• Please review our updated Terms and Rules here

Osborne 1 keyboard interface?

So, going all the way pack to posts 1 and 2 of this thread you stated that when you did the <RETURN> boot trick with CP/M without the keyboard/cable connected and then connected the keyboard/cable you got a load of lower case 't's displayed. Is this still the case? This would indicate a problem with D4 (check the key matrix) but this is connected to UE12 pins 16 (input) and 15 (output) and nowhere near pins 7 or 8. This is inconsistent.

Did you perform the test I asked back in post 3? If there is a faulty UE12 (or associated logic) this *** SHOULD *** show up with this test.

I agree that if you have performed the test with UE12 pins 7 and 8 three times with the same result then it is valid. However, in my experience, measuring a datas bus line with a voltmeter (which is effectively what UE12 pin 7 is) shouldn't cause the problem you are stating. Otherwise, I am stumped - unless we have a problem with the output of UE12 pin 7 - and this should show up with the test asked for in post 3 again.

UE12 pin 7 hosts keys 7 O U J M and L.
UE12 pin 8 hosts keys 2 left-cursor W S X down-cursor.

We may (of course) be chasing multiple faults - each with their own symptoms. This is a possibility.

The 'damage' (if that is what has occurred) could have happened with the unpolarised cables you used - and then we have subsequently observed that symptom. The original problem could have been a faulty keyboard 'T' switch?

I would guess that replacing the main board should work. But there are still no guarantees and (until you know for sure) this could be an expensive way to find out it doesn't cure your problem.

I am off to bed...

Dave
 
Last edited:
So, going all the way pack to posts 1 and 2 of this thread you stated that when you did the <RETURN> boot trick with CP/M without the keyboard/cable connected and then connected the keyboard/cable you got a load of lower case 't's displayed. Is this still the case?

* * * yes, same thing happens. Lots of lower case t's.

This would indicate a problem with D4 (check the key matrix) but this is connected to UE12 pins 16 (input) and 15 (output) and nowhere near pins 7 or 8. This is inconsistent.

Did you perform the test I asked back in post 3? If there is a faulty UE12 (or associated logic) this *** SHOULD *** show up with this test.

* * * That's the good news. As I reported earlier, if I boot, then drop to cp/m, then shunt various pin pairs, I do get the correct character expected.

I agree that if you have performed the test with UE12 pins 7 and 8 three times with the same result then it is valid. However, in my experience, measuring a datas bus line with a voltmeter (which is effectively what UE12 pin 7 is) shouldn't cause the problem you are stating. Otherwise, I am stumped - unless we have a problem with the output of UE12 pin 7 - and this should show up with the test asked for in post 3 again.

UE12 pin 7 hosts keys 7 O U J M and L.
UE12 pin 8 hosts keys 2 left-cursor W S X down-cursor.

We may (of course) be chasing multiple faults - each with their own symptoms. This is a possibility.

The 'damage' (if that is what has occurred) could have happened with the unpolarised cables you used - and then we have subsequently observed that symptom. The original problem could have been a faulty keyboard 'T' switch?

* * * That's what we thought when I first started these threads. So I took the keyboard apart, broke the key base connector points for the T, and following instructions, slit the membrane under the T key. No change (other than the keyboard membrane was now ruined ...). Thus, the replacement keyboard.

I would guess that replacing the main board should work. But there are still no guarantees and (until you know for sure) this could be an expensive way to find out it doesn't cure your problem.

I am off to bed...

* * * Good idea! I'll do the same in a few hours. Thanks.

Dave
 
A good nights sleep!

There are a few things in the post above - so I would like to take them one at a time if I may.

I think measuring the voltages on UE12 pins 7 and 8 is a bit of a 'red herring'. Measuring something will, invariably, change the parameter we are measuring in some way. I don't personally think that what is observed is correct behaviour - but I don't think it is related to your 'T' problem either or (if it is) it is via an obscure route.

I therefore propose that we 'file' this issue until we have sorted out the keyboard and then come back to it later.

Are you in agreement with this?

Dave
 
Sounds good to me. But I won't be cutting and slicing things any more. I can't afford to kill the patient again, just to do a post mortem!

One point to remember: the "repeated t's" issue existed with the previous keyboard also.

T.
 
>>> One point to remember: the "repeated t's" issue existed with the previous keyboard also.

That was what I was going to get onto next...

>>> Sounds good to me. But I won't be cutting and slicing things any more. I can't afford to kill the patient again, just to do a post mortem!

Agreed :)!

>>> * * * That's what we thought when I first started these threads. So I took the keyboard apart, broke the key base connector points for the T, and following instructions, slit the membrane under the T key. No change (other than the keyboard membrane was now ruined ...). Thus, the replacement keyboard.

Now, I seem to remember that we diagnosed a faulty 'T' switch by measuring the resistance between the various pins of the keyboard connector itself when disconnected from the cable and computer. We came to the conclusion that there should only be two (2) pins connected together on this connector (these were the two 0V/ground pins) but (when you tested the keyboard connector with your multimeter) you found another set were shorted together. We both deduced that the two pins were associated with the 'T' switch. The post mortem was conducted based upon those measurements.

So, putting the old keyboard to one side for the moment.

You purchased a replacement keyboard and you still observe 't's on the screen when CP/M is forced to boot and you then connect the new keyboard. Yes?

Question: How do you know that the replacement keyboard hasn't got the same fault as your original keyboard? Unlikely, but...

Test proposal:

Disconnect your new keyboard from the cable and computer and (using your multimeter) test the resistance between all of the keyboard pins.

Let's ignore any numbering scheme that may be in existence and give the pins of the keyboard connector letters. Looking at the keyboard connector:

Code: 
[FONT=Courier New]A B C D E F G H I J
K L M N O P Q R S T[/FONT]

I would use the multimeter (on resistance) to check for continuity between A and B, then A and C, then A and D,... A and T. Next for B-C, B-D, ..., B-T and so forth up to the last test which should be S-T.

We are assuming here that there are no semiconductor devices on the keyboard (which is the case as it should only consist of passive switches) and, therefore, it doesn't matter which way we connect the multimeter probes to the connector. We can reduce the number of tests by one pin every time because we have tested that combination once.

When you have finished, you should ONLY have two (2) pins connected together. Anything else is a fault.

This should absolutely rule out a short circuit in the replacement keyboard you have purchased.

Does this step sound logical to you and are you happy with the test process that I have described and why we are performing it?

Dave
 
You purchased a replacement keyboard and you still observe 't's on the screen when CP/M is forced to boot and you then connect the new keyboard. Yes?

*** yes ***

Question: How do you know that the replacement keyboard hasn't got the same fault as your original keyboard? Unlikely, but...

*** I don't know. But the replacement keyboard worked correctly on the O1 it was previously connected to. Seems very unlikely it would develop the same "t" problem as my original, but ... you never know ... ***

Test proposal:

Disconnect your new keyboard from the cable and computer and (using your multimeter) test the resistance between all of the keyboard pins.

*** OK, here's another area I don't understand. Do you mean the keyboard pins on the keyboard? or the keyboard pins on the motherboard? It's things like this that really get me frustrated; if I spend a couple hours doing these tests, then find out I was testing the wrong "thing" ... well, it's not pretty. ***

Let's ignore any numbering scheme that may be in existence and give the pins of the keyboard connector letters. Looking at the keyboard connector:

*** Again, the keyboard connector on the keyboard? or the keyboard connector on the motherboard? ***

Thanks.
Tom M.
 
The keyboard pins on the keyboard itself (for both questions).

What we are trying to demonstrate/prove is that the physical keyboard itself doesn't have any unintended short circuits that shouldn't be there. I don't expect it will - as you have stated that it does work on another O1 - but I want to absolutely rule it out.

Dave
 
Here we go - the only "closed circuit" is between J and K. No other pair of pins shows continuity. This is the same result as on the old keyboard - and I'm pretty sure we already said that those two pins are GND.

Onward!
 
>>> Here we go - the only "closed circuit" is between J and K. No other pair of pins shows continuity.

Excellent news, that was what we were expecting with a fully-functional keyboard though wasn't it!

>>> This is the same result as on the old keyboard - and I'm pretty sure we already said that those two pins are GND.

Partially correct. We did have two of the extreme pins connected together (GND as you state) but in post #21 of the original thread (http://www.vcfed.org/forum/showthread.php?67552-Osborne-I-no-disk-drive-response&p=550756#post550756) you identified a picture of the keyboard connector on the original keyboard and identified that two further pins were also shorted together. We both diagnosed these pins to be the letter 'T' key and this is where we came to the conclusion that the original keyboard you had was faulty.

Do you agree?

Dave
 
" ... on the original keyboard and identified that two further pins were also shorted together. "

Yes, completely forgot about that. Thanks. But at least the replacement keyboard doesn't have that problem. Does seem strange that both keyboards had the "t" problem however. Remember that the replacement keyboard does work correctly one key at a time, if I just lightly touch a female terminal to the male pins, then remove it. Push down all the way, and the problem returns.

Tom M.
 
Excellent.

The next thing (as you have correctly pointed out) is that your problems seem to start when you connect the physical keyboard up to the ribbon cable. What I want to try next is to see whether plugging the ribbon cable into the keyboard connector on the keyboard does or doesn't create an inadvertent short circuit.

So, to do this proceed as follows:

1) Select a ribbon cable. I would use the ribbon cable that has the polarising lugs at both ends so we can't accidentally plug it the wrong way round.

2) Connect the keyboard end of the ribbon cable to the keyboard connector of the physical keyboard.

I need to know two things from you:

Q1) What is the manufacturer, make and model of your multimeter? and

Q2) What resistance measurement ranges have you got on your multimeter?

What I want to do is to select a resistance reading range that will reliably detect anything of 1 kOhm or less. I would like a range with (say) a maximum reading of 10 kOhm.

Dave
 

Q1) What is the manufacturer, make and model of your multimeter? and

I have two. Best DT-9205M Ohm ranges: 200 2k 20k 200k 2M 20M 200M and "continuity beep"

Radio Shack 22-109 Ohm ranges: x10 x100 x1K and "continuity beep"
 
The 'Best' sounds the best (!) set to a 20k range I think.

With the polarised keyboard ribbon cable connected at one end to the keyboard connector on the keyboard - and with the other end free - you need to repeat the test that you previously did back in post #45 on the free end of the ribbon cable (if this makes sense).

The O1 main board takes no part in this test.

The problem that you will have is that you have a female connector to contend with rather than the male connector you did previously on the keyboard connector on the keyboard.

The way I deal with this is to get a couple of stout pieces of single-core copper wire, strip the insulation away and use those attached to my multimeter probes. You can practice ensuring you get a good contact on the two pins that we know for sure are connected together (0V).

What you are going to do (yes again) is to test every one of the 20 pins to each and every other one looking for an unintended short circuit somewhere - but this time induced by the keyboard-to-cable connection.

Have I explained this OK for you or not?

Dave
 
Sounds clear cut to me. I use small sewing needles to test female connectors.

I should have results within 30 minutes.

Tom M.
 
Interesting. Polarised cable plugged into keyboard only. Testing the free female end, the usual J-K pair shows continuity, BUT, I also have continuity at N-R. No other connections evinced.

(I use the same alpha terminology to match the male pins we started with on the m/board connector. Thus, m/board A pin connects to cable A socket.)

To double check, I disconnected the keyboard end of the cable, and tested the other end again. No N-R connection. Plugged keyboard end of cable back onto keyboard male pins, and N-R connection reappeared.

I then disconnected the cable from the keyboard again, and tested the keyboard male pins. The J-K "common" shows continuity, but the pins corresponding to the N-R lines show continuity also. To me this indicates a keyboard fault.

This gets me closer to jumping off the Brooklyn Bridge. The replacement keyboard now has the same problem as the old keyboard?
 
Just tried to Edit my post #56. Screen froze and lost my edits (additions). Here's what I meant to add:

This gets me closer to jumping off the Brooklyn Bridge. The replacement keyboard now has the same problem as the old keyboard - different pins, but same "result". Replacement keyboard booted up the source O1 by hitting the Return key. Looks like I should have done this extensive pin test on the replacement keyboard!

This may be the time to throw in the towel.
 
Hmmmmm...

This solves why this problem is being such a pain!

So, (to sum up) you have two keyboards. The original keyboard had (we believe) a faulty T key. The replacement keyboard an hour or so ago seemed to be OK but now has a faulty key. This happened without the keyboard being physically plugged into an O1 at any time between the two tests (without and with the cable)?

However, the new keyboard still appears to work on the original O1?

Have I summed up correctly?

Dave
 
OK - third attempt at making reply. Previous two attempts gave me "Message too short. Please write message longer than five characters".
Only option was to cancel, and try to post my reply again.

Really folks, this is getting tedious. Not your fault, but I'm about to go nuts here.

daver2 said:
Hmmmmm...

This solves why this problem is being such a pain!

So, (to sum up) you have two keyboards. The original keyboard had (we believe) a faulty T key. The replacement keyboard an hour or so ago seemed to be OK but now has a faulty key. This happened without the keyboard being physically plugged into an O1 at any time between the two tests (without and with the cable)?

*** no, replacement keyboard never worked on my O1. Soon as I plugged in the keyboard and powered up, I got the CRT blinking and inability to use Return to boot the system.

However, the new keyboard still appears to work on the original O1?

*** can't tell that, as I no longer have access to that replacement keyboard's O1 ***

Have I summed up correctly?

Dave
Click to expand...
 
You must log in or register to reply here.
Back
Top