Model B - Issue 7 - Beep of doom

for bbc micro/electron hardware, peripherals & programming issues (NOT emulators!)
Post Reply
MadeGray
Posts: 75
Joined: Wed Dec 27, 2017 4:44 pm
Contact:

Model B - Issue 7 - Beep of doom

Post by MadeGray » Thu Oct 10, 2019 3:51 pm

Hi all,

I wondered if one of our learned friends could offer a little advice on a poorly Issue 7.

The machine exhibits the Caps Lock and Beep of doom with no screen display.

I've done quite a bit of fault finding so far as follows:

Checked all PSU voltages and at all locations. Tested a known working PSU as well.
Swapped all socketed chips with known good one at a time. Still no joy and the chips swapped all work perfectly in the second issue 7
tried alternate known working keyboard
Tested "limited configuration" removing all non essential chips to see if I could get "Language?"

Followed this trail from a previous post to check 6502 was not in continuous reset or not resting etc

Pin 40 is the Reset pin. When you press (and hold) break on the keyboard, the voltage on this pin should be a low logic level (0V - 0.8V). When you release it, it should go high (2V - 5V).

Pin 37 is the Phi 0 clock input. This should be a 2MHz square wave, but your multimeter can't show you that. It can show you the average voltage, though. It should be around 2V - 2.5V (the wave will be high half the tie, and low half the time).

The CPU outputs two clocks for the rest of the board to use. They're Phi 1 and Phi 2 on pins 3 and 39. They should give a similar average voltage to pin 37.
Pin 7 (SYNC) is normally low, but goes high whenever the CPU fetches and decodes an instruction. If you get an average voltage of ~0V, then your CPU isn't fetching and executing instructions any more.

All the above work as expected

Checked the ULA to ensure 1,2,4 and 8Mhz are accurate and all check out ok.

Swapped out OS for known working still the same

compared every jumper and link to known working
ensured anything non standard removed

changed the memory links from north, south and open - no change

I then burnt trick's test rom - Got total white screen, tried in known working got the same - then remembered its a 16k image and have to double it up for a 32k eprom but found a 16k one. test rom works fine in working machine. still the same test did not work in the failed machine.

To me this looks like an issue with ROMSEL or similar as the test rom should have at least kicked in.

I have a basic scope Chinese LCD but does enough and a decent multimeter that also measure frequency so any advice what to test next would be really appreciated. I've a few fully working issue 7's that i could go through the service manual and test each chip etc in turn but wondered if anyone had experienced such a thing. I have even cleaned and changed the ROMSEL jumpers and tried an IFEL ROM board to try to pass the main OS Rom to no effect.

It really seems to me that the machine can't see the OS chip (or Tricky's test chip) which leads me to believe the ROMSEL area is playing up but i might be totally wrong of course.

Any help greatly appreciated

John Gray
John Gray

BBC Master x2 with BeebSCSI, PI copro etc and lots of Goteks. A few Model B’s and a trusty Elk. Won’t mention the ZX80 and ZX81!

User avatar
marcusjambler
Posts: 670
Joined: Mon May 22, 2017 11:20 am
Location: Bradford
Contact:

Re: Model B - Issue 7 - Beep of doom

Post by marcusjambler » Thu Oct 10, 2019 3:55 pm

Hi John

Interesting problem :D
Have you tried buzzing out the OS socket?

Marcus

User avatar
1024MAK
Posts: 9236
Joined: Mon Apr 18, 2011 4:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Model B - Issue 7 - Beep of doom

Post by 1024MAK » Thu Oct 10, 2019 4:04 pm

Yes, inspection of, and continuity testing of the pins of the ROM chip (in the socket) to either the underside, or relevant CPU pins is the next logical step. As a single contacts failure can cause a world of pain...

Mark

MadeGray
Posts: 75
Joined: Wed Dec 27, 2017 4:44 pm
Contact:

Re: Model B - Issue 7 - Beep of doom

Post by MadeGray » Thu Oct 10, 2019 4:39 pm

Hi all,

I've checked all connections from OS chip side to underneath and reflowed them all to be sure. Checked between each rom socket and the expected connections are made. I need to test from the OS Chip back to processor etc but that will be one for tomorrow it seems as wife want's the kitchen back.......

Interestingly using Tricky's test ROM - after several power cycles I will quite often get the cassette relay on for a second and off for a second - Then the occasional Blue Screen - Some music and once I got the Hopper Screen - (changing memory banks doesnt make a difference) tried leaving on for a while - still the same randomness.

So - I think as I got the Hopper screen once I have probably got a logic chip not being happy or something - I will do a full trace back from OS socket tomorrow but if anyone has any more ideas i'm listening!

Thanks again
JOhn
John Gray

BBC Master x2 with BeebSCSI, PI copro etc and lots of Goteks. A few Model B’s and a trusty Elk. Won’t mention the ZX80 and ZX81!

User avatar
1024MAK
Posts: 9236
Joined: Mon Apr 18, 2011 4:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Model B - Issue 7 - Beep of doom

Post by 1024MAK » Thu Oct 10, 2019 5:43 pm

I still think a poor connection (DIL socket, link, power connector etc...) or dry solder joint is the most likely. Try gently tapping one corner of the board with the plastic handle of a screwdriver.

Tricky’s Test ROM does not care if there is any working RAM, it will run anyway, so it is best to leave link S25 north.

Mark

User avatar
marcusjambler
Posts: 670
Joined: Mon May 22, 2017 11:20 am
Location: Bradford
Contact:

Re: Model B - Issue 7 - Beep of doom

Post by marcusjambler » Thu Oct 10, 2019 6:03 pm

Check the power connector solder joints.
I've had some that seem solid but on closer inspection it was possible to rotate the connector within the solder joint.
CPU socket could be suspect also.

Marcus

MadeGray
Posts: 75
Joined: Wed Dec 27, 2017 4:44 pm
Contact:

Re: Model B - Issue 7 - Beep of doom

Post by MadeGray » Sat Oct 12, 2019 12:38 pm

Good afternoon all,

A few more checks, unfortunately the same.

Checked all address lines from all ROM sockets to 6502 all check ok

Checked all data lines from all ROM sockets to 6502 all check OK.

Noticed slight hot spot where the resistor next to ULA was laying on the board - checked and reflowed all joints in that area.

Checked Pin 20 on all ROMs back to relevant s21 etc links - checked these back to relevant pins on ic21 and 20.

Checked all power connections.

I think my thought flexing the board and sometimes I get a bit of trickys test Rom working is a red herring.

I can at times get the cassette on and off and at others the leds flash in turn as per the start of the Test Rom at which point all dies.

Not seen the hopper screen again but that I think was just one of those things, just look of the drawer.

I do have one of those cheap 8 port logic checkers and el cheap Chinese lcd oscilloscope so I think I’ll perhaps start comparing various outputs against a working one?

It’s a strange one for sure - it’s like she wants to work but just doesn’t kick in!

Any more thoughts welcomed!
John Gray

BBC Master x2 with BeebSCSI, PI copro etc and lots of Goteks. A few Model B’s and a trusty Elk. Won’t mention the ZX80 and ZX81!

User avatar
marcusjambler
Posts: 670
Joined: Mon May 22, 2017 11:20 am
Location: Bradford
Contact:

Re: Model B - Issue 7 - Beep of doom

Post by marcusjambler » Sat Oct 12, 2019 6:24 pm

Hi John

Is it possible to post a photo of the board?

Are these in sockets? IC4 IC5 IC7 IC73

If not it may be worth removing them ( starting with IC 7 and IC73 ) and putting sockets in so you can swap for known good stock.

Post Reply