Customer Story : Techie uses Amberroot REhub to DIY his Solar install

Do it Yourselves (DIY) : Adding Solar PV in your house !

I have below a detailed blog from a ‘Techie’ Ajeet Sharma who set up a Solar plant in his house – all by himself. He used Amberroot’s REhub MPPT for the same. Like a true engineer, he is continuing with his experiments with tracking the sun on the roof top.

Thank you for choosing Amberroot, Ajeet and best wishes on your further experiments!

Link to his Blog is here 


Saturday, April 30, 2016

Powering Up!

In January, I had posted the last update on my solar power project, that I have been working for quite some time now. Over time, working in the evenings on workdays and almost through the day on weekends/holidays, I have finally started my own solar power generation generation system, that is now functional, partially. The tracker that had been in testing for around three months now, finally got to host what it had been designed for, and all this while, was carrying a really small solar module.
Just a few days back (20th of April, 2016), I finally installed the solar modules on the tracking platform, and hooked it up to the existing power backup system, I had installed some time back for my home. And the result, free power for my home, which, for now, will service the loads partially, thereby resulting in a reduced power bill for the utility power I have been using. Over the period, I have planned to upscale the capacity of my solar power system, which, will then be self sufficient to provide power for all the needs of my house.
Now, the system has been installed and is functioning, what’s left is to begin the tracking system, which, for now is non functional, because the tracking system appeared to be ill equipped to handle such a large load, and thus, I am tweaking it a bit to make it capable enough to do what is intended to, without breaking a sweat. Since this is something I am doing for the first time, and without any theoretical or practical knowledge, there was no way I could’ve anticipated what was in store for me, once I had the solar modules installed on the tracker, which, for the time it was working with that small module, appeared pretty capable enough to efficiently complete its task without a problem. This, I know, is a part and parcel of DIY, for you learn with what you do, and over time, there comes a point where you achieve perfection, and what you wanted to do, and this is what I have been trying to achieve.
Over time, I had to change a lot of things in the planned tracking system. To say the least, the entire idea of the tracking system we had planned to take up to work on, as we moved our application in the Mahindra Rise Prize’s Solar Challenge, has changed. The use of linear actuators has replaced one of the gear based mechanisms, again, this actuator based mechanism is up for an overhaul as we build ourselves a better, more capable actuating mechanism, just because we need to have a better one, and buying this one off the shelf is out of question. Why? Firstly, because these cost a bomb, and secondly, we do not have too many options to look for, because looking for parts like these needs a massive hunt for places which deal in such speciality parts, which, is a very great impediment in doing almost anything. Also, I had to improve the light sensor setup, improve the overall reaction to light changes over the sensors, and am required to add some more ways to make the tracker work better, be a completely autonomous system that is capable enough to do the tracking throughout the day, and reset itself back to east at dusk, to begin a new day of tracking, all on its own. What more in the store is a web based realtime monitoring system that lets me check out on vital stats of the system, from anywhere in the world with the help of a website that will be fed by a variety of sensors reading some vital stats of the system, viz Voltage and Current (both of which will give me the total output in watts), Cell Temperature, Location (in terms of direction of the tracking platform), a planned anemometer, which will be helpful in case of a storm, to keep the modules relatively safer.
Now the good part is, I have enquiries pouring in from neighbours and whoever can see my solar modules on the tracker, facing the sun, on my rooftop. Appears, this is also a chance where I could cash in the advantage I have, over the regular solar power system integrators, most of whom, just do a static installation, for a cost that is be more than double of what I had to shell out for this system.

Did I make you curious? In case you would like to take a look at the specifications to get to know this better, you can take a look below:

The components/parts I bought off the shelf:

Part Qty Make/Specs
SOLAR MODULES 2 Nos. Canadian Solar 265 watt @ 24v (Another 40 watt 12v solar module to provide power supply for the tracking electronics)
SOLAR CHARGE CONTROLLER 1 No. Amberroot ReHUB MPPT (12/24v, 40A)
INVERTOR 1 No. Luminous 1500VA Pure Sine wave
BATTERY BANK 2 Nos. Luminous 150Ah @12V Flooded Lead Acid Batteries, connected in series to make this a 24v configuration.
The components/parts I have built myself, again with parts I have procured to be able to build the parts I needed:
MOUNTING PLATFORM 1 No. Dual Axis Tracker (that I have built myself)
DRIVE MECHANISM Self designed, with the core of the system being controlled by the awesome open source Arduino Platform running with Atmel’s ATMEGA 328 running in its heart.


The platform works with the help of an array of light sensors designed such that once the sun is near perpendicular to the light sensor array setup, it’s rays fall on all the sensors uniformly. Once the sun deviates a little, one or more sensors experience shadow because of the divider that has been put between the four light sensors. The MCU (Arduino, in this case) is used to feed in the data from the light sensor array which then makes the comparision based on the software code we have put inside the Arduino to take decisions based on the realtime information provided by the variety of arrays of sensors we will be feeding the data from, in the Arduino. Once the MCU takes a decision regarding the direction of the movement, it sends the relevant instruction to the H-Bridge Motor Driver based on the powerful L298N motor driver which can handle both of my 12v DC geared motors, having a stall torque of 45kgcm. The motor driver sends a signal, based on the input from the arduino for the relevant motor to move, in the direction, the MCU told the motor driver to move the motors in. Once this happens, and all the four sensors are again aligned such that the sun’s rays are again falling uniformly on all of them, with no shadow on any of the four sensors, the MCU tells the Motor driver to stop the motors till the time one of the light sensors again experiences shadow, which makes the aforementined process, start again.
As for the updates, they are not yet over. I have a lot of work yet to be completed. No, not about the functioning, as the system is up and running fine for around ten days now, since I first hooked up the panels to the charge controller, ten days back. What needs to be done is more of the stuff related to additional functionalities, that, I expect will add to the system’s geekiness, a lot more. Some of these things, that I can hgive you a glimpse, though, textually, here are – 1. Adding the Gyro sensor to: a) limit the movement of the panels and the tracker, at the extreme ends. b) realign the panel to east, in the morning after the earlier day’s tracking aligns the panels in the west. Right now, this is being done manually, though, I don’t have any issue in doing this manually, because, afterall, I do visit the rooftop daily, to take a stroll around my plants, and water them. 2. Redeploying the solar tracking functions. As it turned out to be, the actuator we had worked upon, for the test rig, was not capable enough of single handedly handing the extreme pressures of the solar panels, and if adding the additional stress exerted upon, by, say, a strong gust of wind blowing directly into the panels’ large surface area of around 3.2 square meters, hanging on, 30 feet above the ground, it is not much far from disaster. So, I decided to look for linear electric actuators, only to give up again, seeing the costs. And finally, I have thought up of a new design to make this happen, properly, and work on this new linear actuator is yet to start, as I am waiting for the parts. For the time being, I will just mount the new threaded rods I had bought, to replace the previously used 6mm threaded rods, which buckled down due to the extreme pressure of the modules and the frame. This new 10mm threaded rod in stainless steel, should ideally be capable enough to take care of the load, itself, so, I’ll go ahead and give it a try, for the time being.
For now, I will leave you guys with a couple of pictures of what’s happening these days and how the system, I have been posting updates on, for quite some time now, has transformed over time, to being the final product, something, that I had only envisaged, but had never visualised, the way it looks, and works, today.
The new mounting frame for the two solar modules. With the hand drill, and too less an experience, it was quite hard to get through this thing, especially, when one is working with steel.
Both the solar modules, ready to be mounted on the tracker. The size, did appear to be intimidating, but then, I had to do this, at all costs, and so this was done, finally.
Powering up! It felt great to see my dream realize.
While I ultimately powered up, it was evening, and since there was still power, being generated, the charge controller chose solar power over the mains power, and the above shot is a result of the inverter batteries getting discharged, because of the loads were still extracting power, but there was not much being supplied to charge the batteries, so, this one shows the batteries being charged with mains power, after the power from mains to the inverter was restored after the solar out put was quite less.
The modules facing the morning sun. For now, I have tied them (using the frame) up to make them stationary, and to check their movement under the influence of winds. Once the actuator is finally ready, the ropes will all be gone, atleast for the time the weather is good. I had to make separate arrangement for the shade net for my plants that are placed around this part. of the roof.
The modules facing the morning sun.
The peak output I could see/capture.
Finally, the cumulative power generation by the solar modules, since I powered up the system up around ten days back.
Since, the work on this is system is still underway, I will be coming up with a few more posts updating here, the work in progress, the status of the project and how it ends up, once the work is complete. And, maybe, my power bill as well, at the end of this month, to compare what’s the difference since I installed the system.

Adding Solar PV to Existing Inverters

How to get real savings in Electricity bills and keep the batteries safe!

Electricity supply in India has improved enormously in the recent past. The Southern region is expected to have an high Peak shortage (Power cuts !). The Eastern region is expected to have an energy surplus in 2015-161. With increasing power availability, consumers are looking to add Solar PV to help in reducing electricity bills. In the absence of grid tied options, adding Solar to existing Inverter based back up makes economic sense. Adding Solar to existing Inverters allows the reuse of the investment already made in Inverters & Batteries.

With this approach, the Solar PV panels & a Charge Controller that controls the charging from both the existing Inverter and Solar PV is retrofitted to the existing Inverter. The combined system is expected to work as a single unit, maximizing the use of Solar PV and providing sufficient power back up in case of power cuts.

Most retrofit charge controllers control the Mains input to the Inverter, switching off the Mains supply, when Solar PV is available so that the loads connected to the Inverter are powered from Solar PV (and the batteries).

A good ‘retrofit’ charge controller MUST control the Charge Rate to the batteries if they aim to really save electricity bills with the addition of Solar PV. To get real savings in electricity bills, the Solar PV addition must generate enough energy to completely service the day time loads connected to the Inverter. For example, in a typical home, the day time energy consumption might be 2-3 Units (kWHr). A kWp of Solar PV panel would generate about 4-4.5 Units per day. A Solar array suitably sized would be capable of generating the required energy. The conflict arises becase of the size of battery back up available. Most existing Inverters would have Lead acid batteries of ~ 150Ah capacity. Lead acid batteries must not be charged at a rate higher than 10 -15% of their battery Ah. When charged at an higher rate, the batteries reach gassing Voltage before they are fully charged.However, a Solar PV array to take care to day time loads might deliver Current that is higher than the optimal Charge rate. Installing larger batteries is expensive & not very useful. Installing smaller PV capacity is counter productive – in the end, the energy deficit would be bridged by the use of Grid supply and will result in increased bills. Not what the user wants either. The retrofit charge controller Must monitor the net current delvered to the batteries and Must control this current to the optimal charge rate – so that Loads get the required power and the batteries are used safely as an effective buffer for the variations in load and generation.

Therefore a good solution to add Solar to existing Inverters to be of real benefit to users must allow sufficient Solar capacity to take care of day time loads and keep the batteries safe at the same time.

1Page i.. Anticipated All India Power Supply Position for the Year 2015-16. Load Generation Balance Report 2015-16. Central Electricity Authority.


An argument for Solar Part – II

In my earlier post ( I discussed how in the places where power cuts are common, Solar makes immediate sense and how REhub makes it even more easier to decide on Solar. What about places where Power cuts are not so common but the electricity tariffs are high ?

15TVKSEB_eps_2059993eTake the state of Kerala . Kerala has power tariff slab that not only increases with the number of units consumed per month but also penalizes the customers exceeding a certain number of units consumed. Once the units exceed a limit, the consumers are charged the higher rate for all units consumed and not just for the extra units. Let us say that 334 units were consumed in a given month. On exceeding 250 units consumed, the charge is Rs 5 for all the units I.e for 334 units and not for the extra 84 units beyond 250. In a normal telescopic rate based tariff, the consumer would have paid Rs 1696/- versus as per the non telescopic tariffs he would have to pay Rs. 1904/- . Imagine the benefit if Solar PV helps in reducing the units so that the total falls into a lower slab. 22 + states increased the electricity tariffs last year and we are headed to a situation where the cost of electricity from Solar is already lesser than alternatives – what experts call Grid parity. As more and more states allow net metering or reverse metering, the case for Solar becomes even more stronger. In the case of net metering (or reverse metering) the Grid becomes the storage and the consumer is paid for the amount of units put into the Grid. In this scenario, the constraints are not the storage or the day time load in a house. The constraint is now shifted to the amount of shade free roof top available and the money available for investment. And I will show in a future post how much of a good investment that is with Karnataka and Tamilnadu’s reverse metering policy for small rooftops. For now, let us stick to adding Solar to existing back up and go into the mechanics of the addition in some detail. That’s in the next post of the series.

Image from The Hindu article on power tariff revision in Kerala - link here.

An argument for Solar on your rooftop – Part I

Take a typical home in Hyderabad or Coimbatore or for that matter most cities and towns in India. The uncertainty surrounding power cuts and the need to tide over them has spawned a whole industry around Inverters and Batteries & everyone seems to have one in their home for power back up. Yet the use of Inverters has perversely increased the demand for electricity – leading to more black outs and at an individual level increase in electricity bills. The round trip efficiencies of the Inverter and batteries are so poor that for every two units of energy that is put-in, one gets barely 1 unit of useful energy. Effectively the consumer is paying double for the unit of energy used during blackout hours. Even if one is willing to pay more for the convenience of always-on availability of power, the power cuts are so long that batteries just do not have sufficient time to recharge. A typical lead acid battery takes more than 10 hours to recharge completely under ideal charge rates. Most Inverters in the market are undersized relative to the size of the batteries and would take longer to recharge the batteries.


With REhub – stands for Renewable Energy hub – one can now add Solar to existing Inverters, effectively upgrading to Solar without having to invest in a completely new system again. Amberroot’s Solar chargers efficiently extract power from the Solar photovoltaic panels (Solar PV) and keep the batteries charged.

Care must be taken to size the Solar panels: install too little and Solar becomes nothing more than a token and worse can end up increasing the electricity bills, install too much and extractable energy is wasted because there are not enough loads or storage to utilize the energy that can be generated. Ideally, Solar installation must match the day time energy requirement in the absence of option of grid export. In battery based storage systems an additional factor comes into play. When Solar is added it becomes the second charger to the existing set up and depending on the AH (Ampere hour – the storage capacity of the batteries) the combined charge rate of both the Solar PV and Inverter can exceed the ideal charge rate to the batteries. If charged at a rate (speed of charging – higher current put into the batteries in a shorter time would mean higher charge rate) higher than what is ideal, the batteries end up getting ‘surface charged’ with diminished capacity to hold charge and eventually shortening the expected life of the batteries.

This is where REhub makes a big difference. By virtue of monitoring the currents going in and out of the batteries and dynamically adjusting charge rate (output of REhub to the batteries), REhub is the only device that adds Solar in way to maximize the extraction of Solar without exceeding the recommended charge rate to the batteries. Power cuts do not mean compromises anymore. Switching on the extra fan or light in the daytime or watching a Soap in the afternoon need not lead to an anxiety on the availability of power in case of further power cuts in the late evenings or night time. Adding Solar enhances the convenience that was promised by the addition of the inverter – at a cost that is lower month on month both to the consumer and to the environment.

What about cities and places where blackouts are not common ?

The next post in this series will cover that.

Installing REhub MPPT -A Customer Video.

My Eco Day is a Solar Startup run by Ifthikar Javed. My Eco Day  has an online Solar store and recently they have also opened a brick and mortar Solar shop in Calicut, Kerala.Yesterday a  customer called me up asking for REhub after seeing a video done by My Eco Day on Youtube. MyEco Day promotes Solar DIY in India.

I have posted a link to the video below. Ifthikar is doing the voice over: Watch his explanation of how REhub reacts to different scenarios of Solar and Battery availability. Since REhub  is all about adding Solar to existing inverters, the question frequently asked is ” Is Solar going to be used the maximum ? ” If so how ? Check this out.


System Integrator Question Series – What Panel should I use with my PWM charger?

The other day, in a meeting with a system Integrator from Trivandrum (He has been a great supporter of our products and has given us tons of valuable feedback and some good business too) we were asked if it was OK to use a 60 Cell panel with 24 V battery system and a PWM charger.  We had to tell him that it would be better to use a 72 Cell Panel with a 24V battery system esp with a PWM charger. (Similarly a 36 Cell panel with a 12V battery system).

The reason is simple.The Current – Voltage curves of the 60 Cell Panel seem to indicate that the MaxiEldora250Pmum power point is at 30V. Given that the 24V batteries are going to operate at a maximum of 28V and therefore the Maximum power Voltage is higher than the highest operating battery Voltage, (See the I-V curve on the left – this one is from the Panel data sheet of Vikram Solar  the PWM charger ought to have no difficulty charging the batteries at all the various Battery voltages of a 24V battery system. However – these curves are at STC (Standard Test conditions ) and even at STC,  at lower Solar availability, the maximum power point has shifted to a lower Voltage value. And real life conditions are very different from STC . For one, the ambient temperatures when we get 1000W/m2 of Solar power is likely to be  much higher than 25 deg C that the STC uses. The actual cell temperature is probably close to 65- 70 deg C. Panel data sheets refer to the drop in Voc and the Vmpp with a temperature coefficient. (e.g the same Panel data sheet the Temperature coefficient is mentioned as -0.31% per deg C. Apply that to the STC values and we find that the actual maximum power point is now nearer 24-26V and not the 30V that we see in the standard I-V curves. Now with battery voltage at 24V or more – the PWM charger is forced to operate to the right of the knee instead of the left of the knee.  ( in the curve above , the maximum power point is close to the knee of the curves).

The problem with operating at the right of the knee instead of the left of the knee is obvious once we look at the curves above. On the right of the knee, we see that the curve is dropping sharply and that means that a small shift to the right on Voltage will mean a large drop in available Current – and therefore the Power ( Power: multiply I and V). Whereas if we operate to the left of the knee- the value of Current (I) does not change much and we end up getting more power in the exact same conditions. We may even be operating on the knee and getting the maximum power with the PWM charger if the knee is ‘right’ enough. Take a look at a similar IV curve for a 72 Cell panel below. Even under real life conditions, we are likely to operate at the left of the knee and that is why it is advisable to use a 72 cell panel with 24V battery systems and 36 cell panels with 12V battery systems. Eldora200P








Will I save Money If I Install Solar ?




Yes that is the bottom line isn’t it ? Should be straight forward to answer if – the Ifs and but(s) appear and the issue becomes too confusing.  Let me explain why.

Installed Wp Vs What we actually get on field. 

Wp rating of the Solar Panel is what you pay for. However the Wp rating is based on Standard Test Conditions (STC) – which differ widely from the conditions in your installation. For example STC assumes 25 deg C for the Solar Cell temperature. What you have on your roof top is an ambient temperature of say 35 deg C and the Solar cell temperature operating at closer to 70 deg C ! As the output of the Solar module drops with increased Cell temperature -the maximum that you can get on the roof top is already lesser than what the name plate suggests. A good rule of thumb is about 75% of the rated Wp that you can get as the maximum.

Now that you have calculated the maximum Wp that can be extracted from the Solar Panel- you need to know how many units of electricity the panel would extract per day.

In India – you can assume 5.5 hrs equivalent of Wp – roughly for the calculation of Units generated. The extract-able energy from the Solar PV panels varies through the day and there would be a time close to noon when this is close to the maximum that can be extracted from the Solar Panels. Let us say Sun is available for 10 Hrs. When we say 5.5 hrs of Wp equivalent – what that means is that the amount of energy that can be extracted in Ten hours is equivalent to having 5.5 Hours of maximum Wp extracted.

Based on the thumb rule above and the 5.5 hrs Wp equivalent – an one kWp module will deliver approximately – 5.5 X 1 X 0.75 = 4.1 kWhr Or about 4 units per day .

The energy generated is only one side of the equation. It is useful only if the energy generated can be consumed !

Ex-tractable energy  Vs What is actually consumed

In the absence of Grid tie option – where a solar installer can export excess energy into the local grid and can get paid for the same- offgrid installations can end up wasting extractable energy unless Solar Panels are undersized. There would be periods when the batteries are full and loads not enough to consume all the available energy from the installation. In this case – the charger ends up extracting lesser than the maximum power available – thus reducing the returns from the installation. Its difficult to match the load timing to the Solar power availability. Note however – this author feels that if the objective is to obtain sufficient back up – then that needs to be addressed first. That need would drive the panel sizing and the ‘wasted’ energy is not such a bad thing.

Yes – Installing Solar PV does save money. 

Solar energy when used – saves on the power otherwise drawn from the grid for charging the batteries or powering the loads. However most offgrid systems still have a payback period of > 8 Years and compares well with alternatives like Diesel and not having power at all. In states like Kerala where the rate per Unit of electricity for increased consumption is much higher – Solar PV would deliver a shorter payback.

* Picture of the electricity Meter is from BESCOM site.