Okra Modular Microgrid Baseline Pilot

by | Sep 19, 2017

Is there a need for small scale microgrids in rural off-grid communities? What’s the potential efficiency gains over standalone solar home systems?

At Okra we’ve been running theoretical calculations for a while, and they all seem to suggest that a smart connected network of solar panels & batteries are almost always going to be more efficient in providing off-grid electrification than standalone solar home systems. That’s because it’s hard to predict exactly how much power a household will consume day to day, add to that the intermittency of sunshine and the fact that batteries, the most expensive part of the system can only hold a finite amount of power before becoming full. This combination of factors means a standalone solar home system usually has to be oversized to guarantee sufficient power and there’s usually a non-trivial amount of power that’s wasted when there’s nowhere to store it.

We decided to back up our calculations with some empirical evidence so we went to Prey Pdao in Takeo Province, Cambodia and installed 5 solar home systems for the community. The Levelised Cost of Electricity in Cambodia from charging batteries with diesel is approximately $4.5/KWh, so we decided to charge households $1/kWh of electricity consumed, very steep compared to what we’re used to in urban areas, but still a fraction of the cost these households were paying. With the help of our implementation partners NRG Solutions and Camsolar, we installed 150Wp solar home systems with a 100Ah battery in each of the 5 households – which turns out was self-extended by the community to serve 7 households (will get to that later).

Our objectives were to let these households connect their old or new DC appliances to these systems, and observe the energy consumption behaviour, and importantly find out if there was any excess generation that could potentially be utilised in a smart connected microgrid network.

The pre-pilot was running since the start of August and the results are really encouraging (at least in terms of opportunities to improve!)

Summary (90 day sample)

Total kWh Consumed

Total no. Houses Currently Powered

Estimated Cost of Consumption w/ Diesel


Total kWh Generated

Potential no. Houses Powered w/ Okra

Cost of Consumption w/ Okra

House A

This is the village chief’s house. There’s 6 people who live at this house, they’re rice farmers who also run the equivalent to the village convenience store when there’s nothing to harvest. They also make fishing nets while unoccupied because as you can imagine there’s not too much traffic through their store. They run a couple of lights, a fan, and a small T.V using most of their electricity at night time.

House B
This is Grandma Yu’s house, she lives with her granddaughter who she takes care of. Her children live in South Korea and bring back cool gadgets every time they come back. Even though Grandma Yu is off-grid she uses an Samsung S6 edge when making facebook calls.

House C
This family never had electricity before Okra, they used kerosene for lighting and cooking. They don’t have much in terms of money so can’t really afford appliances, asides from the few high efficiency lights they purchased when we installed the system.

House D
We don’t have complete data for this house – the latest information we have is that the homeowners have moved to Phnom Penh, a shame because the electricity is not being used by anyone. We plan on utilising the power from this system once we connect the houses into an Okra network, which is the next phase of our pilot!

House E
This one is a funny case. We had originally installed the system for a single household, but when we came back 2 weeks later to conduct some monitoring and evaluation, we found that there were actually 2 additional houses connected to the power supply. Turns out the gentleman who owned the house felt the same way about us as there being excess power, and wanted to share it with his neighbours. He said he’s happy to pay for their power consumption, but makes them turn their T.V off early so the battery doesn’t run out.

Only 5 households were connected in our experiment, evidently the 150Wp solar home systems (~$500/system) were probably sized a bit too large for these houses. We do expect over time as the households purchase more appliances their wasted energy portion will reduce. But importantly the data shows that even in the best of cases, a substantial amount of power that could be used is wasted from standalone solar home systems. Additionally, there is a lot of variation between household consumption in the same village, meaning some waste a lot, others waste more, and they waste power during different parts of the day.

This is really encouraging data for us because with our smart, connected microgrid technology, Okra will be able to distribute the aggregate amount of power between the households more efficiently – this means a reduced cost of power, more reliable power and importantly more households electrified from the same amount of generation and storage assets.


  1. Anjal Niraula

    Good one mate ! Quick question though – how does the calculation for the generation part of the ‘Power utilization’ chart work ? I initially assumed that it was a theoretical calculation spread over 7-8 days. But, the numbers seem to be different for different houses. Since these were all 150 watt systems, I started thinking that these were measured off of something. Then I noticed that House D still showed 4.5 kWh of generation, even when they had only consumed 6 Watts for a fraction of the time ?

  2. Afnan

    Hey @Anjal. So much attention to detail;) The generation is from NREL SAM (System Advisor Model) data as you had assumed it’s theoretical. The reason why house D is different to the rest is because after a period of no-consumption, the monitoring had turned off. The other houses all have the same generation. Consumption is measured as controller-out. And Usable wastage is measured as Generation – controller-in. If you want to check out the data in more detail I’ve put together a snapshot for you and you can zoom in on the timescales too. Definitely interested to hear if you have any other thoughts! http://grafana.okrasolar.com/dashboard/snapshot/M5m9l6MbX9cwVlS986J2dhO3YMnKzJ3P

  3. Leo Msanga

    The system is much applicable where houses are dense and not qualify for min-grid that need much investment.

  4. Felix Lawrence


    How do we get your products in Liberia, West Africa.


Submit a Comment

Your email address will not be published. Required fields are marked *