Advanced search FAQ Register Login

It is currently Tue Jan 23, 2018 12:55 am

 

Postby TerryJC » Thu Aug 24, 2017 5:13 pm

Hi,

Last November I bought two RASClock modules; one for keeping time for our Model Town Minster's chimes and one as a spare. After initial minor glitches, the chimes have kept perfect time since we switched them on around March 31st this year.

We have now moved on to another project and last weekend I borrowed the spare for use in the prototype of the new system. I have had major timekeeping problems with this and cannot work out why. The battery is new and measures 3.1 V, so I'm fairly sure it can't be that.

What is different to last time is as follows:

  • We are running Raspbian Stretch and the Afterthought Setting Up Instructions only cover Wheezy and Jessie. I followed the instructions for Wheezy.
  • I admit to making a mistake initially, when I set the RASClock up, in that I forgot to comment out the two lines which contain --systz. That has now been done.
  • Our Pi Zero is connected via a ribbon cable to a PCB with various interfacing devices on it. The IDC connector on the end of the ribbon prevents the RASClock from being connected directly to the Pi, so I laid out six pins on the board to replicate the first six GPIO pins and wired the corresponding ends of the ribbon to these pins. The RASClock therefore sits on the PCB about 2-3 cm from the GPIO. I've checked the voltages; they are correct and hwclock works fine.
I first set this up on Monday afternoon and on Tuesday morning the hwclock was about 15 mins slow. On Tuesday afternoon, I reset the clock and it was accurate on Wednesday morning. This morning I checked again and the clock was nearly an hour slow. I rechecked the installation instructions and corrected the error that I had made and the clock was about 1-2 mins fast after being shut down for one hour.

After each of the activities described above, the system was shut down and left for the periods indicated.

Can you suggest what might be wrong here?

TerryJC
Initiated

Posts: 36

Joined: Mon Nov 07, 2016 2:23 pm

Location: Wimborne Minster, Dorset

Terry

Postby Dave » Fri Aug 25, 2017 10:46 am

Hi Terry,

Are you able to swap the RASClocks over to see if the problem is unique to one device?
Also, just to double check the battery is in the right way round :)

I've contacted After Thought Software and they are hoping to have play with the new Rasbian Stretch to see if there is anything that needs tweaking. I'll keep you posted on how that goes.

Dave
ModMyPi

User avatar

Posts: 2083

Joined: Wed Jul 22, 2015 11:36 am

Forum Administrator & Technical Support

Postby TerryJC » Sun Sep 10, 2017 8:46 am

Dave wrote:Are you able to swap the RASClocks over to see if the problem is unique to one device?
Also, just to double check the battery is in the right way round :)

Sorry abou the delay in responding, I've only just discovered that the forum S/W doesn't automatically subscribe a user to a topic that they create....

The system only has one RASClock in it and the other is working full time keeping the chimes going! However, I did buy another RASClock and this exhibited the same symptoms.

What I have also done is to remove the IDC connector from the Pi and connected the RASClock directly to the GPIO Pins; same problem. Last night, I put the old Jessie SD Card into the new Pi's slot and this morning the time was correct; implying a stretch issue or something else that we are doing for the new project. To eliminate as much as possible, I have also removed the Python programs that are executed at boot-up; one at each of the two Pis in the system. That made no difference, although I didn't expect them to; they are simply reading patterns from GPIO pins and writing out results. They only access the time to write the date stamp next to the measurement.

The only other thing that we are doing is to have the 'master' Pi run an NTP server, so that the time can be synchronised on the two Pis. I'm currently thinking that this may be the issue.

TerryJC
Initiated

Posts: 36

Joined: Mon Nov 07, 2016 2:23 pm

Location: Wimborne Minster, Dorset

Terry

Postby Dave » Mon Sep 11, 2017 10:33 am

Hi Terry,

We've spoken to the manufactures of the RasClock just to ask them to double check compatibility with the latest Stretch OS, and they have confirmed that everything is working as it should.

You mentioned following the setup instructions for Wheezy, I would suggest using the Jessie instructions as Jessie is the most recent OS after stretch, Wheezy is pretty old now.

Are you able to test the RasClock on a completely fresh install of Stretch to confirm it works before adding anything else hardware/software related

Dave
ModMyPi

User avatar

Posts: 2083

Joined: Wed Jul 22, 2015 11:36 am

Forum Administrator & Technical Support

Postby TerryJC » Mon Sep 11, 2017 12:09 pm

Dave wrote:We've spoken to the manufactures of the RasClock just to ask them to double check compatibility with the latest Stretch OS, and they have confirmed that everything is working as it should.


OK. Thanks.

Dave wrote:You mentioned following the setup instructions for Wheezy, I would suggest using the Jessie instructions as Jessie is the most recent OS after stretch, Wheezy is pretty old now.

Actually that was my mistake when I typed the original message; I actually followed the instructions for Jessie.

Dave wrote:Are you able to test the RasClock on a completely fresh install of Stretch to confirm it works before adding anything else hardware/software related

That was going to be my next step. However, yesterday, I removed the code in ntp.config to set up an NTP server on the 'master' pi and a few minutes ago the time was accurate to within 1 second. Obviously, I need to monitor it for a bit longer, but it looks like something is interfering with the clock when the server is running (man hwclock --hctosys is quite voluble on this point, although it took me a while to get my head around what it was telling me).

I think that it's likely I've found the problem (with the RTC) and I need to work out how to prevent the NTP server affecting the time on it's own machine.

If I reach any conclusions, I'll report back here in case anyone else ever tries to do this.

TerryJC
Initiated

Posts: 36

Joined: Mon Nov 07, 2016 2:23 pm

Location: Wimborne Minster, Dorset

Terry

Who is online

Users browsing this forum: Bing [Bot] and 6 guests

Board index

The teamDelete all board cookies • All times are UTC [ DST ]