Troubleshooting my latest BBC B

for bbc micro/electron hardware, peripherals & programming issues (NOT emulators!)
adrm
Posts: 41
Joined: Sun Sep 10, 2017 12:07 pm
Location: Norway

Troubleshooting my latest BBC B

Postby adrm » Sun Sep 10, 2017 8:44 pm

Hello all.

A quick background:
Having kicked myself mentally on a regular basis over the last 25(-ish) for selling off my beloved Beeb, I finally purchased another unit from RetroClinic. Not cheap, but seeing the work they put into these, I feel they are well worth it.

Playing Toccata and Fugue using BBC BEEP-BEEEP-BEEPs, literally brought tears of joy to my eyes :D

Anyway, I then bought another model B on eBay, mostly to have a spare available. This unit is in excellent condition, showing a nearly brand new casing colour and having a clean issue 7 main board.

However, there is an issue with it.
I have searched various troubleshooting guides, many describing similar, but not quite same, symptoms:

When I power on OR do a hard reset, the normal BBC logo appears, but the speaker sounds a continuous tone, Caps Lock is no and the keyboard does not respond to anything except Break.
A soft reset makes the keyboard operate normally, until the next hard reset.


I have replaced the usual caps in the PSU and used a DMM and a oscilloscope to check some basics (My skills in electronics are ... basic).
My finding, so far:
1 - The PSU delivers a stable +/- 5V
2 - The RAM CAS / RAS signals appear to be ok.
3 - The 6502 reset pulse seems to behave normally when break is pressed, i.e. low, then permanently high

What I have done:
A - I have removed all non-essential (socketed) ICs
B - I have swapped the 6522s
C - I have done basic RAM testing using Mark's (I believe) program posted elsewhere on these forums

So far, nothing to indicate a source of problem

Other notes:
a - The continuous tone from the speaker is a lot less "intense" today, but I don't have a clue a to why
b - On the first day, I did get a strange corruption after running a simple looping BASIC program. When Break was pressed, the Beeb responded with a prompt that looked something like "GOTO10>", no matter what I did. I suspected a RAM and/or heating problem, but the issue has not resurfaced later, no matte what I do.

After doing a soft reset, the machine seems to work fine, as far as I can test it, although I haven't been able to stress test it with a game or some such.

I will continue searching for things to test and I'll start reading the Service Manual tomorrow, but if one of you experts would be kind enough to drop a hint as to when I should start checking, I'd be very grateful.

Cheers,
Tore
-------
Tore

User avatar
DutchAcorn
Posts: 1584
Joined: Fri Mar 21, 2014 9:56 am
Location: Maarn, Netherlands

Re: Troubleshooting my latest BBC B

Postby DutchAcorn » Sun Sep 10, 2017 9:21 pm

Welcome to Stardot, Tore! :D

The symptoms you describe very much suggest an issue with the system VIA; the 6522 just north of the keyboard socket on the main board.

Since you already tried swapping the 6522 (also tried the known good from the working beeb?), you could have a look for issues with the IC socket and / or dry joints.

You can use the circuit diagram and a multi meter to check if all the pins are making contact.

Another thing to check would be if all the power connectors are making good contact.
Paul

adrm
Posts: 41
Joined: Sun Sep 10, 2017 12:07 pm
Location: Norway

Re: Troubleshooting my latest BBC B

Postby adrm » Sun Sep 10, 2017 9:36 pm

Thanks, Paul.

I'll check those tomorrow.

The connectors running from the PSU should be well connected, but I'll test input voltage to various components around the board, in case there's an issue on the MB itself.

I'll reflow the solder joints on the 6522 socket if they look suspicious, although I dread having to remove that IC again, due to the annoying Watford 1770 controller board blocking access to it. I'll also probe it with the scope, although I don't really know what to look for.
-------
Tore

adrm
Posts: 41
Joined: Sun Sep 10, 2017 12:07 pm
Location: Norway

Re: Troubleshooting my latest BBC B

Postby adrm » Mon Sep 11, 2017 11:00 am

Hi again.

Checking with a VMM uncovered no problems with voltage distribution.
I found no apparent breaks leading from the pins of IC3 (6522).

BUT .....
when I probed IC3 with my scope, I noticed that unlike, say, the 6502 and the other 6522 (IC59), the RESET pin doesn't go LOW when the BREAK key is pressed.

Questions:
1) Is this expected behaviour for IC3? (I'm thinking not)
2) If this is the culprit, is there another immediate step in troubleshooting, apart from checking the tracks from pin 34 (RST[i])?
-------
Tore

dominicbeesley
Posts: 396
Joined: Tue Apr 30, 2013 11:16 am

Re: Troubleshooting my latest BBC B

Postby dominicbeesley » Mon Sep 11, 2017 12:11 pm

It's meant to do that, the system via only resets on power up (simple diode/resistor/capacitor to +5V), this is how the beeb detects a warm vs cold start.

First try swapping system and user vias whilst cleaning the pins ( a gentle scrape off of oxide with a knife and a squirt of servisol or similar). Also worth doing the same for both ends of the keyboard connector. With the IC's it might be worth *very slightly* bending the pins out so they make better contact with the socket pins, half a mm at most. A ruler and very gentle pressure on all pins at once is a good idea.

I've had all manner of issues with the sockets used on the beeb, it seems the springiness of the pins fails after a while leading to bad contacts, on a couple of motherboards they required all sockets replacing, hopefully yours isnt that bad.

Don't throw away any chips at this stage, they are probably ok, most likely a mechanical/oxidisation issue.

johnkenyon
Posts: 105
Joined: Wed Jul 20, 2011 2:21 pm
Location: Coventry

Re: Troubleshooting my latest BBC B

Postby johnkenyon » Mon Sep 11, 2017 12:18 pm

adrm wrote:Hi again.

Checking with a VMM uncovered no problems with voltage distribution.
I found no apparent breaks leading from the pins of IC3 (6522).

BUT .....
when I probed IC3 with my scope, I noticed that unlike, say, the 6502 and the other 6522 (IC59), the RESET pin doesn't go LOW when the BREAK key is pressed.

Questions:
1) Is this expected behaviour for IC3? (I'm thinking not)
2) If this is the culprit, is there another immediate step in troubleshooting, apart from checking the tracks from pin 34 (RST[i])?


There are effectively two hardware resets on the BBC B and Master - "power on" and BREAK key.
The system VIA only gets the power on reset. Everything else gets both power on and Break key.
The reset code in the MOS reads the interrupt enable register of the system VIA.
If it's all zeros then the MOS goes on to handle the event as a "power on reset".

(Have a look at http://mdfs.net/Docs/Comp/BBC/OS1-20/D940 and scroll down to D9CD)

dominicbeesley
Posts: 396
Joined: Tue Apr 30, 2013 11:16 am

Re: Troubleshooting my latest BBC B

Postby dominicbeesley » Mon Sep 11, 2017 1:15 pm

Rough location marked on the bbc schematics...it took me a while to find a few weeks ago when I discovered the same thing!
Attachments
beeb-rsta.png

adrm
Posts: 41
Joined: Sun Sep 10, 2017 12:07 pm
Location: Norway

Re: Troubleshooting my latest BBC B

Postby adrm » Mon Sep 11, 2017 2:43 pm

Silly me.

I knew this 25 years ago AND I read it a few days ago. Yesterday I'd forgotten again. Internal memory is not what it used to be.

I'll try the cleaning suggestion next.

I'm still bothered by the fact that a Hard Reset causes problems, while a Soft Reset "fixes" it, but we'll see what comes from a good cleaning :D
-------
Tore

User avatar
DutchAcorn
Posts: 1584
Joined: Fri Mar 21, 2014 9:56 am
Location: Maarn, Netherlands

Re: Troubleshooting my latest BBC B

Postby DutchAcorn » Mon Sep 11, 2017 7:27 pm

I had a BBC B+ with a faulty capacitor in the system VIA reset circuit. It may be worth checking C10.

BTW; is the reset pin really 5V or something a bit lower?
Paul

adrm
Posts: 41
Joined: Sun Sep 10, 2017 12:07 pm
Location: Norway

Re: Troubleshooting my latest BBC B

Postby adrm » Mon Sep 11, 2017 7:30 pm

So far, no luck.

What I have tried today:

- Swapped the original 6522 back into IC3 socket (I'd been using the one from IC59 for the last few days), after cleaning + scraping the legs. No difference.
- Cleaned the connectors of 6522 socket as best I could (the dratted DFS board blocks everything!). No difference
- Measured from the top of IC legs through to the soldering on the underside of the PCB. I feel this would check both for broken pins and legs spring connection in the socket. All showed 0.3 - 0.5 Ohm resistance, i.e 0 on my DMM)
- Measured all leads through the keyboard cable, using the connector pins, i.e. not the cable's leads. All show ~0 Ohms
- Checked conection from some of the 6522 pins to other parts of the board. What I didn't check were: adress and data lines, both because they seem to work fine (after Soft Break), and a few other pins I had some difficulty locating through the circuit diagram. I will do all of these again tomorrow, if that's the recommended course of action

Question:
If I recall correctly, the Beeb queries all ROMs on Power On and HR. Can this hang be caused by one of them failing to respond correctly?
(The only ones left in sockets are System and BASIC)
-------
Tore

User avatar
1024MAK
Posts: 6676
Joined: Mon Apr 18, 2011 4:46 pm
Location: Looking forward to summer in Somerset, UK...

Re: Troubleshooting my latest BBC B

Postby 1024MAK » Mon Sep 11, 2017 8:01 pm

I can't think of anything right now, but your symptoms make me think that someone else also may have had this problem in the past. Try searching the forum (or use Google to search the forum, start the search with:

Code: Select all

site:http://stardot.org.uk/
followed by a space and then your search term...

Can you confirm that everything works after a soft break. That is all keys on the keyboard, large BASIC programs, machine code games, etc.

Also can you take a photo of what happens after a power on, then post the photo here please.

It may be a problem with the normal reset system (that resets the CPU etc.).

Mark
For a "Complete BBC Games Archive" visit www.bbcmicro.co.uk NOW!
BeebWiki‬ - for answers to many questions...

adrm
Posts: 41
Joined: Sun Sep 10, 2017 12:07 pm
Location: Norway

Re: Troubleshooting my latest BBC B

Postby adrm » Mon Sep 11, 2017 8:34 pm

Hi Mark.

Will search as suggested.

---After Power On, everything looks normal, i.e I don't believe a picture would help:
- The screen says "BBC Computer 32K, BASIC, >" (3 lines), and a flashing cursor. In Mode 7.
- The Caps Lock LED is on
--- What's not normal is:
- The speaker emits a loud continuous note. Actually it's changed to a buzzing at moderate volume since yesterday.
- The only keys on the keyboard the machine responds to are Break and Control+Break. The Caps Lock LED does not switch off if CAPS LOCK is pressed

When Break (only) is pressed:
- It resets and displays "BBC Computer, BASIC".
- The speaker buzzing changes slightly to a ... umm .. slightly warbling buzz
- ALL keys on the keyboard work as expected, including the function keys.

Unfortunately, I can't load any machine code programs, as I don't have any storage medium connected

I can write a basic program that allocates most of the memory to arrays and write/read those, if that will serve. I have already run a RAM test program I found here, though. I believe that did the same thing.
-------
Tore

cmorley
Posts: 219
Joined: Sat Jul 30, 2016 7:11 pm
Location: Oxford

Re: Troubleshooting my latest BBC B

Postby cmorley » Mon Sep 11, 2017 8:40 pm

Sounds similar to the !RSTA problem I had on an issue 7. Dry joint presumably from the factory.

http://stardot.org.uk/forums/viewtopic.php?f=3&t=13313#p173697

The sys via reset pin would charge up after a few seconds & it would work after a BREAK press.

User avatar
hoglet
Posts: 6359
Joined: Sat Oct 13, 2012 6:21 pm
Location: Bristol

Re: Troubleshooting my latest BBC B

Postby hoglet » Mon Sep 11, 2017 8:43 pm

What happens if you power up with the break key held down and then release it?

adrm
Posts: 41
Joined: Sun Sep 10, 2017 12:07 pm
Location: Norway

Re: Troubleshooting my latest BBC B

Postby adrm » Mon Sep 11, 2017 9:00 pm

Cmorley: I will study that thread. Sounds like a promising lead.

Hoglet: The screen remains black and the Caps Lock LED remains off while the Break key is held down. When it's released, it continues into a normal power on / hard reset with the problems described
-------
Tore

adrm
Posts: 41
Joined: Sun Sep 10, 2017 12:07 pm
Location: Norway

Re: Troubleshooting my latest BBC B

Postby adrm » Mon Sep 11, 2017 9:11 pm

Cmorley:
On my machine the Sys VIA RES goes high the moment I turn the machine on, so that's different, i.e. no delay.
-------
Tore

philb
Posts: 88
Joined: Sat Aug 05, 2017 6:05 pm

Re: Troubleshooting my latest BBC B

Postby philb » Mon Sep 11, 2017 9:19 pm

When you say "hard reset", do you mean that Ctrl-Break also puts it into the broken state (but Break on its own works as expected)?

adrm
Posts: 41
Joined: Sun Sep 10, 2017 12:07 pm
Location: Norway

Re: Troubleshooting my latest BBC B

Postby adrm » Mon Sep 11, 2017 9:25 pm

Philb: Exactly
-------
Tore

User avatar
1024MAK
Posts: 6676
Joined: Mon Apr 18, 2011 4:46 pm
Location: Looking forward to summer in Somerset, UK...

Re: Troubleshooting my latest BBC B

Postby 1024MAK » Mon Sep 11, 2017 9:29 pm

adrm wrote:Cmorley:
On my machine the Sys VIA RES goes high the moment I turn the machine on, so that's different, i.e. no delay.

It maybe worthwhile checking that circuit anyway. If C10 is faulty, or there is a poor/broken connection or dry joint, the reset line will rise very quickly behind the power rail instead of being slightly delayed.

Mark
For a "Complete BBC Games Archive" visit www.bbcmicro.co.uk NOW!
BeebWiki‬ - for answers to many questions...

User avatar
1024MAK
Posts: 6676
Joined: Mon Apr 18, 2011 4:46 pm
Location: Looking forward to summer in Somerset, UK...

Re: Troubleshooting my latest BBC B

Postby 1024MAK » Mon Sep 11, 2017 9:32 pm

One other thing, check what the level is on the /NMI line (pin 6 on the CPU).

Mark
For a "Complete BBC Games Archive" visit www.bbcmicro.co.uk NOW!
BeebWiki‬ - for answers to many questions...

philb
Posts: 88
Joined: Sat Aug 05, 2017 6:05 pm

Re: Troubleshooting my latest BBC B

Postby philb » Mon Sep 11, 2017 9:42 pm

If Ctrl-Break and power-on both make it fail, but regular Break works, that sounds as though the machine is correctly identifying the type of reset but that some operation in the startup code (which is performed for a cold boot and skipped for a warm boot) is making it go wrong. But, that said, I can't think of any very likely culprits. The difference between a warm and cold boot is not all that much and it's a bit hard to think of things in the cold boot path that would cause the sort of delayed symptoms you're seeing (i.e. it completes the boot sequence and enters BASIC but then dies). Other than, as Mark says, stuck interrupts but it's not entirely clear to me why those would be happening in the cold boot case but not the warm one.

If you press BREAK to get into the "mostly working" state and then type Ctrl-G to make it beep, what happens?

You mentioned a disk interface board at one point. Can you unplug that or is it soldered in?

adrm
Posts: 41
Joined: Sun Sep 10, 2017 12:07 pm
Location: Norway

Re: Troubleshooting my latest BBC B

Postby adrm » Mon Sep 11, 2017 9:50 pm

1024MAK wrote:One other thing, check what the level is on the /NMI line (pin 6 on the CPU).

Mark



NMI seems stuck low, no matter what. It fluctuates between -110mV and +380mV.
(Not sure if that +380mV peak I see constitutes a HIGH?)

Additional info
I attached a "Logic Analyzer" to pin 6.
About 1ms after power on there are 10-25 HIGH pulses registered, and after that nothing, i.e. constant low
I'm guessing those pulses may be some kind of power ripples ...
Last edited by adrm on Mon Sep 11, 2017 10:14 pm, edited 1 time in total.
-------
Tore

adrm
Posts: 41
Joined: Sun Sep 10, 2017 12:07 pm
Location: Norway

Re: Troubleshooting my latest BBC B

Postby adrm » Mon Sep 11, 2017 10:03 pm

philb wrote:
If you press BREAK to get into the "mostly working" state and then type Ctrl-G to make it beep, what happens?

You mentioned a disk interface board at one point. Can you unplug that or is it soldered in?


Ctrl-G does nothing apparent, i.e. the "warbling buzz" that results from a soft reset (as opposed to the constant buzz resulting from a hard reset) does not change.

Unfortunately the blasted thing is soldered in. I started unsoldering it from the MB, but gave up for lack of a desoldering gun. I'm worried about damage to the MB.
What I can do try (if strongly adviced), is unsolder the elevated board containing the WD1770 and a few other components.
-------
Tore

User avatar
1024MAK
Posts: 6676
Joined: Mon Apr 18, 2011 4:46 pm
Location: Looking forward to summer in Somerset, UK...

Re: Troubleshooting my latest BBC B

Postby 1024MAK » Mon Sep 11, 2017 10:51 pm

WAIT...
Check IC27 inputs and outputs, as shown in the following schematic diagram extracts:-
IMG_5252.JPG
Disk Drive Interface

IMG_5251.JPG
Econet Interface


Of course, the add in board may alter things a bit...

Mark
For a "Complete BBC Games Archive" visit www.bbcmicro.co.uk NOW!
BeebWiki‬ - for answers to many questions...

User avatar
1024MAK
Posts: 6676
Joined: Mon Apr 18, 2011 4:46 pm
Location: Looking forward to summer in Somerset, UK...

Re: Troubleshooting my latest BBC B

Postby 1024MAK » Tue Sep 12, 2017 12:05 am

If no disk Drive Interface is fitted, or if a disk Drive Interface is fitted that does not use interrupts, then link S9 must be made.
If however, there is a disk Drive Interface that uses interrupts, then link S9 must be broken / open circuit.

It's similar for the Econet interface with link S2.

Mark
For a "Complete BBC Games Archive" visit www.bbcmicro.co.uk NOW!
BeebWiki‬ - for answers to many questions...

User avatar
1024MAK
Posts: 6676
Joined: Mon Apr 18, 2011 4:46 pm
Location: Looking forward to summer in Somerset, UK...

Re: Troubleshooting my latest BBC B

Postby 1024MAK » Tue Sep 12, 2017 6:58 am

adrm wrote:NMI seems stuck low, no matter what. It fluctuates between -110mV and +380mV.
(Not sure if that +380mV peak I see constitutes a HIGH?)

Logic levels are supposed to be TTL, so a logic high is 2V to 5V.
Logic low should be between 0V and 0.8V inclusive.

More here :wink:

Mark
For a "Complete BBC Games Archive" visit www.bbcmicro.co.uk NOW!
BeebWiki‬ - for answers to many questions...

adrm
Posts: 41
Joined: Sun Sep 10, 2017 12:07 pm
Location: Norway

Re: Troubleshooting my latest BBC B

Postby adrm » Tue Sep 12, 2017 7:22 am

1024MAK wrote:WAIT...
Check IC27 inputs and outputs, as shown in the following schematic diagram extracts:-

Of course, the add in board may alter things a bit...

Mark


Pin 10 = permanently HIGH (As it should be?)
PIN 9 = permanently LOW
resulting in:
PIN 8 = permanently HIGH

I fitted the Watford DFS ROM back in and issued *DISC, *CAT, with no changes to PIN 8 state.
Does it require a physical drive to be fitted to alter states?
(Now, it has to be said that my previous attempt at removing the Watford board may have left one or more pins without a connection to the main board. I decided to leave it as it was, while deciding whether to reverse my desoldering attempt or to proceed)
-------
Tore

adrm
Posts: 41
Joined: Sun Sep 10, 2017 12:07 pm
Location: Norway

Re: Troubleshooting my latest BBC B

Postby adrm » Tue Sep 12, 2017 7:26 am

1024MAK wrote:If no disk Drive Interface is fitted, or if a disk Drive Interface is fitted that does not use interrupts, then link S9 must be made.
If however, there is a disk Drive Interface that uses interrupts, then link S9 must be broken / open circuit.

Mark


On this unit S9 has been left open.
The DFS board appears to have been fitted by a professional, and I'm guessing that the previous owner had a working floppy drive connected in the past.
-------
Tore

adrm
Posts: 41
Joined: Sun Sep 10, 2017 12:07 pm
Location: Norway

Re: Troubleshooting my latest BBC B

Postby adrm » Tue Sep 12, 2017 7:28 am

1024MAK wrote:
adrm wrote:NMI seems stuck low, no matter what. It fluctuates between -110mV and +380mV.
(Not sure if that +380mV peak I see constitutes a HIGH?)

Logic levels are supposed to be TTL, so a logic high is 2V to 5V.
Logic low should be between 0V and 0.8V inclusive.

More here :wink:

Mark


Thanks, Mark.
I had read that a while back, but I wasn't sure about the exact figures.
-------
Tore

adrm
Posts: 41
Joined: Sun Sep 10, 2017 12:07 pm
Location: Norway

Re: Troubleshooting my latest BBC B

Postby adrm » Tue Sep 12, 2017 9:05 am

Not sure if it's a relevant clue or not, but I just scoped the positive speaker pin on the MB to get a look at the wave generating the buzzing sound.

According to my scope there's a 85 (-ish) KHz fluctuating signal coming from that pin.
-------
Tore


Return to “hardware”

Who is online

Users browsing this forum: No registered users and 6 guests