Effects of clouds on your array? I'm considering SMA's Secure Power Supply inverters.

[nomadh]

Don't know how real any of the info is but it seems plausible. So you are saying the optitrac is pretty much what most of the other string inverters are already doing? It seems sunnyboy is claiming the technology gets them into the ballpark of power op modules or even microinverters. The info I'm looking for now is a head to head of sunnyboy optitrac vs microinverter vs solar edge.

[inetdog]

More info on optitrac

http://forums.whirlpool.net.au/forum...m?t=1697128#r9


In summer, I have no shading, however now the sun is lower, I get shading until about 9:30am.

I initially had OptiTrac off, as I had no shading. However I turned it on and you can see the difference in the following graph.

http://pvoutput.org/comparelive.jsp?sid=1006&df=20110601&dt=20110602

Both days were sunny in the morning. Here is a direct comparison:
Without Optitrac @ 8:20am, 1/06/2011: 384W
With OptiTrac @ 8:20am, 02/06/2011: 971W


or:

I also noticed a small improvement, but I haven't quantified it.
My panels are shaded 2-3pm in winter. Before before enabling OptiTrack, the worst I got was about 10% of the unshaded output. After enabling OptiTrack it looks like it gives up on the shaded panels and the MPTT follows the unshaded string, giving closer to 50% of the unshaded output.

Thanks for the experimental results nomadh.

One thing that readers should be aware of is that there are a great number of ways to implement MPPT, and some of them (e.g. full width scan) do what optitrac does as part of their basic operation and do not need any additional code.
The basic MPPT algorithm used by SMA is subject to capture by a local maximum on the high voltage/low current end of the curve, so for them Optitrac (TM) does add a significant benefit. It does not necessarily translate into a competitive advantage over anything but their own older inverters.

[nomadh]

More info on optitrac

http://forums.whirlpool.net.au/forum...m?t=1697128#r9


In summer, I have no shading, however now the sun is lower, I get shading until about 9:30am.

I initially had OptiTrac off, as I had no shading. However I turned it on and you can see the difference in the following graph.

http://pvoutput.org/comparelive.jsp?sid=1006&df=20110601&dt=20110602

Both days were sunny in the morning. Here is a direct comparison:
Without Optitrac @ 8:20am, 1/06/2011: 384W
With OptiTrac @ 8:20am, 02/06/2011: 971W


or:

I also noticed a small improvement, but I haven't quantified it.
My panels are shaded 2-3pm in winter. Before before enabling OptiTrack, the worst I got was about 10% of the unshaded output. After enabling OptiTrack it looks like it gives up on the shaded panels and the MPTT follows the unshaded string, giving closer to 50% of the unshaded output.

[nomadh]

The secure power feature seems to be about as good as hoped first glance. It does seem to put out 1500 watt and gracefully shutsdown when overtaxed and can function is less than ideal conditions. Turns on around 800 watt dc and needs little headroom to be functional. Now to see how well their power pt tracking works.

some info from the sma thread quoted below



Testing on a rainy day, the total wattage of my panel system was about 670 watts on the inverter screen (grid-tied) and was slowly rising as the clouds were moving out. I shut off the grid disconnect and started the SPS at this wattage. The charger kicked in at about 700 watts registering on the “stand-alone ops” status – with a reading of 13 DC amps, quickly settling at 11 amps on the charger. The inverter continued to show 700 watts and started cutting back as the DC amps started coming down. When I shut down the SPS and reconnected the inverter to the grid, the wattage reading on the inverter screen was around 800 watts. This indicates to me that the SPS can start a golf cart type charger with little “surge” capacity required and can consistently charge the batteries with anything above 800 watts available from the PV panels. This would of course vary with the type of charger, but again appears that the computer controlled type charger of the typical club car will work very well with the Sunny Boy SPS. - See more at: http://www.smainverted.com/2013/05/1....3VR8Ql8K.dpuf


Follow-up on SPS testing;
Full sun testing – Cart charging 12 amps (48 volts) – 715 standalone watts
Added Low fan hair dryer – 795 watts
low fan, warm hair dryer – 964 watts
high fan, warm hair dryer – 1383 watts (9 amps on charger)
low fan, high hair dryer – SPS kicks off (this is a 1850 watt hair dryer)
Comments; Under cloud cover, the cart charger will lower it’s charging amps and corresponding “standalone Ops” watts to as low as 460 watts, then would come back up to 523 watts as the cloud cover lifted. Cannot explain this, but it consistently happens. Also, the charger will click off on occasion, but comes back on within a few seconds – not the 45 seconds or so when the SPS is overloaded.

Again, appears the SPS is very well suited to charge a rather large set of 4-12V deep cycle batteries in the Club Car Golf Cart using the charger designed for the cart. Due to continuing downward amperages as the batteries get more fully charged, it could take most of the day to fully recharge depleted batteries – especially if there is substantial cloud coverage.

- See more at: http://www.smainverted.com/2013/05/1....3VR8Ql8K.dpuf

[home4]

A few tidbits are available from SMA in the comments of their product launch page. Greg Smith works for SMA.

Thanks a lot for that thread. There is a lot of good info there.

I heard back from SMA. They claim that although there is no buffer built into the inverter, the DC->AC conversion abilities match the normal grid-tied efficiency of the inverter, so for example a 500W load would require about 512W from the PV.

They also mention that a UPS could possibly aid in surge loads beyond the 1500W. That other thread mentions that someone had been successful in using this UPS, connected to the Secure Power Supply, to allow starting up a newer-model refrigerator.

[nomadh]

This off grid secure power seems to be a near zero cost feature on an already competitively priced inverter. Just a few bucks to mount an extra ac socket in the garage and a 100 ft extension cord it seems. At least I can keep the koi pond breathing for a few hours during the day or pumping water if a flood or the mini fridge or small ac unit if hot. Hell if its just an easy way to charge the cell phones and laptop it would be nice. If this feature is near the 97% efficiency of the main inverter most people should be able to get their 1500 watts during much of the daytime. How effiecient it is is a great question.

I was sold on enphase but ran across this feature on the sunnyboy. After talking to a solar sales company they are working up a quote with 1/2 enphase for the shaded part and half sunnyboy for this feature. He thinks they would be very compatible together and should be a pretty good price.

[Mike90250]

on a cloudy day, my KW of arrays produce 300-500 watts. It's a pretty drastic cut off when clouds come in, power drops to zilch.

[sensij]

Is this true in your experience? This is the exact question I have been seeking to get answered in this thread. I have seen this repeated elsewhere, that if a small cloud comes in the output from the Secure Power Supply would go off. But it doesn't seem to me like this would be the case. Say I have a 6 kW PV array and the Secure Power Supply, there is an outage and I'm drawing 500W powering some loads. The Secure Power Supply would not go off unless it is no longer able to produce that 500W AC, which means that the array would need to be producing maybe 800W DC (or not even). In your experience, if a cloud goes over your array will its output drop by 85% (6kW to 800W)? I'm not trying to argue this point -- I really do not know if it would, as I do not currently have a solar PV system, but I'm looking to find out that answer because this seems to be key in how well this system would work. When I read online about the effect of cloud cover on PV it seems as though the entire sky would have to be quite cloudy for it to drop that much.

The effect of clouds on your power production is hard to predict. There are two components... how much light is available to be converted, and how well your panels perform in low light conditions.

NREL makes daily irradiance data freely available from their test stations here.

Data from the LA station on September 28 looks pretty cloudy, and the power from your panels would be intermittent, even if they were perfectly efficient.

On the other hand, September 29th looks clear.

Since all clouds (and cloudy days) are not created equal, it is really difficult to predict how your system would perform under the exact combinations of events that would lead to the SPS function being necessary. If you want to take a chance that the zombie apocalypse will only occur on a bright sunny day, that's your choice. I would choose something more reliable than solar to depend on.

[sensij]

Yes, that would be good to know. I will attempt to contact SMA and see if they can provide more information.

A few tidbits are available from SMA in the comments of their product launch page. Greg Smith works for SMA.

Hi Ralph,
There is no surge capacity for the Secure Power Supply (SPS) – if you pull more than the available power from the unit, the inverter will disconnect the SPS, wait 20 seconds and then attempt to reconnect.

It will continue to do this until:
1. Loads are disconnected
or
2.More sun becomes available to support the connected loads (up to 1500W).

Thanks,
Greg

Ralph
09/30/2013 at 12:58 pm

Hi Greg,
Thanks for the information. Since all of the units could output > 1500W, is there a reason why 1500W was selected as the cutoff?

Greg Smith
10/02/2013 at 4:29 pm

Hi Ralph,

You’re welcome! Since most electronic equipment is sensitive to voltage fluctuations, it was an engineering decision to choose 1500W, that way the inverter can utilize the capacitor bank to create a stable voltage for the SPS socket.

Best,
Greg

This exchange suggests that the conversion efficiency for the SPS is the same as for normal grid tie, since the power available on the "inverter screen" matched the power available from the SPS.

Mark Claussen
09/01/2014 at 1:57 pm

Justin – this is a copy of what I have reported on the Northern Arizona Sun and Wind forum – seems to be really good one!

Additional Sunny Boy 4000 TL-US inverter data;
48 volt Club Car charger test;
Charger consistently starts on the SPS and registers about 13 DC amps and about 700+ watts on the inverter screen. Normally the DC amps will drop to 8 amps or less fairly quickly depending upon state of charge of the batteries
Testing on a rainy day, the total wattage of my panel system was about 670 watts on the inverter screen (grid-tied) and was slowly rising as the clouds were moving out. I shut off the grid disconnect and started the SPS at this wattage. The charger kicked in at about 700 watts registering on the “stand-alone ops” status – with a reading of 13 DC amps, quickly settling at 11 amps on the charger. The inverter continued to show 700 watts and started cutting back as the DC amps started coming down. When I shut down the SPS and reconnected the inverter to the grid, the wattage reading on the inverter screen was around 800 watts. This indicates to me that the SPS can start a golf cart type charger with little “surge” capacity required and can consistently charge the batteries with anything above 800 watts available from the PV panels. This would of course vary with the type of charger, but again appears that the computer controlled type charger of the typical club car will work very well with the Sunny Boy SPS.

I don't think anyone here is saying that the SPS doesn't do what it says it should do. It is just that the conditions in which the SPS is useful are limited. As I see it, all of the following must be true:
1) You think it is worth the extra expense to wire in the appropriate switches and UPS systems (The cost of the SPS feature itself is negligible, but integrating it is not).
2) You want to insure against long term total power loss in which the gas required for a standard generator exceeds your storage and is otherwise unavailable (but the sun is mostly shining).
3) You are unwilling to spend the money on a true hybrid inverter or see a need for more than 1500 W for a few hours a day.

[home4]

Yes, that would be good to know. I will attempt to contact SMA and see if they can provide more information.

[inetdog]

The one thing missing in the SMA specification is the ratio of DC panel output (not DC input to the SPS) to the AC output of the ACS.

If the SPS does not contain any significant (multiple cycles at 60Hz) energy storage, then the peak available panel output power would have to be about twice the needed AC output power at any given time.
(The peak current from the panels would have to match the peak of the AC waveform, which is 1.414 times the average current for a linear load.)
Depending again on the degree of energy storage at the input of the SPS, the SPS might also have problems with low power factor loads or peaky loads like rectifier inputs in power supplies. Not to mention motor starting surges.

That is consistent, FWIW, with the 1500W SPS being installed with the SBx000-TL-US where the smallest size in the family takes no more than 3000W of panel.

So as long as the panels are capable of 800+ watts at the worst moment of cloud coverage, the SPS should be able to deliver 400 watts reliably.
And, of course, if the SPS is forced to shut down it will automatically try again in, I believe, five minutes, in the hope that the clouds have passed and it is not night yet.

[home4]

Even if you could use 100% of the solar panel output when the grid is down it wouldn't be a continuous flow of electricity. Even a small cloud will drop a pv system output so your lighting and tv and ac would be going on and off depending on how good the sun is shining. The only why to get the system to be continuous is to use batteries which are terribly expensive.

Is this true in your experience? This is the exact question I have been seeking to get answered in this thread. I have seen this repeated elsewhere, that if a small cloud comes in the output from the Secure Power Supply would go off. But it doesn't seem to me like this would be the case. Say I have a 6 kW PV array and the Secure Power Supply, there is an outage and I'm drawing 500W powering some loads. The Secure Power Supply would not go off unless it is no longer able to produce that 500W AC, which means that the array would need to be producing maybe 800W DC (or not even). In your experience, if a cloud goes over your array will its output drop by 85% (6kW to 800W)? I'm not trying to argue this point -- I really do not know if it would, as I do not currently have a solar PV system, but I'm looking to find out that answer because this seems to be key in how well this system would work. When I read online about the effect of cloud cover on PV it seems as though the entire sky would have to be quite cloudy for it to drop that much.

For me personally, I'm interested in this system because I want SOME sort of backup in case of a long-term power outage. I have a generator, and that is very useful, but I have friends who went without power for 2 weeks a couple of years ago, and it would be really really nice to have even a little power for things like lights, radio, phone, etc, during a time like that. During Hurricane Sandy, gas was nearly impossible to come in certain areas. A PV solution would not provide power during the storm, but in the aftermath, it could be very useful.

It would be great to have a fully islanding system with batteries during a time like this, but unfortunately it seems like these systems are really really expensive. Unless there is an inverter I haven't read about, even without batteries they are expensive because the inverters cost so much more, and they usually restrict the panels you can get.

The Secure Power Supply seems to me like it provides some of the benefits of an islanding system, with a lot less cost, and also more restrictions. But with a UPS and a generator, it seems like it could be a good choice, and that's why I'm interested in finding out more about it. Thanks.

[SunEagle]

I understand the safety aspect of not feeding power back into a broken grid. A simple sensor and a relay at the grid portal are hardly the most complex or expensive parts of a pv inverter system. And if a $79 battery backup can have its own freewheeling inverter it seems most of the complexity is already included in a grid tied inverter. I have a hard time believing most people don't understand the benefit here. We will all have 4 or 7 or even 11 Kw of power generating on our rooftops for 5 hours a day but in an emergency no one is even interested in tapping it? Does everyone think these oversized overextended big government / power company systems will always swoop in and get things working again in 1-2 days? Sunnyboy states that the limited islanding of their system was a requirement to sell in Japan after a few of their natural disasters left people without power for a very long time. Haven't we had people here is the US without power for weeks? What about earthquakes or terrorist attacks or economic meltdown or even zombie apocalypse ? There are experts that think our power infrastructure is especially teetering on collapse add a well placed terrorist strike and it could be months. I think this would work very well with a very small generator for night/storms. Also sunnyboy is supposed to be a very good brand, right?
I can't afford to double the price of my system but if it is a very small increase or even a decrease from a microinverter and the power optimization can be within 1% of a microinverter. I find that very tempting. I'm still trying to figure out details. It seems it could be about $1500 less than enphase on a 5.7 kw system but I'm not sure if that included the power optimizer modules.

Even if you could use 100% of the solar panel output when the grid is down it wouldn't be a continuous flow of electricity. Even a small cloud will drop a pv system output so your lighting and tv and ac would be going on and off depending on how good the sun is shining. The only why to get the system to be continuous is to use batteries which are terribly expensive.

It is much cheaper to run a generator for those long outages than to have a battery system of comparable kw size to provide you emergency power.

Most cost affective plan (for most people) is for a solar pv grid tie system and a backup generator for the long outages. For others just use the Utility grid power and then a generator during an outage.

[nomadh]

I understand the safety aspect of not feeding power back into a broken grid. A simple sensor and a relay at the grid portal are hardly the most complex or expensive parts of a pv inverter system. And if a $79 battery backup can have its own freewheeling inverter it seems most of the complexity is already included in a grid tied inverter. I have a hard time believing most people don't understand the benefit here. We will all have 4 or 7 or even 11 Kw of power generating on our rooftops for 5 hours a day but in an emergency no one is even interested in tapping it? Does everyone think these oversized overextended big government / power company systems will always swoop in and get things working again in 1-2 days? Sunnyboy states that the limited islanding of their system was a requirement to sell in Japan after a few of their natural disasters left people without power for a very long time. Haven't we had people here is the US without power for weeks? What about earthquakes or terrorist attacks or economic meltdown or even zombie apocalypse ? There are experts that think our power infrastructure is especially teetering on collapse add a well placed terrorist strike and it could be months. I think this would work very well with a very small generator for night/storms. Also sunnyboy is supposed to be a very good brand, right?
I can't afford to double the price of my system but if it is a very small increase or even a decrease from a microinverter and the power optimization can be within 1% of a microinverter. I find that very tempting. I'm still trying to figure out details. It seems it could be about $1500 less than enphase on a 5.7 kw system but I'm not sure if that included the power optimizer modules.

[SunEagle]

In my opinion, the more important reason that a Grid Tie solar system needs to be shut down when the grid goes down is that the design of the inverter simply does not allow standalone operation.
The SPS inside a SunnyBoy is a separate inverter, and one which is designed slightly differently from a typical off grid inverter so that it can operate directly off the panels without needing batteries as an energy buffer.
A single inverter design which can both operate independently AND synchronize its output to a grid connection, as found in some hybrid inverters, is more expensive, and that is reflected in the price of those units.

I understand that the main inverter has the anti islanding controls to protect people and the SPS does not. It is a nice package. But while that feature gives you access to a small power source (when the sun is shining) is it worth the manually intensive hassle of using it.

A hybrid inverter would do all of the controls automatically but as you indicated at a much higher cost.

Still if someone needs a backup power source it is worth the expense and hassle of the 1500 watt SPS system or just get an 2000w inverter style generator that will work even at night when the power goes out.