Smart meter causing damage to electronics?

[Sierrasparky]

What in these new smart meters with solid state relays would cause damage to electronics on remote turn off.
I have had a couple of querry on the issue.
What could be going on with the solid state relay. Could the break be very slow with diminished voltage or current? ( slow as in too many a/c cycles before complete off)

 

[gar]

130515-1313 EDT

Sierrasparky:

Why do you think there is a solidstate relay in the Smart Meter to break the hot lines? Who told you this, or where did you see one? How much power does 1 V at 200 A equal?

Whether the relay is solidstate or not how does its operation affect loads that are connected that is different than opening the main breaker?

Also you should note that SCR or Triac devices under normal circumstances only open at a current zero-crossing.

.

 

[GoldDigger]

What in these new smart meters with solid state relays would cause damage to electronics on remote turn off.
I have had a couple of querry on the issue.
What could be going on with the solid state relay. Could the break be very slow with diminished voltage or current? ( slow as in too many a/c cycles before complete off)

Some of the smart meters are spec'd by the POCO to include a remotely actuated contactor to allow them to shut off the customer without leaving the office. But whether it is solid state or mechanical I do not know.
The smart meter itself will put an RF/pulse communication signal onto the wires periodically to report its information to the central office. I suppose that RF noise could accidentally activate some RF remotes, but not IR remotes. There could also be problems with interference in either direction between the smart meters and WiFi devices.

I do not know to what extent timing of turn off of the service interrupter in the meter might cause problems, but there is as great a likelihood that damage would happen at turn on as at turn off IMHO.

 

[Sierrasparky]

These are Itron called openway Meters. Apparently when a call when out to remove a lock ring at a condo complex instead they sent out remote shut off info over the network.
Several electronic stuff had fried power supplies. All the stuff were plugged into APC UPS. The stuff plugged into the UPS survived, however the monitor plugged into the UPS no Batt , surge only fried. It was just the power supply on cord.

No electronics were plugged in prior to turn on. All components in the units were physically unplugged.

 

[GoldDigger]

These are Itron called openway Meters. Apparently when a call when out to remove a lock ring at a condo complex instead they sent out remote shut off info over the network.

For those who have not followed the other thread here, the consensus is that nobody would trust the remote shutoff to make the panel safe to work on! Sounds like POCO needs to train their service dispatchers better.
I would look on the Internet to see if there are any similar experiences, since I have not seen any other reports here yet.

I would be more inclined to suspect contact bounce than slow disconnect. Capacitor input power supplies are very vulnerable if there are also inductive loads on the load side that could cause spikes on disconnect. Since wall warts cannot be powered off, they should be unplugged instead.

 

[Sierrasparky]

I would never rely on the remote turn off instead of pulling the meter.

 

[gar]

130515-2344 EDT

Sierrasparky:

You did not answer my question on how much power 1 V at 200 A represents. That is 200 W. Since there could be 200 A flowing thru each leg in the meter that is a total of 400 W. How hot would it get inside a package 6" in diameter, and possibly 6" high? Very high. There can be no solid-state relay in a Smart Meter. There is no place to heat sink that amount of power. Thus, you will find an electromechanical contactor in the meter.

The opening of an ordinary contactor is not going to be much different than opening a breaker, pulling fuses, opening a switch, or a tree falling on and breaking a service line.

Don't blame remote turn off for the problem. The problem is no transient protection and poor design of the devices that failed. Or possibly turn off did not cause the failure.

However, your description does seem to imply turn off may have been the cause. But to get large voltage transients with a lot of energy at turn off I would expect that a substantial inductive load was present at turn off.

.

 

[hurk27]

These are Itron called openway Meters. Apparently when a call when out to remove a lock ring at a condo complex instead they sent out remote shut off info over the network.
Several electronic stuff had fried power supplies. All the stuff were plugged into APC UPS. The stuff plugged into the UPS survived, however the monitor plugged into the UPS no Batt , surge only fried. It was just the power supply on cord.

No electronics were plugged in prior to turn on. All components in the units were physically unplugged.

Sounds like there was a problem with the electric that required the power to be shut off, I suspect that maybe a bad neutral connection was the problem that the person called to turn off the electric and repair or was still in the troubleshooting stage, I can not believe that the turning off the smart meter would have caused any damage like this, if it could then there would have been many other complaints like this, and a neutral loss is about the only way you can get an over voltage to electronics that can cause this kind of damage in a home, and a home meter doesn't even have a connection for the neutral to the plug in meter so there is no way that it could possibly cause a neutral failure, did you have any first hand testing at the site? what was the reason they needed the meter to be removed?.

Even if the equipment was undervoltaged because of a slow disconnect I have never seen this kind of damage done to that many pieces of equipment, most electronics would not have a problem with undervoltage, this is more of an over voltage problem which is why I suspect that it was a neutral problem.

I have electronics running on inverters that have a soft start that slowly raises the voltage up to full voltage, if this would be a problem then I would have had fried electronics a long time ago.

 

[Sierrasparky]

We were out in the are doing inspections of the old equpement as over the winter there had been main breaker burns at another building in the complex. I was hired to inspect as preventaive measures. We needed lock rings removed just to get to the Main breakers. I was at the managers unit when this happend the lock rings were supposed to be pulled the day before. In the past this has not been an issue.
So if these are not Solid state then the Itron guy at the trade show did not know what he was talking about. They sure looked solid state to me.

 

[GoldDigger]

We were out in the are doing inspections of the old equpement as over the winter there had been main breaker burns at another building in the complex. I was hired to inspect as preventaive measures. We needed lock rings removed just to get to the Main breakers. I was at the managers unit when this happend the lock rings were supposed to be pulled the day before. In the past this has not been an issue.
So if these are not Solid state then the Itron guy at the trade show did not know what he was talking about. They sure looked solid state to me.

If they are solid state, by using appropriate FET and/or SCR circuitry you can get a voltage drop a lot less than 1 volt across the relay. But if it is solid state, it would just about have to be the kind that only turns off as the current goes through zero and so will not have any cycles of delay in the turn off.
There could still be inductive current flowing on the load side of the relay at the time it opens, so damage might still occur unless snubbers or surge protectors were in place.

 

[gar]

130516-1507 EDT

Sierrasparky:

Look at this datasheet http://www.pwrx.com/pwrx/docs/T627--20.pdf and price http://www.newark.com/powerex/t627062064dn/thyristor-scr-200a-600v-to-200ab/dp/09B4181. Look at the voltage drop curve. At 200 A about 1.6 V. Look at the dimensions. About 1.5" in diameter and 0.5" high without any heat sink. If SCRs are used it will take four. Triacs two. Triac voltage drop is in the same ballpark.

None of this begins to cover the huge heat sink required.

Also a solid-state device will require some energy to keep it turned on. A latching mechanical relay requires no energy except to set or reset it.

No way can you put a solid-state relay (contactor) in a smart meter package of the present size, nor would you want to do so. Cost, size, heat, and no useful technical advantage. An electromechanical relay is probably more reliable in this application than a solid-state relay. That said, the relay in smart meters is probably rather wimpy for the job, and may be the cause of some of the reported meter fires. I have not seen the insides of a smart meter, nor a fire damaged meter. But there are some things with a little common sense and some background that identify unworkable concepts.

.

 

[gar]

130516-1530 EDT

GoldDigger:

See http://www.st.com/st-web-ui/static/active/en/resource/technical/document/datasheet/DM00066266.pdf Table 7. Shows 1.5 V at 138 A.

There could still be inductive current flowing on the load side of the relay at the time it opens, so damage might still occur unless snubbers or surge protectors were in place.

Consider an SCR with an inductive load including internal resistance in the inductor. Thus, a series LR load on the SCR.

Turn the SCR on and current flows. Assume a sine wave voltage source. When the source voltage goes to zero there is still current flowing in the load. The inductor creates a voltage that attempts to keep current flowing at the same magnitude and direction. This puts a negative voltage on the SCR cathode that keeps the SCR turned on. This continues until the stored energy in the inductor is dissipated. At that point current is zero and the SCR turns off. There is no damaging voltage transient. One reason to use an SCR switch.

You could have a more complex load circuit with capacitors and inductors such that the SCR sees zero current while some inductors still have current flowing. But in this case there won't be any serious voltage transients because the energy will be shifting around between the capacitors and inductors.

.

 

[Jraef]

I agree with everything gar has said. In fact, here is an interior view of an Itron smart meter (from a company touting their comm board used inside).


Unfortunately because the purpose was not to show the power disconnect, you can't see it clearly. But in the description it says:

200A contactor driver (pulsed magnetic latching type)

That pretty much closes the book on that part of this issue, it is NOT a solid state switching device, it is electro-mechanical.

Now that said, that is one heck of a small box to put even an electro-mechanical 200A contactor into. Sure, they don't need UL wire bending space, but still, I have done heat run tests on 200A contactors, no way that would have passed. I personally think that in the rush to get these put in place in the installed base and the cost cutting needed to have them backward compatible with existing non-switching meter sockets, they have over stepped the prudent boundaries of equipment design standards to which all other equipment mfrs are held to. More likely, in my guesstimate, these contactors are over heating under full load, the resistance is increasing as a result, which drops the voltage, possibly on only one of the 2 poles and causes issues with downstream devices.

 

[GoldDigger]

... More likely, in my guesstimate, these contactors are over heating under full load, the resistance is increasing as a result, which drops the voltage, possibly on only one of the 2 poles and causes issues with downstream devices.

Or, more generally, the contactors, or one pole of the contactor, may fail catastrophically or even simply open one pole without all the other poles opening.

The description of magnetically held indicates that if the contactor changes state on its own, without a control signal (failure), it will stay in that new state until it receives a new command or fails again, but in the opposite direction.

 

[Sierrasparky]

That is great info.
I looked into the Itron website and I did see that the comm board has a battery that lasts only 12 years. Wow so they will be srvicing these in 12 years or so.
I have seen meters in place for 30-50 years and more without replacement.
Who is going to pay that bill.

ME

 

[GoldDigger]

I have seen meters in place for 30-50 years and more without replacement.

I bet they replaced the meter reader a few times during that period though.

 

[Jraef]

Or, more generally, the contactors, or one pole of the contactor, may fail catastrophically or even simply open one pole without all the other poles opening.

The description of magnetically held indicates that if the contactor changes state on its own, without a control signal (failure), it will stay in that new state until it receives a new command or fails again, but in the opposite direction.

It would only need 2 poles, presuming a single phase service, and if one failed to operate it would just kill power to half of your circuits in the house.

Yes, in the past a latching contactor would have had a "Set" coil and a "Reset" coil. I have seen people using DC electronic coil designs now though where there is one bi-directional coil and they change the polarity of the pulse to make it set or reset, like a toggle switch with a plunger. Not sure how they are doing it here though. It's also not clear if the 2 poles would be mechanically linked so that only one operating coil was necessary, or if they would use two separate coils. I would guess one coil and mechanically linked contacts.

I'd like to get my hands on one and take it apart...

 

[Jraef]

Ah, just thought of something. :slaphead:
Vacuum contactor! Could it be?
They are usually expensive, but in a high enough volume (like EVERY household?) it might be worth it because they could be made very small and mechanical latching is a very common strategy for those.

 

[hurk27]

I don't know, I still am having a hard time beliving that a meter or any disconnect device that is only connected between the hots in the meter base could cause the failure of so much electronics, it just doesnt make any sense, meters are pulled every day under load and this does not happen, even if there were many motors on the circuit maybe one or two electronic items that didn't have any surge protection, but I don't see several, and the fact that there were UPS units inplace that would have surge protection on its line side which can protect up stream as well as down stream for a short distance, if the meters were wiggled in trying to remove them maybe a bad neutral bus connection.

We installed some T&B meter stacks that were so poorly design that even just moving the meter could cause a neutral loss, the screws that held buss jumpers together were steel and the bus were aluminium, T&B actully used self taping steel screws into the aluminium bus that would become loose over time and strip out, the small aluminium neutral conductor strips that were screwed to the main neutral bus was also connected to the meter insert so moving the meter would cause this neutral strip to loose connection at the screw connection to the main bus, our liability company had to shell out a few bucks to replace these peoples equipment then they went after T&B, we even had a surge strip burn the carpiting as the MOV's in it got so hot that the plastic case it was made of melted and started burning, one thing that threw me off was the damaged equipment was on both phases, and it wasnt till the home owner told me she was using the microwave at the time, and it popped and went off when the fridge tried to start up so when the microwave failed the surge seesawed from one leg to the other, she said the lights in the kitchen which was on two circuits on oppsite legs went from one set being really brite and the other set dim, to switching to the dim set going brite while the bright set went dim when the micorwave popped and went out.

First time I had ever had a bad neutral damage electronics on both legs of a panel.

Other then a lightning strike the only time I have ever seen multible equipment failures are long term failures cause either by a bad neutral connection or when a higher voltage power line came into contact with a lower voltage local line, in two of these cases it was a 69kv line that came down on a 7200 local line and surge the whole east side of our city, and one where a transformers primary lost a few turns as several turns shorted together lowering the turn ratio, but it only raised the voltage to 156 volts so only a few electronics was damaged like the furnace control board a some micro waves and a stoves controls.

In all of these the common thing was the fact that all these are what I call a long term surge or over voltage, and is most likely the cause when lightning can be ruled out as the cause.

I have had many bad main breakers that were in the procces of failing as they were arcing on the bus connection or in the contacts or at the terminals and still not have electronics damage like this, just doesn't add up.:?

 

[kwired]

I agree with everything gar has said. In fact, here is an interior view of an Itron smart meter (from a company touting their comm board used inside).

View attachment 8542
Unfortunately because the purpose was not to show the power disconnect, you can't see it clearly. But in the description it says:

That pretty much closes the book on that part of this issue, it is NOT a solid state switching device, it is electro-mechanical.

Now that said, that is one heck of a small box to put even an electro-mechanical 200A contactor into. Sure, they don't need UL wire bending space, but still, I have done heat run tests on 200A contactors, no way that would have passed. I personally think that in the rush to get these put in place in the installed base and the cost cutting needed to have them backward compatible with existing non-switching meter sockets, they have over stepped the prudent boundaries of equipment design standards to which all other equipment mfrs are held to. More likely, in my guesstimate, these contactors are over heating under full load, the resistance is increasing as a result, which drops the voltage, possibly on only one of the 2 poles and causes issues with downstream devices.

Maybe the POCO's are betting on the fact that many of them will never be loaded to 200 amps very often.