UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Computers

Notices

Vintage Computers Any vintage computer systems, calculators, video games etc., but with an emphasis on 1980s and earlier equipment.

Reply
 
Thread Tools
Old 2nd Jan 2018, 10:01 pm   #21
Sean Williams
Dekatron
 
Sean Williams's Avatar
 
Join Date: Feb 2005
Location: St.Ippolyts, Hitchin, Hertfordshire QRA IO91UW
Posts: 3,454
Default Re: 8085 processor questions

Quote:
Originally Posted by cmjones01 View Post
They have 18 pins which makes me think they might be 2114 static RAMs, which are amongst the most unreliable chips ever made! Check what part they actually are.
Chris, they are AM92L44DPC.

No Idea

Sean
__________________
Engineers make things work and have spare bits when finished
Sean Williams is offline   Reply With Quote
Old 2nd Jan 2018, 10:56 pm   #22
RogerEvans
Pentode
 
Join Date: Oct 2014
Location: Wiltshire, UK.
Posts: 197
Default Re: 8085 processor questions

AMD (advanced micro devices) 4k x 1 bit 250ns static RAM.

I think that may be a 4114 near equivalent, maybe more reliable than the earlier parts but still a likely failure. Are they socketed?

I think the typical failure is that the output bit sticks at either 0 or 1 so you might be able to find a dead chip with a scope on the data out pin. I suspect Chris is more expert than I am, it is 30 years since I played with these devices.

Roger
RogerEvans is offline   Reply With Quote
Old 3rd Jan 2018, 7:21 am   #23
Sean Williams
Dekatron
 
Sean Williams's Avatar
 
Join Date: Feb 2005
Location: St.Ippolyts, Hitchin, Hertfordshire QRA IO91UW
Posts: 3,454
Default Re: 8085 processor questions

No, sadly not, so if they are dodgy, it is going to mean an awful lot of desoldering
__________________
Engineers make things work and have spare bits when finished
Sean Williams is offline   Reply With Quote
Old 3rd Jan 2018, 11:13 am   #24
cmjones01
Octode
 
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 1,974
Default Re: 8085 processor questions

I don't have any experience with the AM92L44 static RAMs. At least they're not 2114s, so I wouldn't write them off just yet! However, as Roger says, RAM faults are very common.

A good next move would be look at the address and data buses with a scope or logic probe. Are any bits stuck? Is there a brief burst of activity when the board is reset (if there's a reset button somewhere or you can bodge one with a bit of wire)?

It's good that the display hardware is working. Is that on a separate board to the CPU, or the same one? How are they interconnected? It could also be that the CPU is running but can't write anything to the display.

You say a green light comes on: is there any other sign of life, such as indicators responding to keypresses, that sort of thing?

Chris
__________________
What's going on in the workshop? http://martin-jones.com/
cmjones01 is offline   Reply With Quote
Old 3rd Jan 2018, 1:24 pm   #25
RogerEvans
Pentode
 
Join Date: Oct 2014
Location: Wiltshire, UK.
Posts: 197
Default Re: 8085 processor questions

Sean,

Very early on you said the power supply was OK. Have you looked at the 5V on pin 40 of the 8085 with a scope? If the cpu is continuing to run you should see a lot of noise on the 5V line but staying within say 4.7V - 5.3V. If the power supply regulation is letting the voltage wander further then the CPU may not work reliably or may reset itself. As Chris says it is well worth a lash up to reset the CPU with the power on in case the power is a bit unstable as the CPU comes out of reset. All it needs is a test lead grounded at one end and briefly touch the probe on pin 36 (add a 100R series resistor if you want to be very safe). This resets the CPU with the power lines already up and hopefully stable.

A quick check with the scope on pin 31, WR and pin 32 RD will show if the CPU is trying to do anything. If they are always high then the CPU has probably encountered a HALT instruction after finding something wrong during its self checks (or the code in the EPROM has got bit rot and produced some other error). If the code is running in a tight loop waiting for something to happen then you might be lucky enough to get the scope to sync on a repeating pattern on either RD or WR (you might need to play with the holdoff control to get the timing right).

You should also probe pin 36 (RESET) with the scope to see if anything is trying to reset the CPU after powerup and also pin 6 (TRAP) but that is harder to interpret without a circuit diagram.

Without a circuit or a logic analyser this could become an uphill task!

Best of luck,

Roger
RogerEvans is offline   Reply With Quote
Old 3rd Jan 2018, 6:30 pm   #26
Sean Williams
Dekatron
 
Sean Williams's Avatar
 
Join Date: Feb 2005
Location: St.Ippolyts, Hitchin, Hertfordshire QRA IO91UW
Posts: 3,454
Default Re: 8085 processor questions

Chris,

As far as I can tell, there is another card that deals with the display - I have some information from the German clone of this machine (SEL Lo3000) but the German service manual has no diagrams, which is not so helpful .

Roger,

Thanks, will fire the scope up again a little later

Cheers
Sean
__________________
Engineers make things work and have spare bits when finished
Sean Williams is offline   Reply With Quote
Old 3rd Jan 2018, 7:41 pm   #27
Sean Williams
Dekatron
 
Sean Williams's Avatar
 
Join Date: Feb 2005
Location: St.Ippolyts, Hitchin, Hertfordshire QRA IO91UW
Posts: 3,454
Default Re: 8085 processor questions

I have this information from the German Lo3000 Manual - the Cheetah is a clone of this, both produced by ITT-Creed.

The screen receives its image signals via a ribbon cable from the printed circuit board monitor control (VDCB).
The task of the VDCB is to convert the data of the CPU (*) into control signals.
The data of the CPU reach the 2K RAM of the VDCB via the RAM access control. On
Address multiplexer switches either internal, or the addresses of the CPU to the RAM memory. The memory
is divided into 16 blocks of 128 bytes. The first block (0) contains the character data, such as number of characters per
Line, Scroll (**) information, number of the selected RAM block. Blocks 1-15 each contain the
Character data of a line. The character data is composed of character and feature code. Of the
Feature code contains information about the way a character is represented, e.g. underlined, inverse, bright
etc. 14 lines are displayed on the monitor. The lines are divided into 16 Raster grid lines®. The horizontal
Division of a character is either 9 or 10 pixels per character (80 or 72 characters per line;
5.17).
Fig. 5/17 .: Character representation

When a line is displayed, the line start information (line length, RAM block selection, etc.) is first entered in
Screen RAM address generator stored.
The character clock generator clocks a counter that generates the RAM addresses. About the address multiplexer and the
Feature address generation, an address is selected in the RAM memory. The row and row address generator
generates part of the address for the character PROM (grid line addresses). The characters and feature codes
The RAM is cached and synchronized. In addition, the feature code is still
decodes, since a feature code contains the information for two characters. The character code forms the remaining one
Part of the character PROM address. The data of the EPROM is converted into serial information. Last
The serial character information is synchronized with feature generation. At the exit of the VDCB
then arises the image signal VIDEO +.
Furthermore, the synchronization signals are generated in the screen control. Horizontal synchronization
(LSYNC -) and vertical synchronization (FRSY +). Other control signals are READY (VDCB Ready), CSVDC
(Selection of CPU), IRVDC (interrupt to CPU) and TIVDC (flashing).
The character generator is normally populated with a 2K byte EPROM (e.g., 2716). As an option, a 2.
Character set possible. A 4K EPROM (e.g., 2732) must then be used. The 7th bit of the character code
then selects the character set. Normally (2 Kbytes), the 7th bit contains the flashing status when using a LO 3000
4 kbyte memory no flashing of characters possible.
The screen consists of three functional units:
From the cathode ray tube, the deflection system and the display driver VDD (Visual Display Drive
Board). The VDD generates the necessary voltages for the beam generation of the picture tube. In addition, the
Synchronization signals of the VDCB used for horizontal and vertical beam deflection (by coils). The
Video signal controls the cathode of the tube via amplifier (light-dark keying). The 11 kV anode voltage will
generated by a transformer. The brightness can be pre-set on the VDD and by an external potentiometer of the
Be reset. The power supply provides a direct power supply (12V DC)
Display driver board.
*) (central control electronics)
**) Scroll: Image shift.
Hardscroll = gradual shift
Softscroll = even shift


Of course I think some context is missing due to machine translation.

Sadly no schematics exist it seems
__________________
Engineers make things work and have spare bits when finished
Sean Williams is offline   Reply With Quote
Old 3rd Jan 2018, 7:43 pm   #28
Sean Williams
Dekatron
 
Sean Williams's Avatar
 
Join Date: Feb 2005
Location: St.Ippolyts, Hitchin, Hertfordshire QRA IO91UW
Posts: 3,454
Default Re: 8085 processor questions

Forgot to mention, yes, according to the manual, and layout of the machine, the display board is a separate entity
__________________
Engineers make things work and have spare bits when finished
Sean Williams is offline   Reply With Quote
Old 3rd Jan 2018, 8:05 pm   #29
Sean Williams
Dekatron
 
Sean Williams's Avatar
 
Join Date: Feb 2005
Location: St.Ippolyts, Hitchin, Hertfordshire QRA IO91UW
Posts: 3,454
Default Re: 8085 processor questions

Roger,

pin 31, high pulses.
pin 32, always low.
pin 36, always high.
pin 6, some odd "chatter" that changes when the reset button is released.

THis, to my untrained eye looks like some form of binary code, or counting.

Cheers
Sean
__________________
Engineers make things work and have spare bits when finished
Sean Williams is offline   Reply With Quote
Old 3rd Jan 2018, 9:24 pm   #30
cmjones01
Octode
 
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 1,974
Default Re: 8085 processor questions

A quick test would be to locate the ribbon cable joining the CPU board and display board and disconnect it, and see if the behaviour of the display changes. The 'junk' visible on the screen may be the random data which is in the display board's RAM at power up, or it may be getting written there by the CPU. In any case, a change in the display contents when switching on with and without the display cable connected gives us a clue that the CPU is at least trying to do something.

Chris
__________________
What's going on in the workshop? http://martin-jones.com/
cmjones01 is offline   Reply With Quote
Old 3rd Jan 2018, 9:30 pm   #31
Sean Williams
Dekatron
 
Sean Williams's Avatar
 
Join Date: Feb 2005
Location: St.Ippolyts, Hitchin, Hertfordshire QRA IO91UW
Posts: 3,454
Default Re: 8085 processor questions

Ah, this will be a problem - the cards are all plugged into a mother board.

If you try to run the machine up with either the video board, or the cpu board unplugged, the monitor does not power up

The rubbish on the screen does change if I unplug either the keyboard driver/printer driver board, or the message storage boards
__________________
Engineers make things work and have spare bits when finished
Sean Williams is offline   Reply With Quote
Old 3rd Jan 2018, 10:09 pm   #32
Records
Pentode
 
Records's Avatar
 
Join Date: Mar 2017
Location: Harwich, Essex, UK.
Posts: 111
Default Re: 8085 processor questions

The junk on screen could also be a bad keyboard.
Records is offline   Reply With Quote
Old 4th Jan 2018, 9:33 am   #33
RogerEvans
Pentode
 
Join Date: Oct 2014
Location: Wiltshire, UK.
Posts: 197
Default Re: 8085 processor questions

Sean,

Your measurement that the RD pin 32 is always low shows that something very odd is happeninng. Just to be doubly sure, logic low is less than 0.8V, logic high is greater than 2V and I use the term 'active' to mean that changes between low and high are occurring. Your cpu clock speed should be half the crystal frequency, so about 2.5MHz, and you need a sweep speed of around 1usec/div to observe activity on the control pins and cpu busses. If the scope has a holdoff control then set it to minimum to minimise the period between sweeps.

You observe that RD is always in the low state. Is there a short period after releasing the reset button when RD is active?

The only legal condition I can find for RD to stay low (strictly speaking it is tri-state) is that another device has requested control of the data and address bus by taking pin 39 (HLD) high, in which case pin 38 (HLDA) should go high within a couple of clock cycles. So can you put the scope probe on pin 39 and see what happens when you press and release reset? If you find that HLD is being taken high then the next steps are to find which device is asserting HLD and why HLD is not being released. A list of all the 'large' chips ie with 0.6" pitch, would help but simce we are dealing with multiple boards and a backplane it might mean checking all the other boards as well.

If pin 39 is always low then something else, apart from a HLD request, is causing RD to be pulled low and in the absence of circuit diagrams it may be a case of trying to follow PCB tracks!

Roger
RogerEvans is offline   Reply With Quote
Old 4th Jan 2018, 11:22 am   #34
RogerEvans
Pentode
 
Join Date: Oct 2014
Location: Wiltshire, UK.
Posts: 197
Default Re: 8085 processor questions

Sean,

Apologies for missing this in the previous post - can you check ALE pin 30 with the scope and pressing and releasing the reset button. If the 8085 is trying to control the address and data busses then there will be a pulse on ALE every few clock cycles.

Roger
RogerEvans is offline   Reply With Quote
Old 7th Jan 2018, 10:24 am   #35
Sean Williams
Dekatron
 
Sean Williams's Avatar
 
Join Date: Feb 2005
Location: St.Ippolyts, Hitchin, Hertfordshire QRA IO91UW
Posts: 3,454
Default Re: 8085 processor questions

Progress has rather slowed at present - the PSU has decided to throw it's hand in, so I am going to be occupied trying to sort that out for a while.

This is proving to be a pain in the proverbial!
__________________
Engineers make things work and have spare bits when finished
Sean Williams is offline   Reply With Quote
Old 7th Jan 2018, 3:20 pm   #36
Sean Williams
Dekatron
 
Sean Williams's Avatar
 
Join Date: Feb 2005
Location: St.Ippolyts, Hitchin, Hertfordshire QRA IO91UW
Posts: 3,454
Default Re: 8085 processor questions

Righto, Power supply now repaired - a 5v regulator had decided to act as a short circuit.

A list of "Big" ICs on each board.

CPU Board

D8251AC
D8259AC
D8253C-5
Program Eprom

Printer Drive Board/Decoder Board

D8243C
D8741AD
D8039HL
2 Eproms

D8259AC
D8251AC
D8243C
D8253C-5
D8039HL
1 Eprom

Keyboard

D8039HL
22-950-38
FX1 059
1 Eprom

There are also 2 memory boards - from the description of the German machine that is similar, these two cards are only used for message storage functions, so as far as I can tell these will have little bearing on the basic start up of the machine.

The display driver board has a large selection of 74 series logic chips, which I guess must take the data from the CPU and convert it to something that the monitor understands.

Roger, I will have a rearrangement in the workshop to get the machine and scope closer together, so that I can take some screenshots of what is happening.

Cheers
Sean
__________________
Engineers make things work and have spare bits when finished
Sean Williams is offline   Reply With Quote
Old 7th Jan 2018, 5:23 pm   #37
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 3,486
Default Re: 8085 processor questions

I will add after each IC what it does...



Quote:
Originally Posted by Sean Williams View Post
Righto, Power supply now repaired - a 5v regulator had decided to act as a short circuit.

A list of "Big" ICs on each board.

CPU Board

D8251AC - USART -- serial chip, maybe serial port to the line
D8259AC - Interrupt controller
D8253C-5 - Counter.timer -- maybe used for baud rate generation as the 8251 doesn't include that function.
Program Eprom

Printer Drive Board/Decoder Board

D8243C - I/O expander for the 8048 family. More ports to drive motors, printhead,etc ?
D8741AD - Slave microcontroller
D8039HL - ROMless microcontroller.
2 Eproms - For the 8039 I guess

Is this another PCB?

D8259AC - Interrupt controller
D8251AC - USART serial chip
D8243C - I/O expander for the 8048 family
D8253C-5 - Counter timer -- again maybe baud rate generator for the 8251
D8039HL - ROMless microcontroller
1 Eprom - Firmware for the 8039 I guess

Keyboard

D8039HL - ROMless microcontroller
22-950-38 - I think that's one possible interface chip for a Keytronics capacitive keyboard. How many pins?
FX1 059 - No idea, How many pins?
1 Eprom - firmware for the 8039

There are also 2 memory boards - from the description of the German machine that is similar, these two cards are only used for message storage functions, so as far as I can tell these will have little bearing on the basic start up of the machine.
Unless there's RAM on one of the other boards, I would assume it needed at least one good memory board to start up.

I think if it were mine I would now consider tracing out the full schematics. It will take a long time, particularly if you are not used to doing it, but then you would know what is going on and how it should work.
TonyDuell is offline   Reply With Quote
Old 7th Jan 2018, 10:14 pm   #38
Sean Williams
Dekatron
 
Sean Williams's Avatar
 
Join Date: Feb 2005
Location: St.Ippolyts, Hitchin, Hertfordshire QRA IO91UW
Posts: 3,454
Default Re: 8085 processor questions

Hi Tony,

Thanks for the above. There are a dozen RAM chips on the CPU board.

The printer/decoder board is in two completely separate halves of the same PCB - electrically isolated other than connections via the motherboard.

I think drawing this out will be a mammoth task! there are 8 double sided PCBs and a double sided motherboard - not impossible, but a serious task!

I am hoping to find something in the BT archives that may help - I am planning a trip down there in the very near future.

Cheers
Sean
__________________
Engineers make things work and have spare bits when finished
Sean Williams is offline   Reply With Quote
Old 8th Jan 2018, 12:28 am   #39
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 3,341
Default Re: 8085 processor questions

The chip with the 'FX' prefix doesn't also have a 'CML' logo on it as well does it?

CML made a series of dedicated tone encoder / decoder ICs all with 'FX' prefix numbers. I mainly encountered them in two way radio selcall / signalling systems, but they may have had other uses or made other ICs dedicated to modem-type signalling.
SiriusHardware is online now   Reply With Quote
Old 8th Jan 2018, 12:41 pm   #40
RogerEvans
Pentode
 
Join Date: Oct 2014
Location: Wiltshire, UK.
Posts: 197
Default Re: 8085 processor questions

Sean,

You have got a very complex system on your hands, not only the 8085 that has been the main focus so far but a number of 8039 microcontrollers with their own EPROMs and an 8741 with on-board EPROM and several other chips with internal registers that determine their behaviour. Fault finding without any further information is likely to be very hard work or very good luck. Even if you trace out the circuitry you are faced with not knowing what should happen on power up so it is hard to find the first event that the processor finds to be wrong rather than the one that finally causes it to hang. If the fault turns out to be EPROM data that has been corrupted over time then you are faced with not knowing what it should be in order to fix it.

Do you know anyone with the same or similar machine in working condition? That would give you information on the timing of events after start up eg how long before the screen clears and what happens with status lights. If you were really lucky you might get to do board swaps to isolate the fault down to a single board.

One thing that is worth a try is to remove all boards apart from the 8085 board and the display board and see if the start up proceeds as far as clearing the display and then maybe some sort of start up message or error message.

Beyond that I guess it is up to you how much time, effort and money you want to invest and how much pleasure you get out of the different activities (I doubt many of us enjoy tracing out circuitry from PCB tracks but there may be no other choice). If you want to track down problems on the 8085 board then a logic analyser (at least 16 channel) is a sound investment, it would be interesting to hear if anyone has experience of the USB logic analysers that sell in the £40 - £80 range on eBay. The 8039s can be partly monitored with the logic analyser but their RAM is on-board so the contents don't appear on the data bus. The 8741 has its program memory on-board as well so trying to debug that would be quite a callenge!

Regards,

Roger
RogerEvans is offline   Reply With Quote
Reply

Thread Tools



All times are GMT +1. The time now is 10:26 pm.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2020, vBulletin Solutions, Inc.
Copyright ©2002 - 2020, Paul Stenning.