HP-71b CMT-64E Eprom - Whats under the black goop ?

[coitboy2000]

I have a few of these Eproms and decided to open one up to have a look. Sylvain has been kind enough to re-program some for me in the past. I was wondering what lies beneath the black dob of goop ? I guess an integrated circuit to account for the lack of 32 pins (13 only on module, 32 on PLCC32 eprom) ? I am going to try and program the eprom as a bit of a learning exercise with an Arduino Mega. Just curious whats underneath and how you might remove the black material.

Cheers

Richard

[J-F Garnier]

(03-14-2020 07:00 AM)coitboy2000 Wrote:  I was wondering what lies beneath the black dob of goop ? I guess an integrated circuit to account for the lack of 32 pins (13 only on module, 32 on PLCC32 eprom) ?

There is a HP specific integrated circuit that makes the interface between the standard 8-bit wide EPROM and the 4-bit HP71 bus. The circuit reference is 1LJ4, it is also used in packaged form in the card reader memory modules both from CMT and HHP.

A few years ago, Paul Berger and I attempted to document it and identify the pinout, see this document:
http://www.jeffcalc.hp41.eu/misc/files/hhp.pdf

I know that one member of this forum once (long ago) said he has the datasheet of the 1LJ4 circuit, but I never succeeded to get it, unfortunately.

J-F

[coitboy2000]

(03-14-2020 08:56 AM)J-F Garnier Wrote:  

(03-14-2020 07:00 AM)coitboy2000 Wrote:  I was wondering what lies beneath the black dob of goop ? I guess an integrated circuit to account for the lack of 32 pins (13 only on module, 32 on PLCC32 eprom) ?

There is a HP specific integrated circuit that makes the interface between the standard 8-bit wide EPROM and the 4-bit HP71 bus. The circuit reference is 1LJ4, it is also used in packaged form in the card reader memory modules both from CMT and HHP.

A few years ago, Paul Berger and I attempted to document it and identify the pinout, see this document:
http://www.jeffcalc.hp41.eu/misc/files/hhp.pdf

I know that one member of this forum once (long ago) said he has the datasheet of the 1LJ4 circuit, but I never succeeded to get it, unfortunately.

J-F

Thanks J-F

[coitboy2000]

I guess the CMT eprom programmer must know exactly how this HP integrated circuit works in order to read and write with only 13 contacts.

[Sylvain Cote]

(03-17-2020 09:33 AM)coitboy2000 Wrote:  I guess the CMT eprom programmer must know exactly how this HP integrated circuit works in order to read and write with only 13 contacts.

I you read this thread, you will find that one of the founder of CMT in 1984 was David Lin an ex.HP employee.
I am pretty sure he had access to extensive internal documentation and to old friends still at HP to get that information.

[KimH]

There were IDS documents describing most all of the 71.

One was the HW IDS - which can be found in several places including the MoHP ThumbDrive.

The bus is 4 bit wide with control signals like Command/Data, Strobe, Halt, Interrupt, DaisyChain in/out, plus 3 odd ones IR14 and OD (Output Disable - or Tri-State Bus) and HALT

The CMT Rom Burner Manual can also be found in several places CMT-71-P01. It uses port 4 where pin 9 is NOT connected to anything

The thing I would like to know is WHERE (which pin) the Program Pulse of 14V for the EPROM goes in - i suspect Pin9.

If anyone has the answer to that question, I would be excited to hear about it.

(03-17-2020 09:33 AM)coitboy2000 Wrote:  I guess the CMT eprom programmer must know exactly how this HP integrated circuit works in order to read and write with only 13 contacts.

[Dave Frederickson]

(03-14-2020 08:56 AM)J-F Garnier Wrote:  A few years ago, Paul Berger and I attempted to document it and identify the pinout, see this document:
http://www.jeffcalc.hp41.eu/misc/files/hhp.pdf

On p.4 of the document is a picture of an HHP card reader port EPROM module. How was this module programmed?

Note: There is currently one of these modules on eBay.

Dave

[J-F Garnier]

(03-18-2020 05:48 AM)Dave Frederickson Wrote:  

(03-14-2020 08:56 AM)J-F Garnier Wrote:  A few years ago, Paul Berger and I attempted to document it and identify the pinout, see this document:
http://www.jeffcalc.hp41.eu/misc/files/hhp.pdf

On p.4 of the document is a picture of an HHP card reader port EPROM module. How was this module programmed?

I did program HHP card-reader EPROM module back at the time, some for professional applications. The EPROM was programmed on a standard EPROM programmer, the only thing to know is that the bytes are complemented, I guess to make an erased EPROM (all FF) looks as zero bytes from the HP71. This is not true for CMT front port EPROM module, an erased module disturbs the HP71 operations.

J-F

[KimH]

About a cleared EPROM disturbing the 71B

I just erased 2 64kb front port EPROM modules from CMT.

I wanted to see if the erase was completed properly. Placed them in Port 3 & 4

PEEK as provided in the 71B OS ROM won't PEEK as the EPROM is seen as a protected ROM. [ERR:File Protect]

SYSEDIT from the JPC E2 was my friend, found the ROMs and they were all "F".

ROMs seem to be placed underneath the Card Reader area sort of Top Down regardless of which port they go into.

I suspect - as J-F suggests - that the cleared ROMs would disturb the machine, so I removed them once done checking.

& J-F: Thanks for sharing that the 32k EPROMs for HHP are complemented! Programming one of those is my next project - when it rains again...

(03-18-2020 08:29 AM)J-F Garnier Wrote:  

(03-18-2020 05:48 AM)Dave Frederickson Wrote:  On p.4 of the document is a picture of an HHP card reader port EPROM module. How was this module programmed?

I did program HHP card-reader EPROM module back at the time, some for professional applications. The EPROM was programmed on a standard EPROM programmer, the only thing to know is that the bytes are complemented, I guess to make an erased EPROM (all FF) looks as zero bytes from the HP71. This is not true for CMT front port EPROM module, an erased module disturbs the HP71 operations.

J-F

[coitboy2000]

(03-18-2020 09:55 AM)KimH Wrote:  About a cleared EPROM disturbing the 71B

I just erased 2 64kb front port EPROM modules from CMT.

I wanted to see if the erase was completed properly. Placed them in Port 3 & 4

PEEK as provided in the 71B OS ROM won't PEEK as the EPROM is seen as a protected ROM. [ERR:File Protect]

SYSEDIT from the JPC E2 was my friend, found the ROMs and they were all "F".

ROMs seem to be placed underneath the Card Reader area sort of Top Down regardless of which port they go into.

I suspect - as J-F suggests - that the cleared ROMs would disturb the machine, so I removed them once done checking.

& J-F: Thanks for sharing that the 32k EPROMs for HHP are complemented! Programming one of those is my next project - when it rains again...

(03-18-2020 08:29 AM)J-F Garnier Wrote:  I did program HHP card-reader EPROM module back at the time, some for professional applications. The EPROM was programmed on a standard EPROM programmer, the only thing to know is that the bytes are complemented, I guess to make an erased EPROM (all FF) looks as zero bytes from the HP71. This is not true for CMT front port EPROM module, an erased module disturbs the HP71 operations.

J-F

I assume you are lucky enough to have an official CMT Eprom Programmer to erase them ?

[KimH]

Actually I used a standard UV eraser for 15-20 minutes, that part is standard.

The programmer is the hard one to find, i know of 3 in existence, two owners of them are communicating here on this forum.

[/quote]

I assume you are lucky enough to have an official CMT Eprom Programmer to erase them ?
[/quote]

[Dave Frederickson]

I know of at least 7 CMT EPROM programmers among forum members and three of the members who own them are participating in this thread.

The CMT EPROM Programmer won't erase chips, a UV light is used for that.

To test for an erased EPROM, use the CMT tool, PTBLANK$.

(03-18-2020 08:29 AM)J-F Garnier Wrote:  I did program HHP card-reader EPROM module back at the time, some for professional applications. The EPROM was programmed on a standard EPROM programmer, the only thing to know is that the bytes are complemented, I guess to make an erased EPROM (all FF) looks as zero bytes from the HP71.

I figured an external programmer was used because the card reader port can't provide the necessary programming voltage. I was more curious about how to create the EPROM image for the programmer. It sounds like all it takes is to dump an image from a 71, invert all the bits using one of the many tools available for manipulating EPROM images, and burn the chip and see what happens.

(03-18-2020 09:55 AM)KimH Wrote:  ROMs seem to be placed underneath the Card Reader area sort of Top Down regardless of which port they go into.

Not sure what you're saying here, but memory allocation for the 71 is well documented.


Dave

[J-F Garnier]

(03-18-2020 01:49 PM)Dave Frederickson Wrote:  I figured an external programmer was used because the card reader port can't provide the necessary programming voltage. I was more curious about how to create the EPROM image for the programmer. It sounds like all it takes is to dump an image from a 71, invert all the bits using one of the many tools available for manipulating EPROM images, and burn the chip and see what happens.

The HP71 doesn't provide programming voltages to any port; the CMT front-port EPROM modules are programmed by being plugged into the CMT programmer, which DOES provide the programming voltage.

To make the ROM image from a IRAM port, I recommend to use the ROMCOPY function to first create a ROM file on a mass-storage unit, then to load it back (again with ROMCOPY).
This may not be needed for ROMs with only LEX files, but it is mandatory for ROMs that include BASIC files, to make sure the chaining/compilation process is correctly done. This is something we discussed in the past, but I can't find the reference right now.
Then, a small program can PEEK into the IRAM port and send the bytes (complemented) to an external programmer.

J-F

[rprosperi]

I have programs (FORTH & BASIC) for generating the proper image, sent to a PC to burn the EPROM chip, just waiting for confirmation to post them here, as it was a collaboration with someone else. Did not think there would be much of an audience, as it's somewhat of a niche within a niche.

While not real complex, it's not as straightforward as it seems.

[Ken S]

(03-18-2020 05:48 AM)Dave Frederickson Wrote:  

(03-14-2020 08:56 AM)J-F Garnier Wrote:  A few years ago, Paul Berger and I attempted to document it and identify the pinout, see this document:
http://www.jeffcalc.hp41.eu/misc/files/hhp.pdf

On p.4 of the document is a picture of an HHP card reader port EPROM module. How was this module programmed?

Note: There is currently one of these modules on eBay.

Dave

I saw that and will bid on it. A while back I picked up one with the 32k ram and a half dozen burned eproms so could use those with this on one of my 71s.

[rprosperi]

(03-18-2020 06:10 PM)Ken S Wrote:  

(03-18-2020 05:48 AM)Dave Frederickson Wrote:  Note: There is currently one of these modules on eBay.

Dave

I saw that and will bid on it. A while back I picked up one with the 32k ram and a half dozen burned eproms so could use those with this on one of my 71s.

Never, ever admit that publicly. You've just notified essentially the entire pool of potential bidders that you'll be bidding too, so now they're off sharpening their pencils and setting reminders to jump in and bid... lol.

BTW, I have a CMT-71-P01 now too, so there are more of them around within the community now than just a few years ago.

[Ken S]

(03-18-2020 07:18 PM)rprosperi Wrote:  

(03-18-2020 06:10 PM)Ken S Wrote:  I saw that and will bid on it. A while back I picked up one with the 32k ram and a half dozen burned eproms so could use those with this on one of my 71s.

Never, ever admit that publicly. You've just notified essentially the entire pool of potential bidders that you'll be bidding too, so now they're off sharpening their pencils and setting reminders to jump in and bid... lol.

BTW, I have a CMT-71-P01 now too, so there are more of them around within the community now than just a few years ago.

Not a worry, last time I bid I managed to miss the decimal point and it could happen again. No one outbid me on that item with a max bid of 5187 euros! Fortunately I managed to win the item at a somewhat saner 33.04 euros.

[Dave Frederickson]

(03-18-2020 03:30 PM)J-F Garnier Wrote:  To make the ROM image from a IRAM port, I recommend to use the ROMCOPY function to first create a ROM file on a mass-storage unit, then to load it back (again with ROMCOPY).
This may not be needed for ROMs with only LEX files, but it is mandatory for ROMs that include BASIC files, to make sure the chaining/compilation process is correctly done. This is something we discussed in the past, but I can't find the reference right now.

The CMT tool Port Generate Checksum, PTGENC$ described in Step 10 on p.11 of the CMT-71-P01 manual will chain all files that need to be chained.

Dave

[rprosperi]

(03-18-2020 04:38 PM)rprosperi Wrote:  I have programs (FORTH & BASIC) for generating the proper image, sent to a PC to burn the EPROM chip, just waiting for confirmation to post them here, as it was a collaboration with someone else. Did not think there would be much of an audience, as it's somewhat of a niche within a niche.

While not real complex, it's not as straightforward as it seems.

I created a new thread (here) to post the programs for creating the image files, to avoid further hijacking of the original Black goop topic.

[mfleming]

(03-18-2020 06:10 PM)Ken S Wrote:  

(03-18-2020 05:48 AM)Dave Frederickson Wrote:  On p.4 of the document is a picture of an HHP card reader port EPROM module. How was this module programmed?

Note: There is currently one of these modules on eBay.

Dave

I saw that and will bid on it. A while back I picked up one with the 32k ram and a half dozen burned eproms so could use those with this on one of my 71s.

Everyone here who doesn't have this on their watch list raise their hand so I know who the other 14 of you are. Kidding aside, how difficult would it be to reproduce such a module? Though there is surely some custom logic involved, microprocessors and FPGAs have become quite small, if somewhat allergic to 5V.