Please explain "Harvest Efficiency"

qwertinsky

I work for a small company beginning its first endeavors in to alternative energy. I have been tasked with evaluating several solar charge controllers. One of the comparisons I an supposed to make is "Harvest Efficiency" but I can not get or find a simple explanation of what that means. We have had several guesses and don't really want to look like we don't know what we are doing.

Cariboocoot

Re: Please explain "Harvest Efficiency"

It can mean several things as there is no absolute definition for the term.

In the off-grid system it refers to the difference in the potential power the array can produce in a day compared to the actual power realized. This tends to be 52% end-to-end for all power going in to and coming out of the batteries.

In other words a 1kW array over 4 hours of equivalent good sun should produce 4kW hours of power, but due to the inefficiencies inherent in the system it is more like 4000 Watt hours * 0.52 or 2080 Watt hours AC.

This can vary due to local conditions that affect panel operation (hot panel = lower Voltage, cold panel = higher Voltage, high elevation increases insolation, et cetera). It can also be increased by how the power is utilized: one of the big factors is that once the batteries are recharged the panels could produce more power but won't because there is little or no demand on them. Thus the reason why we use 'opportunity' loads to increase the 'harvest' of potential power.

Grid-tie systems don't have this problem because as long as there is a grid to sell to there is someplace for the power the panels can produce to go. They are thus more efficient in harvesting the power potential of the array. Most of the GTI's are in the 95%+ conversion efficiency range too.

You could have some comparison between a micro-inverter system and a central inverter system as well. In general, if the all panels face the same way and are clear of shadow it would be a wash. But if there is the need for panels facing in different directions due to mounting space available and/or there are issues with potential shadows then the over-all daily harvest from the micro-inverters may exceed that of the central inverter for the same size installation.

Was that any help?

qwertinsky

Re: Please explain "Harvest Efficiency"

Yes, no.. maybe... That was helpful and is pretty much on the same lines as we were thinking.

We are renting a Solar Array Simulator and I can put in the array parameters and environmental conditions, array temperature, solar radiation levels (W/M^2), etc.

If I was to test each controller using the same SAS settings would the "Harvest Efficiency" be the same as the units efficiency? %Efficiency=(Wout/Win)*100

BB.

Re: Please explain "Harvest Efficiency"

Can you describe your loads, where the system (near major city) is to be installed, emergency backup or full off grid or Grid Tied (to reduce utility bills), etc.?

In general, we can do some back of the envelope calculations that will get you to within 10% accuracy... And being more accurate is really not worth the efforts. Long term weather can give you +/- 10% changes in output, measurement errors, how you choose to operate your battery bank (charging methodologies), etc. all add to the variability of the system output. And your loads will usually vary by much more than +/-10% too.

If you are off grid, I would suggest that your daily base loads are only ~66 to 75% or so of your predicted system capacity. This allows for some load growth, days with higher energy usage (wash day, working at home, occasional shop tool usage), bad weather, etc... You can run closer to 100% of predicted output--But that usually takes active management of loads and battery charging/generator usage. Most of us would rather be doing something else.

And other site related issues... Lots of shade--Systems will not generate very much (if any) useful power when even partially shaded (again, depends on system size/configuration/amount of shade). System in an area with heavy coastal marine layer (like coast near San Francisco)--You could be looking at 4 days of clouds and 2 days of sun--Will kill system performance by 1/2 or more.

If you have utility power--Then you simple will use the utility power to make up for when loads exceed system capacity. Frequently at about the same costs as the solar power you are using (places with Time of Use and demand/reservation pricing--Solar can save money or even cost you money--It all depends on the details of the tariffs).

For commercial systems--The billing/tariff plans are typically more important in the return on investment than the 10% better with vendor A vs vendor B system (best case differences between hardware).

There are many good reasons for good instrumentation and understanding of system overall operation--But you need to do a simple back of the envelope set of calculations for a sanity check first--It will help you understand were your strengths and weaknesses are.

-Bill

qwertinsky

Re: Please explain "Harvest Efficiency"

Tell you the truth I do not know the exact system they are going to be used in. From my point of perspective it is the charge controllers it's self that is being tested. As I explained before I am using a solar array simulator and can simulate any array and environmental conditions.

Would a watt hour measurement of the power produced during a 12 hour "Sunrise to Sunset sunny day" irradiation profile versus the watt hours charger output be a better indication of harvest efficiency?

BTW: I am using a Yokogawa WT230 with 50mV/50A current shunts for my measurements. All of our equipment including the current shunts are calibrated yearly.

So I am gathering that harvest efficiency is a system level measurement and is irrelevant when focusing on just the charge controller, but the efficiency of the charge controller will effect the overall system harvest efficiency.

Cariboocoot

Re: Please explain "Harvest Efficiency"

Any good charge controller has an efficiency of 95% +/- 1% or so. It is not particularly significant in the over-all efficiency of the system. A standard grid-tie system will not have a charge controller or batteries and will run in the same high 90 percentile efficiency range.

Frankly I think you're over-thinking the design. The day-to-day variations in conditions you have no control over will make a bigger difference in the functionality.

zoneblue

Re: Please explain "Harvest Efficiency"

Cariboocoot wrote: »

Any good charge controller has an efficiency of 95% +/- 1% or so. It is not particularly significant in the over-all efficiency of the system. A standard grid-tie system will not have a charge controller or batteries and will run in the same high 90 percentile efficiency range.

Frankly I think you're over-thinking the design. The day-to-day variations in conditions you have no control over will make a bigger difference in the functionality.

Yeah probably quite true. However if the OP is really interested I would add the following:

- each charge controller will have a set of baisc efficiency curves, for varying nominal input voltages and throughput power. These are derived use lab power supplies, and are more theoretical in nature. You can see a full set of these in the FM80 manual for example. And btw the efficiency is not always as high as 95%. For high input voltages and low output voltages (mmpt), efficiency drops off.

-however energy yeild is also determined by how effective the power point tracking algorithym is. The more accurately, and responsively it tracks changing irradiance conditions the more the overall yeild will be. As an example midnite tested various controllers, and found that the tracking makes a difference, and claimed that the classic yeilded more power per day off identical arrays than any other controller on the market. However they never published the data.

Yeild is considerably less relevent for PWM controllers, and other factors like how well matched the array is to the controller, wire length and guage, product reliability, aspect and shading etc etc etc will dominate the outcome, much as they do in general.

vtmaps

Re: Please explain "Harvest Efficiency"

I think zoneblue came closest to answering your question... the MPPT algorithm matters and the input voltage and current matter. A less efficient controller with a better MPPT algorithm may harvest more power, or vice versa.

qwertinsky wrote:

So I am gathering that harvest efficiency is a system level measurement and is irrelevant when focusing on just the charge controller, but the efficiency of the charge controller will effect the overall system harvest efficiency.

Yes. There is no way to compare two controllers without setting up two identical arrays and two identical battery banks. And even then, you might find that the winner on one day is not the winner the next day... no two days are the same temperature, clouds, etc.

Even creating identical arrays and batteries is not easy... you may have to put each controller on each array to do a fair trial.

That said, "harvest efficiency" is not the most important factor in designing a system... for example: if your array is far from your controller you will have some losses in the cable between them. If you raise the solar panel voltage to reduce cable losses, you will lower the efficiency of the controller. If you use a larger gauge cable, the cost of the extra copper might be more than the cost of just adding another panel to the array.

All things being equal, higher efficiency is better... but all things are never equal.

--vtMaps