BBC B+ 128K: Fault after warming up.

discuss both original and modern hardware for the bbc micro/electron
User avatar
1024MAK
Posts: 10549
Joined: Mon Apr 18, 2011 5:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by 1024MAK »

Sounds like IC40 (74LS20) is faulty.

Mark
stephen_usher
Posts: 143
Joined: Sun Apr 10, 2016 2:47 pm
Location: Oxford, UK.
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by stephen_usher »

I've just realised that I've been a complete and utter idiot and have been mis-counting chip pins all along! (Instead of counting anti-clockwise I was counting down the left side and then down the right side. Should have read more/better.)

OK, back to the drawing board then...

Pin 8 is pulsing, pin 9 is always high, pin 10 is the inversion of pin 8, pins 12 and 13 are always high.

So, maybe it is doing the right thing after all.
Last edited by stephen_usher on Sat Mar 30, 2019 1:56 pm, edited 1 time in total.
stephen_usher
Posts: 143
Joined: Sun Apr 10, 2016 2:47 pm
Location: Oxford, UK.
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by stephen_usher »

I'm still puzzled how there are video glitches in Mode 7, which is being generated purely by the SAA 5050 and the video fed into the VC2069, only when the processor is running. As soon as it halts (which you can hear from the speaker) the display becomes totally clear and non-glitchy.

e.g. https://youtu.be/uyyFZn76DAo
Last edited by stephen_usher on Sat Mar 30, 2019 2:14 pm, edited 1 time in total.
stephen_usher
Posts: 143
Joined: Sun Apr 10, 2016 2:47 pm
Location: Oxford, UK.
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by stephen_usher »

Oh, it reset into Mode 0... That was interesting. Lots of corruption when the CPU is running but also when it had halted.

See this video... Most of the time the CPU had crashed and halted. Notice the right-hand side of the A character?

https://youtu.be/4AKk6vu5nTk
Last edited by stephen_usher on Sat Mar 30, 2019 4:10 pm, edited 1 time in total.
User avatar
1024MAK
Posts: 10549
Joined: Mon Apr 18, 2011 5:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by 1024MAK »

stephen_usher wrote:
Sat Mar 30, 2019 2:12 pm
I'm still puzzled how there are video glitches in Mode 7, which is being generated purely by the SAA 5050 and the video fed into the VC2069, only when the processor is running. As soon as it halts (which you can hear from the speaker) the display becomes totally clear and non-glitchy.

e.g. https://youtu.be/uyyFZn76DAo
For all modes the 6845 CRTC (cathode ray tube controller) generates the video address in main RAM. The 6845 CRTC also generates the cursor.
For modes 0 to 6 the data from the main RAM is sent to the videoproc (VC2069) which then converts the data to RGB video outputs.
For mode 7, the data from the main RAM is sent to the SAA 5050 which then converts the data to RGB video, and is then sent on to the videoproc (VC2069) which then outputs RGB video to the rest of the video circuitry.

You can halt the CPU at any time by pressing and holding the BREAK key.

Mark
stephen_usher
Posts: 143
Joined: Sun Apr 10, 2016 2:47 pm
Location: Oxford, UK.
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by stephen_usher »

I'm getting a bit disheartened now.

I feel as if I'm not really getting anywhere with the fault finding other than it seems to be bad data on the data lines when the CPU is running and some mis-reading of memory when it's not (mode 0 character edge "wobbles", which could be a dodgy memory chip I guess, but only on read as the bits reappear correctly later).

I don't have a working reference system to use to see what's "normal" as the BBC Model B is quite different.
Last edited by stephen_usher on Sat Mar 30, 2019 4:10 pm, edited 1 time in total.
User avatar
1024MAK
Posts: 10549
Joined: Mon Apr 18, 2011 5:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by 1024MAK »

If it getting to you, have a break.

Fault finding can be extremely frustrating at times.

Mark
stephen_usher
Posts: 143
Joined: Sun Apr 10, 2016 2:47 pm
Location: Oxford, UK.
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by stephen_usher »

It's just knowing where to start. I don't s'pose you've come across these symptoms before have you?
Kazzie
Posts: 1793
Joined: Sun Oct 15, 2017 8:10 pm
Location: North Wales
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by Kazzie »

I'm coming back to this thread after a while away, so bear with me while I catch up.

After replacing IC49, you said the Beeb is "slightly more stable". Could you add a bit more detail to that: what is it now doing, vs no longer doing?

You said recently that the machine booted to Mode 0 unexpectedly. The Beeb can be configured to boot in a particular Mode by the use of three of the optional links on the keyboard. These behave like extra keys on the keyboard, and are read at boot. So it's possible that the same errors causing phantom keypresses also caused the state of these links to be mis-read. (Though going from Mode 7 to Mode 0 would require all three screen-related links to be mis-read.) Do you still get phantom keypresses (or is the system not booting far enough for you to tell)?

Comparing your first two videos with the latest (Mode 7 and Mode 0) ones, you don't seem to have any characters substituted any more. (It seems that Bit 1 was stuck low, with all the faulty characters being replaced with one 2^1=2 values lower. (e.g. @AQIA 1550 DDQ: the italic characters should have bit 1 set, to read ACORN 1770 DFS.) Did this fault go away when you replaced IC49, or do you still see some characters substituted occasionally?
stephen_usher wrote:
Sat Mar 30, 2019 2:12 pm
I'm still puzzled how there are video glitches in Mode 7, which is being generated purely by the SAA 5050 and the video fed into the VC2069, only when the processor is running. As soon as it halts (which you can hear from the speaker) the display becomes totally clear and non-glitchy.
Pressing (and holding) the Break key pulls the /RESET (/RS) line low. This resets the CPU, but is also used by some other chips, namely (as far as I can see) the external VIA (IC10), the 68B54 Econet chip (IC81), and the 8271 Floppy Controller (IC15, using RS rather than /RS). You don't have IC 15 fitted (you've got the 1770 controller instead), and I'm going to guess that you don't have the Econet chips fitted either. That'd leave the external VIA as the only chip apart from the CPU that responds directly to the reset signal.

It's worth noting that the /RS line is also exported on the 1MHz and Tube connectors on the underside of the Beeb. It'd be worth checking to make sure they're not bent out of position or anything like that.

The internal VIA has it's own resistor-capacitor-based reset circuit to reset it on a cold boot, and it doesn't get reset again by the break key.

---

A thought has come to me as I research/ramble through this post: IC11 (a 74ls374) latches A0-A7 for the 1MHz peripherals. Latched A4-A7 are only used by the 1MHz expansion connector, but latched A0-A3 are used for the two VIAs, and I see on the circuit diagram that the latched LA0 is used for the register select pins on the 6845 CRT controller and the 6850 serial interface (IC82) (confusingly labelled as RS, but nothing to do with resetting!). That means that IC11 could be a single point of failure for both the keyboard interface and the glitches on the video circuitry.

:arrow: I'd suggest taking a closer look at IC11, and see if it's latching the address lines correctly. The eight input (D) and output (Q) pins should match each other when the 1MHz signal (pin 11) is high, but the Q pins shouldn't change value while pin 11 is low.
Last edited by Kazzie on Sat Mar 30, 2019 9:50 pm, edited 1 time in total.
BBC Model B 32K issue 7, Sidewise ROM board with 16K RAM
Archimedes 420/1 upgraded to 4MB RAM, ZIDEFS with 512MB CF card
RiscPC 600 under repair
Acorn System 1 home-made replica
stephen_usher
Posts: 143
Joined: Sun Apr 10, 2016 2:47 pm
Location: Oxford, UK.
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by stephen_usher »

By "slightly more stable" I mean more predictable. It always seems to crash in the same/similar manner.

The substituted characters went away by themselves. The fault when I saw it originally, when I picked up the machine in '97, was always that symptom with an almost immediate crash.

After being left in the loft for almost 20 years the machine would boot normally but then after about a minute start misbehaving. (Something had partially self-healed over that time.) Initially I could write a short BASIC parogram and run it, but it would start getting errors whilst reading the code and sometimes crash, but a Break and OLD would recover it, meaning that it was most likely to be read errors from memory rather than the memory itself failing.

After leaving the machine on for an hour to see what happened the machine got into the effectively the state it is now where it's immediately having problems, as seen in the videos. i.e. the component has fully failed again, in a slightly different manner.
stephen_usher
Posts: 143
Joined: Sun Apr 10, 2016 2:47 pm
Location: Oxford, UK.
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by stephen_usher »

That's interesting... I've been freeze spraying chips again. IC54 is sensitive now and the system won't start if it's too cold. That's interfacing the data lines with the SAA 5050.
stephen_usher
Posts: 143
Joined: Sun Apr 10, 2016 2:47 pm
Location: Oxford, UK.
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by stephen_usher »

Pin 4 on IC54 is behaving strangely in that putting the probe on it will not allow the CPU to run for long at all. None of the other data lines are sensitive. That's D1, controlling the flip-flop used by the SAA 5050... Which could be the potential cause for the strange characters.
Last edited by stephen_usher on Sat Mar 30, 2019 10:42 pm, edited 2 times in total.
Kazzie
Posts: 1793
Joined: Sun Oct 15, 2017 8:10 pm
Location: North Wales
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by Kazzie »

stephen_usher wrote:
Sat Mar 30, 2019 10:01 pm
By "slightly more stable" I mean more predictable. It always seems to crash in the same/similar manner.

The substituted characters went away by themselves. The fault when I saw it originally, when I picked up the machine in '97, was always that symptom with an almost immediate crash.

After being left in the loft for almost 20 years the machine would boot normally but then after about a minute start misbehaving. (Something had partially self-healed over that time.) Initially I could write a short BASIC parogram and run it, but it would start getting errors whilst reading the code and sometimes crash, but a Break and OLD would recover it, meaning that it was most likely to be read errors from memory rather than the memory itself failing.

After leaving the machine on for an hour to see what happened the machine got into the effectively the state it is now where it's immediately having problems, as seen in the videos. i.e. the component has fully failed again, in a slightly different manner.
Thanks. Sorry to be pedantic/dense, but all four of the videos represent current faulty behaviour since swapping out IC42, yes? Phantom characters being "typed in", Mode 7 characters substituted sometimes, failure to boot to a prompt when warm, and screen corruption in Mode 7 and Mode 0 (which can change in nature every reset).
stephen_usher wrote:
Sat Mar 30, 2019 10:11 pm
That's interesting... I've been freeze spraying chips again. IC54 is sensitive now and the system won't start if it's too cold. That's interfacing the data lines with the SAA 5050.
That could be very interesting. A fault with IC54 could do all sorts of things in Mode 7, including substituting characters, and even generating characters of its own, possibly from reading the RAM Data Bus when it shouldn't be (i.e. when the RAM's being accessed by the CPU). I can't think how it could affect the contents of the RAM and cause a crash, though, unless the chip is seriously borked.
Last edited by Kazzie on Sat Mar 30, 2019 10:52 pm, edited 1 time in total.
BBC Model B 32K issue 7, Sidewise ROM board with 16K RAM
Archimedes 420/1 upgraded to 4MB RAM, ZIDEFS with 512MB CF card
RiscPC 600 under repair
Acorn System 1 home-made replica
Kazzie
Posts: 1793
Joined: Sun Oct 15, 2017 8:10 pm
Location: North Wales
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by Kazzie »

stephen_usher wrote:
Sat Mar 30, 2019 10:38 pm
Pin 4 on IC54 is behaving strangely in that putting the probe on it will not allow the CPU to run for long at all. None of the other data lines are sensitive. That's D1, controlling the flip-flop used by the SAA 5050... Which could be the potential cause for the strange characters.
Recalling that the character substitution we've seen is a "stuck-low" fault on D1, yes, It could be. But remember that putting the probe there means it's also directly connected to RAM chip IC56, IC49 (replaced) and the video chip (IC53). But if they're not sensitive to the freeze spray, whereas IC54 is, I'd say that it's worth replacing.
BBC Model B 32K issue 7, Sidewise ROM board with 16K RAM
Archimedes 420/1 upgraded to 4MB RAM, ZIDEFS with 512MB CF card
RiscPC 600 under repair
Acorn System 1 home-made replica
stephen_usher
Posts: 143
Joined: Sun Apr 10, 2016 2:47 pm
Location: Oxford, UK.
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by stephen_usher »

Kazzie wrote:
Sat Mar 30, 2019 10:46 pm
Thanks. Sorry to be pedantic/dense, but all four of the videos represent current faulty behaviour since swapping out IC42, yes? Phantom characters being "typed in", Mode 7 characters substituted sometimes, failure to boot to a prompt when warm, and screen corruption in Mode 7 and Mode 0 (which can change in nature every reset).
The last two video were taken today. (I swapped IC49 last night.)

The first video ("Coolish") is what it's currently doing consistently, except the first "phantom" typed character is almost always 'a' followed by the dashes.

As for the corruption of the data on the bus???

How is the CPU activity perturbing the display of the characters generated by the internal character generator on the SAA5050????
Kazzie
Posts: 1793
Joined: Sun Oct 15, 2017 8:10 pm
Location: North Wales
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by Kazzie »

stephen_usher wrote:
Sat Mar 30, 2019 11:01 pm
Kazzie wrote:
Sat Mar 30, 2019 10:46 pm
Thanks. Sorry to be pedantic/dense, but all four of the videos represent current faulty behaviour since swapping out IC42, yes? Phantom characters being "typed in", Mode 7 characters substituted sometimes, failure to boot to a prompt when warm, and screen corruption in Mode 7 and Mode 0 (which can change in nature every reset).
The last two video were taken today. (I swapped IC49 last night.)

The first video ("Coolish") is what it's currently doing consistently, except the first "phantom" typed character is almost always 'a' followed by the dashes.

As for the corruption of the data on the bus???

How is the CPU activity perturbing the display of the characters generated by the internal character generator on the SAA5050????
The SAA5050 (via IC54) and/or the Video Processor (IC53) should only be reading the RAM when the CPU isn't using it, (i.e. when the CPU is executing an instruction). If, for whatever reason, it was reading at other times, it would get extra characters for the screen from whatever the CPU was reading or writing at the time.

As for corrupting the contents of the RAM (via the RAM data bus), Pin 4 of IC54 should be a data input to a flip-flop. If, for whatever reason, the chip has been damaged such that pin 4 was now connected to ground (possibly through a low resistance), that would conflict with bit 1 of whatever the CPU was trying to read or write. (Obviously in your case this doesn't happen constantly, as the machine is able to boot and function to some extent.)

I'd still suspect IC54 at this point (unless RAM chip IC56 is found to be temperature-sensitive too).

If you want to test whether IC54 is to blame without waiting for a replacement, one possible method would be to fit keyboard links so that the machine boots into a non-teletext mode, then compare how the system behaves before and after you desolder and remove IC54. (That's assuming that you can remove IC54 without resorting to cutting its pins off...)
BBC Model B 32K issue 7, Sidewise ROM board with 16K RAM
Archimedes 420/1 upgraded to 4MB RAM, ZIDEFS with 512MB CF card
RiscPC 600 under repair
Acorn System 1 home-made replica
stephen_usher
Posts: 143
Joined: Sun Apr 10, 2016 2:47 pm
Location: Oxford, UK.
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by stephen_usher »

I've a Beeb in the loft which is in poor condition and was given to be "broken" I can use as a donor machine for parts. I'll take a look to see if this chip is socketed, if so I can borrow it from there. This part of the circuitry didn't change between the B and B+.
Kazzie
Posts: 1793
Joined: Sun Oct 15, 2017 8:10 pm
Location: North Wales
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by Kazzie »

stephen_usher wrote:
Sun Mar 31, 2019 8:47 am
I've a Beeb in the loft which is in poor condition and was given to be "broken" I can use as a donor machine for parts. I'll take a look to see if this chip is socketed, if so I can borrow it from there. This part of the circuitry didn't change between the B and B+.
My model B has that ls273 soldered in place, but you might be lucky. Here's hoping. ;)
BBC Model B 32K issue 7, Sidewise ROM board with 16K RAM
Archimedes 420/1 upgraded to 4MB RAM, ZIDEFS with 512MB CF card
RiscPC 600 under repair
Acorn System 1 home-made replica
stephen_usher
Posts: 143
Joined: Sun Apr 10, 2016 2:47 pm
Location: Oxford, UK.
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by stephen_usher »

Darn! I'll take a look anyway. :-)
stephen_usher
Posts: 143
Joined: Sun Apr 10, 2016 2:47 pm
Location: Oxford, UK.
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by stephen_usher »

OK, I've now ordered a replacement 74LS273 (plus a set of 4164 DRAM chips and sockets). I might as well replace the RAM just in case.
stephen_usher
Posts: 143
Joined: Sun Apr 10, 2016 2:47 pm
Location: Oxford, UK.
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by stephen_usher »

OK. I can boot into Mode 5 (jumper 7 on the keyboard) and get phantom key presses (mostly right-arrow) but I can type and run BASIC. All the other modes cause crashing with display corruption.

At one point I had the bit 5 data line being high most of the time and being pulled low by bits, it seemed.

Anyway, whilst running in Mode 5 the RAM, video processor and ICs 13 and 14 are getting hot. The CRTC is getting warm.
Last edited by stephen_usher on Sun Mar 31, 2019 3:20 pm, edited 1 time in total.
stephen_usher
Posts: 143
Joined: Sun Apr 10, 2016 2:47 pm
Location: Oxford, UK.
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by stephen_usher »

Seeing as I have a replacement on the way I decided to be radical and remove the 74LS273 and see what happened.

Of course, the machine doesn't boot, but the data bus did look cleaner. Something else changed also, IC56's data input/output looked wrong. All the other chips had nice square pulses, e.g. IC55
IMG_1397.JPG
However, IC56 looked like this, and was almost random:
IMG_1398.JPG
Last edited by stephen_usher on Sun Mar 31, 2019 3:44 pm, edited 1 time in total.
Kazzie
Posts: 1793
Joined: Sun Oct 15, 2017 8:10 pm
Location: North Wales
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by Kazzie »

That's interesting that phantom keypresses are showing up outside Mode 7. The 74ls273 (IC54) couldn't be causing that in Mode 5. But given that we know something's wring with Bit 1 in that area, it's probably best to sort that issue out before looking elsewhere.

I think there's a cloud of suspicion looming around IC56 now. With the IC54 lifted out, and IC49 freshly replaced, the only other chip that's connected to its data line is the video processor (IC53). It's a good thing you've got some RAM on order. :)

Did you happen to scope out IC56's data output before removing the ls273? I can't make out the scale on your scope, but those peaks look they're scraping toward 2.0V, which would just about be within tolerance for a high TTL-level input. When the ls273's load was on IC56 as well, the poor RAM chip's voltage output would presumably have been slightly lower again.

(When the CPU is connected to the RAM bus, the Video Processor and ls273's inputs should have been high impedance, and the RAM chips would only have been feeding the CPU's buffer chip (IC49). That could explain how in the past, your Beeb would run programs correctly, but with corruption on-screen.)

As for IC13 and IC14, they're only buffering values for the tube interface. I can't think why they'd be running hot all on their own (unless some of the tube connector pins are bent and shorting).
BBC Model B 32K issue 7, Sidewise ROM board with 16K RAM
Archimedes 420/1 upgraded to 4MB RAM, ZIDEFS with 512MB CF card
RiscPC 600 under repair
Acorn System 1 home-made replica
stephen_usher
Posts: 143
Joined: Sun Apr 10, 2016 2:47 pm
Location: Oxford, UK.
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by stephen_usher »

I had scoped RD1 on IC56 before removing it and the data line didn't look too bad, which is weird. Maybe the flip-flop was helping in some way?

I think the value is just about getting up to 2.5V, sometimes.

I'll check the TUBE connector.
stephen_usher
Posts: 143
Joined: Sun Apr 10, 2016 2:47 pm
Location: Oxford, UK.
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by stephen_usher »

The tube connector is fine.

Having had confirmation that the memory is on its way I snipped out IC56 and put the scope on RD1, which now only has the video processor on it plus the 64K RAM expansion board.

There are still seemingly random voltage glitches, but only on this data line.

Could this be caused by a duff video processor?
Kazzie
Posts: 1793
Joined: Sun Oct 15, 2017 8:10 pm
Location: North Wales
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by Kazzie »

It could be caused by the video processor, but hopefully not, as they're a custom chip. Plus, because the video chip generates the CPU clock, you can't really remove it to see if the fault goes away.

Another explanation would be that with the RAM disconnected from D1, only the CPU buffer (IC49) is driving the D1 line, and that's only while the CPU is writing to memory. When the CPU is reading, or when the video processor is reading, without a RAM chip to generate a signal (and no pull-up/down resistors) you'd expect to see noise on the line.

In other words, those glitches may well disappear when the new RAM chip is fitted.
BBC Model B 32K issue 7, Sidewise ROM board with 16K RAM
Archimedes 420/1 upgraded to 4MB RAM, ZIDEFS with 512MB CF card
RiscPC 600 under repair
Acorn System 1 home-made replica
stephen_usher
Posts: 143
Joined: Sun Apr 10, 2016 2:47 pm
Location: Oxford, UK.
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by stephen_usher »

Success!

After replacing the 74LS chip and the one DRAM the machine is alive again.

Phew!

Now to do a RAM test somehow.
IMG_1402.JPG
IMG_1399.JPG
Last edited by stephen_usher on Tue Apr 02, 2019 7:45 pm, edited 1 time in total.
User avatar
Sniffer
Posts: 179
Joined: Tue Aug 11, 2015 10:30 pm
Location: Colwyn Bay, North Wales
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by Sniffer »

Good to see you were successful.

Does anybody know which ram ICs map to which addresses?
Kazzie
Posts: 1793
Joined: Sun Oct 15, 2017 8:10 pm
Location: North Wales
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by Kazzie »

Each RAM chip corresponds to one bit in every one of either the lower or upper 16k of addresses:

Code: Select all

   /CAS0 /CAS1
D0= IC61  IC53
D1= IC62  IC54
D2= IC63  IC55
D3= IC64  IC56
D4= IC65  IC57
D5= IC66  IC58
D6= IC67  IC59
D7= IC68  IC60

With S25 north, /CAS0 provides &0000-&3FFF, with /CAS1 providing &4000-&7FFF.


Scratch that, I forgot we're looking at a B+ :oops:
Last edited by Kazzie on Tue Apr 09, 2019 7:10 am, edited 2 times in total.
BBC Model B 32K issue 7, Sidewise ROM board with 16K RAM
Archimedes 420/1 upgraded to 4MB RAM, ZIDEFS with 512MB CF card
RiscPC 600 under repair
Acorn System 1 home-made replica
User avatar
MartinB
Posts: 5364
Joined: Mon Mar 31, 2008 10:04 pm
Location: Obscurity
Contact:

Re: BBC B+ 128K: Fault after warming up.

Post by MartinB »

Kazzie wrote:Each RAM chip corresponds to one bit in every one of either the lower or upper 16k of addresses:

Code: Select all

   /CAS0 /CAS1
D0= IC61  IC53
D1= IC62  IC54
D2= IC63  IC55
D3= IC64  IC56
D4= IC65  IC57
D5= IC66  IC58
D6= IC67  IC59
D7= IC68  IC60
That's the Model B answer, have we switched from a B+?
Post Reply

Return to “8-bit acorn hardware”