Failed RTC chip in the Master 128?

discuss both original and modern hardware for the bbc micro/electron
Post Reply
User avatar
simonm
Posts: 315
Joined: Mon May 09, 2016 3:40 pm
Contact:

Failed RTC chip in the Master 128?

Post by simonm » Wed Jul 31, 2019 2:30 pm

I've got a Master 128 that won't preserve CMOS RAM settings and I'm wondering if it's the RTC HD146818AP chip that might have failed.

The machine has a replacement 4.5V AA battery pack with inline diode & resistor, brand new duracells and 4.7V measured voltage from the batteries, and this voltage is also present at the RTC chip Vss/Vcc pins.

The symptoms are that powering up the machine after some time without power results in the CMOS settings being lost.

I don't mind trying a new RTC chip, but didn't want to do that if its unlikely to be the chip itself.

Thanks!
Last edited by simonm on Sun Aug 04, 2019 11:42 am, edited 2 times in total.

User avatar
danielj
Posts: 8410
Joined: Thu Oct 02, 2008 5:51 pm
Location: Manchester
Contact:

Re: Failed RTC chip in the Master 128?

Post by danielj » Wed Jul 31, 2019 3:29 pm

And the voltage is there when you're powered down? (I think that's how it reads but just want to be sure).

d.

User avatar
simonm
Posts: 315
Joined: Mon May 09, 2016 3:40 pm
Contact:

Re: Failed RTC chip in the Master 128?

Post by simonm » Wed Jul 31, 2019 3:33 pm

danielj wrote:
Wed Jul 31, 2019 3:29 pm
And the voltage is there when you're powered down? (I think that's how it reads but just want to be sure).
Yes.

User avatar
1024MAK
Posts: 10287
Joined: Mon Apr 18, 2011 5:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Failed RTC chip in the Master 128?

Post by 1024MAK » Wed Jul 31, 2019 4:18 pm

simonm wrote:
Wed Jul 31, 2019 2:30 pm
The machine has a replacement 4.5V AA battery pack with inline diode & resistor, brand new duracells and 4.7V measured voltage from the batteries, and this voltage is also present at the RTC chip Vss/Vcc pins.
Are you measuring across the pins of the chip for both Pin 24 (+V/VDD) and Pin 12 (VSS/GND) with the meter probes on both pins at the same time?
Also what do you get on pins 13 (/CS), 14 (AS), 16 (/STBY), 18 (/RESET) and 22 (PS) both with the mains off, and with the computer powered on?

Mark

User avatar
simonm
Posts: 315
Joined: Mon May 09, 2016 3:40 pm
Contact:

Re: Failed RTC chip in the Master 128?

Post by simonm » Wed Jul 31, 2019 6:15 pm

Thanks Mark, so full measures are as follows:

Battery pack output - 4.85V
Out of the PL connector - 4.5V

ON = powered on, OFF = powered off
Across Pin 12/24 Vss/Vcc - 4.59V (ON) 4.18V (OFF)
Pin 13 /CS - 4.55V (ON) 4.18V (OFF)
Pin 14 /AS - 63mV (ON) 0V (OFF)
Pin 16 /STBY - 0.3mV (ON) +/- 0.02mV (OFF)
Pin 18 /RESET - 4.9V (ON) 21-45mV (OFF)
Pin 22 /PS - 4.9V (ON) 48mV (OFF)

Is the 4.18V unusual? Presumably there's some components between the PL connector and the chip causing the voltage drop?

I noticed that with power off and the battery pack unplugged there's still a residual voltage across pins 12/24, starts at 1.3V and degrades to around 0.68V quite quickly then slowly reducing to < 0.5V over 20 minutes. Powering the machine up after 20 mins still retains the configuration so I'm going to leave it for longer to see how long it takes the chip to fully "forget" the config.

Thanks

Pic for reference:
2019-07-31 17.24.21.jpg
Last edited by simonm on Wed Jul 31, 2019 6:41 pm, edited 1 time in total.

User avatar
1024MAK
Posts: 10287
Joined: Mon Apr 18, 2011 5:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Failed RTC chip in the Master 128?

Post by 1024MAK » Wed Jul 31, 2019 7:37 pm

The +5V supply to IC11 (HD146818) has a silicon diode (D8) to prevent the battery from trying to power up the whole machine. Hence when on mains, the +v supply to IC11 will be approximately 0.4V to 0.7V less than the +5V from the PSU.

The battery pack also contains a silicon diode to prevent the PSU from trying to recharge the (non-rechargeable) primary cells. So when the mains is off, there will be about 0.4V to 0.7V less than the battery voltage across the power pins of IC11.

The supply to IC11, whether from the PSU, or from the battery, is smoothed by capacitor C11. The charge that this holds will allow IC11 to retain data and to run the RTC for a short amount of time.

The voltages you measured all look about right.

Are you resetting the ‘CMOS’ memory and setting a valid configuration after you find invalid data? You also should ensure that you reset the time and date to a valid time/date.

Mark
Last edited by 1024MAK on Wed Jul 31, 2019 7:39 pm, edited 1 time in total.

User avatar
simonm
Posts: 315
Joined: Mon May 09, 2016 3:40 pm
Contact:

Re: Failed RTC chip in the Master 128?

Post by simonm » Wed Jul 31, 2019 9:11 pm

1024MAK wrote:
Wed Jul 31, 2019 7:37 pm
The +5V supply to IC11 (HD146818) has a silicon diode (D8) to prevent the battery from trying to power up the whole machine. Hence when on mains, the +v supply to IC11 will be approximately 0.4V to 0.7V less than the +5V from the PSU.

The battery pack also contains a silicon diode to prevent the PSU from trying to recharge the (non-rechargeable) primary cells. So when the mains is off, there will be about 0.4V to 0.7V less than the battery voltage across the power pins of IC11.
That explains the 4.18V volts then, thanks.

1024MAK wrote:
Wed Jul 31, 2019 7:37 pm
The supply to IC11, whether from the PSU, or from the battery, is smoothed by capacitor C11. The charge that this holds will allow IC11 to retain data and to run the RTC for a short amount of time.

The voltages you measured all look about right.
Hmm.. I was afraid you might say that. Perhaps there is a fault in the chip, that despite the power being applied it won't persist stable CMOS RAM values.
Are you resetting the ‘CMOS’ memory and setting a valid configuration after you find invalid data? You also should ensure that you reset the time and date to a valid time/date.
Yes, power on +R at the start of the test session.
*conf. lang 12
*conf. file 9
*conf. noboot
*conf. mode 7
TIME$="Wed, 20 Aug 1990,12:00:00"

When it loses the config, it goes back to random TIME$ and settings. Took about 90 minutes to lose config with no power or battery backup. I'm just timing it again with backup battery reconnected.
Last edited by simonm on Wed Jul 31, 2019 9:12 pm, edited 1 time in total.

User avatar
1024MAK
Posts: 10287
Joined: Mon Apr 18, 2011 5:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Failed RTC chip in the Master 128?

Post by 1024MAK » Wed Jul 31, 2019 10:19 pm

Test the voltage on the chip (IC11) after 90 minutes. Just to check that the voltage from the battery is actually getting through okay.

If the chip can remember the settings while running only on the charge from C11, then it should be able to run on standby from the battery for at least 6 months.

Mark

User avatar
simonm
Posts: 315
Joined: Mon May 09, 2016 3:40 pm
Contact:

Re: Failed RTC chip in the Master 128?

Post by simonm » Wed Jul 31, 2019 11:25 pm

Still got 4.18V on IC11 after 2 hours, powered up and config is still ok so far.
Gonna leave it overnight now and see what happens.

User avatar
simonm
Posts: 315
Joined: Mon May 09, 2016 3:40 pm
Contact:

Re: Failed RTC chip in the Master 128?

Post by simonm » Sun Aug 04, 2019 11:51 am

Update on this. Annoyingly, it's held settings for over 4 days on battery backup power now, even though nothing has changed in the machine, so I'm pretty confident the RTC chip is ok now.

I now have a horrible and sneaking suspicion one of the batteries wasn't fully inserted since when I was setting up the soak test I pushed the battery pack down into place and felt one of the batteries move down just a few millimetres. Can't be certain that was the cause of the issue, because I'd measured the voltage from the pack, but an insecure/intermittent connection would be a reasonable diagnosis if indeed it was the problem.

Thanks for the help with this, much appreciated. At least maybe this thread might help someone in the future if their RTC is playing up! :lol:

User avatar
1024MAK
Posts: 10287
Joined: Mon Apr 18, 2011 5:46 pm
Location: Looking forward to summer in Somerset, UK...
Contact:

Re: Failed RTC chip in the Master 128?

Post by 1024MAK » Sun Aug 04, 2019 1:39 pm

Some AAA/AA battery holders have plastic reinforcements that cause poor, intermittent or no contact if cells that don’t have a very high button top are used :(
The problem is that the plastic reinforcements catch the edge of the top of one of the cells. So the metal contact, although it looks like it is in contact with the positive button top of the cell, and makes contact when you poke it with a meter probe, the rest of the time, it is either making a poor contact, or there is a very small air gap.

Not in a Beeb, but in a LED Christmas fairy / party light set, that problem sent me round the bend several times before I worked it out :evil:

Mark
Last edited by 1024MAK on Sun Aug 04, 2019 1:39 pm, edited 1 time in total.

Post Reply

Return to “8-bit acorn hardware”