Odd electrical issue after solar PV installed...

[Ian S]

One of our digital clocks - the alarm clock in the bedroom - no longer keeps time properly i.e. it gains 20 to 30 minutes a day. At first I thought it was just starting to fail after a long (>17 years) trouble free life. But then I realized the problem coincided with the solar installation. Coincidence? Apparently not because it keeps perfect time when plugged in at my office. Apparently others have noticed such a problem but it's only some digital clocks. The oven, microwave, and cable box clocks all are fine. I even got an older digital clock dating back to the 1980's and it seems OK. I'll have to check some others. Has anyone else observed this?

[inetdog]

One of our digital clocks - the alarm clock in the bedroom - no longer keeps time properly i.e. it gains 20 to 30 minutes a day. At first I thought it was just starting to fail after a long (>17 years) trouble free life. But then I realized the problem coincided with the solar installation. Coincidence? Apparently not because it keeps perfect time when plugged in at my office. Apparently others have noticed such a problem but it's only some digital clocks. The oven, microwave, and cable box clocks all are fine. I even got an older digital clock dating back to the 1980's and it seems OK. I'll have to check some others. Has anyone else observed this?

I assume that you have a grid-tie system?
A digital clock that synchronizes to AC line frequency can be designed to count the voltage peaks (pretty robust) or to count the number of times the voltage passes through zero. If done badly, the latter approach is very sensitive to noise pulses when the voltage is close to zero. I suspect that your grid-tie inverter is putting just enough noise (in the form of current pulses?) onto the line that even with the grid behind you to stabilize it, the line voltage waveform where the clock is plugged in causes it to double or triple count occasionally.

It may help to put a filter between the clock and the outlet. Or just get a battery powered clock. If the noise is particularly bad and you have side effects on your radio or TV too, there may be a problem with your inverter.

[Sunking]

Well I assume you are off-grid battery? Many of the older digital clocks do use the AC line frequency as the clocking source. Today they do not. So if you are off-grid it is your inverter.

[Mike90250]

is it a transformer based inverter (90#) or transformer-less (30#) Model # would help. I assume grid tie, correct?

[Ian S]

Yes, this is a leased grid-tie system. Sunpower SPR-7000M inverter. 30 x 230 watt Sunpower panels. Been running for three weeks. This is the only odd behavior seen so far.

[Sunking]

Ian do you have a AM radio? Turn it on and try to tune in some stations. I bet you have a lot of trouble and hear the Bug Races. Your inverter it generating a lot of noise and harmonics, and your clock is picking up on it. The fix can be either expensive or cheap, your choice. The expensive way is to use a filtering device like an isolation transformer or get a different inverter

Cheap way is get another clock.

[Ian S]

Ian do you have a AM radio? Turn it on and try to tune in some stations. I bet you have a lot of trouble and hear the Bug Races. Your inverter it generating a lot of noise and harmonics, and your clock is picking up on it. The fix can be either expensive or cheap, your choice. The expensive way is to use a filtering device like an isolation transformer or get a different inverter

Cheap way is get another clock.

Thanks, I will check that out with the offending alarm clock as it also has an am-fm radio. My other radios are either battery powered or run through a wall-wart. I will also check with my Onkyo receiver which has AM capability. I suppose I should compare at night when the PV system is "resting" with daytime around noon. Of course AM reception is also different at night. Oh well, it's more of a curiosity than a real concern as it only seems to affect that one clock. I think I'd like to get one of those alarm clocks that display the time on the ceiling anyway! LOL.

[Sunking]

Yes silly you will have to test during the daylight hours.

Now with the radios or RFI there are two ways for inference to enter the radio. Either over the air via antenna pickup, or induced/injected by the power source. Educated guess is you will have both.

[Ian S]

Yes silly you will have to test during the daylight hours.

Now with the radios or RFI there are two ways for inference to enter the radio. Either over the air via antenna pickup, or induced/injected by the power source. Educated guess is you will have both.

Peak solar production just now and the AM radio works fine with no noticeable noise on a fairly weak local station.

[Sunking]

OK that is a good thing as it is not radiated. It is entering through power line.

[billvon]

One of our digital clocks - the alarm clock in the bedroom - no longer keeps time properly i.e. it gains 20 to 30 minutes a day. At first I thought it was just starting to fail after a long (>17 years) trouble free life. But then I realized the problem coincided with the solar installation.

Clocks use the power line frequency to maintain timing. They look at the zero cross - that place on the waveform where voltage goes from + to - on the AC hot. That happens 120 times a second so the clock divides by 120 to get the "seconds" count.

Many solar inverters cannot run all the way down to the zero cross, and there is a short time during the "switchover" where the solar effectively disconnects. (Power goes into a storage capacitor during this time.) If the reconnection causes a spike near the zero cross, the clock can see an extra cycle or two occasionally and run fast.

One way to fix this is with a power line conditioner. If that's your problem this is pretty much guaranteed to work:

http://www.tripplite.com/en/products/model.cfm?txtModelID=226

It's also way overkill. If you don't mind doing a little wiring this one will also likely work:

http://www.digikey.com/product-detail/en/03DKAG5/603-1653-ND/2744296

[Sunking]

One way to fix this is with a power line conditioner. If that's your problem this is pretty much guaranteed to work:

http://www.tripplite.com/en/products...txtModelID=226

:ROFL:

$100+ to to fix a outdated clock. By a new digital clock for $10. The new models do not use line frequency as the timing source.

[Ian S]

Clocks use the power line frequency to maintain timing. They look at the zero cross - that place on the waveform where voltage goes from + to - on the AC hot. That happens 120 times a second so the clock divides by 120 to get the "seconds" count.

Many solar inverters cannot run all the way down to the zero cross, and there is a short time during the "switchover" where the solar effectively disconnects. (Power goes into a storage capacitor during this time.) If the reconnection causes a spike near the zero cross, the clock can see an extra cycle or two occasionally and run fast.

One way to fix this is with a power line conditioner. If that's your problem this is pretty much guaranteed to work:

http://www.tripplite.com/en/products/model.cfm?txtModelID=226

It's also way overkill. If you don't mind doing a little wiring this one will also likely work:

http://www.digikey.com/product-detai...653-ND/2744296

Easier to get a new alarm clock! Really, the main reason I bought this one years ago was for its large digits. Now I'm thinking I'd prefer one that projects the time onto the ceiling. Anyway, I guess what I found so odd was that other digital clocks in the house aren't affected. Not even the alarm clock that was replaced with the current one. I have a couple of old VCRs that I'll check too.

[billvon]

:ROFL:

$100+ to to fix a outdated clock. By a new digital clock for $10.

Or $9 for the second line filter.

About half of "modern" digital clocks use 60hz as a timing reference, if you include the $9.99 Walmart clocks. It's just plain cheaper than a crystal.