Post Reply 
Sob
07-12-2022, 06:36 AM (This post was last modified: 07-12-2022 06:37 AM by teenix.)
Post: #1
Sob
I was just removing a ROM 0 chip to read. I had it all unsoldered and was loose, but as I was pulling it out with just a little pressure and the case just crumbled and 7 pins broke off - from a HP-92 to make things worse :-(

Obviously the plastic material in these degrades too.

For future reference, even more gentler handling is required.

The IC is bricked for normal use, but luckily the pins that broke were for the display anodes and the chip still functioned and I was able to get the code from it.

I hope now that I find my way with the ROM chip replacements.

cheers

Tony


Attached File(s) Thumbnail(s)
   
Find all posts by this user
Quote this message in a reply
07-12-2022, 10:33 AM
Post: #2
RE: Sob
Ouch!
Find all posts by this user
Quote this message in a reply
07-12-2022, 10:42 AM
Post: #3
RE: Sob
Ow, my condolences.

Is that package what I believe was called a "phenolic"? I recall the IC's with the smooth plastic, and the others that had the more matte finish and were less durable. I seem to recall they would crumble a bit like bakelite given the chance.

Mahalo

HP-45/65/67,HP-10/11/12/15/16c,HP-21/25/25c/29c,HP-31e/32e/33e/34c,HP-41C/V/X,HP-22s/27s/32sii/42s,HP-48s/gx,HP-35s,more
Find all posts by this user
Quote this message in a reply
07-12-2022, 12:23 PM
Post: #4
RE: Sob
Sorry to hear this happened Tony, but also somewhat relieved the "Sob" in your title wasn't the other kind... Wink

--Bob Prosperi
Find all posts by this user
Quote this message in a reply
07-12-2022, 12:28 PM
Post: #5
RE: Sob
I made the samething to two of my ACTs chip from woodstock.

When I remove them from the PCB they cracked.

From this point of view, the ceramics packaging chip is better than plastic packaging.

HP35 HP45 HP55 HP65 HP67 HP21 HP25 HP29C HP19C HP97 HP10C HP11C HP12C HP15C HP16C HP31E HP32E HP33E HP33C HP34C HP37E HP38E HP38C HP41
Find all posts by this user
Quote this message in a reply
07-12-2022, 04:57 PM
Post: #6
RE: Sob
Nightmare!

Some consolation, at least you managed to get the code.
Find all posts by this user
Quote this message in a reply
07-12-2022, 05:58 PM
Post: #7
RE: Sob
Nightmare!

Some consolation, at least you managed to get the code.
Find all posts by this user
Quote this message in a reply
07-13-2022, 12:56 AM
Post: #8
RE: Sob
(07-12-2022 12:23 PM)rprosperi Wrote:  Sorry to hear this happened Tony, but also somewhat relieved the "Sob" in your title wasn't the other kind... Wink

It was more like look on Steve Martins face in the scene in Planes Trains and Automobiles when John Candy reversed the burnt out car into the motel room front window :-)

cheers

Tony
Find all posts by this user
Quote this message in a reply
07-13-2022, 03:22 AM
Post: #9
RE: Sob
Good enough scene that I just went and watched it. Yeah, that seriously Ooops! feeling...

--Bob Prosperi
Find all posts by this user
Quote this message in a reply
07-15-2022, 11:16 AM
Post: #10
RE: Sob
(07-12-2022 06:36 AM)teenix Wrote:  I was just removing a ROM 0 chip to read. I had it all unsoldered and was loose, but as I was pulling it out with just a little pressure and the case just crumbled and 7 pins broke off - from a HP-92 to make things worse :-(

Oh my, sorry to hear that.. Sad

I'm not sure, but it seems to be an epoxi case, and these can absorb humidity from the air. And when this happens, and the IC is heated, water will expand and potentially create micro fractures in the epoxi.

That's why components are "pre baked" before soldered in soldering ovens, to evaporate the water before the higher soldering temperatures hit the part. It's a long shot, but may be one of the possible causes!

Also, there's a small possibility to save the IC, by using a rotary tool with a suitable bit to eat away enough epoxi to expose what's left of the legs. The you can plug the chip in a regular IC socket, and then plug the loose leads to the same socket, just to hold everything in place, so you can try resoldering the legs back. And then, if all that works well, use some epoxi adhesive to mechanically reinforce the joints. It's a lot of work, but I managed to do that on some rare ICs from old 8 bit computers.

Eduardo
Find all posts by this user
Quote this message in a reply
07-15-2022, 01:53 PM
Post: #11
RE: Sob
(07-15-2022 11:16 AM)Marcovecchio Wrote:  Also, there's a small possibility to save the IC, by using a rotary tool with a suitable bit to eat away enough epoxi to expose what's left of the legs. The you can plug the chip in a regular IC socket, and then plug the loose leads to the same socket, just to hold everything in place, so you can try resoldering the legs back. And then, if all that works well, use some epoxi adhesive to mechanically reinforce the joints. It's a lot of work, but I managed to do that on some rare ICs from old 8 bit computers.

I don't think it is home repairable, the pins are well and truly detached.

cheers

Tony


Attached File(s) Thumbnail(s)
   
Find all posts by this user
Quote this message in a reply
07-15-2022, 03:02 PM
Post: #12
RE: Sob
(07-12-2022 06:36 AM)teenix Wrote:  The IC is bricked for normal use, but luckily the pins that broke were for the display anodes and the chip still functioned and I was able to get the code from it.

I hope now that I find my way with the ROM chip replacements.

At least some good may still come out of this.

73
Bill WD9EQD
Smithville, NJ
Find all posts by this user
Quote this message in a reply
07-15-2022, 03:33 PM
Post: #13
RE: Sob
The old HP would mail you a replacement.
Find all posts by this user
Quote this message in a reply
07-16-2022, 12:20 AM
Post: #14
RE: Sob
(07-15-2022 03:33 PM)Sukiari Wrote:  The old HP would mail you a replacement.

I'd post them a letter and get down and pray, but I think I'll only get sore knees :-)

cheers

Tony
Find all posts by this user
Quote this message in a reply
07-16-2022, 12:44 AM
Post: #15
RE: Sob
(07-16-2022 12:20 AM)teenix Wrote:  
(07-15-2022 03:33 PM)Sukiari Wrote:  The old HP would mail you a replacement.

I'd post them a letter and get down and pray, but I think I'll only get sore knees :-)

cheers

Tony

I eagerly await your 97 board.
Find all posts by this user
Quote this message in a reply
07-16-2022, 12:16 PM
Post: #16
RE: Sob
Maybe it is time to consider new HP92 board.

HP35 HP45 HP55 HP65 HP67 HP21 HP25 HP29C HP19C HP97 HP10C HP11C HP12C HP15C HP16C HP31E HP32E HP33E HP33C HP34C HP37E HP38E HP38C HP41
Find all posts by this user
Quote this message in a reply
07-17-2022, 12:51 AM (This post was last modified: 07-17-2022 12:55 AM by teenix.)
Post: #17
RE: Sob
(07-16-2022 12:16 PM)tangdfx Wrote:  Maybe it is time to consider new HP92 board.

In the (somewhat congested) pipeline as is the 91. Both should work from a single design.

I have been redoing my 92 simulator and now trying to use the microcode I got from the ROM reader and turn it into an emulator. It took half a day just to figure out the key codes and I have to do the same with the printer codes. Something is amiss though, as the code is not processing some of the financial functions properly.

cheers

Tony
Find all posts by this user
Quote this message in a reply
07-17-2022, 05:03 AM
Post: #18
RE: Sob
Hello,tony

Maybe combine with HP95C which should be shared same platform.

And We can transfer an HP91 or 92 to an HP95C.

HP35 HP45 HP55 HP65 HP67 HP21 HP25 HP29C HP19C HP97 HP10C HP11C HP12C HP15C HP16C HP31E HP32E HP33E HP33C HP34C HP37E HP38E HP38C HP41
Find all posts by this user
Quote this message in a reply
07-17-2022, 09:33 AM
Post: #19
RE: Sob
(07-17-2022 05:03 AM)tangdfx Wrote:  Hello,tony

Maybe combine with HP95C which should be shared same platform.

And We can transfer an HP91 or 92 to an HP95C.

There is no HP code for that model as far as I know. I did a simulator for it, but you never know I may write my own microcode for it like I did for the HP-10.

cheers

Tony
Find all posts by this user
Quote this message in a reply
07-18-2022, 12:50 PM
Post: #20
RE: Sob
Boy what fun trying to get the 92 debugged.

The financial functions would not work. I had to snap shot the execution flow from the real 92 and trace instruction by instruction until I found the error, which was go to $92 instead of go to $97. It turned out a problem with a decoding file that I use for all emulators which placed erroneous code in the 92 memory. This took most of the day.

There is still an anomaly with the print codes. printing the numeral zero has the same print code as capital "0" which is $03. I even checked this with the real 92 working by capturing the data as it was fed to the PIK chip, so not sure what is going on there.

More fun to go.

cheers

Tony
Find all posts by this user
Quote this message in a reply
Post Reply 




User(s) browsing this thread: 1 Guest(s)