exxos's Atari, Amiga & retro forum

I've also been toying with getting a beeb - I'd probably go for a master to be honest because that's what I used to do my A-Level in comp sci back in the late 80's - lot's of fond memories of coding on it.

I later went on to work in a school and ran a L3 Econet file server setup and oversaw it getting upgraded to archimedes (of various kinds) before moving it to thin wire ethernet.

Very happy memories - I wish I had access to all that kit now.

Good luck with the repair, I'm enjoying following along.

stephen_usher wrote: 16 May 2021 16:58 It's probably the bus buffer chips, they're always a bit flakey on all machines. Without a decent oscilloscope though it's rather a case of flying blind.

@stephen_usher Which specific buffers? The 74LS245, IC 14 has been replaced.

I just went back and pulled the CPU and I get the same symptoms. Permanent tone from the speaker, caps lock led on and no video.

With the CPU in, /RESET is high and I have what seems to be clocks in the right places. Pulling IC6 stops all the symptoms as it removes all the clocks. I think the clocks are probably ok, but IC 40 & 43 could be bad and the clocks at the CPU are off.

I have ordered replacements for IC 44 & 45, which deal with the cas/ras logic, but I don't think they are the problem.

I just don't think the CPU is executing any code at all, if it was, then removing it would make a difference to the symptoms. Also, dropping in another CPU would make a difference if the CPU was bad, and it doesn't.

Running in minimal configuration doesn't help either.

That leads me to believe:

1. The clocks might not be correct. No decent scope makes this very tricky to check.
2. There is an address or data bus problem, meaning the ROMs cannot be correctly accessed and therefore code cannot be retrieved and executed, or the wrong code is executed.
3. The ROM is bad causing the CPU to stall.
4. A combination of the above.

I think I need to go and buzz out the data and address buses, make sure they go where they are supposed to go.

I can't test the ROM, as I have no way to read it and compare to a good one for now.

I suppose the sound chip could be bad and hanging the system, but it seems unlikely.

Well, it's certainly proving a challenge, but worse case scenario I don't have too long to wait as our house now has a roof, windows, doors and is starting to be fitted out internally. We could be in in 6-8 weeks and then I'll have all my lovely test kit back again :)

Then we'll see who's broken!!

BennehBoy wrote: 16 May 2021 18:20 I've also been toying with getting a beeb - I'd probably go for a master to be honest because that's what I used to do my A-Level in comp sci back in the late 80's - lot's of fond memories of coding on it.

I later went on to work in a school and ran a L3 Econet file server setup and oversaw it getting upgraded to archimedes (of various kinds) before moving it to thin wire ethernet.

Very happy memories - I wish I had access to all that kit now.

Good luck with the repair, I'm enjoying following along.

Thanks, there do seem to some Masters out there and not too expensive. It sounds like I may have been a little earlier than you as we had Model B's in our computer lab. Wish my parents could have afforded one, but they couldn't stretch that far and I had a ZX81 then Spectrum at home, followed by an Atari 800XL.

Good look finding one and glad you're enjoying following along.

What type roms are in 2k mask 8k mask

Roms

The problem with the Beeb is that if it doesn't get far enough in the boot process to initialise the sound chip and the chip controlling the keyboard LEDs you get the same symptoms for many faults. Hence the debug ROM, sort of like the C64 dead test cart.

Tomswork wrote: 16 May 2021 19:26 What type roms are in 2k mask 8k mask

Roms

16384 word x 8bit

@rubber_jonnie

Something like a 27128

Tom

stephen_usher wrote: 16 May 2021 20:54 The problem with the Beeb is that if it doesn't get far enough in the boot process to initialise the sound chip and the chip controlling the keyboard LEDs you get the same symptoms for many faults. Hence the debug ROM, sort of like the C64 dead test cart.

Yep, the troubleshooting guide I have shows a LOT of faults, all with very similar symptoms.

I'm trying to get my hands on a diagnostic ROM via stardot, but unfortunately the guy who was going to help is having some issues burning a ROM and only has one spare, that he can't spare if that makes sense.

I am so kicking myself at not bringing more tools with me to the rental. But hey, ho what can you do, I didn't plan on deep diving a BBC B when we got here, but I got the machine super cheap compared to a lot of others on eBay at the time.

I have been looking more closely at the logic surrounding the CAS/RAS lines. There is what looks like normal activity on the CAS lines through IC 45, but not RAS. It's just a flat line. The inverter it runs through is also part of the clock circuit, IC 43. I figure I may as well change it as I have spares, haven't got anything to lose and bad RAS would prevent boot I think.

Tomswork wrote: 16 May 2021 21:21 @rubber_jonnie

Something like a 27128

Tom

That exactly as it happens, though I'm told you can also use a 27C256 or 27C512 as long as you place the ROM contents correctly.

rubber_jonnie wrote: 16 May 2021 21:43

Tomswork wrote: 16 May 2021 21:21 @rubber_jonnie

Something like a 27128

Tom

That exactly as it happens, though I'm told you can also use a 27C256 or 27C512 as long as you place the ROM contents correctly.

Yeah, I've used 28C256 EEPROMs and they work fine. You just have to put the data starting at 0x4000 as the top bit of the address is always high.

Actually, I could program one of my spares and send it to you. I can probably find a link to the ROM image on Stardot.