Faulty ROM sockets

discuss both original and modern hardware for the bbc micro/electron
Post Reply
Footie
Posts: 87
Joined: Wed Aug 02, 2006 12:13 am
Location: Auckland, New Zealand
Contact:

Faulty ROM sockets

Post by Footie » Sun Jul 12, 2020 6:27 am

With the faulty Model B I have, I have now replaced the RAM and the decoupling capacitor, however there is still a problem with accessing ROMS in IC52 and IC88.
The ROM IC is good and the other 3 ROM sockets work fine.
I've checked voltages on the all the pins and they are same across all sockets.

What should I check next to see what is causing IC52 and IC88 to not work? IC20?

cmorley
Posts: 1319
Joined: Sat Jul 30, 2016 8:11 pm
Location: Oxford
Contact:

Re: Faulty ROM sockets

Post by cmorley » Sun Jul 12, 2020 8:21 am

You may well have answered your own question with the thread title. Have you checked the sockets? Acorn fitted some fairly poor quality sockets to some revisions of model B which are prone to contacts breaking.

edit: You could try wiggling them when you check the continuity in case a fracture is making partial contact. Also check all the jumpers are fitted correctly S20, S21 (x2) and S22 off the top of my head.

User avatar
vanekp
Posts: 907
Joined: Thu Nov 30, 2000 7:09 am
Location: The Netherlands
Contact:

Re: Faulty ROM sockets

Post by vanekp » Sun Jul 12, 2020 8:35 am

there are a bunch of jumper for configuring the ROM sockets I take it you have checked all of those (you can find the settings in the service manual)
was there some sort of ROM/RAM board fitted that may have damaged the pins in the socket.
Peter.

Footie
Posts: 87
Joined: Wed Aug 02, 2006 12:13 am
Location: Auckland, New Zealand
Contact:

Re: Faulty ROM sockets

Post by Footie » Sun Jul 12, 2020 10:13 am

Good point about the jumpers S20-S22. The machine has a turned socket fitted so must have had a SWR board at some stage.
I'm getting voltages in all the pins so possibly not a bad solder joint but I'll recheck that too.

User avatar
KenLowe
Posts: 1380
Joined: Mon Oct 18, 2004 5:35 pm
Location: UK
Contact:

Re: Faulty ROM sockets

Post by KenLowe » Sun Jul 12, 2020 11:35 am

Just be careful when using a meter to check for continuity / voltage. Sometimes just the pressure of holding the meter probe onto an IC pin is enough to make the connection good again, fooling you into thinking there isn't a problem with the pin.

User avatar
1024MAK
Posts: 10294
Joined: Mon Apr 18, 2011 5:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Faulty ROM sockets

Post by 1024MAK » Sun Jul 12, 2020 2:09 pm

See this topic for more about problems with ROM sockets.

Image
BBC Model B sideways ROM socket signals

Pin 1 is not connected (N/C).
Pin 20 (/CE - chip enable) is also known as /CS (chip select).
Pin 22 (/OE) is output enable.

Apart from pin 1 which is not connected, pin 20, which is an individual chip select control circuit (from IC20) and pin 26 which goes via selection jumper shunts/links, all the other pins are connected together on all four sideways ROM sockets.

The selection jumper shunts/links for pins 26 are S32 and S33.

Test with the power switched off using the 200 ohm range on your multimeter. You are looking for a resistance of less than 1 ohm. Always test with a ROM chip fitted in the sockets and test the very top of the IC leg/pin.

Mark

johnkenyon
Posts: 248
Joined: Wed Jul 20, 2011 3:21 pm
Location: Coventry
Contact:

Re: Faulty ROM sockets

Post by johnkenyon » Sun Jul 12, 2020 4:00 pm

1024MAK wrote:
Sun Jul 12, 2020 2:09 pm
Test with the power switched off using the 200 ohm range on your multimeter. You are looking for a resistance of less than 1 ohm. Always test with a ROM chip fitted in the sockets and test the very top of the IC leg/pin.
Pinging off at a tangent (as is my wont)...

I'm a bit paranoid about using the resistance range to do tests which might result in current from my meter flowing though "silicon" on the piece of kit I'm working on. I use the "diode test" on my meter, which tells me what the voltage drop is across the "device" between my meter probes.

On my meter:
A reading of "1." (with no trailing digits) = "infinity" which means open circuit
A reading of 0 = "short" - zero voltage drop between meter probes
Any other reading (usually between 0 and 1 volt) means that theres something between my two probes, and it isn't a short and it isn't open circuit.

The resistance range (+continuity tester on the 200ohm range) is saved for testing parts of circuits things that I know for sure aren't connected to anything that could be damaged by stray voltages.

User avatar
1024MAK
Posts: 10294
Joined: Mon Apr 18, 2011 5:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Faulty ROM sockets

Post by 1024MAK » Sun Jul 12, 2020 5:14 pm

John, have you ever measured the test current from a digital multimeter when switched to the 200 ohms (or equivalent) resistance range?

The vast majority of digital multimeters actually use a low test current and hence they have a low test voltage, normally chosen so that normal silicon semiconductor junctions are not turned on. If the junctions did turn on, in circuit test results would be seriously affected.

Diode test ranges however have to produce a voltage high enough to turn on semiconductor junctions of all types. Some digital multimeters have a high enough test voltage to dimly light up red LEDs (normal forward voltage of around 1.8V to 2V)

Mark

Footie
Posts: 87
Joined: Wed Aug 02, 2006 12:13 am
Location: Auckland, New Zealand
Contact:

Re: Faulty ROM sockets

Post by Footie » Mon Jul 13, 2020 4:38 am

Looks you you guys are right. The jumpers were in the right places, but I took them off and cleaned them, then wiggled the ROM sockets and now it's working OK. So, just need to track down which pin(s) have the bad connection and re-older them.

Thanks all.

User avatar
vanekp
Posts: 907
Joined: Thu Nov 30, 2000 7:09 am
Location: The Netherlands
Contact:

Re: Faulty ROM sockets

Post by vanekp » Mon Jul 13, 2020 4:08 pm

=D> glad you got it working and hopefully able to pin point exactly where the problem is.
Peter.

Post Reply

Return to “8-bit acorn hardware”