• Please review our updated Terms and Rules here

Osborne 1 keyboard interface?

NutmegCT

Experienced Member
Joined
Jul 24, 2009
Messages
153
Location
Connecticut
Good morning all!

If you managed to read through my earlier topic, you know how I've been troubleshooting. With keyboard plugged into m/board, power up and CRT flickers with opening screen asking for system disk and <return>. There's no response to any keyboard input.

http://www.vcfed.org/forum/showthread.php?67552-Osborne-I-no-disk-drive-response

If keyboard cable is unplugged, system starts up normally, no CRT flicker, and you can "trick" the system to boot from A by shunting two keyboard pins.

Took the keyboard apart - really apart - and found the membrane was discolored, so figured the keyboard is toast.

So yesterday I got a replacement keyboard. Keyboard and original computer work fine.

Put the new keyboard onto my O1 - and exactly the same problem. CRT flickers, no keyboard response to system. Unplug cable, system is fine. I tried this with two other keyboard cables - same result. With no cable plugged in, system doesn't hang. With only the cable plugged in (no keyboard at other end of cable), system doesn't hang. Soon as you plug cable into keyboard, system hangs (CRT flickers).

To me this would indicate the problem is in the m/board keyboard circuit. Nothing seems loose, wiggling and flexing the m/board changes nothing.

Edit: if I boot by shunting the two terminal pins, then drop to cp/m prompt, connect the cable and keyboard, the screen fills with capital T's. That's exactly what would happen with the original keyboard that started my troubleshooting.

Any thoughts as to how I should proceed?

Thanks.
Tom M.
O1 s/n A14649
 

NutmegCT

Experienced Member
Joined
Jul 24, 2009
Messages
153
Location
Connecticut
Sorry. That Edit line should read: Edit: if I boot by shunting the two terminal pins, then drop to cp/m prompt, connect the cable and keyboard, the screen fills with lower case t's.
 

daver2

10k Member
Joined
Jun 19, 2012
Messages
14,402
Location
UK - Worcester
Let me have another think tonight...

Is it at all possible that the original keyboard (or the cable) has managed to get crossed-over somehow? Please, however, don't just reverse the cable to test this out!!!

If you disconnect the keyboards altogether from the O1 and boot up CP/M with the <RETURN> jumper trick - what happens if you use your jumper to short out the row and column for each of the other keys in turn? Do you get the correct character displayed at the CP/M A:> prompt corresponding to what you think you should?

Dave
 

daver2

10k Member
Joined
Jun 19, 2012
Messages
14,402
Location
UK - Worcester
The Commodore PET computer keyboard can have a 'fit' sometimes - but this is due to the 6820/6821 LSI PIA chip that drives the keyboard matrix. If this device goes faulty, continuous interrupts can result.

The O1 just uses very simple TTL gates - so I can't see that would be a problem.

Just ruling out what it is unlikely to be...

Dave
 

NutmegCT

Experienced Member
Joined
Jul 24, 2009
Messages
153
Location
Connecticut
Thanks. Let me give that a try. I'll see if shunting pins gives the correct screen result.

Tom M.

Let me have another think tonight...

Is it at all possible that the original keyboard (or the cable) has managed to get crossed-over somehow? Please, however, don't just reverse the cable to test this out!!!

If you disconnect the keyboards altogether from the O1 and boot up CP/M with the <RETURN> jumper trick - what happens if you use your jumper to short out the row and column for each of the other keys in turn? Do you get the correct character displayed at the CP/M A:> prompt corresponding to what you think you should?

Dave
 

NutmegCT

Experienced Member
Joined
Jul 24, 2009
Messages
153
Location
Connecticut
OK - before I go stark raving mad ... please remind me of how to interpret the keyboard matrix diagram:

matrix.jpg

I see the intersection for "Return". Pin 3 and Pin 15.

I assume the pins are arranged on the m/board's terminal as:

upper row: 1 3 5 7 9 11 13 15 17 19
lower row: 2 4 6 8 10 12 14 16 18 20

(There is no pin #19.)

So, shunting 3 and 15 should give me a Return signal. But it doesn't.

To get a Return signal to the system, I have to shunt 5 and 17.

Am I totally looney? (be careful how you answer that ...)

Thanks.
Tom M.
 

daver2

10k Member
Joined
Jun 19, 2012
Messages
14,402
Location
UK - Worcester
If I renumber your connector thus:

Code:
19 17 15 13 11  9  7. 5. 3. 1
20 18 16 14 12 10. 8. 6. 4. 2

Does that help?

It all depends upon whether you are looking at the connector on the keyboard or looking at the connector that plugs into it...

Or have we found a(nother) problem?

I would suggest drawing everything out and testing at least one key on each row and one key on each column.

I would then be tempted to test the keyboard out with the same keys (using your multimeter) to double check that the correct key matches up with the appropriate row and column.

A methodical working practice is required here...

Dave
 

NutmegCT

Experienced Member
Joined
Jul 24, 2009
Messages
153
Location
Connecticut
Wow - I never knew you counted from right to left on the m/board keyboard terminal pins! I'd always assumed you'd count the pins the way they appear when you're looking at them. Unless you're from the 13th Egyptian Dynasty, and read your heiroglyphics right to left.

Yep - the screen does show the correct character corresponding to the pin pairing.

I also notice that if I have the keyboard cable plugged into the m/board terminal, and *just barely* touch keyboard end of the cable to the keyboard terminal pins, the keys work correctly from the keyboard. But if I push the connector fully into position, I get the flashing CRT and/or system hang. Lift the keyboard end of the cable up, and the system returns to normal. Same thing happened with the old keyboard.

The mind boggles!
 

daver2

10k Member
Joined
Jun 19, 2012
Messages
14,402
Location
UK - Worcester
Never assume!

Welcome to the wonderful world of vintage computers :)!

I am about to go out for the remainder of the evening - so I will give your interesting problem a little more thought...

Dave
 

NutmegCT

Experienced Member
Joined
Jul 24, 2009
Messages
153
Location
Connecticut
Quick question: to boot the O1 from the B *right* drive (if the A *left* drive isn't working), is the correct procedure as follows:

System disk in B drive.
Close door.
Press right-shift key and the +/= key.

Just want to check.
Thanks.
Tom M.
 

ef1j95

Experienced Member
Joined
Oct 10, 2013
Messages
145
Location
Philly, Yo
Well, dagnabbit. My brown case OCC-1 worked a year ago, but it's showing similar keyboard issues now. Unresponsive keyboard, won't boot. The screen isn't necessarily flickering all the time, but did when I hit reset. It's been stored in the retrolab. Hrrrmmm...
 

NutmegCT

Experienced Member
Joined
Jul 24, 2009
Messages
153
Location
Connecticut
Keep following this thread, but since your symptoms don't exactly match mine ... I'd suggest you start a new thread on your problem, and grab a cheese steak!

Tom M.
 

daver2

10k Member
Joined
Jun 19, 2012
Messages
14,402
Location
UK - Worcester
Sorry,

It's been hectic at work today!

I will post a simple 'taster' now and follow it up (if necessary) with more detail tomorrow.

If there is a problem when you plug the keyboard in, we should be able to detect some sort of change with the keyboard disconnected and connected.

My plan is as follows:

Disconnect the keyboard AND ribbon cable from the main board.

Measure the voltages (with a multimeter) from the inputs of IC UE12 (81LS95) to ground/0V. There should be 8 inputs - each with a pull-up resistor to +5V. I can provide pin numbers tomorrow if required.

Plug the ribbon cable into the main board - but leave it disconnected from the keyboard for the moment.

Measure the 8 voltages again.

Connect the keyboard to the ribbon cable.

Measure the voltages yet again.

At some point I am expecting you to get a low voltage reading instead of a high voltage reading...

I see we have another machine to fix... The more the merrier... Just worked out the reference to the cheese steak!

Dave
 

NutmegCT

Experienced Member
Joined
Jul 24, 2009
Messages
153
Location
Connecticut
OK - pardon my ignorance (again!).

In the Tech Manual, Figure 2.3 is the logic board.

I see nothing labeled UE12 or 81LS95.

Difficult to measure the voltages at the item, if I can't find the item itself! Could you indicate on Fig. 2.3 where I'll find it?

Thanks.
Tom M.
PS - glad you got my cheese steak reference. Probably would have been clearer if I said Cornish Pasty? :D
 

VERAULT

Veteran Member
Joined
Jan 30, 2012
Messages
8,004
Location
Connecticut, USA
Well, dagnabbit. My brown case OCC-1 worked a year ago, but it's showing similar keyboard issues now. Unresponsive keyboard, won't boot. The screen isn't necessarily flickering all the time, but did when I hit reset. It's been stored in the retrolab. Hrrrmmm...

I dont think thats abnormal. Mine is doing it as well. I did adjust the CRT pots this past summer but it almost seems like I am running it without the RF shield.. some wobble. Im not too worried.
 

daver2

10k Member
Joined
Jun 19, 2012
Messages
14,402
Location
UK - Worcester
Can you point me to an online link for a technical manual that matches the machine you have? I have located an online manual - but it may be a different revision to the machine you have.

I have found this before...

Failing that, can you locate the schematic with the keyboard interface on it and post that somewhere for me to look at please?

I had a Philly cheese steak for lunch once when I was in the States. I didn't need to eat anything in the evening after that!

EDIT: Just rechecked the manual at bitsavers again. IAI0063 sheet 8 of 9 contains UE12 which (according to the parts list on sheet 1 of 9 is an 81LS95.

Dave
 
Last edited:

NutmegCT

Experienced Member
Joined
Jul 24, 2009
Messages
153
Location
Connecticut
Can you point me to an online link for a technical manual that matches the machine you have? I have located an online manual - but it may be a different revision to the machine you have.

OK - I'm way out of my depth. I dont' even know where to look for a tech manual which matches my O1. Here's the only manual I know:

https://drive.google.com/open?id=1rhT0AgpITFOxiehVqEsSoD-YP_hwJf8X

That's where I found the logic board: Fig. 2.3.




I have found this before...

Failing that, can you locate the schematic with the keyboard interface on it and post that somewhere for me to look at please?


The manual I linked also has the keyboard info, beginning at page 41.

I had a Philly cheese steak for lunch once when I was in the States. I didn't need to eat anything in the evening after that!

Philly cheese steaks are known for powering the entire city, in case the mains fail.

EDIT: Just rechecked the manual at bitsavers again. IAI0063 sheet 8 of 9 contains UE12 which (according to the parts list on sheet 1 of 9 is an 81LS95.

Sorry - where is the manual at bitsavers? And I'm afraid I don't know what "IAI0063 sheet 8 of 9" even means!
 

daver2

10k Member
Joined
Jun 19, 2012
Messages
14,402
Location
UK - Worcester
OK - so a little explanation is required…

>>> Sorry - where is the manual at bitsavers? And I'm afraid I don't know what "IAI0063 sheet 8 of 9" even means!

The manuals for the Osborne 1 are located on bitsavers at http://www.bitsavers.org/pdf/osborne/osborne1.

<plug>Bitsavers is an excellent resource for vintage computer ‘nuts’ and would be the ‘go to’ site for any documentation requirements in trying to resurrect vintage computers. If you have any documentation for vintage computers (even different revisions of documents) then it would be worth scanning the manual, putting it on a file sharing website and either sending AL Kossow (the documentation wizard) an e-mail – or posting on a VCFED forum a link to the document and it should magically appear on the bitsavers website for future use by the community. </plug>…

For your information “IAI0063” is the drawing number allocated by Osborne to the schematics for the Osborne 1 and “sheet 8 of 9” is the 8th drawing of a series of 9 drawings for the main logic board.

The document you posted and that on bitsavers is the same, so I will use the references from the document you posted so we are all on the same page (no pun intended).

Figure 2.3 on page 5 of the document (PDF page 15) is the assembly drawing NOT the schematic drawing. If you skip down to Section 12.0 of your document (PDF page 123) you will find the schematic diagram “IAI0063 sheet 8 of 9” that I referred to in my post. This is the schematic for the keyboard interface logic. You should be able to see on that schematic the integrated circuit I referred to “UE12”. The input pins of this IC (connected to the keyboard interface connector P4) are pins 4, 8, 14, 18, 16, 12, 6 and 2.

If you now skip backwards a bit to PDF page 116 – you will find UE12 described in a little table as an “81LS95”. If you now skip forwards a bit to PDF page 125 (the Bill of Materials) you will also find UE12 described as an “81LS95”.

EDIT: It's even marked on the schematic as an 81LS95 Dooooo!

How do we find this IC on the main logic board?

Skip back to your figure 2.3 (assembly drawing) on page 5 of the document (PDF page 15).

What is the meaning of UE12? The letter ‘U’ stands for an integrated circuit (don’t ask me why). E12 refers to the location on a grid. This is Column E Row 12. If you look at the main logic board you should see letters on one side of the board edge (shorter) and numbers on the other edge (longer) at the intersection of E and 12 is the IC we are interested in. I can just make out the text “E” and “12” and at the intersection (right near the keyboard connector P4) I can just make out a 20 pin integrated circuit ‘hand marked’ on the assembly drawing as “81LS9?” (I can’t quite make out the last digit). This is the device we are looking for.

The markings on the device itself should align with the 81LS95 in some way, shape or form – but it may be subtly different. For example (depending on the manufacturer) they may add a prefix or a postfix to the number (for example: DM81LS95ABC). This is still the same device we are looking for. The DM indicates a manufacturer and the ABC will refer to speed, packaging, temperature limit capability and possibly a host of other things.

Does this make things a little clearer?

Your first job (armed with new information) is to hopefully locate this pesky IC (UE12 = 81LS95) on the main logic board!

EDIT: Have a look here https://www.techrepublic.com/pictures/cracking-open-the-osborne-1-the-original-portable-computer/34/. I can just make out an AMD DM81LS95N... where UE12 should be.

Dave
 
Last edited:
Top