HP86 problem

[Alexkumt]

A festive greeting to all.

HP86 problem

I purchased an old HP86-B as the first known problem that the text of the screen is missing to the left of about 7-8 characters, and to the right of another 10-11 characters.

I connected the video output to the RCA input of a TV (MultiSistem NTCS-PAL-SECAM Etc)

as a second problem the system displays both the characters from the keyboard, but any command is rejected with "Error 88 SMT Bad"
example "3 + 3" instead of answering 6 gives me the same error
"Error 88: BAD STMT"

it appears that the correct operating system is not present
By pressing the "test" button the computer performs the test correctly.

But I can't get him to digest anything

[smp]

(07-04-2019 01:15 PM)Alexkumt Wrote:  ... as a second problem the system displays both the characters from the keyboard, but any command is rejected with "Error 88 SMT Bad"
example "3 + 3" instead of answering 6 gives me the same error
"Error 88: BAD STMT"

When the system powers up, you are in BASIC, so, indeed, you are entering a bad statement. Try: Print 3+3<CR>

You have to enter proper BASIC statements in immediate mode, or enter a line number for each statement if you want to build up a multi-line program.

I know that my 86 sent its display off a little to the left side. I had to find a monitor that would allow me to "Under Scan." That is a control button on some monitors. That is what fixed it for me.

Good luck!

[jackrubin]

There is a mailing list dedicated to the HP Series 80 machines here - https://groups.io/g/hpseries80 .

Come take a look!

[rprosperi]

(07-04-2019 03:39 PM)smp Wrote:  When the system powers up, you are in BASIC, so, indeed, you are entering a bad statement. Try: Print 3+3<CR>

You have to enter proper BASIC statements in immediate mode, or enter a line number for each statement if you want to build up a multi-line program.

Actually, HP BASIC (and likely many others, but I don't have access to them to check) allows commands such as "3 + 3 [END LINE]" as valid, as there is an implied DISP command inserted by the OS. So typing that should return 6. Such lines are not illegal and should not produce any errors, so it's definitely something else in your system.

I know this is true for Series-70 and Series-80 machines, but don't know for sure about the 90x0 and 90x5 machines, though I'd guess they're all the same in this way.

[vassilisprevelakis]

(07-04-2019 07:12 PM)rprosperi Wrote:  Actually, HP BASIC (and likely many others, but I don't have access to them to check) allows commands such as "3 + 3 [END LINE]" as valid, as there is an implied DISP command inserted by the OS. So typing that should return 6.

Yes this is true. In fact the Series 80 keyboard has a key (RESLT) which brings up the result of the previous calculation.

So 3+3 ENTER
gives 6
if you then press RESLT you will get 6 inserted at the cursor position.
Very useful for quick calculations.

I believe that the 9825 also had this facility with the added bonus that you could do calculations while the computer was running a program (live keyboard).

Anyway back to the HP-86. What prompt do you get? Have you tried running other commands to see if BASIC is present? Maybe there is some other ROM that is interfering with the boot process. Is there anything connected to the expansion slots in the back?

Some pictures of the unit will definitely be useful.

**vp

http://www.series80.org

[Alexkumt]

This is Mainboard HP-86B
there is a strange Dip-Switch

[Alexkumt]

Is there a possibility to have the wiring diagram of the motherboard of the HP-86B?

[teenix]

(07-10-2019 10:22 AM)Alexkumt Wrote:  Is there a possibility to have the wiring diagram of the motherboard of the HP-86B?

This may help

http://www.hpcc.org/cdroms/schematics5.0.../hp86b.pdf

cheers

Tony

[Alexkumt]

teenix thank you very much

[Alexkumt]

now I have news about this problem. in fact, after connecting the computer to an HP monitor, I could see the first missing line. when I turn on the computer it displays nothing, even waiting for several minutes. however, when I press a button, a Warning "Warning 101: LANG UNDEFINED" message appears, then any command I issue always leaves the error "Error 88: BAD STMT" pressing the Run button, this appears instead "Error 48 on line 99999: End

[Paul Berger (Canada)]

(07-19-2019 05:44 PM)Alexkumt Wrote:  now I have news about this problem. in fact, after connecting the computer to an HP monitor, I could see the first missing line. when I turn on the computer it displays nothing, even waiting for several minutes. however, when I press a button, a Warning "Warning 101: LANG UNDEFINED" message appears, then any command I issue always leaves the error "Error 88: BAD STMT" pressing the Run button, this appears instead "Error 48 on line 99999: End

I wonder if the "LANG" the warning message is referring to might be the national language ie English, French, German, etc. There is a switch on the board S2 that sets this and there are lots of potentially invalid combinations. Even if it appears to be set correctly you may want to verify switches that are supposed to be closed really are, it is not unheard of for these switches to go bad. The setting for English should be all switches in SW2 set to 0 which is likely the closed position it should be marked on the board.

If my guess above is correct then the error 88 may be coming up because it is unsure of what you are typing however I do not believe any of the BASIC statements are translated.

The error 48 may be normal when there is no end statement in memory it may search to the max line number looking for a end statement when you are trying to execute a program.

For some monitors you may need to adjust the horizontal and vertical size to get everything on the screen. The video signal is similar to NTSC but not exactly the same.

I can't verify my guess about LANG right now as my 86B is packed away for moving.

Paul.

[Alexkumt]

this is the motherboard, in red I highlighted the dip-switches

https://imge.to/i/5oR86

[Paul Berger (Canada)]

(07-19-2019 09:43 PM)Alexkumt Wrote:  this is the motherboard, in red I highlighted the dip-switches

https://imge.to/i/5oR86

You can find the service guide for 86B at http://www.akso.de/fileadmin/Series_80/A...ce_86B.pdf on page 4-24 of the manual the settings for SW2 are defined, I cannot tell from you picture which way the switches are set, however in my experience the rocker type can be especially flaky you may want to rock them back and forth a few times to make sure they are closed, and verify the connections using a meter. Unfortunately the scan of the schematics in that manual are pretty poor but you can see that on one side of that switch is connected to non-inverting buffer outputs and the inputs are grounded. There is another signal connected which is likely an enable signal, so my guess is that when the 86B starts up the buffers are enabled and the switch read and then the buffers are turned off. Near as I can tell from the schematics the switch is connected to some of the column lines, so that would be the reason for the non-invert buffers, so it could be read and then get out of the way. Best guess is the non-invert buffer chip is the 1820-1402 (74LS365) right behind the switch but you could confirm with a meter.

If the enable signal never arrives or one of the buffers is defective it could lead to the LANG warning. That switch is unfortunately missing from Tony's schematic, perhaps it only appeared in later revisions of the 86B, and the schematic section of the service guide for the switch is pretty blurry.

I also happened upon a manual "Introduction to 86B Local Language" on hpmuseum.org and found your message in there "101 (Warning) LANG UNDEFINED Computer hardware is not properly configured for local languages - the computer requires servicing"

Paul.

[Alexkumt]

In my opinion. must have changed some internal eprom otherwise I don't explain why the tests pass them, respond to the function keys etc.but then always responds with the error message "Error 88: BAD STMT"for me there is some devilry in the system ROM.

this HP86 computer comes from germany,and perhaps the system was composed differently with somespecial module like the one for CPMsorry for the delay in replying.but I was unable to verify the computer first.

thank you againsorry for the delay

[Alexkumt]

(07-20-2019 01:50 AM)Paul Berger (Canada) Wrote:  

(07-19-2019 09:43 PM)Alexkumt Wrote:  this is the motherboard, in red I highlighted the dip-switches

https://imge.to/i/5oR86

You can find the service guide for 86B at http://www.akso.de/fileadmin/Series_80/A...ce_86B.pdf on page 4-24 of the manual the settings for SW2 are defined, I cannot tell from you picture which way the switches are set, however in my experience the rocker type can be especially flaky you may want to rock them back and forth a few times to make sure they are closed, and verify the connections using a meter. Unfortunately the scan of the schematics in that manual are pretty poor but you can see that on one side of that switch is connected to non-inverting buffer outputs and the inputs are grounded. There is another signal connected which is likely an enable signal, so my guess is that when the 86B starts up the buffers are enabled and the switch read and then the buffers are turned off. Near as I can tell from the schematics the switch is connected to some of the column lines, so that would be the reason for the non-invert buffers, so it could be read and then get out of the way. Best guess is the non-invert buffer chip is the 1820-1402 (74LS365) right behind the switch but you could confirm with a meter.

If the enable signal never arrives or one of the buffers is defective it could lead to the LANG warning. That switch is unfortunately missing from Tony's schematic, perhaps it only appeared in later revisions of the 86B, and the schematic section of the service guide for the switch is pretty blurry.

I also happened upon a manual "Introduction to 86B Local Language" on hpmuseum.org and found your message in there "101 (Warning) LANG UNDEFINED Computer hardware is not properly configured for local languages - the computer requires servicing"

Paul.

however I have solved the language problem.by correctly setting the switches
Thank you.

[Alexkumt]

I Have this news

Checking in the disassembler part System_P6 http://www.akso.de/fileadmin/Series_80/S...tem_P6.pdf
I noticed that this 88 BAD STMT error exists

Item Location OP Code SCR=ORGFXX OBJ=GEXCFX
============================================================
444 021 063 316 377 377 ERR88 JSB = ERROR + BAD STMT
445 021 066 130 OCT 98D
446 021 067 ***************** PARSER END HERE ****************
447 021 067

Look
I saw this information checking the list of disassembler files:

here
http://www.akso.de/index.php?id=hp_serie...r_listing0


I think it's a Statement error "ERR88 BAD STMT" BAD STATEMENT
but i have a lot of trouble understanding the assembler code of this computer

I Check in the "HP Series 80 Bug List
http://www.akso.de/fileadmin/Series_80/S...g_List.pdf
I don't find this error ?

[Martin Hepperle]

- these error messages are described in the user manuals (look at the appendices).

- the bug lists describe bugs found in hardware and firmware and whether they had been fixed in later revisions or whether workarounds exist. These lists are not relevant for your problem.

- if you have an external storage medium or the HPDRIVE emulator you could use one of the emulators to write a small AUTOST program, store it on the disk and then boot the system from this disk to check whether the system core (ROMs, RAMs, CPU) works (an AUTOST program is executed when the system finds it on a disk at boot time).
If this is the case maybe the keyboard is not properly connected or there is a problem with the keyboard controller.

[Alexkumt]

I thought of using the software emulator for HP87 to trace the cause of the problem through Disassembler.

everything I type on the keyboard appears on the screen correctly. in my opinion this hardware seems to work properly.

when the command line has been written and the enter key is pressed. the system does not digest what is written. ho is not intended correctly.

even if you wrote incorrect things, the system should respond with a syntax error. instead that damned mistake always appears.

I thought I would find help from an expert person in interpreting the disassembler part.

probably if there is some electrical noise in the hardware in that part. the problem could perhaps be traced back to them.

perhaps the previous owner of this computer installed a particular ROM. (but I don't seem to notice any tampering in the motherboard, which seems perfect with no changes or anything)

I thought that even if there is this problem. you could insert a module for CPM with Z80 and check carefully if the system works. in this way the main ROM is bypassed.

or get me a self test module to insert in the drawerRom

[J-F Garnier]

(02-08-2020 05:25 AM)Alexkumt Wrote:  I thought of using the software emulator for HP87 to trace the cause of the problem through Disassembler.
...
I thought I would find help from an expert person in interpreting the disassembler part.

probably if there is some electrical noise in the hardware in that part. the problem could perhaps be traced back to them.

According to the HP86 manual, error 88 BAD STMT occurs when COM, a user function or INPUT is used in "calculator mode" (that is in immediate mode, not programming). Also when trying to enter older HP85 statements in a program.

However, playing with the emulator, typing an incorrect statement by for instance starting with a ) or ? or $ also produce the BAD STMT error.
It may be that there is a (possibly invisible) bad keystroke at the beginning of the line buffer, or the first (maybe all) characters are corrupted in an intermediate memory during the line processing.
Where is the cursor position when the error happens? It should be at the offending character, the ) or ? or $ character in my tests.

I'm afraid that attempting to trace the code in an emulator will not give you more information than that.

[Edited:]
You may try this test: type several space characters then ENDLINE. On a good system, there is no error and the blank line is just ignored.

J-F

[Alexkumt]

(02-08-2020 09:14 AM)J-F Garnier Wrote:  You may try this test: type several space characters then ENDLINE. On a good system, there is no error and the blank line is just ignored.

J-F

So you mean that if you edit a series of [Space] characters and press ENDLINE, the system should not declare any errors.

but if instead the error appears, it means that a strange character like [)], [?], [$]

When I am present I will do this test directly on the computer.

For now can I simulate this on the HP87 Emulator?

I have tried it on the emulator and this error actually appears.
It seems that the mysterious character [!, ", $,%, &, /, (,), =,?, ....]
is the first on the command line.
because if it were the last it would give a different error

example
(Q => ERROR 88 BAD STMT
Q) => ERROR 89 INVALID PARAM

for this reason I suggest that it is the first character of the command line to be indicted.
Obvious to Video with the broken computer I don't see any strange characters in the monitor.

I think this is the way forward



thank you very much J-F Garnier