Non booting BBC/B Issue 7 - Continuous beep Mk 2

discuss both original and modern hardware for the bbc micro/electron
User avatar
1024MAK
Posts: 9969
Joined: Mon Apr 18, 2011 5:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Non booting BBC/B Issue 7 - Continuous beep Mk 2

Post by 1024MAK » Sat May 04, 2019 7:33 pm

Kazzie wrote:
Sat May 04, 2019 7:16 pm
Edit: Great minds think alike Mark! :lol:
Yep, two heads are better than one, well, unless they are both trying to fit in the same space at the same time :lol:

Mark

User avatar
1024MAK
Posts: 9969
Joined: Mon Apr 18, 2011 5:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Non booting BBC/B Issue 7 - Continuous beep Mk 2

Post by 1024MAK » Sat May 04, 2019 7:41 pm

Could you please set your ‘scope to a more sensitive range and re-do the signal on pin 37 please. A photo with the waveform nearly filling the screen is best.

Are you putting your probe on the pins of the CPU?

Mark

User avatar
Sniffer
Posts: 173
Joined: Tue Aug 11, 2015 10:30 pm
Location: Colwyn Bay, North Wales
Contact:

Re: Non booting BBC/B Issue 7 - Continuous beep Mk 2

Post by Sniffer » Sat May 04, 2019 8:08 pm

Nothing changes with a known good cpu.

User avatar
Sniffer
Posts: 173
Joined: Tue Aug 11, 2015 10:30 pm
Location: Colwyn Bay, North Wales
Contact:

Re: Non booting BBC/B Issue 7 - Continuous beep Mk 2

Post by Sniffer » Sat May 04, 2019 8:18 pm

Could you please set your ‘scope to a more sensitive range and re-do the signal on pin 37 please. A photo with the waveform nearly filling the screen is best.
Please bear in mind I'm a novice with a scope. The scope is not hooked upto a pc, the images are directly from the scope.

SDS00005.png

This is pin 37, is this any better?
Are you putting your probe on the pins of the CPU?
Yes, I've been using a single probe at 10X, gnd connected to 0v by vcc3 and probe directly onto cpu pin(s).

Kazzie
Posts: 1544
Joined: Sun Oct 15, 2017 8:10 pm
Location: North Wales
Contact:

Re: Non booting BBC/B Issue 7 - Continuous beep Mk 2

Post by Kazzie » Sat May 04, 2019 8:28 pm

Sniffer wrote:
Sat May 04, 2019 7:32 pm
Can you please post images of the waveform on the following pins: 37, 39 and 3.
Sure. Inorder 3,37,39;


3-37-39.png
I see something interesting on the pin 37 signal.

The on-screen stats (of your earlier image) state that the square wave is 2.40V peak-to-peak, with a mean of 3.36V. As it's got a 50% duty cycle, the mean will also be the midpoint, so the "low" will be at ~2.16V, and the "high" will be at ~4.56V.

(Edit: for the zoomed-in image, 2.76V PtP and 3.24V mean gives a low of ~1.86V and high of 4.62V)

The Rockwell datasheet for the 6502 (page 15) states that for Phi0, a high input must be between 2.4V and VCC (which it is), but the low input must be between -0.3V and 0.4V (which it isn't).

To paraphrase, the low voltages of the clock signal are too high for the CPU to recognise, so the CPU doesn't think it has a clock input.

I suggest back-tracking from the Phi0 input on pin 37. This signal is generated by an OR gate on IC29 (74ls32).
BBCB_PHI0_INPUT.png
:arrow: Please probe pins 11, 12, and 13 of IC29, and show us what you see.
Last edited by Kazzie on Sat May 04, 2019 8:36 pm, 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
1024MAK
Posts: 9969
Joined: Mon Apr 18, 2011 5:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Non booting BBC/B Issue 7 - Continuous beep Mk 2

Post by 1024MAK » Sat May 04, 2019 8:34 pm

@Kazzie

An out of specification clock (phase) input signal is what I was thinking of as well. That’s the only explanation of why the CPU clock output pins are static.

Mark

User avatar
Sniffer
Posts: 173
Joined: Tue Aug 11, 2015 10:30 pm
Location: Colwyn Bay, North Wales
Contact:

Re: Non booting BBC/B Issue 7 - Continuous beep Mk 2

Post by Sniffer » Sat May 04, 2019 9:30 pm

Please probe pins 11, 12, and 13 of IC29, and show us what you see.
Sure. This was I think the first IC I replaced. Order 11,12,13.

SDS00006.png

Kazzie
Posts: 1544
Joined: Sun Oct 15, 2017 8:10 pm
Location: North Wales
Contact:

Re: Non booting BBC/B Issue 7 - Continuous beep Mk 2

Post by Kazzie » Sat May 04, 2019 9:59 pm

Righty-ho...

The first image, pin 11, doesn't have the superimposed statistics, but in the bottom right it suggests that the peak-to-peak voltage is 2.82V. That fits with the measurement at the other end of the wire, on pin 37 of the CPU.

The second image's statisitics indicate a 18.8V mean, 36V PtP(!) signal. Did you perhaps knock the x10 switch by mistake? Assuming that they are actually 10 times smaller than indicated, that's make for a Vlow of 0.08V and a Vhigh of 3.68V, both of which are within tolerance for a TTL chip. (See the link in 1024MAK's signature on TTL levels.)

The third image doesn't look as nice, and the numbers aren't good either. A mean of 1.4V and PtP of 680mV makes for a Vlow of 1.06V and Vhigh of 1.74V, which aren't valid values for a TTL chip. And this signal comes from pin 6 of IC34, a 74ls74 dual flip-flop.

The question is: is IC34 giving invalid output because it's getting an invalid input, or is there a fault inside IC34 itself?

:arrow: Firstly, check that the 5V and GND supplies for IC34 (pins 14 and 7) are good. Then, could you use your scope to look at the values on pins 1-6 of IC34, please?
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
Sniffer
Posts: 173
Joined: Tue Aug 11, 2015 10:30 pm
Location: Colwyn Bay, North Wales
Contact:

Re: Non booting BBC/B Issue 7 - Continuous beep Mk 2

Post by Sniffer » Sat May 04, 2019 10:44 pm

Did you perhaps knock the x10 switch by mistake?

Sorry, yep. Here is IC29 11,12,13 again.

IC29-11-12-13.png
Firstly, check that the 5V and GND supplies for IC34

They're good.

Then, could you use your scope to look at the values on pins 1-6 of IC34, please?
IC 34, pins 1-6 in order;

IC34-1-6.png
Thanks Kazzie.

Kazzie
Posts: 1544
Joined: Sun Oct 15, 2017 8:10 pm
Location: North Wales
Contact:

Re: Non booting BBC/B Issue 7 - Continuous beep Mk 2

Post by Kazzie » Sun May 05, 2019 7:30 am

Sniffer wrote:
Sat May 04, 2019 10:44 pm
Did you perhaps knock the x10 switch by mistake?

Sorry, yep. Here is IC29 11,12,13 again.


IC29-11-12-13.png

Firstly, check that the 5V and GND supplies for IC34

They're good.

Then, could you use your scope to look at the values on pins 1-6 of IC34, please?
IC 34, pins 1-6 in order;


IC34-1-6.png

Thanks Kazzie.
Thanks.

Looking at the corrected readings for IC29, the inputs on pins 12 and 13 are valid TTL input values (lows below 0.8V, highs above 2.0V) but the output on pin 11 isn't a valid TTL output (should be lows below 0.5V, highs above 2.7V). It appears that IC29 isn't able to generate a valid output when given valid inputs. It should be noted that as well as driving the CPU's Phi0 clock input, it also drives an input on IC25, so a fault on IC25 could be causing the Phi0 problems.

Moving on, let's look at that flip-flop on IC34. Pin 1 is a CLEAR input, and pin 4 is a PRESET. They should both be high (except pin 4 when pressing Reset), which they are. There's a bit on noise on pin 4, but it's voltage stays well above the 2V level required for a TTL high value. The clocking input, pin 3, has a nice square wave of appropriate values: it's the same signal as on pin 12 of IC29.

A D-type flip-flop is triggered by the rising edge of its clock signal. At this point, it copies the input value (D, pin 2) to its output (Q, pin 5), and inverted output (/Q, pin 6). These output values are held (latched) until the next rising edge of the clock. Our input is fixed at ~4.56V, so it's perfectly valid that Q should be fixed high (pin 5 is 4.08V) and /Q fixed low (pin 6 is ~148mV).

(IC34 is used to generate a 1MHz clock signal (that's in sync with the main 2MHz clock signal), for the slower devices in the Beeb that can't operate at 2MHz. The two possible clock signals are combined by the OR gate of IC29 before being fed into the CPU. These slower devices include the VIA used to interface with the keyboard and speaker, so a 1MHz clock signal should be input at some point as the machine boots. Of course, because the CPU doesn't recognise the off-voltage clock signal it's given, it doesn't run the OS far enough to do that.)

-----

So to recap: the output of IC29 at pin 11 is invalid for its inputs. I think this is most likely due to a fault on IC29, but it could be due to a different fault on IC25, overriding IC29's output. Options at this point include:
  • Measuring the voltage supplied to IC29 and IC25 (pins 7 and 14 on each), to check they're getting a good power supply
  • Scoping out the other pins on IC29 and IC25, to check if the other logic gates on them are operating correctly. Faults on other gates would suggest that chip is to blame, but logic gates can fail independently of each other.
  • Replacing IC29, or IC25, and seeing if the fault goes away
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
Sniffer
Posts: 173
Joined: Tue Aug 11, 2015 10:30 pm
Location: Colwyn Bay, North Wales
Contact:

Re: Non booting BBC/B Issue 7 - Continuous beep Mk 2

Post by Sniffer » Sun May 05, 2019 10:04 am

Some success :) Socketed a replacement IC25 and Trickys rom is off and running.

I think I have some ram issues which I'll now look at but I just wanted to report the success =D>

Thanks guys.

Kazzie
Posts: 1544
Joined: Sun Oct 15, 2017 8:10 pm
Location: North Wales
Contact:

Re: Non booting BBC/B Issue 7 - Continuous beep Mk 2

Post by Kazzie » Sun May 05, 2019 12:13 pm

Excellent! Good news. =D>
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
Sniffer
Posts: 173
Joined: Tue Aug 11, 2015 10:30 pm
Location: Colwyn Bay, North Wales
Contact:

Re: Non booting BBC/B Issue 7 - Continuous beep Mk 2

Post by Sniffer » Sun May 05, 2019 4:54 pm

Excellent! Good news.
Yep, good news. As mentioned above, I might have ram issues so I intend to run a memory test prog that I have on disc (listing from this site) so I have installed the 1770 disc upgrade but can't access the attached drive. The setup is good, it works well with other beebs.

Code: Select all

*cat
just appears to hang forever. So I installed instead the 8271 upgrade and that results in the same situation.

I'll start a new thread though as I now have a 'booting' beeb :)

Thanks again all.

User avatar
Sniffer
Posts: 173
Joined: Tue Aug 11, 2015 10:30 pm
Location: Colwyn Bay, North Wales
Contact:

Re: Non booting BBC/B Issue 7 - Continuous beep Mk 2

Post by Sniffer » Thu May 09, 2019 2:00 pm

Some success :) Socketed a replacement IC25 and Trickys rom is off and running.
Well I spoke too soon. Having socketed a couple of ICs relating to my disk problem I thought I had fried IC28, leaving me back at sqaure one. However, I replaced the IC and it didn't resolve the issue, I still had a beeb that just 'beeeeeeeeeeeeeeeped'.

In a minimalist configuration, with Tricky's test rom installed, it does sometimes boot but the beeeeeeeeeeeeeeeeeep continues. All very, very odd.

It will randomly boot, probably one in 10 power ups, in it's minimalistic state, with the test rom.
Last edited by Sniffer on Thu May 09, 2019 2:01 pm, edited 1 time in total.

Post Reply

Return to “8-bit acorn hardware”