Threaded Mode | Linear Mode

teenix · (This post was last modified: 06-30-2019 10:38 AM by teenix.)

Hi all,

I posted a PDF file with some Classic display notes I threw together if anyone is interested.

They are not as comprehensive as Jarques Laporte's excellent works but may be useful just the same.

http://www.teenix.org/ClassicDisplay.pdf

cheers

Tony

TomC · 06-30-2019, 02:53 PM

Tony:

As always, great work! Thank you again for your excellent engineering preservation activities!

TomC

teenix · 07-01-2019, 01:17 AM

Hi all,

Caution: Techy stuff follows :-)

Has anyone figured out how the Classic cathode driver is actually stepped to the next digit? I've looked around and no info. I know it's just a simple pulse, but how that pulse is generated is confusing.

I know there are errors shown on the HP45 patent document. One in particular (I think) is the RCD line that resets the cathode driver scan. My scope sees the pulse as 20uS during T4, whereas the HP diagram (Fig 18) shows a 1.25 uS pulse on T1.

The C CL pulse is supposed to step the cathode driver to the next display digit on its rising pulse edge and is shown doing this in Fig 18. It occurs just as the first LED segment of the digit is about to be excited (segment E).

I'm not sure how this works as it suggests that after a scan reset, the first digit is stepped over before it gets displayed. This obviously doesn't happen so I'm a bit confused.

This pulse is also tied to the anode driver [E] line from the ARC chip which seems to suggest that the LED G segment being lit or not affects the stepping also, which to me doesn't make sense.

PS Not expecting an answer, but you never know :-)

cheers

Tony

teenix · 07-03-2019, 06:31 AM

Hi all,

Doing a heap of research for the Classic displays paid off.

What's even more cool, the whole circuit based on a PICmicro is running off a 3.7V supply.

:-)

cheers

Tony

Harald · 07-03-2019, 08:39 AM

(07-03-2019 06:31 AM)teenix Wrote: Hi all,

Doing a heap of research for the Classic displays paid off.

What's even more cool, the whole circuit based on a PICmicro is running off a 3.7V supply.

:-)

cheers

Tony

Wow! How did you do that?
I did not see a chance of getting it to work without the DMA. Thats why I went for the renesas controller.

Cheers,
Harald

teenix · 07-03-2019, 10:16 AM

(07-03-2019 08:39 AM)Harald Wrote: Wow! How did you do that?
I did not see a chance of getting it to work without the DMA. Thats why I went for the renesas controller.

Cheers,
Harald

It was a bit of a surprise to me also. Except for the power supply everything is bread boarded with flying leads. It was squinty eyes and a nervous finger on the power switch Smile

The PIC can collect and output the Anode data in 375nS plus a small time needed to detect when to do it. I suspect the inductor charge time is eaten into a little bit but the display seems bright enough.

I am yet to hook up the scope to see what it happening on the inductors. Even though the LEDs light up ok, it might be doing it just barely, but I think I have another trick up my sleeve to all but eliminate the latency.

cheers

Tony

Harald · (This post was last modified: 07-03-2019 07:21 PM by Harald.)

(07-03-2019 10:16 AM)teenix Wrote:
(07-03-2019 08:39 AM)Harald Wrote: Wow! How did you do that?
I did not see a chance of getting it to work without the DMA. Thats why I went for the renesas controller.

Cheers,
Harald

It was a bit of a surprise to me also. Except for the power supply everything is bread boarded with flying leads. It was squinty eyes and a nervous finger on the power switch

The PIC can collect and output the Anode data in 375nS plus a small time needed to detect when to do it. I suspect the inductor charge time is eaten into a little bit but the display seems bright enough.

I am yet to hook up the scope to see what it happening on the inductors. Even though the LEDs light up ok, it might be doing it just barely, but I think I have another trick up my sleeve to all but eliminate the latency.

cheers

Tony

Hi Tony,

I am very keen to see your design.

You have to be careful not to charge the inductors without any of the cathode transistor turned on. That will result in an overvoltage that will sooner or later destroy the anode drive. I have solved this by soldering a green LED across one of the cathode transistors. That dissipates the energy and saves the anode driver. Also it indicates if you still have any glitches (I did at startup) that cause problems. The original HP hardware has a startup problem too, but only a very short pulse occures.

Cheers,
Harald

Edit:
To avoid misunderstandings, the green LED was only a temporary measure during development. Once I was satisfied it never turned on anymore I removed it.

teenix · (This post was last modified: 07-07-2019 03:14 AM by teenix.)

Hi all,

I managed to get a nice shot of the step pulses from the anode driver plus the extra one when the decimal point is displayed. The display is showing 0.00

I've updated the classic display notes to suit.

http://www.teenix.org/ClassicDisplay.pdf

cheers

Tony

Kees van der Sanden · 07-09-2019, 07:18 PM

(07-01-2019 01:17 AM)teenix Wrote: Hi all,

Caution: Techy stuff follows :-)

Has anyone figured out how the Classic cathode driver is actually stepped to the next digit? I've looked around and no info. I know it's just a simple pulse, but how that pulse is generated is confusing.

I know there are errors shown on the HP45 patent document. One in particular (I think) is the RCD line that resets the cathode driver scan. My scope sees the pulse as 20uS during T4, whereas the HP diagram (Fig 18) shows a 1.25 uS pulse on T1.

The C CL pulse is supposed to step the cathode driver to the next display digit on its rising pulse edge and is shown doing this in Fig 18. It occurs just as the first LED segment of the digit is about to be excited (segment E).

I'm not sure how this works as it suggests that after a scan reset, the first digit is stepped over before it gets displayed. This obviously doesn't happen so I'm a bit confused.

This pulse is also tied to the anode driver [E] line from the ARC chip which seems to suggest that the LED G segment being lit or not affects the stepping also, which to me doesn't make sense.

PS Not expecting an answer, but you never know :-)

cheers

Tony

Hi Tony,

This is how I think it works:

The data to be displayed is controlled by the ARC (Arithmetic and Register Circuit).

The ARC has to synchronize the anode driver and cathode driver for displaying of the data in the correct way.

To synchronize the anode driver with the ARC, the anode driver resets itself if input B and D are high and Phi1 is low.
This is an internal reset signal, not accessible from the outside. This pulse is shown in Fig 18.
In this way the ARC knows (forces) the T state of the anode driver.

To synchronize the cathode driver with the ARC, the ARC generates the RCD (Reset Cathode Driver ?) pulse.
This pulse is indeed 5 us long and appears during T4.

The cathode driver gets the C CL pulse from the anode driver to step to the next digit.
(This signal could also be generated by the ARC, but this solution is more simple, I think).
Best regards,

Kees

teenix · (This post was last modified: 07-09-2019 11:16 PM by teenix.)

Hi Kees,

Many thanks for the reply. I agree, in that I suspect a few extra gates in the anode driver creates a simpler system overall.

cheers

Tony

(07-09-2019 07:18 PM)Kees van der Sanden Wrote: Hi Tony,

This is how I think it works:

The data to be displayed is controlled by the ARC (Arithmetic and Register Circuit).

The ARC has to synchronize the anode driver and cathode driver for displaying of the data in the correct way.

To synchronize the anode driver with the ARC, the anode driver resets itself if input B and D are high and Phi1 is low.
This is an internal reset signal, not accessible from the outside. This pulse is shown in Fig 18.
In this way the ARC knows (forces) the T state of the anode driver.

To synchronize the cathode driver with the ARC, the ARC generates the RCD (Reset Cathode Driver ?) pulse.
This pulse is indeed 5 us long and appears during T4.

The cathode driver gets the C CL pulse from the anode driver to step to the next digit.
(This signal could also be generated by the ARC, but this solution is more simple, I think).
Best regards,

Kees

teenix · 11-04-2019, 08:27 AM

Hi all,

If anyone is interested, I added some extra info on the classic displays in my notes PDF.

It shows a few snap shots of the Anode driver step pulses with the ARC RCD pulse to the Cathode driver on a working Classic. It also shows the extra steps for the decimal point(s).

http://www.teenix.org/ClassicDisplay.pdf

cheers

Tony

Danito-Danone · 11-14-2019, 11:31 PM

Hi folks, Are these spare parts useful: CA3081 & CA3082? Thanks.

PANAMATIK · 11-16-2019, 11:34 AM

(11-14-2019 11:31 PM)Danito-Danone Wrote: Hi folks, Are these spare parts useful: CA3081 & CA3082? Thanks.

These are Quad NPN transistor drivers. They were used in HP-19C as printer pixel drivers. As these ICs are obsolete today, there are pin compatible circuits available.

Bernhard

Danito-Danone · 11-16-2019, 05:30 PM

I mean, for an HP-67 1858-0050 display driver: Which is the right replacement: CA3081 or CA3082? Thanks.

teenix · 11-16-2019, 11:17 PM

(11-16-2019 05:30 PM)Danito-Danone Wrote: I mean, for an HP-67 1858-0050 display driver: Which is the right replacement: CA3081 or CA3082? Thanks.

CA3082
It has the common collector configuration.

cheers
Tony

teenix · (This post was last modified: 02-04-2020 03:04 AM by teenix.)

Hi all,

I have added some more info to my Classic notes.

I have snapped some logic traces capturing information on the WS IS and IA busses from my HP-45. I have tried to document how it all works and what the wave forms represent.

I still have to do the RAM chip signals but I cannot make sense of the BCD line yet. Plus it is hard to capture when that bus is active when you press STO 1 for example.

http://www.teenix.org/ClassicDisplay.pdf

I will probably have to slow down with HP stuff for a bit. The flight school that I used to instruct at is opening a new school on the western side of Port Phillip Bay and they want me to manage it. As the school is new and has to get off the ground (pun) it is going to eat into my spare time for awhile. There is not much money at this end of the aviation spectrum so it will mostly be a labour of love - until I fire up my simmering urge to retire :-)

cheers

Tony

Kees Bouw · 02-22-2020, 10:03 PM

(02-04-2020 03:03 AM)teenix Wrote: Hi all,

I have added some more info to my Classic notes.

http://www.teenix.org/ClassicDisplay.pdf

Tony

Hi Tony,

This link does not seem to work anymore. I assume this file has been replaced by ClassicNotes.pdf?

Kees.

teenix · 02-23-2020, 02:49 AM

(02-22-2020 10:03 PM)Kees Bouw Wrote:
(02-04-2020 03:03 AM)teenix Wrote: Hi all,

I have added some more info to my Classic notes.

http://www.teenix.org/ClassicDisplay.pdf

Tony

Hi Tony,

This link does not seem to work anymore. I assume this file has been replaced by ClassicNotes.pdf?

Kees.

Hi Kees,

Yes, Sorry, I should have updated the link.

I changed the name because there is more than just display notes now.

cheers

Tony