[RE-wrenches] Inverters Maximum Input Ratings

Wed Aug 27 06:53:23 PDT 2014

I wrote an open source library called solpy to explore some of these
issues. https://github.com/nrcharles/solpy It's not user friendly but you
can watch the DC power available at the array.  It implements the CEC
inverter model and when I matched it up with Enphase's 5-minute data it
tracked fairly well. (Paper here
https://www.dropbox.com/s/08irv0fb4sky9o5/prediction.pdf?dl=0 it's not peer
reviewed.)

The interesting thing, is how high of a ratio is needed in non-ideal
circumstances for clipping to show up. That said, it often appears that
mismatch and soiling derates are applied to the DC side and that will
reduce the number of hours clipping that shows up in the model. I think it
really get's down to what the Manufacturer warranty will cover, and plan
for an inverter replacement if the warranty is less than the life of the
system.

On Wed, Aug 27, 2014 at 8:08 AM, Matt Partymiller <
matt at solar-energy-solutions.com> wrote:

> Rebekah,
>
> My understanding was that PVsyst modeled inverter nominal power not
> maximum power.  I believe this helps to compensate for inverter
> temperature limitations.  For instance, depending on inverter location and
> power ratio, I would suppose that there may be instances where the
> simulation could clip potential max power production.  This may balance
> those instances where inverter temp derating would result in production
> loss.
>
> I would be curious to see how subhourly data could affect a simulation.
> Admittedly not something I have thought about.
>
> Matt
>
> Matthew Partymiller
> Solar Energy Solutions LLC
> (877) 312-7456
> matt at solar-energy-solutions.com
>
> On Tue, August 26, 2014 7:10 pm, August Goers wrote:
> > Hi Rebekah,
> >
> >
> >
> >
> > This is where checking simulation data versus real output data is key.
> > Our
> > company has hundreds of monitored systems but very few with undersized
> > inverters. So, I donâ€™t have many data points that I can study.
> > Generally
> > our systems meet or beat our simulated as-built production estimates
> > within an amazingly close margin. I think your point about inverter
> > heating and derating is valid for inverters which are significantly
> > undersized. You seem to have a strong opinion that simulators generally
> > underestimate clipping. Do you have any particular sites where you
> learned
> > this from?
> >
> >
> >
> > Best, August
> >
> >
> >
> >
> > *From:* RE-wrenches [mailto:re-wrenches-bounces at lists.re-wrenches.org]
> > *On
> > Behalf Of *Rebekah Hren
> > *Sent:* Tuesday, August 26, 2014 10:54 AM
> > *To:* RE-wrenches
> > *Subject:* Re: [RE-wrenches] Inverters Maximum Input Ratings
> >
> >
> >
> >
> > One of the problems with relying on simulators (PVsyst, Helioscope,
> > whatever) to estimate clipping is that they rely on hourly averaged
> > irradiance from TMY files (with very few exceptions), which may not
> > predict clipping, especially for areas with variable climates.
> >
> >
> >
> > Thinking about partly cloudy days in NC where I live and install systems,
> >  an average hourly value predicts no or a very low percentage clipping
> > for systems with even a fairly high dc to ac ratio (by high I mean
> > anything over about 125%). Without subhourly meteo data, I would be
> > skeptical of simulators estimates of clipping. I believe they are
> > generally underestimated, which can lead to a false financial analysis of
> > the most cost effective dc-ac ratio.
> >
> >
> >
> > Also, I do not think most (any?) of the simulators have the capability to
> >  take inverter internal temperature related power derating (another form
> > of power clipping) into account in the simulations, which could lead to
> > overestimated generation. I realize hot inverter temperatures may not
> > coincide with the colder temps that usually create clipping, but the
> > higher the ratio goes the more often we see an overlap in hot days/high
> > temps/clipping.
> >
> >
> >
> > Cheers,
> >
> >
> > Rebekah Hren
> >
> >
> >
> >
> > --
> >
> >
> > Tel: 336.266.8800
> > http://o2energies.com/
> > Project Engineer
> > NABCEP Certified PV Installation Professionalâ„¢ 091209-85
> > NC Licensed Electrical Contractor
> >
> >
> >
> >
> >
> >
> > On Tue, Aug 26, 2014 at 11:30 AM, August Goers <august at luminalt.com>
> > wrote:
> >
> >
> > Hi Jason and All,
> >
> >
> >
> >
> > Good topic. As far as I know, simple energy estimation programs like
> > PVWatts donâ€™t take inverter size into account when estimating annual
> > production but more robust simulation programs like PVsyst do. I think it
> > is a matter of really being able to trust your simulator to tell the
> > designer how sizing the inverter will affect annual production. Maybe
> > others can chime in on this.
> >
> >
> >
> > Weâ€™ve run different scenarios through PVsyst in house to see how under
> > sizing inverters affects the overall annual production. It is surprising
> > for our area (SF Bay Area) that we can really push an inverter with very
> > little reduction in annual simulated production. Iâ€™m talking about
> > putting maybe 5 kW (or even a little more) of PV on a 3.8 kW inverter or
> > thereabouts. Array orientation also factors in. To Issacâ€™s point, if
> > keeping the inverter small reduces the initial installation cost
> > significantly while possibly only reducing annual production by a little
> > bit then I feel it is a good design.
> >
> >
> >
> > All that said, weâ€™re still fairly conservative with pushing our
> > inverters. Iâ€™ve found that it is very hard to go back to a client after
> > the salesperson has made the sale and try to reduce the inverter size. In
> > other words, we need to nail the inverter size up front during the sales
> > process or the client will oftentimes feel taken advantage of later. It
> is
> > hard to explain why we are only giving a client a 3.8 kW inverter when we
> > are calling their system 5 kW DC.
> >
> >
> >
> > To Isaacâ€™s questions: most of our clients wouldnâ€™t notice about mild
> > to moderate clipping during peak periods. However, some would. This is
> why
> > it is important to setup proper expectations about how we designed the
> > system up front and do our homework to assure that the clipping is
> > accurately estimated and factored in to the production estimates. If we
> > can provide clear logical reasoning for sizing the inverter the way we
> did
> > then we shouldnâ€™t have any problem. I have no idea how array sizing
> > might affect inverter longevity.
> >
> >
> >
> > I think the general points made by others about the declining costs of
> > the modules themselves and increased cost of BOS components means that
> > these types of discussion are very valid.
> >
> >
> >
> > Best, August
> >
> >
> >
> >
> > *From:* RE-wrenches [mailto:re-wrenches-bounces at lists.re-wrenches.org]
> > *On
> > Behalf Of *Isaac Opalinsky
> >
> >
> >
> > *Sent:* Tuesday, August 26, 2014 6:11 AM
> > *To:* RE-wrenches
> >
> >
> > *Subject:* Re: [RE-wrenches] Inverters Maximum Input Ratings
> >
> >
> >
> >
> > Jason,
> >
> >
> >
> >
> > This seems to be a regular topic of discussion in our training classes
> > not only for microinverters, but also for string inverters.  Especially
> > for 3.8 kW units that are optimal for backfeeding 100A service panels and
> > 7.6 kW
> > units that are optimal for 200A panels.  A slightly bigger array can give
> > a higher total yield, maybe some power clipping, without the additional
> > cost of a supply-side connection.  As long as you stay below the maximum
> > VOC and
> > ISC, there isnâ€™t a safety issue.
> >
> >
> >
> >
> > So it really just boils down to economics and the overall value
> > proposition for the customer, which makes it hard to provide a blanket
> > recommendation. Weâ€™ve been training people for years to model PV system
> > performance to determine an acceptable DC/AC ratio on a
> project-by-project
> > basis.
> >
> >
> >
> > The inverter manufacturers pretty much all claim that there is no concern
> >  about overworking or shortening the useful life of their inverters since
> >  limiting operating power limits the operating temperature as well, but
> > that leaves me with two questions:
> >
> > 1.       Does anyone have any evidence that high DC/AC ratios does/does
> > not shorten the life of the inverter?
> >
> > 2.       If there is a small amount of power clipping (say <1% total
> > annual energy), are many customers likely to notice/care?
> >
> > 3.       If they do notice, does the customer service aspect of having to
> >  defend a design decision outweigh the potential economic benefits of a
> > smaller inverter?
> >
> >
> >
> > *Isaac Opalinsky *| Technical Trainer | *SunPower Corporation*
> >
> >
> > Desk 443-569-3476 | Cell 443-277-6286
> >
> >
> >
> >
> > *From:* RE-wrenches [mailto:re-wrenches-bounces at lists.re-wrenches.org
> > <re-wrenches-bounces at lists.re-wrenches.org>] *On Behalf Of *Jason
> > Szumlanski
> > *Sent:* Monday, August 25, 2014 4:41 PM
> > *To:* RE-wrenches
> > *Subject:* [RE-wrenches] Inverters Maximum Input Ratings
> >
> >
> >
> >
> > I had a tough customer recently that grilled me on how we can put a 270W
> > solar module on a 215W inverter. Fortunately, Enphase has a wonderful
> > white paper on the subject. However, it got me thinking... Enphase has
> > demonstrated that higher output panels in many climates (hot SW Florida
> > included) can benefit from modules that far exceed the inverter rating,
> > and even exceed the inverter's "recommended input" rating. Enphase has
> > shown that 270W+ modules can show energy harvest on the M215 where it
> > makes sense to "oversize" the module.
> >
> >
> >
> > I also received a similar query from a rather uninformed plan reviewer in
> >  an area AHJ along similar lines. Fortunately I was within the
> > "recommended
> > input" rating on the spec sheet of 270W with a 265W module, but I wonder
> > what would happen if I had paired the M215 with a 280W module on my
> > plans, which are becoming readily available now in 60 cell modules with
> > 300W
> > modules on the near horizon. I'm pretty sure my plan would have been
> > kicked back for exceeding the manufacturer's recommendation.
> >
> >
> >
> > My question, which applies to string inverters and microinverters, is how
> >  much is too much, what would happen if you paired an array that far
> > exceeded the rating, and how do inverter manufacturers determine the
> > recommended and/or maximum rating of the connected module or array? Also,
> >  why do some manufacturers have a simple recommendation while others have
> > a "maximum" rating?
> >
> >
> >
> >
> > Jason Szumlanski
> >
> >
> > â€‹Fafco Solarâ€‹
> >
> >
> >
> >
> > _______________________________________________
> > List sponsored by Redwood Alliance
> >
> >
> > List Address: RE-wrenches at lists.re-wrenches.org
> >
> >
> > Change listserver email address & settings:
> > http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org
> >
> >
> > List-Archive:
> >
> http://www.mail-archive.com/re-wrenches@lists.re-wrenches.org/maillist.htm
> > l
> >
> > List rules & etiquette:
> > www.re-wrenches.org/etiquette.htm
> >
> > Check out or update participant bios:
> > www.members.re-wrenches.org
> >
> >
> >
> >
> >
> > This message, together with any attachments, is intended only for the
> > named recipient(s).   It may contain confidential or privileged
> > information that may be exempt from disclosure under applicable law.  If
> > you are not the intended recipient(s), you are notified that the
> > disclosure, dissemination, distribution, copying, storing, or other use
> of
> > the contents of this message or its attachments is strictly prohibited.
> > If you receive this
> > message in error, or are not the named recipient(s), please notify the
> > sender at either the email address or telephone number above and delete
> > this email from your computer.  Thank you.
> > _______________________________________________
> > List sponsored by Redwood Alliance
> >
> >
> > List Address: RE-wrenches at lists.re-wrenches.org
> >
> >
> > Change listserver email address & settings:
> > http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org
> >
> >
> > List-Archive:
> >
> http://www.mail-archive.com/re-wrenches@lists.re-wrenches.org/maillist.ht
> > ml
> >
> > List rules & etiquette:
> > www.re-wrenches.org/etiquette.htm
> >
> > Check out or update participant bios:
> > www.members.re-wrenches.org
> >
> >
>
>
>
>
>
> _______________________________________________
> List sponsored by Redwood Alliance
>
> List Address: RE-wrenches at lists.re-wrenches.org
>
> Change listserver email address & settings:
> http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org
>
> List-Archive:
> http://www.mail-archive.com/re-wrenches@lists.re-wrenches.org/maillist.html
>
> List rules & etiquette:
> www.re-wrenches.org/etiquette.htm
>
> Check out or update participant bios:
> www.members.re-wrenches.org
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org/attachments/20140827/cca9ddef/attachment-0002.html>