Faulty BEEB - RAM problem FIXED!

for bbc micro/electron hardware, peripherals & programming issues (NOT emulators!)
User avatar
leenew
Posts: 3291
Joined: Wed Jul 04, 2012 3:27 pm
Location: Doncaster, Yorkshire

Faulty BEEB - RAM problem FIXED!

Postby leenew » Wed Mar 12, 2014 9:48 am

Hi,
I am currently trying to resurrect all my non functioning BEEBS.
I thought it would be useful to record the procedure of a couple that I have just fixed.
This one appeared to have a RAM problem and this seems very common, so this fix may be of use to others.

The BEEB booted up fine, showing BBC Microcomputer 32K

However, switching to modes 6,5,4,3,2,1 or 0 gave a garbled screen.

The BEEB did not boot with link S25 set to SOUTH.

As S25 SOUTH activates the CAS1 bank of RAM only, I deduced a RAM fault in this bank.

I then tried a quick RAM test to determine which chip was faulty.

I referred to this table provided by 1024MAK (Cheers Mark)

Data bits - IC numbers
Code:
/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 a model A /CAS1 controls the only 16k RAM bank, as IC61 to IC68 are not present.

With a model B /CAS0 controls the lower 16k RAM bank and /CAS1 controls the upper 16k RAM bank. /CAS0 is in the memory map from &0000 to &3FFF (0x0000 to 0x3FFF) and /CAS1 is mapped from &4000 to &7FFF (0x4000 to 0x7FFF).

Link S25 is South on a model A (16k) so only /CAS1 is active.
It is north on a model B (32k) and so both /CAS0 and /CAS1 are active.

If you suspect that you have a RAM memory fault, on a model B you can move link S25 South and see if the machine will run with only one 16k RAM bank. The /CAS1 bank will be the active RAM and the RAM controlled by /CAS0 will be disabled.

For testing only, on a model B, link S25 can be removed, then only /CAS0 is active. The /CAS1 RAM bank will be disabled.

If you use a 74LS04 hex inverter to invert the signal from the North pin of S25 and connect the output of the inverter to the middle pin of S25, this will swap the two banks around in memory as far as the MPU is concerned. You can now run a memory test on the suspect RAM from &4000 (0x4000) to &7FFF (0x7FFF). Note that if such a test shows a fault, it is the /CAS0 bank that is faulty.

Mark K.


As the active RAM (CAS1) is from &4000 to &7FFF I tried a quick memory test to determine the faulty chip.
Sometimes this works very well, sometimes not.
Luckily this time it was a breeze.

10 A=0
20 ?&4000=A: IF?&4000<>A THEN PRINT ?&4000
30 GOTO20

At a glance the program seems to do nothing, but it works by checking for stuck bits.

Anyway, the program looped forever which meant that no bits were stuck ON.

Next I changed line 10

10 A=255

This time, the program bombed out immediately, returning 127.

So, location &4000 should be 255, but actually stored 127.

In binary, I stored 11111111 but the computer stored 01111111
in other words, the high BIT 7 was faulty, stuck OFF.

Referring back to Mark's handy table, DataBit7 is controlled by IC's 68 and 60.
As it was the CAS1 bank that I suspected to be faulty (as it did not boot with S25 south), this points to a faulty IC60.

IC60 was swapped, and the beeb is perfect again.

I hope this helps.

Also refer here: http://www.stardot.org.uk/forums/viewto ... f=3&t=5305

Lee.

User avatar
l3k
Posts: 17
Joined: Sat Mar 08, 2014 10:43 pm
Location: Dorset

Re: Faulty BEEB - RAM problem FIXED!

Postby l3k » Wed Mar 12, 2014 7:04 pm

Hi Lee

Thanks for the concise post on troubleshooting RAM issues. I am new to resurrecting Beebs, just picked one up from eBay last week, and after trawling the forums I came across your thread you referred to which matched the symptoms I was seeing. I ran the memory tests yesterday, and after re-reading the thread a few times I finally understood the logic behind being able to identify the exact memory IC that is faulty. This is very weird because at this point I thought I must note down the process and post my thanks, and provide a simple one post guide summarizing the discussions in the thread from last July. But you have done a much better job than I would of done!

My beeb has garbled display when S25 is North (mode0 was clear and is what I used when running the memory test), wont boot when South, and will work in Mode7 with 16K being reported when S25 is off. So I was confident the fault is with CAS1

Ran the basic memory test you listed, although I had to run in Mode 0 (and make sure I printed a couple of blank lines at the start otherwise the results were missing off the top of my LCD TV)

Running the first test a few times writing/reading 0, it would taken anything from 3 mins to 30 mins to get an error, with the result always 1 for me... or "00000001" meaning databit 0, and with with Marks ready reckoner D0/CAS1 points to IC53 as the culprit

Running the test to write/read 255 always came back with an immediate fail with the result of 254... or "11111110" which luckily points to the same databit 0 failing

I also ran the tests writing/reading address &3f00 and left this running for ages to to put my mind at rest that CAS0 was ok

Hopefully my logic above does indeed point to IC53 being faulty for me as I have a replacement HM4816AP-3 on order and will be de-soldering IC53 when the replacement arrives tomorrow!

Again thanks for this and the original thread, and especially Marks assistance which helped me understand the problem

cheers

Andy

User avatar
leenew
Posts: 3291
Joined: Wed Jul 04, 2012 3:27 pm
Location: Doncaster, Yorkshire

Re: Faulty BEEB - RAM problem FIXED!

Postby leenew » Wed Mar 12, 2014 7:56 pm

Hi Andy,
I am glad it was of help.

Also, if you have experience in desoldering chips, you can skip this bit, but if not......

Good advice I received from someone on here with regard to swapping RAM chips is to use some small snips to snip the legs from the chip, as close to the chip as possible, so the leg is left protruding from the board.

Turn the board over and, using a solder sucker and / or desoldering braid, remove as much solder as you can from the hole.

Then, lightly grip the remanant of the pin with a small pair of snipe nose pliers and heat from the back whilst ever so gently pulling on the pin. (I wedge the board between my knees when I am doing this!!)
When all 16 pins have been removed, I use a solder sucker to remove the rest of the solder from the hole.
Using the knee wedge method, I sometimes stand the board up, place the solder sucker on the top side, heat with the soldering iron from the back and then suck the molten solder out. This gives nice clean holes.
If you have got this far, fitting the socket is a breeze.... :D

I have my fingers crossed for you! [-o<
Lee.

User avatar
hoglet
Posts: 6275
Joined: Sat Oct 13, 2012 6:21 pm
Location: Bristol

Re: Faulty BEEB - RAM problem FIXED!

Postby hoglet » Wed Mar 12, 2014 8:01 pm

leenew wrote:When all 16 pins have been removed, I use a solder sucker to remove the rest of the solder from the hole.


Desoldering braid is also a good option at this stage, and less likely to damage PCB traces. Perversely, adding a bit of fresh solder at this point can help the braid suck up what's already there in one go.

Dave

User avatar
1024MAK
Posts: 6593
Joined: Mon Apr 18, 2011 4:46 pm
Location: Looking forward to summer in Somerset, UK...

Re: Faulty BEEB - RAM problem FIXED!

Postby 1024MAK » Thu Mar 13, 2014 12:39 am

I'm glad that the info I posted a while back is still useful. Even better if more dead Beebs are bought back to life (even if some of them have to be dragged kicking and screaming back to life :lol: ).

Depending on the size of the solder pads (I don't just work on one make/type), I often add a little fresh solder to the top side PCB pad of any cut-off IC pins, then heat with the iron, while very gently pulling the pin out (from the topside of the PCB) with long nose pliers. Then I clean the hole with a solder sucker. If not all the solder goes, add a little fresh solder, then again with the solder sucker. If I'm still not happy, I use desoldering braid. You may need to use desoldering braid on both PCB sides.

To prevent track damage, do not overheat the pad, or leave the iron on it for more than about 2 to 3 seconds. And never try to force a pin or component leg out.

Lee, I gave up holding PCBs between my legs when I dropped my soldering iron on my trainer. It melted and burnt through and made me hopping mad! :!: :twisted: :roll: #-o [-X
Now I have two sets of helping hands (heavy metal base with two clips on an "arm"). Plus a few bits of wood cut to hold various PCBs vertical on my work desk. Also comes in handy when I use a blow torch, sorry paint stripper, whoops, I mean proper hot air gun :^o to desolder DIL chips that I want to reuse...

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

User avatar
l3k
Posts: 17
Joined: Sat Mar 08, 2014 10:43 pm
Location: Dorset

Re: Faulty BEEB - RAM problem FIXED!

Postby l3k » Sat Mar 15, 2014 6:58 pm

Thanks for the tips. Desoldering IC53 went very smoothly, used a solder sucker on the back of the pcb to get out most of the solder, flipped it over and snipped of the legs close to the chip. About half the legs came off the board straight away and the last few just needed a bit of heat and the sucker from the top side and job done!

New HM4816AP-3 onboard now and finally my beed is working with 32K atlast!!

(Next step, replace the external VIA which I had to swap with system VIA when I first started troubleshooting the beeeeeeeeeeeeeeep. On order from sprow.co.uk)

cheers

Andy

User avatar
leenew
Posts: 3291
Joined: Wed Jul 04, 2012 3:27 pm
Location: Doncaster, Yorkshire

Re: Faulty BEEB - RAM problem FIXED!

Postby leenew » Sat Mar 15, 2014 7:12 pm

Good work Andy =D>
Another one lives to beep another day.
Lee.

User avatar
1024MAK
Posts: 6593
Joined: Mon Apr 18, 2011 4:46 pm
Location: Looking forward to summer in Somerset, UK...

Re: Faulty BEEB - RAM problem FIXED!

Postby 1024MAK » Sat Mar 15, 2014 7:29 pm

=D> =D> =D>

Great news, another Beeb back from the dead 8)

Well done :D

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

mrpjevans
Posts: 1
Joined: Tue Aug 04, 2015 9:36 am

Re: Faulty BEEB - RAM problem FIXED!

Postby mrpjevans » Wed Jan 13, 2016 8:42 pm

Hi, just wanted to offer my thanks. Using the information in this thread, I was able to track down a memory fault to IC56 and replaced it. A BBC I thought was done for is now alive and well. Thanks!

User avatar
leenew
Posts: 3291
Joined: Wed Jul 04, 2012 3:27 pm
Location: Doncaster, Yorkshire

Re: Faulty BEEB - RAM problem FIXED!

Postby leenew » Wed Jan 13, 2016 10:29 pm

Excellent work Mr Evans =D>
There is no more satisfying sound than the first double beep when a beeb comes back to life :D

Lee.

User avatar
1024MAK
Posts: 6593
Joined: Mon Apr 18, 2011 4:46 pm
Location: Looking forward to summer in Somerset, UK...

Re: Faulty BEEB - RAM problem FIXED!

Postby 1024MAK » Wed Jan 13, 2016 11:08 pm

leenew wrote:Excellent work Mr Evans =D>
There is no more satisfying sound than the first double beep when a beeb comes back to life :D

Lee.

Wot he said :lol:

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

User avatar
CMcDougall
Posts: 5507
Joined: Wed Feb 02, 2005 3:13 pm
Location: Shadow in a Valley of Scotland

Re: Faulty BEEB - RAM problem FIXED!

Postby CMcDougall » Tue Sep 06, 2016 8:00 pm

one question Lee or anyone else who has done this,
does 'piggybacking' another ram chip on top of 'duffer' made it work, as I just get different things happening :lol:
or would I have to take off either it's 0 &/or 5v leg to make the 'piggybacking' work :-k

not getting any constant answers (with Inverter on) from the MarkH or MartinB prog :? :cry:
so don't want to take a grinder to it until I get a proper BrrrBeep from setting full S25 North
ImageImageImage

User avatar
leenew
Posts: 3291
Joined: Wed Jul 04, 2012 3:27 pm
Location: Doncaster, Yorkshire

Re: Faulty BEEB - RAM problem FIXED!

Postby leenew » Tue Sep 06, 2016 8:21 pm

Hi Col,
Yes.
and No :lol:

Sometimes :D

I have piggybacked (just plopped one on top of another) and had a beeb spring to life.
It is a bit hit and miss though.
What I have done in the past, is find what I *think* is the bad DRAM chip, and maybe piggyback at that point before getting the hammer and chisel out...
The RAM test I use is this: http://www.stardot.org.uk/forums/viewto ... 934#p82202

Lee.

User avatar
CMcDougall
Posts: 5507
Joined: Wed Feb 02, 2005 3:13 pm
Location: Shadow in a Valley of Scotland

Re: Faulty BEEB - RAM problem FIXED!

Postby CMcDougall » Wed Sep 07, 2016 5:17 pm

Cheers Lee,
mine with 255 just spat out lots of 255s all down screen, so changed to 0, after 30mins heat up, it would spit out 128 with a few 0s inbetween more 128s ,so pointed to IC68, piggybacking did not solve , so cut off bottom left & top right legs to side, piggybacking then worked, & so does the now full 32k!!

:D
ImageImageImage

User avatar
leenew
Posts: 3291
Joined: Wed Jul 04, 2012 3:27 pm
Location: Doncaster, Yorkshire

Re: Faulty BEEB - RAM problem FIXED!

Postby leenew » Wed Sep 07, 2016 6:20 pm

I love it when a plan comes together :D
Good work =D>

Lee


Return to “hardware”

Who is online

Users browsing this forum: 1024MAK and 8 guests