Video not displaying properly

for bbc micro/electron hardware, peripherals & programming issues (NOT emulators!)
User avatar
1024MAK
Posts: 9610
Joined: Mon Apr 18, 2011 4:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Video not displaying properly

Post by 1024MAK » Tue Feb 04, 2020 11:16 am

Right, that’s it, otherwise my brain will explode, or maybe implode :shock:

Mark

User avatar
crispy
Posts: 33
Joined: Tue Jan 28, 2020 6:43 pm
Contact:

Re: Video not displaying properly

Post by crispy » Tue Feb 04, 2020 12:26 pm

Thanks Mark - very thorough, and appreciated.

Am I to assume then that if the error every 128 addresses is an A7 problem, then an error every 2 addresses would be an A1 problem and an error every 8 addresses an A3 problem?

The IC identification though comes from the 0 bit being faulty.

User avatar
crispy
Posts: 33
Joined: Tue Jan 28, 2020 6:43 pm
Contact:

Re: Video not displaying properly

Post by crispy » Tue Feb 04, 2020 12:28 pm

1024MAK wrote:
Tue Feb 04, 2020 8:57 am
...the other thing that you can try is Tricky’s test ROM (follow this link).
That would be a good idea, but I don't have the ability to burn a ROM.

I do however have a 8645 in hand ready to test this afternoon - hopefully eliminating one of the variables or issues

User avatar
crispy
Posts: 33
Joined: Tue Jan 28, 2020 6:43 pm
Contact:

Re: Video not displaying properly

Post by crispy » Tue Feb 04, 2020 1:27 pm

Okay, it is going to be one of those days, I can tell...

Got the 6845, booted, and the screen was worse than before, with seemingly no sync. Checked the chip: MC6845P rather than the HD6845SP that it was replacing. Still trying to ascertain whether this should make a difference...but from what I could see through the noise, the text that was being rendered was the same as before; no ROM list on boot, nothing rendered until halfway down the page in MODE 7.

So I guess I am going to have to wait until the replacement RAM gets here.

User avatar
1024MAK
Posts: 9610
Joined: Mon Apr 18, 2011 4:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Video not displaying properly

Post by 1024MAK » Tue Feb 04, 2020 3:04 pm

Alas, the different manufacturers of the 6845 did not produce identical chips...

If you want a known good compatible 6845, obtain one from Mark at RetroClinic (a different Mark).

Mark

Kazzie
Posts: 1072
Joined: Sun Oct 15, 2017 7:10 pm
Location: North Wales
Contact:

Re: Video not displaying properly

Post by Kazzie » Tue Feb 04, 2020 3:24 pm

crispy wrote:
Tue Feb 04, 2020 12:26 pm
Thanks Mark - very thorough, and appreciated.

Am I to assume then that if the error every 128 addresses is an A7 problem, then an error every 2 addresses would be an A1 problem and an error every 8 addresses an A3 problem?

The IC identification though comes from the 0 bit being faulty.
A minor slip-up on my part. An error every 128 addresses would be related to the seventh address line, but as they're typically numbered from A0 upward, that would be A6. But while that interval tells us which addresses are having problems, it doesn't tell us which chip is having trouble with those addresses.

Every 4816 RAM chip has seven address lines as inputs, and one data line as an output. (Technically there's one data input too, but those two data lines are usually connected together.) As 1024MAK said, the arrangement of these address lines is arbitrary, but all eight RAM chips in a bank (such as /CAS0) will have their address lines connected together. This means that it's impossible to use the address lines to identify which chip is faulty (except for identifying the bank it's in).

The way we identify the faulty chip is by looking at what incorrect value we're reading back from the faulty addresses, and comparing it to what it should have been. Your RAM-testing program was writing zeroes to every address, and reporting back not only the addresses it failed to read a zero from, but also the value it did read back. Each and every error line you typed out had a "1" at the beginning, followed by the address.

Looking at the binary values for what we read, and what we expected, we have:

0 = 0b00000000
1 = 0b00000001

Only one bit is different, the least significant bit. Data lines are also counted from zero, so this is D0. Knowing that the faulty chip provides bit 0 to bank /CAS0, we can then look up which IC is responsible.

If there's anything else that's still unclear, let me know.
BBC Model B 32k issue 7, Sidewise ROM board with 16K RAM
Archimedes 420/1 upgraded to 4MB RAM, ZIDEFS with 512MB CF card
Acorn System 1 home-made replica

User avatar
crispy
Posts: 33
Joined: Tue Jan 28, 2020 6:43 pm
Contact:

Re: Video not displaying properly

Post by crispy » Wed Feb 05, 2020 8:42 am

Perfect, thanks - that is what I thought, but did not want to make too many assumptions; its good to understand 'why' as well as 'what'!

Have replacement chips inbound, so on a pause until then.

User avatar
crispy
Posts: 33
Joined: Tue Jan 28, 2020 6:43 pm
Contact:

Re: Video not displaying properly

Post by crispy » Wed Feb 12, 2020 11:06 am

Update.

After finally getting some time today, I replaced the suspect DRAM IC61 - good news is that the memory test that was failing before is now passing.

I also re-ran the beep test for the screen data and no longer get the beep, so CPU is able to access the RAM.

However, I still have the same problem with the display corruption and refusing to show the boot ROM information etc. So, while I have fixed a couple of problems, which is great, it appears that the fundamental issue that I have remains.

I also replaced the 6845, with a new one - no difference whatsoever, so probably not a problem here.

Would I be right in assuming that this pretty much leaves us looking at whatever ICs sit between the 6845 and the DRAM?

Kazzie
Posts: 1072
Joined: Sun Oct 15, 2017 7:10 pm
Location: North Wales
Contact:

Re: Video not displaying properly

Post by Kazzie » Wed Feb 12, 2020 12:22 pm

crispy wrote:
Wed Feb 12, 2020 11:06 am
Update.

After finally getting some time today, I replaced the suspect DRAM IC61 - good news is that the memory test that was failing before is now passing.

I also re-ran the beep test for the screen data and no longer get the beep, so CPU is able to access the RAM.

However, I still have the same problem with the display corruption and refusing to show the boot ROM information etc. So, while I have fixed a couple of problems, which is great, it appears that the fundamental issue that I have remains.

I also replaced the 6845, with a new one - no difference whatsoever, so probably not a problem here.

Would I be right in assuming that this pretty much leaves us looking at whatever ICs sit between the 6845 and the DRAM?
Hi there, good to see you're making some progress. There are a few details I'd like to clarify, though:

:arrow: When you say you "no longer get the beep", is that the continuous beep? The machine should make a "boo-bip" sound on power on when the speaker is connected.

:arrow: What S25 positions have you tried since replacing IC61; do they all behave in the same way? (When S25 is North, the display RAM is in a different bank to the zero-page RAM the CPU needs to boot the OS correctly.)

:arrow: Does the issue still show up in both MODE 7 and MODEs 0-6? If there is a fault between the 6845 and the RAM, there's no chip that is used for both teletext and non-teletext MODEs, so any such failure would be a fairly exotic one.

Looking back at your earlier screenshots, you had a correct MODE 7 screen when booting with S25 South, which corrupted when you switched to MODE 6, and was still corrupted when you tried to switch back to MODE 7. If that still happens, that could help indicate where the fault is.

P.S. Do you have any diagnositc tools such as a logic probe or an oscilloscope? (If not, we may still be able to make some useful measurements with a DC voltmeter, if needed.)
BBC Model B 32k issue 7, Sidewise ROM board with 16K RAM
Archimedes 420/1 upgraded to 4MB RAM, ZIDEFS with 512MB CF card
Acorn System 1 home-made replica

User avatar
crispy
Posts: 33
Joined: Tue Jan 28, 2020 6:43 pm
Contact:

Re: Video not displaying properly

Post by crispy » Wed Feb 12, 2020 1:25 pm

Kazzie wrote:
Wed Feb 12, 2020 12:22 pm
:arrow: When you say you "no longer get the beep", is that the continuous beep? The machine should make a "boo-bip" sound on power on when the speaker is connected.
I get the normal power on beep. I was referring to the vdu7 test listed earlier in the thread.
:arrow: What S25 positions have you tried since replacing IC61; do they all behave in the same way? (When S25 is North, the display RAM is in a different bank to the zero-page RAM the CPU needs to boot the OS correctly.)
Have tried both positions. In the south position the machine boots fine and displays the ROMs properly. In the North, the top of the page does not display unless I scroll down. In 16k mode, mode7 appears to work correctly, but not in other modes. I still get residual corruption moving back to mode 7 after running memory tests.
:arrow: Does the issue still show up in both MODE 7 and MODEs 0-6? If there is a fault between the 6845 and the RAM, there's no chip that is used for both teletext and non-teletext MODEs, so any such failure would be a fairly exotic one.

Looking back at your earlier screenshots, you had a correct MODE 7 screen when booting with S25 South, which corrupted when you switched to MODE 6, and was still corrupted when you tried to switch back to MODE 7. If that still happens, that could help indicate where the fault is.

P.S. Do you have any diagnositc tools such as a logic probe or an oscilloscope? (If not, we may still be able to make some useful measurements with a DC voltmeter, if needed.)
Mode7 is always having the top of screen not showing issue. All other modes have the repeating and shifting pattern. I have a logic probe and a multimeter.

Kazzie
Posts: 1072
Joined: Sun Oct 15, 2017 7:10 pm
Location: North Wales
Contact:

Re: Video not displaying properly

Post by Kazzie » Wed Feb 12, 2020 2:02 pm

crispy wrote:
Wed Feb 12, 2020 1:25 pm
Have tried both positions. In the south position the machine boots fine and displays the ROMs properly. In the North, the top of the page does not display unless I scroll down. In 16k mode, mode7 appears to work correctly, but not in other modes. I still get residual corruption moving back to mode 7 after running memory tests.
Thanks. A few more clarifications, if you could, please:

:arrow: "In 16K mode" , does that apply to both S25 South and S25 Removed?

:arrow: The "residual corruption moving back to MODE 7" , is that in all three configurations of S25?
BBC Model B 32k issue 7, Sidewise ROM board with 16K RAM
Archimedes 420/1 upgraded to 4MB RAM, ZIDEFS with 512MB CF card
Acorn System 1 home-made replica

User avatar
crispy
Posts: 33
Joined: Tue Jan 28, 2020 6:43 pm
Contact:

Re: Video not displaying properly

Post by crispy » Wed Feb 12, 2020 3:11 pm

Yes in both jumper south and jumper removed, mode7 behaves correctly. With the jumper in North, I get the missing top half of the screen until I scroll. To accommodate 16k I was running the tests in mode 6, with &6000 as the lower bound.

The residual corruption only appears to be present when the jumper is in the north position. It doesn’t appear to be consistent. Just now I ran tests and on returning to mode 7 it shows the same issue as on a reboot. Earlier it was showing white boxes; after running the memory test in Mode 0.

Running the tests in mode 0 again (North only) then switching back to mode 7, the top half of the screen is white boxes once more.
image.jpg
(Don’t ask me why this photo is upside down!!!!

Kazzie
Posts: 1072
Joined: Sun Oct 15, 2017 7:10 pm
Location: North Wales
Contact:

Re: Video not displaying properly

Post by Kazzie » Wed Feb 12, 2020 4:16 pm

crispy wrote:
Wed Feb 12, 2020 3:11 pm
Yes in both jumper south and jumper removed, mode7 behaves correctly. With the jumper in North, I get the missing top half of the screen until I scroll. To accommodate 16k I was running the tests in mode 6, with &6000 as the lower bound.

The residual corruption only appears to be present when the jumper is in the north position. It doesn’t appear to be consistent. Just now I ran tests and on returning to mode 7 it shows the same issue as on a reboot. Earlier it was showing white boxes; after running the memory test in Mode 0.
Okay, let me sort that out in my mind (and correct me if I've misunderstood anything):
  • Mode 7 works with S25 south or removed, which are the two configurations that only use one bank of RAM. When using two banks of RAM (S25 North), a portion of the display will be blank at boot. This issue resolves itself as you write lines of code and scroll the screen.
  • Modes 0-6 have a different error, involving portions of the display being repeated in various locations.
  • On changing from one of modes 0-6 back to mode 7, the display appears correct if using one bank of RAM, but a portion of the screen is corrupted if using two banks.

On that evidence, I suspect IC39 (74ls283) or a related chip to be faulty.


IC39 is part of the hardware scrolling circuity. You'll find it below the 6845 (IC2) on this extract of the circuit diagram:
BBCB-IC39.png
IC39 and surrounding area
Hardware scrolling is only applicable in non-teletext modes, and is controlled by two signals, C1 and C0 (see far left). These values are used (along with MA12 from the 6845) to calculate a memory offset for the start of the scrolled display memory. This four-bit value (B1-4) is added to four MA signals from the 6845 (as A1-4), by the 74ls283 4-bit binary adder (IC39). Three of these bits are buffered by IC9 to be used as column values when accessing display RAM in modes 0-6 (as AA0-2), while the fourth bit is sent to IC28.

A fault in the adding up by IC39, or a mistake in its inputs, could result in some duplication of AA0-2's values, manifesting as some areas of the Mode 0-6 screen being duplicated.

That doesn't explain how Mode 7 is affected: IC9 isn't used by Mode 7. To explain this, it's time to explore that signal going to IC28:
BBCB-IC28.png
IC28 and S25
IC 28 (74ls51) is a pair of three-logic-gate circuits on a chip, the half we're interested in is in the dashed box in the upper right. Note the input arrow coming from IC39, and the output from pin 6, which goes to the north pin of S25 in the bottom left.

IC 28 is part of the bank select circuitry. It needs to be involved with the hardware scrolling circuit because some display modes use memory from both banks, and so the start address of a scrolled display could be in either bank. It has no influence on the system if you're operating in a 16K configuration (S25 south, grounded to 0V, or removed, when the middle pin floats up toward a high logic level), it only makes a difference when you have two banks enabled, for the full 32K of RAM.

A malfunctioning IC39 could cause IC28's output to alternate between high and low, even when it's not meant to do so. This would have the effect of showing the video generation circuitry the contents of bank /CAS0 for part of the display, and the contents of /CAS1 for other parts. This could affect all display modes, including mode 7, if S25 is North. The fact that scrolling the display modifies this faulty behaviour (by changing the values of C0 and C1) is another finger pointing in IC39's direction.

:arrow: So the end result of this, assuming that I've not mucked up along the way, is that:
  • Either IC39 is faulty, and needs to be replaced, or
  • one of the chips generating inputs for IC39 is faulty, and needs to be replaced. That would be one of IC27, IC36, or IC40. I'll discount IC2 as you've already replaced it.
For the next step, I can suggest one of two options:
  • You could do further testing with your logic probe and multimeter, to try to identify whether the logic gates on ICs 27, 36 and 40 are operating correctly according to their truth tables. They're all NAND gates, so their output should be low only when all inputs are high. This may be difficult to do identify with a logic probe if the inputs are strobing, but it may be possible to use a DC voltmeter reading on each pin to estimate the duty cycle for each pin (proportion of time at logic high vs logic low), and identify a faulty chip in that way.
  • Or you could remove IC39, replace it, and see if that solves the problem or not. If it does, great! If not, you're down to three suspect chips, rather that four.
Do either of these options appeal to you?
BBC Model B 32k issue 7, Sidewise ROM board with 16K RAM
Archimedes 420/1 upgraded to 4MB RAM, ZIDEFS with 512MB CF card
Acorn System 1 home-made replica

User avatar
crispy
Posts: 33
Joined: Tue Jan 28, 2020 6:43 pm
Contact:

Re: Video not displaying properly

Post by crispy » Wed Feb 12, 2020 6:54 pm

Going to try some tests with the logic probe tomorrow I think. Probably power off the machine and run the chip independently to check the nand gates. Off the back of that I should be in a better position to understand which path to take on replacement.

Kazzie
Posts: 1072
Joined: Sun Oct 15, 2017 7:10 pm
Location: North Wales
Contact:

Re: Video not displaying properly

Post by Kazzie » Thu Feb 13, 2020 9:40 am

crispy wrote:
Wed Feb 12, 2020 6:54 pm
Going to try some tests with the logic probe tomorrow I think. Probably power off the machine and run the chip independently to check the nand gates. Off the back of that I should be in a better position to understand which path to take on replacement.
It might be a little tricky to only operate one chip at a time, as its 5V supply pin will be connected to all the other chips on the board (unless you desolder it to isolate it).

It's occurred to me that it should be fairly easy to check the health of the C0 and C1 signals, as they should be stable whenever you're not in the process of scrolling the screen. If either or these is strobing instead, or at an intermediate voltage level, IC32 (74ls259) is responsible for generating them, by latching signals sent from IC3 (6522 VIA).

And if you remove the 6845, you'll be able to replace the MA8-12 signals that it generates for the logic gates we want to check, even if the rest of the board is powered.
BBC Model B 32k issue 7, Sidewise ROM board with 16K RAM
Archimedes 420/1 upgraded to 4MB RAM, ZIDEFS with 512MB CF card
Acorn System 1 home-made replica

User avatar
crispy
Posts: 33
Joined: Tue Jan 28, 2020 6:43 pm
Contact:

Re: Video not displaying properly

Post by crispy » Fri Feb 14, 2020 9:31 am

Kazzie wrote:
Thu Feb 13, 2020 9:40 am
It's occurred to me that it should be fairly easy to check the health of the C0 and C1 signals, as they should be stable whenever you're not in the process of scrolling the screen. If either or these is strobing instead, or at an intermediate voltage level, IC32 (74ls259) is responsible for generating them, by latching signals sent from IC3 (6522 VIA).
For the sake of completeness, post boot, here is what I am seeing on IC36:
1 H 4.2 C0*
2 ~ 1.9 MA12*
3 L 0.17 MA12
4 L 0.18 C0*
5 H 3.9 C1
6 H 4.3
7 L 0 (GND)
8 H 4.18
9 H 3.91 C1
10 H 3.68
11 L 0.18 MA12
12 ~ 2.18
13 H 4.95 MA12
14 H 4.95 (VCC)

The first thing that jumps out to me here is that CO is meant to be provided to pins 1 and 4 (from the diagram above) - and yet somehow I have these two pins at different values. My assumption here is that this should not be the case in normal operation.

C1 is behaving correctly, with exactly the same voltage, so looks good.

Something strange is happening with the MA12 feed from IC2, which is meant to feed the same values to 3, 11, 2, 13 (is pin 13 shorted to VCC? This is s suspect voltage reading given what I am seeing. Checking with the multimeter, this is indeed the case, pin 13 is shorted to VCC.

Checking the output of IC2 on MA12 (Pin 16) the output is fixed L (0.18), so this is right. Testing with the multimeter, pins 3 and 11 are directly connected with IC2 pin 16, pins 2 and 13 are not.
Last edited by crispy on Fri Feb 14, 2020 10:30 am, edited 1 time in total.

User avatar
crispy
Posts: 33
Joined: Tue Jan 28, 2020 6:43 pm
Contact:

Re: Video not displaying properly

Post by crispy » Fri Feb 14, 2020 10:00 am

Going back to IC32, I can confirm that on boot, pin 9 (C0) is L and pin 10 (C1) is H.

Since I already know that C1 seems to be behaving correctly, Looking at C0, somewhere between the output of IC32 and IC36, Pin 1 is being pulled high.

(Assuming that I am getting the pin numbers right here...)

With the multimeter, checking the connections, I get a direct connection for CO from IC32 and IC36 pin 4, but not for pin 1, so it might be that there is a break in the circuit somewhere here. (although if that is the case, what is putting voltage on pin 1 and pulling it high?)

A bit more poking around, and I can see that the only thing I can find pin 1 connected to is IC37 pin 2...yet to find that on the circuit diagram...

User avatar
crispy
Posts: 33
Joined: Tue Jan 28, 2020 6:43 pm
Contact:

Re: Video not displaying properly

Post by crispy » Fri Feb 14, 2020 10:33 am

crispy wrote:
Fri Feb 14, 2020 9:31 am
Checking with the multimeter, this is indeed the case, pin 13 is shorted to VCC.
Lifting the board out and inspecting, these two pins are connected (13 and 14) which does not appear to map to the circuit diagram listed above.

Questioning my sanity here!

Kazzie
Posts: 1072
Joined: Sun Oct 15, 2017 7:10 pm
Location: North Wales
Contact:

Re: Video not displaying properly

Post by Kazzie » Fri Feb 14, 2020 12:05 pm

A few details that may help your sanity: C0 is connected to pin 4 of IC36, and pin 1 of IC40. Similarly, MA12 is connected to pins 3 and 11 of IC36, pin 2 of IC40, and pin 13 of IC27. ;)

Looking at the two relevant gates of IC 36 (pins 3,4,5,6 and pins 9,10,11,8), the outputs of these two NAND gates are fully consistent with their inputs. And the input values here for C0 and C1 match the values generated on IC32.

Pin 10 of IC36 is connected to (output) pin 3 of IC40, and the inputs of that gate are connected to C0 and MA12. From your existing readings, it looks like that gate's output is correct.

:arrow: As all the relevant bits of IC36 (and IC40) seem to pass the sniff test, could you give IC27 (pins 11,12,13) a quick check, followed by all of IC39?
BBC Model B 32k issue 7, Sidewise ROM board with 16K RAM
Archimedes 420/1 upgraded to 4MB RAM, ZIDEFS with 512MB CF card
Acorn System 1 home-made replica

User avatar
crispy
Posts: 33
Joined: Tue Jan 28, 2020 6:43 pm
Contact:

Re: Video not displaying properly

Post by crispy » Fri Feb 14, 2020 12:51 pm

Thanks for restoring my sanity, I did not see the IC40 on the diagram.

IC 39
1 ~
2 H
3 ~
4 H
5 ~
6 H
7 H
8 L
9 H
10 ~
11 H
12 H
13 ~
14 ~
15 H
16 H

and forgive my ignorance, but I can't find IC27! Did you perhaps mean IC72?

User avatar
marcusjambler
Posts: 724
Joined: Mon May 22, 2017 11:20 am
Location: Bradford
Contact:

Re: Video not displaying properly

Post by marcusjambler » Fri Feb 14, 2020 1:32 pm

IC27 is situated on the left hand side centre of the board

For assistance have a look at :
http://8bs.com/inbbc.htm

Marcus

Kazzie
Posts: 1072
Joined: Sun Oct 15, 2017 7:10 pm
Location: North Wales
Contact:

Re: Video not displaying properly

Post by Kazzie » Fri Feb 14, 2020 1:58 pm

IC27's three other gates are used in the FDC, Econet, and User Port circuity, so it's nowhere near the rest of the video circuit. You'll find it on the far West (left) edge of the mainboard, near the Econet circuit. Based on your other readings, you should find pins 11 and 12 high, with pin 13 low.

Looking at the readings of IC39 (the 74ls283 4-bit adder) things aren't adding up. Following the chip's pinout:
Image
Input A (from the 6845) is 1~~~.
Input B (from the chips you've been probing) is 1111.
The carry input (pin 7) is 1.
The sum (sigma) is ~~~1, and the carry output is 1.

That gives us a sum of:

Code: Select all

Cin:     1
A:    1~~~
B:    1111
  + ------
Sum: 1~~~1
    ------
     1111
(Those are carried ones at the bottom)

Working from right to left, just like long addition...

Bit 1 is the sum of 1, ~, and 1. In binary addition, the internal carry value C1 will definitely be 1 (because we've got two 1s as inputs), and so the S1 sum output should equal the value of A1, which is strobing. But your measurement of pin 4 indicates that it is fixed high.

Bit 2 is the sum of ~, 1, and the internal carry C1, which is 1. Once again, two 1 inputs means the next carry will also be one, and the S2 sum output will equal the value of A2, which is strobing. S2 is indeed strobing, which is correct.

Bit 3 is working identically to bit 2.

Bit 4 is the sum of 1, 1, and the internal carry C3, which is also 1. The binary addition should be a sum of 1 and a carry of 1. C4 is indeed 1, but S4 is strobing.

:arrow: Two outputs of IC39 are incorrect for the measured outputs, S1 and S4. S4 (on pin 10) is used by the bank select circuitry when more than one bank is enabled (S25 North). The fact that it's strobing when it shouldn't be means that the machine is reading display data from the wrong memory bank for part of the MODE 7 display.

My conclusion: IC39 is faulty, and needs 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
Acorn System 1 home-made replica

User avatar
crispy
Posts: 33
Joined: Tue Jan 28, 2020 6:43 pm
Contact:

Re: Video not displaying properly

Post by crispy » Fri Feb 14, 2020 2:11 pm

The readings on 11, 12 and 13 were indeed as predicted - and following your logic, I am now ordering a replacement IC39.

Many thanks for your help with this - and with finding IC27 all the way over there, hidden away (not even shown on the service manual diagram that I was searching)

Kazzie
Posts: 1072
Joined: Sun Oct 15, 2017 7:10 pm
Location: North Wales
Contact:

Re: Video not displaying properly

Post by Kazzie » Fri Feb 14, 2020 3:37 pm

You're very welcome. (Being away from my own Beeb at the moment, the page linked by Marcus is what I used to make sure I was sending you the right way.)

Let us know how you get on when the replacement part arrives. :)
BBC Model B 32k issue 7, Sidewise ROM board with 16K RAM
Archimedes 420/1 upgraded to 4MB RAM, ZIDEFS with 512MB CF card
Acorn System 1 home-made replica

Post Reply