Digital Electricity Meter with 32khz oscillator not keeping time

Thread Starter

Mike Moran

Joined Jan 1, 2019
3
hi guys,
would be interested in your thoughts on this.
we have a customer where the time keeps going out of sync (by several hours) on their khw electricity meter. the meter has a real time clock driven by a 32khz oscillator to facilitate different metering tariffs.
The meter has been replaced with an identical one, and within a few weeks the same issue is apparent (wrong time).
we have several thousand of these meters installed at other sites and never had this issue.
some form of meter tampering is suspected, but it is hard to see how this is happening as the unit is heavily sealed and these are all intact.

Onr possibility is that the frequency of the mains line supply is been altered by the customer to cause malfunction of the meter. possibly by hooking up a variable frequency motor drive to the incomming supply. could supplying the unit and presumably its onboard DC power supply witha frewuency much lower or higher than intended fool the onboard clock?
Appreciate any thoughts.
thanks guys.
mike
 

Alec_t

Joined Sep 17, 2013
10,429
Welcome to AAC!
the time keeps going out of sync (by several hours)
Is the meter time always lagging real time?
Is the RTC oscillator screened well enough to prevent it being affected by external fields?
Is the error always in the customer's favour, tariff-wise? If not, it suggests there is no fraudulent intent.
 

Thread Starter

Mike Moran

Joined Jan 1, 2019
3
One of our guys noticed a motor control variable frequency drive nearby and that’s why we suspected it may be related. It could generate an output of up to 500hz - the meter is designed for use with 50-60hz.

The variation in time could be in the customers favour, although it is entirely possible that there is no malicious intent. Still an unusual occurrence. It is a residential setting and electrically very ‘normal’

What I am wondering is could a vastly different input supply have an effect on the timing oscillator? I imagine this would be siginificantly downstream of the incoming supply. Or what would be the lightly impact otherwise of such a incoming supply frequency? I would have thought the electronics / processer within the meter would be somewhat isolated from the mains supply.

Thanks for the responses guys.
 

Thread Starter

Mike Moran

Joined Jan 1, 2019
3
Welcome to AAC!

Is the meter time always lagging real time?
Is the RTC oscillator screened well enough to prevent it being affected by external fields?
Is the error always in the customer's favour, tariff-wise? If not, it suggests there is no fraudulent intent.
I am not sure how well the oscillator is screened, but the meter has high magnetic field detection as a security feature and this is not being triggered. - it seems unlikely a high magnetic /rf field is being blindly induced into the front of the meter itself.
 

MaxHeadRoom

Joined Jul 18, 2013
19,154
May be an advantage to place a common mode choke in the service supply after the meter, one of the recommendations with VFD instructions is one on either supply in or 3phase out, or both.
Max.
 

Hymie

Joined Mar 30, 2018
710
Obtain an AC power source and see if supplying a meter at higher frequencies (500Hz) causes the clock/time errors you are seeing.
If the unit is using a crystal based time signal, why is the mains frequency upsetting it?

Many moons ago, one of my previous residences had an economy tariff (from midnight – 6am), it was controlled by a mechanical (electrically driven) timer located close to the incoming meter. As a result of power cuts and a deliberate power outage while I replaced the consumer unit – it became completely out of sync, such that my economy usage was similar to the normal tariff.
None of the meter readers who called noted this discrepancy, neither did the power company question how I managed to use so much energy between midnight and 6am.
 

Hymie

Joined Mar 30, 2018
710
It is possible that the electricity meter is using the mains frequency for its primary time keeping, and that the internal crystal is there for when the power fails.

With the mains frequency typically varying by up to 1%, you might think that using the mains frequency would be a dumb idea given the accuracy of quartz crystals – however most national grid systems regulate/control the number of mains cycles over a weekly period, such that clock systems using the mains frequency for their time keeping will actually be very accurate over a long period of time.

If the above explains the situation with regards this meter installation (and the homeowner is deliberately manipulating the meter time in this way), it could be difficult to prove in court – and a simpler option might be to change the meter firmware to ignore mains frequencies above say 65Hz for the purposes of timekeeping.
 
Top