Fault finding index

for bbc micro/electron hardware, peripherals & programming issues (NOT emulators!)
steve3000
Posts: 2089
Joined: Sun Nov 25, 2012 12:43 am
Contact:

Re: Fault finding index

Post by steve3000 » Tue Mar 27, 2018 10:18 pm

Bat wrote:The noise from the speaker has certainly decreased since replacing C19 and C20 so it's improved PSU noise filtering.
Great, that's definitely worth knowing about :)

User avatar
marcusjambler
Posts: 684
Joined: Mon May 22, 2017 11:20 am
Location: Bradford
Contact:

Re: Fault finding index

Post by marcusjambler » Wed Mar 28, 2018 9:59 am

1000uf 16v 105oC axial electrolytic caps
Pretty sure these are radial type required.

Marcus

Bat
Posts: 165
Joined: Tue Mar 13, 2018 1:17 pm
Contact:

Re: Fault finding index

Post by Bat » Tue Apr 03, 2018 8:02 pm

marcusjambler wrote:
1000uf 16v 105oC axial electrolytic caps
Pretty sure these are radial type required.

Marcus
Hi Marcus,
Yes you're right radial type!
Well spotted thank you :)
Cheers :)

Graeme Roberts
Posts: 14
Joined: Thu Mar 15, 2018 8:35 pm
Location: Cardiff, Wales
Contact:

Re: Fault finding index

Post by Graeme Roberts » Tue Apr 17, 2018 5:57 pm

1. BBC Model B+ 64K.
2. Continuous beep, black / blank screen on power up.
3. Removal of all ROMs except OS / BASIC. Then a sequential piggy backing of a known good 4164 RAM chip at each of the eight RAM positions on the motherboard starting at IC55, then IC56, 60, 61, 64, 65, 66, 67 etc.

Procedure identified IC64 as bad as the machine booted when the 4164 was put on top of this chip as shown below.
IMG_1453.JPG
4. I initially tried removing all ROMs and cleaning legs / pins. I then checked the links to make sure all were present. This procedure will work for one bad chip, however I doubt if it would work for multiple RAM chips that might also be faulty. As the B+ motherboard has no link S25, it's difficult to use the 'switch bank' technique to identify if the problem is in the first or second bank of RAM, unless one of you knows a similar feature that can be used on the B+ range?

Hope this helps.

Graeme.

User avatar
marcusjambler
Posts: 684
Joined: Mon May 22, 2017 11:20 am
Location: Bradford
Contact:

Re: Fault finding index

Post by marcusjambler » Tue Apr 17, 2018 7:29 pm

Hi Graeme

If you have access to an eprom programmer you can burn Tricky's test ROM that plugs in the OS socket.
It works in a Beeb, B+
I've used it to diagnose a B+ RAM issue.

The ROM image is here :
http://stardot.org.uk/forums/viewtopic. ... 68#p139467

Marcus
Last edited by marcusjambler on Fri Jun 22, 2018 1:56 pm, edited 1 time in total.

Graeme Roberts
Posts: 14
Joined: Thu Mar 15, 2018 8:35 pm
Location: Cardiff, Wales
Contact:

Re: Fault finding index

Post by Graeme Roberts » Tue Apr 17, 2018 7:54 pm

Hello Marcus,

I don't have access to an EPROM burner, however I do have an EEPROM burner so I was curious to know what types of EEPROM I could use in a Beeb, assuming a 16K size with 24 pin DIL layout. Do any chips of this type exist? I could then update this machine from DFS 2.10 to DFS 2.26 as well as blow some useful software such as Tricky's test ROM image. Thanks for the tip and the link.

Graeme.

User avatar
1024MAK
Posts: 9283
Joined: Mon Apr 18, 2011 4:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Fault finding index

Post by 1024MAK » Wed Apr 18, 2018 1:10 pm

There is a whole thread about using EEPROMs in Acorn computers, so make some lots of time available and go and have a nice long read of this thread :wink:

Dave H. (A member on here) can supply the correct EEPROM chips for a very reasonable price.

Mark

User avatar
Elminster
Posts: 3950
Joined: Wed Jun 20, 2012 8:09 am
Location: Essex, UK
Contact:

Re: Fault finding index

Post by Elminster » Wed Aug 29, 2018 8:46 pm

Just to close down my issue I shall add pointer on here.

Fuzzy corrupts screen in all modes (blind typing of course; using serial you can see that modes 0-4 don't actually work), caused by faulty IC45, with weird background colour screen in modes that do work, except mode 7. Random hangs and all address lines going high also seem to be caused by IC45. A different issue, probably unrelated to IC45 in anyway caused by (once you have fixed IC45 you can tell it is another issue) caused by a misbehaving ULA.

if this sounds like something you have, then details in this thread

viewtopic.php?f=3&t=9438

steve30
Posts: 22
Joined: Fri Sep 02, 2011 4:35 am
Location: Rotherham
Contact:

Re: Fault finding index

Post by steve30 » Tue Apr 09, 2019 1:13 pm

As requested, here is a summary of my floppy disk problems.

Machine: BBC Model B Issue 7.

Problem: Drive motor spins up, head motor steps when required, but no data can be read, and its not clear whether or not it was writing data. Gives a "Disc fault 18 :0 00/00" on Acorn DFS and "Sector not found at 00/02" on Watford DFS etc (these errors are the same).

Solution: Turned out to be a faulty IC 87, which is a 74LS123. This is a monostable timer chip which is connected to the READ DATA pin on the floppy drive. I'm not sure exactly what it does; I'm assuming it cleans up the data pulses or similar. It doesn't appear to be necessary with a 1770/1772 controller. The manual states that you can either install a 74LS123, or install a wire link between pins 9 and 12 (connecting the input and output together). I didn't have any spare 74LS123s, so I just took the old one out and bypassed it with the wire link.

It now works fine :).

Forum topic: viewtopic.php?f=3&t=16844

Kazzie
Posts: 947
Joined: Sun Oct 15, 2017 7:10 pm
Location: North Wales
Contact:

Re: Fault finding index

Post by Kazzie » Sat Jul 06, 2019 10:03 am

(BBC Model B)

Symptoms: Econet becomes jammed when machine is plugged in to network.

Diagnosis: Data lines are both pulled down to 0V. Removing IC93 (75159 differential line driver) confirms that this chip is the cause.

Fault: Pin 9 of the socket for IC93 was poorly soldered when installed, and is floating. Pin 9 is the Enable pin for the line driver.

Solution: Reflow solder on IC93 (and any other suspicious looking pins on the Econet circuitry)
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

Kazzie
Posts: 947
Joined: Sun Oct 15, 2017 7:10 pm
Location: North Wales
Contact:

Re: Fault finding index

Post by Kazzie » Tue Jul 30, 2019 8:46 am

BBC Model B (Issue 7)

Symptoms: when running screen went blotchy, and then blank. On turning off and on again, the machine gives a long beep with no display.

Diagnosis: Power supply tested good, moving S25 has no effect. Removed DFS and fitted S9. Swapped out CPU, 6845, VIAs, SAA5050, MOS and ICs 14, 72, 73, with no change. 1,2,4,8MHz clock lines pulsing, reset signal found at pin 40 of CPU on power up and on pressing break key. Probing CPU pins shows that it has stalled: no activity on pin 7 (Sync), address bus is all high, data bus reads &DC (contents of MOS at address &FFFF).

Fault: Probing IC45 shows pin 11 is not strobing, when its other pins (9,10,12-15) are. This signal controls the gate function of the Address bus buffer for ~A0-A6, IC12.

Solution: replace IC45 (74s139).

Forum thread: viewtopic.php?f=3&t=17530
Last edited by Kazzie on Tue Jul 30, 2019 8:47 am, 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
Acorn System 1 home-made replica

User avatar
Elminster
Posts: 3950
Joined: Wed Jun 20, 2012 8:09 am
Location: Essex, UK
Contact:

Re: Fault finding index

Post by Elminster » Tue Jul 30, 2019 8:53 am

Be interesting if someone had time to go through the whole thread and cross-reference all the issues and solutions.

For example, the above one was replaced by IC45. If you go 3 or 4 posts up you will see my issue, fixed by replacing IC45 and ULA, but symptoms different enough that you might not suspect IC45 was a cause in both.

User avatar
Vyper68
Posts: 43
Joined: Sat Apr 15, 2017 11:42 am
Contact:

Re: Fault finding index

Post by Vyper68 » Thu Aug 15, 2019 1:17 pm

Model:- BBC Model B issue 7

Symptoms:-
Continuous beep on power on, able to get boot screen by pressing break key. Cassette LED continually lit and cassette interface would not work.
Developed into continuous Language? Prompt when pressing break to boot and eventually continuous beep of random tone with Black screen and refused to boot.

After trying various options like replacing 555 timer and resistors on reset circuit. Fault was eventually found to be faulty 74ls139 at IC20

Repair:-
Replaced 74ls139 at IC20 and also IC26 as both work with ROM

Tested and verified working including cassette interface

See link below for thread.
viewtopic.php?f=3&t=17527
Save time… see it my way.

User avatar
anightin
Posts: 334
Joined: Thu Aug 23, 2018 1:03 pm
Location: Cambridge UK
Contact:

Re: Fault finding index

Post by anightin » Tue Aug 27, 2019 11:23 am

Machines: BBC B Issue 4, and Issue 7

Symptoms:
Using a SCART->HDMI converter, loss of sync at random intervals causing screen noise and OSD to be re-displayed.

Complication:
Noisy PSU exacerbated above issue -- feed video output direct into scart/composite TV input to observe any interference.

Solution:
Replaced 16MHz crystal (X1) with a new part. Beware of NOS parts as initial replacement with a NOS X1 exhibited the same symptoms as above.
Recommend low profile version to maintain clearance for add-on boards.

Consider also checking PSU for dry joints and out of spec capacitors.
Also consider replacing IC43.
Last edited by anightin on Tue Aug 27, 2019 11:24 am, edited 1 time in total.

CJS
Posts: 17
Joined: Mon Oct 14, 2019 10:27 pm
Location: Bath, UK
Contact:

Re: Fault finding index

Post by CJS » Fri Oct 18, 2019 9:12 pm

Model:- BBC Model B issue 7

Symptoms: Not all ROMs visible from *HELP. On further investigation - largely by removing all ROMs and swapping some around - identified that only ROM sockets 12 and 14 were working, 13 and 15 were not.

Following some fault finding (see thread linked below), confirmed that 74LS139 at IC20 was faulty.

Solution: Replaced 74ls139 at IC20. *HELP now shows that all ROM Sockets now "visible" to machine.

See link below for thread.
viewtopic.php?f=3&t=18079

Post Reply