Date   

Re: 9825 ROM behavior

 

On Apr 20, 2017, at 8:39 PM, Craig Ruff <cruff@ruffspot.net> wrote:

Just remembered I have a 98217A ROM. Results using one of my 9825T systems:
Oh, and it took a fair amount of force to seat it fully before it responded.


Re: 9825 ROM behavior

 

Just remembered I have a 98217A ROM. Results using one of my 9825T systems:

98217A out:
list says 61670 bytes free
execute ‘drive 8’ gives error 03

98217A in:
list says 60530 bytes free
execute ‘drive 8’ gives error D1


Re: 9825 ROM behavior

Paul Berger
 

On 2017-04-20 12:47 AM, Jack Rubin wrote:

Backing up a step, I'm not really sure how the ROMs should work in my machine. I can't fault Ansgar's disk image if I can't access my disk due to improper ROM function.

I have two machines to work with - a 9825A and a 9825T (by the board set, but it seems to act only as a B at this point, ignoring the top board).
Are you sure it is completely ignoring the top (A25) board? Signals from that board change the behavior of the address selection on A24.

Both machines power up to the 'lazy T' prompt and seem to operate normally within the 'regular' realm. If I type 'list' and press EXECUTE, the A shows 23228 bytes (correct for Option 002 as listed inside the printer cover) and the T shows 31294 (as opposed to the label which claims 61670 bytes; the label lists the chassis number which matches that on the bottom of the machine).

Should the 'list' result change when ROMs are inserted? It doesn't on either of my machines. The manuals for the String/AdvProgramming ROM (98210A) claim that the two functions combined take 56 bytes of RWM (read-write memory). Entering dimA$[8] on either machine gives error 22 (bad dimension), with or without the ROM inserted.
This ROM will not work in the the B/T the equivalent function is built it, but when operating as a B in 32K mode it is mapped out, see memory map in the service manual. The electronics on the A24 board suppress the signal that selects external ROMs except in the areas occupied by Matrix and Mass Memory ROMs, assuming that the address selection logic on A24 is operating correctly.

I've tried several different ROMs (Matrix, 98211A, and Disk, 98217A) and all the slots on both machines without any difference.
The 98211A matrix apparently does not reserve any memory, however you could try keywords specific to that ROM.

On the A, I can pull the Operating System ROM carrier from the base of the machine - this prevents the machine from booting but at least verifies that the ROM connector board is electrically attached to the rest of the system.
There is a difference between the connection for the OS ROM module and the option ROMs. In the service guide you will see they mention chips referred to as "BIBs" These chips are level converting transceivers that drive the Address/Data bus on the ROM side at a +7V logic level instead of the +5V used in the rest of the machine. For the OS ROM module these BIBs are built into the module, and the module connects directly to the memory bus. For the option ROM connectors there is a pair of these BIB chips on the ROM connector board that drive all four slots. These BIBs are a custom chip made by HP and as such are rare.

On the T, I've eyeballed the cable but haven't checked in any other way.

Any thoughts?
On the A the signal that enables external ROMs is generated on CPU card, on the B/T these signal lines are cut and the signals are generated on A24 instead. This signal is labeled -CartOE on Tony's schematics, and his schematic also indicates that there are switches on the system board that if open would disable the 4xxxx and 5xxxx octal ranges of addresses. I don't recall seeing switches on my CPU card, however they may not be actual switches or perhaps they where no populated on the CPU card for a B/T since they would no be used.

The BIBs on the ROM connector board use the -CartOE to control the direction of the transceiver so if it is not not functioning you will never read anything from ROMs. The system also would be used by the ROMs themselves to enable their output. Since the A boots it is reasonable to assume that the connection to the memory bus is OK so the only things unique for the option ROM slots are the BIBs and -CartOE.

The easiest thing would be to verify the ROMs are working in a known good machine. I could do that for you, however I suspect there are people located closer to you that could as well.

Paul.


Re: 9825 ROM behavior

 


On Apr 19, 2017, at 9:47 PM, Jack Rubin <j@...> wrote:

 and the T shows 31294 (as opposed to the label which claims 61670 bytes; the label lists the chassis number which matches that on the bottom of the machine). 

You are right, the T is not working correctly, and that will likely affect the use of the other ROMs if the RAM/ROM selection circuitry on the A25 board is not functioning correctly.  Probably best to test with your A unit.

Should the 'list' result change when ROMs are inserted?

It depends on if the ROM steals RAM for its working area.  The best way to test is to compare the result reported by the bare system with that when one ROM is inserted.  Make sure to do nothing that would allocate memory (i.e. assign a variable, etc) before running the list command.  The 98217 ROM should steal 1140 bytes of memory.  You might see if the “drive” statement is recognized, as I think this would need to be in the 98217 ROM itself rather than in the bootstrap files on disk.


9825 ROM behavior

Jack Rubin
 

Backing up a step, I'm not really sure how the ROMs should work in my machine. I can't fault Ansgar's disk image if I can't access my disk due to improper ROM function.

I have two machines to work with - a 9825A and a 9825T (by the board set, but it seems to act only as a B at this point, ignoring the top board). Both machines power up to the 'lazy T' prompt and seem to operate normally within the 'regular' realm. If I type 'list' and press EXECUTE, the A shows 23228 bytes (correct for Option 002 as listed inside the printer cover) and the T shows 31294 (as opposed to the label which claims 61670 bytes; the label lists the chassis number which matches that on the bottom of the machine). 

Should the 'list' result change when ROMs are inserted? It doesn't on either of my machines. The manuals for the String/AdvProgramming ROM (98210A) claim that the two functions combined take 56 bytes of RWM (read-write memory). Entering dimA$[8] on either machine gives error 22 (bad dimension), with or without the ROM inserted. I've tried several different ROMs (Matrix, 98211A, and Disk, 98217A) and all the slots on both machines without any difference. On the A, I can pull the Operating System ROM carrier from the base of the machine - this prevents the machine from booting but at least verifies that the ROM connector board is electrically attached to the rest of the system. On the T, I've eyeballed the cable but haven't checked in any other way.

Any thoughts?

Thanks,

Jack


Re: 9885 disc on 9825 machine - getting closer

 

If you haven’t seen it, the 9825DiscProgramming-09825-90220-98pages-Aug80.pdf document contains good information about the binary programs needed on a 9885 disc, where they are placed and what they are used for.


Re: 9885 disc on 9825 machine - getting closer

Paul Berger
 



On 2017-04-17 3:20 AM, Ansgar wrote:

Hi Jack,

first of all good you have a working 9885!

Sounds as if the system sectors have not been properly initialized. The HPDir function you mention is experimental (means not yet approved). What I did with HPDir is take the system sectors of a known good 9825 disc, keep the sector data permanently into HPDir, and write the sectors during initialization in single sided mode with the 9895 to the floppy, so that it can be read from a 9825 with a 9885. You should be able to use a DD disc instead of an SD disc if you like in case no SD is at hand.

The 9885 cannot use double sided media, the drives in the 9885 only have one index sensor and it is positioned for single sided media.

Unfortunately I have no 9825 to test on my own, but I will re-check whether the sectors really look like they should with a 9885. I assume that it should not make a difference whether using a 9825A or 9825B, but I am no expert on this.

It would not make any difference.

The msus parameter should be ok with 707: when initializing the disc in the 9895 (no unit specifier needed, if you omit the unit, it should default to 0), otherwise that can be considered as a bug. I'll also check this on my side.

Ansgar



Re: 9885 disc on 9825 machine - getting closer

Paul Berger
 

On 2017-04-17 9:17 AM, Jack Rubin wrote:

Ansgar - re the msus syntax, I don't think it's a bug - it's a feature that you allow more granular addressing, but it wasn't something I was familiar with on my older equipment. I started testing with an HP-85 and talked to the 9895 using msus :770 and :771. When I switched to HPDir, 770: didn't work but 707: did. That's when I realized you had added another addressing digit (that reflects 9845 usage?), thus 7070: and 7071:.
No the 9835/45 use a unique format like nearly every other HP system, it makes it fun switching between them. They use a letter to designate the drive type F=9885 H=9895, the first drive of a 9895 attached to a HPIB adapter with select code 7 and controller address of 7 would be ":H7,7,0" There are also short forms, for instance ":H" = ":H7,0,0"

- re DS vs SS disc/k (even HP goes back and forth on "disk" and "disc" spelling) - I've got the covers back on both drives but can the 9885 read both types? There are index LEDs for both disk types in the 9895 but not sure about the 9885.
The 9885 drive only has the one index sensor and is positioned for single sided media, the drive does however write the data in a double density format.

Craig and others - two questions - first, if I understand your answer, a properly initialized disk should load with the 98217 ROM only and does not require any use of the tape?
The tape utility was only required to write the boot straps onto the diskette, a "properly formatted" diskette for use with the 98217 ROM would need to include these bootstrap, the 98228A ROM does not require these bootstraps to be present. There is a image of the 98228 ROM and suggested electronics available in the file section for this group.

- maybe more fundamental, how do I determine if my ROM is in place and actually working? Again, the system test tape seems to be non-existent but I thought that typing reserved words such as 'drive' or 'cat' should get a D or F error rather than an 03 error. More generally, how do I determine if any ROM is in place and functional? Does typing a reserved word work as a quick functional test?
If the ROM is correctly installed and functional you should get an error other than 3. Beware of the 98218A ROM they have been sold on eBay as the diskette ROM for 9825 but they are really for 9831 (9825A hardware running BASIC) and will not work with the 9825 HPL OS.

Paul.


Re: 9885 disc on 9825 machine - getting closer

Jack Rubin
 

Ansgar - re the msus syntax, I don't think it's a bug - it's a feature that you allow more granular addressing, but it wasn't something I was familiar with on my older equipment. I started testing with an HP-85 and talked to the 9895 using msus :770 and :771. When I switched to HPDir, 770: didn't work but 707: did. That's when I realized you had added another addressing digit (that reflects 9845 usage?), thus 7070: and 7071:.

- re DS vs SS disc/k (even HP goes back and forth on "disk" and "disc" spelling) - I've got the covers back on both drives but can the 9885 read both types? There are index LEDs for both disk types in the 9895 but not sure about the 9885.

Craig and others - two questions - first, if I understand your answer, a properly initialized disk should load with the 98217 ROM only and does not require any use of the tape?

- maybe more fundamental, how do I determine if my ROM is in place and actually working? Again, the system test tape seems to be non-existent but I thought that typing reserved words such as 'drive' or 'cat' should get a D or F error rather than an 03 error. More generally, how do I determine if any ROM is in place and functional? Does typing a reserved word work as a quick functional test?

Thanks,

Jack


Re: 9885 disc on 9825 machine - getting closer

Ansgar
 

Small correction: meant you can use a DS disc instead of an SS disc :-)

I let you know about the checks on my side.

-Ansgar


Re: 9885 disc on 9825 machine - getting closer

Ansgar
 

Hi Jack,

first of all good you have a working 9885!

Sounds as if the system sectors have not been properly initialized. The HPDir function you mention is experimental (means not yet approved). What I did with HPDir is take the system sectors of a known good 9825 disc, keep the sector data permanently into HPDir, and write the sectors during initialization in single sided mode with the 9895 to the floppy, so that it can be read from a 9825 with a 9885. You should be able to use a DD disc instead of an SD disc if you like in case no SD is at hand. Unfortunately I have no 9825 to test on my own, but I will re-check whether the sectors really look like they should with a 9885. I assume that it should not make a difference whether using a 9825A or 9825B, but I am no expert on this.

The msus parameter should be ok with 707: when initializing the disc in the 9895 (no unit specifier needed, if you omit the unit, it should default to 0), otherwise that can be considered as a bug. I'll also check this on my side.

Ansgar


Re: 9885 disc on 9825 machine - getting closer

 

On Apr 16, 2017, at 21:07, Jack Rubin <j@ckrubin.us> wrote:
Is this a Catch-22? Do I still need the Disk System Cartridge (tape) to 'boot' the disk subsystem before new disk related commands become active? Or is it possible that my ROM is not being accessed?
The 98217 ROM only has enough smarts to boot the binary programs written into the system area on the disk, and probably provides I/O utility routines to drive the interface. Those binary programs are swapped into memory as needed to perform various disk operations. If the floppy was initialized properly with the binary programs in the correct locations. It should work.


9885 disc on 9825 machine - getting closer

Jack Rubin
 

A recent trip to Los Angeles resulted in a working 9885M drive and so I got back to work on the disk initialization project. The current version of Ansgar Kueckes' HPDir program allows you to create a 9885 disk on a 9895 drive with the incantation 'hpdir -initialize -9885 -s 9825 <msus>'. Once I realized that Ansgar's syntax added an extra zero to the msus (e.g. with my 9895 addressed as 7, the first drive is 7070: and the second drive is 7071:), I was able to [apparently] initialize a disk for my 9885 drive. 

The 'apparently' part came when I hooked the 9885 drive up to my 9825A with the 98217A Disk ROM in place and attempted to access the drive. I heard contacts close and open in the drive cabinet as I powered the 9825 up and down but any attempt to use disk commands resulted in an 'error 03' - mnemonic unknown - on the 9825 console.

Is this a Catch-22? Do I still need the Disk System Cartridge (tape) to 'boot' the disk subsystem before new disk related commands become active? Or is it possible that my ROM is not being accessed?

Thanks,

Jack


Re: HP 264x tape drive write current for Athana/DC2000 tapes (was HP 5420)

 

Yes I am pretty sure it's somewhere in there (500 pages of service manual!). I was just hoping someone already knew and would save me the digging ;-) . I'll report when I find out.

Marc


Re: HP 264x tape drive write current for Athana/DC2000 tapes (was HP 5420)

Jack Rubin
 

I don’t know the answer but after visiting the treasure trove of 264x service manuals on bitsavers, I’ve got to believe the information you need is in there somewhere. I’m still digesting the cartridge tape manual (13255-91032) that Rik pointed to. I wish HP had continued that level of detail with their other desktop machines.

 

Jack

 

From: VintHPcom@groups.io [mailto:VintHPcom@groups.io] On Behalf Of CuriousMarc
Sent: Monday, March 13, 2017 10:12 PM
To: VintHPcom@groups.io
Subject: Re: [VintHPcom] HP 264x tape drive write current for Athana/DC2000 tapes (was HP 5420)

 

Another trick question for HP264x afficionados. This one is specific to the HP 2647. The HP 2647A came with tapes but no flex disc, and the HP 2647F came with the flex disc but no tapes. Does one know if it is possible to add 2 tapes to an HP 2647F? So it ends up with both tapes and disc, and can copy one to the other? I am sure I can add the tape hardware (there are at least two slots available for the cards and space for the tape drives), but does anyone know:

- Would I need to add a tape ROM firmware support top the F? Or is it included in the 2647F base set, inherited from the A?

- Would the firmware allow to copy tape to disc and vice versa?

If that were possible, that would be quite convenient to archive 264x tapes to 5.25" disc, then make dump images of that on a PC.

Marc


Re: HP 264x tape drive write current for Athana/DC2000 tapes (was HP 5420)

 

Another trick question for HP264x afficionados. This one is specific to the HP 2647. The HP 2647A came with tapes but no flex disc, and the HP 2647F came with the flex disc but no tapes. Does one know if it is possible to add 2 tapes to an HP 2647F? So it ends up with both tapes and disc, and can copy one to the other? I am sure I can add the tape hardware (there are at least two slots available for the cards and space for the tape drives), but does anyone know:

- Would I need to add a tape ROM firmware support top the F? Or is it included in the 2647F base set, inherited from the A?

- Would the firmware allow to copy tape to disc and vice versa?

If that were possible, that would be quite convenient to archive 264x tapes to 5.25" disc, then make dump images of that on a PC.

Marc


Re: HP 264x tape drive write current for Athana/DC2000 tapes (was HP 5420)

 

Hi Marc,

 

I haven’t looked in the schematics, just looked it up 😉

But I think your conclusion is the right one.

I quick look, I think the zeners are for protection of the 74145 (Vce max is 15V).

The RIP signals seems to be some kind of feedback of the write signal.

 

-Rik

 

 

 

Van: VintHPcom@groups.io [mailto:VintHPcom@groups.io] Namens CuriousMarc
Verzonden: maandag 13 maart 2017 6:26
Aan: VintHPcom@groups.io
Onderwerp: Re: [VintHPcom] HP 264x tape drive write current for Athana/DC2000 tapes (was HP 5420)

 

Thanks, very helpful. I see, it's on the main control board, not the PCB of the tape drive itself. This circuit below, R42 is the resistance in the emitter of Q16. If I follow this right, When Q15 is conducting, due to the Zener CR4, it should bring the base of emitter follower Q16 to ~4V, and the emitter of Q16 to ~4.5V. Then the resistor R42 sees 7.5V across it. So that would provide a 10 mA current source, not dependent on the head impedance. To increase it by 20% I would have to make the resistor 625 Ohms. Does this sound reasonable to you?
What's the funny business with all the diodes between the Collector and Emitter of Q16? Clamping the signal that goes to Q17's base both ways? What's the RIP signal?
Marc
 


Re: HP 264x tape drive write current for Athana/DC2000 tapes (was HP 5420)

 

Thanks, very helpful. I see, it's on the main control board, not the PCB of the tape drive itself. This circuit below, R42 is the resistance in the emitter of Q16. If I follow this right, When Q15 is conducting, due to the Zener CR4, it should bring the base of emitter follower Q16 to ~4V, and the emitter of Q16 to ~4.5V. Then the resistor R42 sees 7.5V across it. So that would provide a 10 mA current source, not dependent on the head impedance. To increase it by 20% I would have to make the resistor 625 Ohms. Does this sound reasonable to you?
What's the funny business with all the diodes between the Collector and Emitter of Q16? Clamping the signal that goes to Q17's base both ways? What's the RIP signal?
Marc
 




Re: HP 5420

 

Hi Marc,

 

I was on a business trip, so it took a little time to answer you.

When I look at the diagram from the manual:

http://bitsavers.trailing-edge.com/pdf/hp/terminal/264x/service/13255-91032_Cartridge_Tape_Module_Aug76.pdf

It seems to me R42(750 ohm)  in the write current circuit limits the write current.

So decreasing the value of this resistor should increase the write current.

A good approach could be; First measure the resistance of the head so you can determine the write current and then increase it with 10-20%.

Keep in mind the 74145 BCD decoder has open collector outputs (80mA at 15V breakdown) so is capable to fry your head coils..

 

 

-Rik

 

Van: Marc Verdiell [mailto:marc.verdiell@...]
Verzonden: donderdag 9 maart 2017 9:33
Aan: Rik Bos <hp-fix@...>
Onderwerp: Re: HP 5420

 

Rik,

I have a question regarding HP 264x tape drives. Do you know how to increase the write current for DC2000 and Athana tapes on this mechanism? Board is different from a HP 9825 drive.

Marc


9825T project files added

 

I've uploaded two zip files containing the current version of my 9825T firmware disassembly project.  The dis9825.zip file is the disassembler source and Makefile, along with the disassembler command files used to control the disassembler to produce the disassembly listings.  The disassembler is a C++ 11 program, but just uses basic C++ features and C++ standard I/O to do its work.

The doc9825.zip file contains the Sphinx documentation sources that describe the disassembler and things I have learned so far about the hardware and firmware.  Just realized I forgot to write some notes about how the 98228A Disk ROM works.  That will wait for the next release.

If anyone starts digging in, I've only done preliminary work on documenting the option ROMs.  The mainframe and General I/O ROM are fairly well documented based on the patent firmware listings.

If you want to feed me updates to either the disassembler command files or the documentation sources, I'll be happy to coordinate periodic update releases.

7001 - 7020 of 7153