RAM chips. Just because I really like these ceramic ICs with their gold plated lids ;-)

Cheers,
Harald

[Brad Barton]

(09-21-2015 03:48 PM)Harald Wrote:  Edit: The issue has somewhat changed. I now have a non-programmable 25C. I have checked a couple of functions and they work, but whatever I enter in program mode, all I get is "13 00".
Is program memory located in one of the chips and the other one is for all the rest?

I get exactly this response with a 25 (non-C) that Geoff helped me repair at HHC2015. It will register the shift keys in PRGM mode, but as soon as I finalize the program step I get 13 00. Reviewing the program lines shows the same code, although all 49 lines are visible. Also LastX returns zero, just like yours.

It sounds like a bad RAM chip like yours. I may have to get one of PANAMATIK's ACT's to get it back to working condition, unless I find a bad solder joint or similar.

EDIT: Mine doesn't have the same RAM chips as yours being the non-C model. A bit of a different layout.

Brad