Costs for Small MW-Scale Grid Connected Power Plants in India – LCOE and Optimal Tariff

This blog post focusses on the cost of generation (LCOE or levelized cost of electricity) and the optimal bid for small scale solar projects in India (1-5 MW)
For our analyses, solar projects with and without trackers have been considered.

The analyses was done for three scenarios – pessimistic, likely (acceptable) and optimistic

We have not considered the accelerated depreciation benefit for this analysis. The main reason being, a good % of the first time developers putting up small-medium scale solar power plants are first time entrepreneurs, at least they do not have a profitable enough company right now to take advantage of the AD.

HIGHLIGHTS & INFERENCES

LCOE

Not surprisingly, the LCOE and the tariff for good project IRRs, are far more favourable for projects with trackers than for projects without.

The LCOE is the lowest for projects with trackers with reasonably optimistic assumptions. At Rs 4.77/kWh, this probably represents the lowest generation cost of solar power in the country, for solar power plants in the sub 5 MW scale. For projects without trackers, the lowest LCOE is about Rs 5.2/kWh.

The takeaway? Do seriously consider the use of trackers for your solar power plants. Sure, use of trackers increase the extent of maintenance, and as a result the maintenance expenses, but the additional returns more than compensate for the extra costs.

Bids

The lowest bid that fetches a decent IRR (project IRR of about 13%) for developers at sub 5 MW scale is approximately Rs 5.5/kWh for projects with trackers, and about Rs 6/kWh for projects without trackers. But these projects need to be executed with consummate skill so that the optimistic assumptions are satisfied.

For those who wish to play it safe because they are not sure the optimistic assumptions are too optimistic in their cases, can look at the lowest bids of Rs 6.25/kWh (with trackers) and Rs 6.75/kWh (without trackers)

Generation

With trackers, one can expect generation upto 19.3 lac units per annum per MW. Without trackers, one can go upto 16.6 lac units / MW/ year. Some power plants have reported upto 20 lac units / MW/year with trackers.

INTRODUCTION

Most of the limelight for grid connected solar projects has been on large-scale projects.

In fact, the recent SunEdison tariff bid at Rs 4.63/kWh made everyone starting to wonder if such a tariff is indeed competitive, and if yes, has the levelized cost of solar power dropped to really low levels. Well, for Sun Edison to make decent profits at Rs 4.63/kWh, the LCOE needs to be lower than Rs 4/kWh!

Is the cost of generating solar power really low for large solar power plants of sizes 50 MW and above? (Sun Edison’s total size was 500 MW, expected to be implemented in parcels of 50 MW each) Now, that is a question the answer to which is only of academic interest to most of us because rarely will one of us get a chance to put up a 50 MW power plant and above.

What is of interest to most small/medium developers in the country is the answer to the following question:

For solar power plants of size range 1-5 MW, what could be the lowest tariffs that will still fetch a decent return?

This is the question we decided to answer in detail in this post. Fortunately, we have just completed a couple of assignments at Solar Mango in which we guided developers new to solar to put up solar power plants in the range 2-5 MW. We decided to estimate the LCOE (levelized cost of solar power) and the optimal tariffs using real-life data we collected during these recent assignments.

What we found was surprising: Even for a relatively smaller power plant in the scale 1-5 MW, if the developer is able to optimally undertake all activities including financing, the LCOE is in the range 4.8-5.2 Rs/kWh, and the returns (project and equity IRRs) are at acceptable levels at a tariff as low as Rs 5.5/kWh (with trackers) and Rs 6/kWh (without trackers).

You might think this is not a big deal when we are hearing bid numbers like Rs 4.63/kWh, but we at Solar Mango feel it is indeed a revelation, as the bids for small-size (1-5 MW) have usually been upwards of Rs 6/kWh or at least close to Rs 6 if it is sub-6. But our analysis shows that it is just about possible to get a Rs 5.5/kWh bid that returns a reasonable IRR. This, without the benefit of accelerated depreciation!

I am sure you will be eager to know more details – what are our assumptions, and how do the LCOE and optimal tariff vary for optimistic, pessimistic and acceptable scenarios.

Here we go.

The key to achieving these costs and the tariff bids is an optimal design and construct for the solar power plant. What indeed are the components of such an optimal construct?

Key aspects of the optimal construct

Not surprisingly, such an optimal construct has to focus on two main aspects:

  • Maximising power generation
  • Minimizing costs

Main drivers to maximise solar power generation

  • Design – Optimal design and implementation of the power plant
  • Trackers – Use of trackers
  • EPC – Ensure you go for the right EPC

Main drivers to minimize costs

  • Minimize capital costs – Simple Bargaining – Bargain well to get the premium brands for each of the components at the best possible price
  • Minimize lending rates – Go for the lowest possible lending rates
  • Minimize O&M expenses – Keep the O&M costs at an optimal level
MAXIMISING POWER GENERATION – DETAILS

Well, two of the aspects mentioned above are well-known: paying attention to every important of power plant design, and choosing the right EPC who will ensure he does not cut corners while cutting costs.

What is not so prevalent in India is the use of trackers. Solar Mango estimates that less than 100 MW of solar power plants in India (of the 4500 MW grid-c0nnected power plants installed as of Dec 2015), that is, only about 2% of solar power plants use trackers, compared to much higher %s in the US and Europe. Well, our analysis clearly shows that the use of trackers has the potential to significantly bring down the LCOE of solar power.

MINIMIZING COSTS – DETAILS

Minimizing capital costs

Capital costs for solar power plants are the hottest topic today, as the LCOE depends hugely on this number. We have seen developers trying to reduce this by indiscriminately compromising on quality. We all know this is a bad idea; what we would suggest is to focus on certain components within capital costs that could be reduced without affecting the overall quality of the power plant infrastructure and without any negative effects on the generation. Bargain hard with the EPC, but bargain sensibly!

Minimize lending rates

This is where a good amount of homework can give you significant returns.

Banks in India are lending in the range 11.5-12.5% for small-scale solar plants in the range 1-5 MW. IREDA, the government’s renewable energy lending agency, starts at 10.8% and even for lower credit rated entities, lends at 11.4%

Do your homework, and it is possible you can get the lending rate to 11.5. If you are able to provide 100% collateral, and if you are able to really convince your bank in a big way, 11.5% might just be possible. 

If you are able to get IREDA lend to you, and do your homework well, a 11% interest rate might just be possible for you.

A 11% interest rate could really make an awesome difference as you will see in the analysis that follows.

Minimize O&M expenses

O&M expenses account for 5-7 lacs/MW/year for small scale solar power plants. This plays a minor role overall, so don’t be too bothered. But if you can save a couple of lacs per MW, why not. But if you feel the O&M quality could be compromised by the EPC if you ask for a lower O&M price, don’t bother trying to reduce it. That would really be penny-wise, pound-foolish.

More details on specific aspects

CUF & Generation

Now, this is obviously one of the key levers.

For solar PV power plants without trackers, if these are installed in regions with good DNI (higher than 5.5 kWh/m2/day), a generation of 15.5-16 lacs/annum (about 18% CUF) in the first year and a 0.5-1% decrease year-on-year until the 25th year are acceptable numbers. Optimistically, one can perhaps go up to 16.5 lacs per annum / MW.

For solar PV power plants with trackers, an optimistic CUF is 21% (some might say even 22%; there are power plants that have achieved upto 22% CUF with trackers, but these still seem to be the outliers in terms of high generation) and an acceptable CUF is 19%, if these are installed in regions with good DNI (higher than 5.5 kWh/m2/day). These correspond to generations of 19 lac units/MW/year optimistic and 18 lac units/W/year acceptable. Similar to the one without trackers, a 0.5-1% decrease year-on-year until the 25th year are acceptable numbers.

Total Capital Cost

With today’s prices, you can get total capital cost of a 1 MW solar power plant with trackers at just above Rs 6.2 crores per MW. We have taken this as Rs 6.2 crores.

Types of Panels Used

We are looking at using panels from tier 1 solar module makers, such as Trina, JA Solar etc.

Inverters

We are looking at using inverters from top tier makers such as ABB.

O&M Costs

We have considered annual O&M costs to be around Rs 7 lacs / MW for O&M of tracker based solar power plants. For those without trackers, the O&M costs could be much lower, perhaps about Rs 4-5 lacs/MW/year

Transformers, Cables and other Electricals

All these are from prominent and reputed makes/brands

Trackers / Mounting Structures

Again, we are considering trackers from prominent Indian tracker + mounting structure makers. The mounting structure/tracker market is indeed relatively unorganized compared to inverter and panel markers, and hence getting standard prices for these is not easy. However, we are basing our data based on the real time work we had done in the past few months and actual quotations received.

Loan & Interest Rate

This is a critical factor. Today, many conventional banks (nationalised and private) are willing to give loans to 2-5 MW scale solar power plants at about 12%.

Agencies such as IREDA are willing to give these at 11.5% and less too, though they are keen on lending for scales above 5 MW – that should however not deter you, as IREDA is willing to consider sub 5 MW solar power plants too.

It is however becoming increasingly possible for financing rates to be sub-12%, with the overall decreases in base rates from the RBI, as banks get more comfortable with solar, and with the government really pushing banks to give a better deal to solar power developers.

For small-medium solar power plants (1-10 MW), you can optimistically expect 11% if you are really lucky, 12% in most cases, and in the pessimistic scenario, 12.5%.

We have considered the debt:equity to be 75:25, this is again something many banks today are willing to give.

Loan Term & Moratorium

Loan term

We have taken the loan term to be 12 years, including a year of moratorium. Note that banks are today willing to give up to 15 years tenure.

Moratorium

We have taken a moratorium of 1 year.

Taxes

We have considered a tax holiday for the first ten years, during which only the minimum alternative tax is applicable. Beyond the 10th year (years 11-25), we have taken the corporate tax rate of 35%

All the other cost considerations have been taken as well, including

  • Approval and liaison costs
  • Service room
  • Interest costs during construction
  • Working capital in the beginning stages, equal to 2 months’ revenues
  • Suitable boundary wall
  • Lighting for the entire power plant
  • Roads inside the power plant
  • Grid extension and substation bay extension

Things not considered

  • We have considered normal depreciation with the WDV method, and NOT accelerated depreciation
  • Cost (if any) for securing right of way from solar power plant to substation

Let us look at what we are getting with all the above data, inputs and assumptions

Let’s look at the estimates for the following, with the above data and assumptions.

  • Equity IRR (EIRR)
  • Project IRR (PIRR)
  • DCSR

We have done the above estimates for the following sets of circumstances

  1. With and Without Tracking
  2. For three different sets of assumptions – Optimistic, Acceptable, Pessimistic

Solar Mango can provide expert assistance in putting up small and medium scale MW solar Power Plants. Click here to know more or send a note to ramya@solarmango.com

 

WITH SOLAR TRACKERS

PESSIMISTIC ASSUMPTIONS

  • Capital Cost: 7 crores/MW (all inclusive)
  • Generation: 20% CUF (approx 17.5 lac units/annum/MW)
  • Interest rate: 13%
  • Generation Loss: 4%
  • Accelerated Depreciation Benefit Applicable?: No

LCOE: Rs 5.97/kWh

Tariff EIRR PIRR DSCR
6 9.8 10.8 0.98
6.25 11.2 11.4  1.02
6.5 12.6 12.04 1.07
6.75 14.7 12.1 1.11
7 15.7 13.3 1.16
7.25 17.4 13.9 1.20
7.5 19.1 14.5 1.25
7.75 20.9 15.1 1.29
8 22.7 15.7 1.34
8.25 24.6 16.3 1.38

 

ACCEPTABLE ASSUMPTIONS

  • Capital Cost: 6.75 crores/MW (all inclusive)
  • Generation: 21% CUF (approx 18.4 lac units/annum/MW)
  • Interest rate: 12
  • Generation Loss: 3%
  • Accelerated Depreciation Benefit Applicable?: No

LCOE: Rs 5.33/kWh

Tariff EIRR PIRR DSCR
5.3 (LCOE) 9.8 10.3 0.99
5.5 11.1 10.8 1.03
5.75 12.7 11.5 1.08
6 14.5 12.2 1.13
6.25 16.3 12.9 1.19
6.5 18.1 13.6 1.24
6.75 20.1 14.3 1.29
7 22.1 14.9 1.34

 

OPTIMISTIC ASSUMPTIONS

  • Capital Cost: 6.5 crores/MW (all inclusive)
  • Generation: 22% CUF (approx 19.3 lac units/annum/MW)
  • Interest rate: 11%
  • Generation Loss: 2%
  • Accelerated Depreciation Benefit Applicable?: No

LCOE: Rs 4.77/kWh

Tariff EIRR (%) PIRR (%) DSCR
4.75  10.3  10  1.02
5  12.2 10.8  1.08
5.25  14.1  11.5  1.13
5.5  16.1  12.5  1.2
5.75  18.2  13  1.25
6  20.4  13.8  1.31
6.25  22.7  14.5  1.37
6.5  25  15.7  1.43

 

WITHOUT SOLAR TRACKERS

Now, let’s do the same calculations as above, but this time, without trackers. Thus, an overall cost of Rs 5.75 crores/MW is acceptable, and a generation of about 16 lac units is acceptable.

PESSIMISTIC ASSUMPTIONS

  • Capital Cost: 6.6 crores/MW (all inclusive)
  • Generation: 17% CUF (approx 14.9 lac units/annum/MW)
  • Interest rate: 13%
  • Generation Loss: 4%
  • Accelerated Depreciation Benefit Applicable?: No

LCOE: Rs 6.62/kWh

Tariff EIRR PIRR DSCR
6.6 9.3 10.5 0.97
6.75 10.1 10.9 0.99
7 11.4 11.5 1.03
7.25 12.7 12 1.07
7.5 14.0 12.6 1.11
 8  17  13.7  1.19

ACCEPTABLE ASSUMPTIONS

  • Capital Cost: 6.35 crores/MW (all inclusive)
  • Generation: 18% CUF (approx 15.8 lac units/annum/MW)
  • Interest rate: 12%
  • Generation Loss: 3%
  • Accelerated Depreciation Benefit Applicable?: No

LCOE: Rs 6.08/kWh

Tariff EIRR PIRR DSCR
6.1 11.3 10.9 1.04
6.25 12.2 11.3 1.06
6.5 13.7 11.9 1.11
6.75 15.4 12.5 1.16
7 17.0 13.2 1.21
7.25 18.8 13.8 1.25
 5.2  20.6  14.4  1.3

OPTIMISTIC ASSUMPTIONS

  • Capital Cost: 6.1 crores/MW (all inclusive)
  • Generation: 19% CUF (approx 16.6 lac units/annum/MW)
  • Interest rate: 11%
  • Generation Loss: 2%
  • Accelerated Depreciation Benefit Applicable?: No

LCOE: Rs 5.18/kWh

Tariff EIRR PIRR DSCR
5.2 10.4 10 1.02
5.5 12.4 10.8 1.08
5.75 14.2 11.5 1.14
6 16.1 12.2 1.19
6.25 18 12.9 1.25
6.5 20 13.6 1.3

 

SUMMARY for SMALL SCALE MW SCALE POWER PLANTS

All data in Rs/kWh

  With Trackers Without Trackers
LCOE Optimal Tariff for Project IRR of about 13% and DSCR of 1.30 LCOE Optimal Tariff for Project IRR of 13-13% and DSCR of 1.30
Pessimistic Assumptions 5.97 6.75-7 6.62 7.5-7.75
Acceptable  Assumptions 5.33 6.25-6.5 6.08 6.75-7
Optimistic  Assumptions 4.77 5.5-5.75 5.18 6-6.25
HIGHLIGHTS & INFERENCES – REPEATED FOR EASE OF READING!

LCOE

Not surprisingly, the LCOE and the tariff for good project IRRs, are far more favourable for projects with trackers than for projects without.

The LCOE is the lowest for projects with trackers with reasonably optimistic assumptions. At Rs 4.77/kWh, this probably represents the lowest generation cost of solar power in the country, for solar power plants in the sub 5 MW scale. For projects without trackers, the lowest LCOE is about Rs 5.2/kWh.

The takeaway? Do seriously consider the use of trackers for your solar power plants. Sure, use of trackers increase the extent of maintenance, and as a result the maintenance expenses, but the additional returns more than compensate for the extra costs.

Bids

The lowest bid that fetches a decent IRR (project IRR of about 13%) for developers at sub 5 MW scale is approximately Rs 5.5/kWh for projects with trackers, and about Rs 6/kWh for projects without trackers. But these projects need to be executed with consummate skill so that the optimistic assumptions are satisfied.

For those who wish to play it safe because they are not sure the optimistic assumptions are too optimistic in their cases, can look at the lowest bids of Rs 6.25/kWh (with trackers) and Rs 6.75/kWh (without trackers)

Generation

With trackers, one can expect generation upto 19.3 lac units per annum per MW. Without trackers, one can go upto 16.6 lac units / MW/ year. Some power plants have reported upto 20 lac units / MW/year with trackers.

Solar Mango can provide expert assistance in putting up small and medium scale MW solar Power Plants. Click here to know more or send a note to ramya@solarmango.com

Other web resources from Solar Mango

IIT Club

Green Expo

RLS

164 thoughts on “Costs for Small MW-Scale Grid Connected Power Plants in India – LCOE and Optimal Tariff

    1. Narasimhan Santhanam Post author

      Hi Nawal

      Thanks for asking.

      Solar trackers usually come as part of the mounting structure.

      On the cost side, mounting structures for a 1 MW solar power plant could cost around Rs 40 lacs, a mounting structure along with trackers could cost an additional 40 lacs per MW.

      So, effectively, the cost of having a tracker results in an additional cost of about Rs 40 lacs per MW.

      While this might seem a lot (we are talking about 7-8% added to project cost for a single axis tracker), trackers also result in 15-20% higher generation (for single axis trackers, again). You can see from the analysis made in the post above that, because the higher returns more than compensate for the higher cost, the LCOE, project IRR and equity IRR for tracker-based solar power plants are better than those for power plants without trackers

      Hope I made sense!

    2. Narasimhan Santhanam Post author

      Just to add one more angle Nawal, the cost of trackers depend on whether these are single axis (track the sun on one axis) or dual axis (track on both x & y axes).

      Dual axis trackers could cost 25-30% higher than single axis trackers, and also provide 30-35% more than single axis trackers. But beware, many locations in India (especially mid and south India) will be more suited to single axis trackers – in the locations, single axis trackers give the best value for money, while locations in north India could do with dual axis trackers.

      It is a fairly intricate subject, but I hope I have been able to throw light on the key dimensions for decision making

      Thank you once again for your question and interest

  1. Pingback: Quality of India's Solar Installations Compromised in the Race to Low Tariffs - Solar Mango – #1 guide for solar

  2. Raghava charyulu T V

    Sir ,
    A very useful and informative article .
    The cost escalation with solar tracker is not mentioned .
    Inder the heading of Minimising o&m cost , in second line it is mentioned shave instead of save .
    Please correct if it is ok .

    1. Narasimhan Santhanam Post author

      Dear sir

      Many thanks indeed for your kind comments

      On the cost escalation with solar trackers – while it has not been specifically mentioned, you can see that the cost added per MW for all three options with trackers is Rs 40 lacs/MW higher than the equivalent options without trackers. Rs 40 lacs/MW is thus the additional cost we have taken for trackers.

      Shave has now been saved as “save”, thanks for pointing it out.

    1. Narasimhan Santhanam Post author

      Well, Neha, what is a good IRR I would say to a certain extent depends on the individual investor, but in general, I would say project IRRs of 13% and above can be considered to be good for projects with long lives such as that of solar. Would like to know others’ opinions on this, thanks

    2. Narasimhan Santhanam Post author

      Well Neha, I was reviewing my earlier comments, and thought I’d add this: What is a good project IRR also depends on how much the investor is getting from his other projects.

      Remember, many large solar projects involve project investments of upwards of $100 million, and there might not be too many alternative investments for these investors that give super returns.

      Hope I was able to make a point here

      Thanks once again for your question, all the best!

  3. Alex

    You have said yourself that Rs 5.5/kWh is the lowest bid possible for good returns? How did a prominent co such as Sun EDison quote almost 1 Re less? Are they losing money in this project?

    1. Narasimhan Santhanam Post author

      Well Alex, I am not sure if the Sun Edison project is similar to this power plant, for multiple reasons.

      One, Sun Edison/ capacity is much larger than the small scale power plants we are talking about – they are talking about 500 MW in all, in parcels of 50 MW each. While solar is indeed modular, there still are reasonable economies of scale

      Two, Sun Edison might be getting incentives in terms of land and a few other things such as approvals, grid extension and bay extension etc. These could add to some real useful savings in capital costs

      Three, we do not know what the rate of interest is for Sun Edison’s loan component. It could be lower than the 11% we have assumed here under the optimistic scenario, especially if they are getting international funding.

      Finally, some of the conservative assumptions we have made in terms of generation loss etc might not either be there or it could be even lower. Even these small %s count…

      As a result of all the above, it is quite possible that Sun Edison’s LCOE is a lot less than the best LCOE we have arrived at for this – which is, as you have rightly pointed out, Rs 5.5/kWh

  4. Monica Chodhury

    What is minimum Dscr banks want in India? Some are asking for very high – up to 1.5, which is not possible for any solar project…

    1. Narasimhan Santhanam Post author

      Monica – thanks for your comment

      Yes, I have also heard of banks asking for DSCR of 1.4 and above. Now, it is very difficult to get those kinds of DSCRs unless the tariff offered is super high.

      From what many civilized banks do, I think DSCRs of 1.25 and above should be considered safe. But then, I am not a bank! It is pretty much a bank’s privilege to decide what DSCR they want, but at benchmark of 1.5, I am sure that they will clear very few loans for solar power projects!

  5. Ravi Sundaram

    From what I know, solar pv does not have economies of scale, as solar panels are absolutely modular. If this is case, how can larger power plants from companies such as Sun edison get lower LCOEs? Will the lowest for them also only be Rs 5.5/unit, which is the case for the 1-5 MW scales that you have analysed above?

    1. Narasimhan Santhanam Post author

      Hi Ravi – pl see the reply I have given to Alex above.

      (BTW, I think you meant solar pv does not benefit from economies of scale…)

      Solar PV power plants might not have great economies of scale, but they do certainly have some economies of scale.

      How?

      Sure, modules are modular, but there still is the aspect of logistics and other dimensions involved with the installation of modules which follow the economy of scale curve.

      Second, modules form only 45-50% of total cost of a solar pv power project. The rest are for inverters, cables, mounting structures, AC side stuff like switchgear and transformer, and costs for installation. All these dimensions benefit from economies of scale.

      Hope I have been able to answer your question…thanks again for asking

    2. Narasimhan Santhanam Post author

      Another point I might want to add here is that, while solar pv in concept might be modular, scale still weighs in for something when it comes to financing as well.

      For instance, if you wish to get international loans which are cheaper than Indian loans, these international loans are more easily available to larger power plants than smaller ones, as the international banks surely would not be interested in lending to small projects as their servicing costs are fairly high and are kind of fixed, whatever the size of the project is

      This is yet another example where scale can be shown to matter for costs in the case of solar pv

  6. Raful Akhtar

    Dear Mr Narasimhan – thank you very much for the really comprehensive and detailed analysis. V useful and I have not found such an analyses any where else. Thank you sir once more…

    Only question: All the costs and other data you have used, are these valid for the next 3 months at least?

    Also want to know, what do you think could a similar analysis reveal if it is done one year later? What could be the lowest LCOE and the lowest feasible bids be?

    1. Narasimhan Santhanam Post author

      Dear Raful

      Many thanks for your positive and appreciative comments.

      On validity of data and inputs: Well, the data used were based on projects we had worked with as late as November 2015, so we can safely say that these data will be valid to a large extent until the of financial year 2015-16, that is until Mar 2016.

      On our prediction for costs, returns and tariffs one year on from now (end of 2016), well you know as much as I do that prediction is a difficult business. So whatever I am forecasting needs to be with a disclaimer 🙂

      However, as you asked, these are my predictions:
      For small solar power plants (1-5 MW scale), the lowest LCOEs (which will happen for solar projects with trackers) will be about Rs 4.5/kWh (instead of 4.77 now) and the lowest bid that can get you a decent return will be about Rs 5.25/kWh (for projects with trackers once again), down from Rs 5.5/kWh.

      What makes me predict the above numbers?
      You can see that I am predicting a 5% decrease; I see this mainly coming from two sources: Decrease in bank interest rate by perhaps 1% and some decrease in costs of modules and trackers. A 1% decrease in interest rate results in about 10 paise reduction per kWh in LCOE and I expect an equivalent decrease to happen (about 10-15%) from small decreases in costs of panels, inverters and trackers.

      Once again, thanks for asking!

  7. Ravi

    This is a very informative and useful article which ideally should be made available as a downloadable pdf for reference. Well written and good job!

    1. Narasimhan Santhanam Post author

      Dear Ravi

      Many thanks for the kind appreciation…

      Your idea of a PDF is indeed a useful thought, will work on this soon, thanks again

  8. Gopal.VMK

    Dear Narasimhan,
    This is a very detailed and useful article and appreciate your attention to pertinent details in your analysis.

    From my personal experience with at the least three nationalized banks, all of them demanded an overall DSCR of minimum 1.3 and one of the banks asked for 1.5. None of the banks I dealt with, both nationalized banks and Private banks, were ready to offer more than 10 years loan repayment period. Even with 100% collateral, banks are willing to provide a max of 10 year repayment period only, for small size (1 to 5 mw) projects.

    I completely agree with you that using trackers (Single, dual, etc based on the latitude) makes business/financial sense and we need to be installing more projects in India with trackers.

    The idea of smaller size solar parks wherein smaller size developers can still benefit from the savings that comes with having their plant inside a solar park (ease of evacuation/right of way and the savings that comes with that, ease of land acquisition, O&M cost, etc) should gain more prominence in India. The way things are moving in the solar domain now in India, it looks like only the Wellspuns, SunEdisions and ACMEs can do Utility scale Solar PV projects.

    As you said, it is indeed heartening to see the Solar filed take off in India. We have all been waiting a long time for this.

    Thanks again for your excellent, incisive analysis. Looking forward to more such articles from you where facts and numbers are presented and elucidated.

    Gopal

    1. Narasimhan Santhanam Post author

      Thank you Gopal, and it is nice to say Hi here once more.

      On DSCR, I agree. While some banks insist on 1.5, they themselves know it is pretty high and I am sure most will eventually agree for 1.3

      On repayment period, I am surprised you say that banks are not willing to give you over 10 years. Because, at lease for some of the banks we got in touch with for our clients, they were OK upto 15 years (including the 1 year moratorium). Which banks did you check with, by the way?

      Smaller solar parks is indeed a good idea. Perhaps 25 MW each, is that the scale you have in mind? Well, I guess this can be private sector initiative as well, though most folks I have met who are putting up solar parks are looking at 100 MW and above.

      The apprehension you have about solar becoming a Big Boys game is shared by many of us. Well, not sure how the future will move, but let us also realise that it might not be that easy for large scale solar power plants to be erected all over India. On paper, a few 500 MW solar parks facilitating the move to satisfy the government’s target of 60 GW looks like a good idea, but when it comes to actually getting it done, in terms of land procurement, having enough evacuation infrastructure, it might be more difficult than we think. We will come to know of how this transpires in 2016.

      I am only hoping that small businesses also have a role to play in the MW scale, grid connected solar power plants in India. Time will tell.

    2. Narasimhan Santhanam Post author

      Gopal – another quick question.

      What is the best interest rate you have been able to get from any of the banks so far?

        1. Narasimhan Santhanam Post author

          Thanks Gopal

          Any chances that we could get it rates slightly lower than 11.5% for 1-5 MW power plants, put up by individuals and not large companies?

  9. PRAFULLA

    You said O&m expenses will be in range Rs 5-8 lacs Per annum per MW. Can this be reduced? What are the main divisions for these O&m expenses?

    1. Narasimhan Santhanam Post author

      Dear Prafulla

      Thanks for your question

      The indicative cost I have given is the typical fees EpCs who take O&M services also, charge – about Rs 5 lacs/MW for farms without trackers and about Rs 7-8 lacs/MW for farms with trackers

      The main O&M for solar farms without trackers is the periodic cleaning of solar panels. Apart from this, if any of the panels get damaged, these need to be replaced through the use of warranty. Besides, a regular check of all the electrical connectivity will be another O&M job.

      Besides the above, a regular monitoring of the SCADA/monitoring system for signals of panel of array non-functioning could be considered to be another regular maintenance job – in this context, I presume the better O&M providers also have trained their engineers to do a better analytics of the data that are received from the monitoring system. While these monitoring systems do have some basic analytics (charts etc) themselves, we have found that with the data from a simple inverter, a lot more useful analytics could be performed – at Solar Mango for instance, we have shown how specialised analyses of data from the inverters of solar farms could unravel more than what are brought out by basic analytics available with the current monitoring systems.

      For farms with trackers, there is an additional maintenance required, as trackers have moving parts (actuators, motors and torque tubes) as well as sensors with IC chips. O&M companies typically charge about 2 lacs/MW/annum more for farms with trackers.

      Now, coming to your question of cost reduction for solar farms: Yes, it is definitely possible because these are early days and EPCs themselves are on a learning curve on what the actual costs and cost components are. For Indian conditions, a semi automated O&M system could be the one that results in optimal costs.

      One specific O&M cost reduction mechanism might be to simply follow what the conventional manufacturing and process industries follow: Have an excellent preventive maintenance framework and follow it to the last dot. I am not entirely sure if such a practice has already taken root in the Indian solar power sector. A related move will be to do constant reviews and analytics of data that are available from SCADA/monitoring system to identify any possible problems before they actually happen.

      1. Brahatheeswaran

        O&M costs can be significantly reduces if we include high quality monitoring systems on the DC side which constantly monitor and inform us of the small problems as and when these problems happen instead of having to wait until the problem is magnified

  10. Babukumar

    Some of the EPCs I talked to were not in favour of having trackers becos they feel that trackers can result in too many maintenance breakdowns. What are your perspectives on the maintenance Problems of trackers as you seem to recommend trackers for solar PV Power plants?

    1. Narasimhan Santhanam Post author

      Babukumar

      Many thanks for your comment

      There is no denying that trackers will increase the maintenance cost, as these present perhaps the only moving parts inside a large solar farm.

      But these do not significant increase the maintenance, in spite of having moving parts, and here’s why

      A tracker comprises mainly 3 components – the Sensor that has sun’s position data for the location, the Actuator/Motor that converts the sensor data into a mechanical movement, and the Torque Tubes that rotate and change the angle of the solar panels.

      As you can see, the sensor and the related components are mainly software and non-,moving hardware (location data are already stored in the sensor so it need not actually collect these everyday!); the actuator/motor surely has a mechanical movement but these are slow movements that are not for 24 hours, but for possibly a few minutes, happening at intervals – so these are fairly slow, discrete motions. The torque tubes also rotate only slowly.

      So, as you can see, while mechanical motions are indeed there, these are very slow and discrete and as a result are unlikely to result in significant maintenance requirements / breakdowns which could be the case with fast moving motors.

      For the sensor and the related software itself, if there is any malfunctioning, most times all it requires is to flip a switch on and off – not unlike what we do when there is something wrong with our computers – a simple reboot!

      Overall, hence, my submission is that trackers do add an additional component for maintenance, but it is not a large addition.

      And given that trackers can increase output by upto 25%, this addition to the maintenance is more than compensated for by the significant increase in output.

  11. Mohandas Menon

    Hi Narasimhan

    Thanks for a very insightful and detailed post.

    Can you also give us an idea of how these LCOEs have fallen over the past 5 yrs, starting with perhaps 2015?

    1. Narasimhan Santhanam Post author

      Hi Mohandas

      Many thanks for your question

      It is tough to answer this question, as there were few reliable benchmarks available for costs of each component of the solar power plant in the initial years (from 2010 to 2012), but I will try – these are the approx LCOE estimates for small scale (1-5 MW ) solar power plants, and are for solar power plants without trackers:

      2010: Rs 14 / kWh
      2011: Rs 11
      2012: Rs 10
      2013: Rs 8.5
      2014: Rs 6.5
      2015: Rs 5.5

      Mind you, these are the LCOEs, not the optimal tariffs. A review of the post above will show you the difference between the two.

    1. Narasimhan Santhanam Post author

      Hi Khushwant.

      You are possibly a very man, if you are indeed having these as the inputs to your solar power plant.

      Anyway, at capex of Rs 5 crores/MW and interest rate of 10.5% and with the additional inputs: CUF 19% and generation loss: 3%, you get a really good LCOE of Rs 4.25/kWh.

      Just in case you are curious, for the above LCOE and other inputs, you get a decent IRR (about 17% equity and 12.5% project IRR) at a tariff of Rs 5.5/kWh

      Hope it helps

    2. Narasimhan Santhanam Post author

      BTW, my first sentence in the earlier comment should read: “You are possibly a very LUCKY man, if you are indeed having these as the inputs to your solar power plant.”!

  12. Sushil Prasad

    Informative article Mr. Narshiman. Thanks for the same.

    Was wondering if trackers add complexity as far as O&M is concerned. Are these trackers manually operated or are they automated / programmed?

    Where do you see the LCOEs stabilizing at? The prices cannot go southwards forever. Many of the customers hesitate in adopting the technology because they assume that they would get a better deal in the future so why commit to PPAs at the current prices. This is creating an inertia at retail level.

    Best,

    Sushil Prasad

  13. Pingback: Dos and Donts for Indian MW Solar Power Plant Developers - Solar Mango – #1 guide for solar

  14. Ashok

    Thanks Narasimhan for the wonderful stuff. Very informative indeed..

    Could you please help in bifurcating Rs, 6.2 Crore, the cost of 1 MW solar pv plant with trackers, into its components such as modules, inverters, cables, mounting structures with trackers, AC side stuff like switchgear and transformer, and costs for installation.

    Thanks.

    1. Narasimhan Santhanam Post author

      Hi Ashok

      Thanks for the question

      Firstly, Rs 6.2 crores / MW is the very best price you can get as of Jan 2016 with single axis trackers.

      This price does not include the cost of taxes (which could amount to Rs 10-15 lacs /mw if you had taken care of all the exemptions) and also costs for approvals etc ( which could be a few tens of thousands to a coupla lacs depending on how “good” and transparent the state is).

      We have not considered the cost of land as this could vary significantly from one place to another, and actually many folks who come to Solar Mango for consulting already own the land on which they plan to have the solar power plant.

      With these disclaimers made, the cost breakup is roughly as follows (lacs/mw):

      Cost of modules: 325
      Inverter: 35
      Mounting structures + trackers: 85
      Other BOS (cables, other electricals, monitoring systems…): 80
      Civil construction (materials + labor): 100

      The above is the rough breakup, hope it helps

      1. SP

        The taxes mentioned above are the tax during installation of plant or these are taxes every years in books ?

        15 lacs is 2.5% of 6Crores .

      2. Dinesh Shah

        Well, I think Mr Narasimhan is referring to the taxes (customs and excise) to be paid on the various components you are purchasing for the solar power plant.

        If this is the case, these are one time taxes

    2. Narasimhan Santhanam Post author

      BTW Ashok, I would also request you to get in touch with a couple of good EPCs and get their quotes on similar lines

      I can tell you one thing for sure: You will be surprised with the difference in the quotes – both overall and breakups.

      After negotiations, they are all likely to fall in line with the above

      Just in case you find that numbers given by others are quite different (after negotiations) from what I have mentioned above, do let me know by perhaps mentioning that here? Thanks

  15. Satya Priya

    Dear sir,

    we are looking for a development of 2MW on a roof top of one of our customer in India.. Looking for BOT based tariff bidding. Could you guide us on the rate of tariff prevailing and its calculations..

    1. Narasimhan Santhanam Post author

      Dear Satya

      Many thanks for your interesting question

      I am presuming that you are referring to a model where you would be operating a power plant for 20-25 years on your customer’s rooftop and that there would be no transfer of ownership during this period – that is, I am assuming you are referring to a BOO (build own operate) and not BOT (build own transfer), as the latter could be a more challenging task to estimate the price for, as there more variables at play here

      For a BOO model, as you will know the market is still fairly unorganized so there is nothing like a benchmark price I can suggest. It would be a smarter move for you to rather first calculate the LCOE for the rooftop solar based on the amount of generation you expect to have and your contract period for the BOO model, and based on this, arrive at a tariff that provides you the IRR you are looking for. (you can also directly calculate the tariff without having to necessarily calculate the LCOE as long as you know all the input costs and the returns you are looking for; the LCOE is more for academic benefit)

      That said, the typical prices I hear from the market from vendors offering BOO model are in the range Rs 6.3-6.5/kWh with a 2-3% annual escalation from reliable vendors. Now. I have heard of vendors offering much less than this too, but not sure how they would be able to get good IRRs. Kindly note that, even at Rs 6.5/kWh, most investors get a good returns only because they also get the accelerated depreciation benefit from this investment, as LCOE for rooftop solar (100 kW+ installation) are still about Rs 6.5/kWh without the accelerated depreciation benefit, implying that the prices charged (without annual escalation) will need to be over Rs 7/kWh, and even for those with annual escalations of 2-3%, it might need to be closer to Rs 7/kWh than to Rs 6.5/kWh.

      Hope the data I have provided above was of some help, thank you once again for your question.

    1. Narasimhan Santhanam Post author

      Hi Meeran

      Thanks for your query

      You might want to see this analysis too – http://www.solarmango.com/blog/2015/12/05/solar-power-price-in-india-rs-4-2-kwh-by-2020-says-kpmg-and-rs-3-69kwh-by-2025/

      Anyway, in 2015 $, I think we can expect power from large scale (50+ MW) to cost about Rs 3.8-4 per kWh by 2020 (from 4.5-4.6 Rs/kWh currently); for smaller power plants (less than 5 MW), I would estimate it to be in the range of Rs 4.5-4.7 per kWh from the current Rs 5/kWh. I am factoring use of trackers for both these.

      Essentially, I am expecting a drop of about 10-15% in cost over the next 3-4 years; you might consider this a bit conservative, as solar power cost has been decreasing at a much higher rate in the past 5 years, but I feel cost reductions in the next 3-4 years are unlikely to be steep as there could even be some minor upward revisions in the cost of panels during this period.

      Let me know your thoughts on this. and many thanks once again for asking

  16. CA.S.Krishnan

    Dear Mr.Narasimhan thank you very much for enlightening people like me who are evincing interest in solar projects to light the
    better future for India where many villages are without lights. thank you once again for the input Sir.
    with best regards
    CA.S.Krishnan

    1. Narasimhan Santhanam Post author

      Thanks to you too Krishnan for your appreciation

      Do let me know if there is any other way that Solar Mango can assist you in your laudable social endeavours

      All the best!

  17. Rodriguez

    These are all data you have provided for India? How much will these be for South American countries like Chile and Brazil?? Will they be higher sir?

    1. Narasimhan Santhanam Post author

      Dear Rodriguez

      Many thanks for your query

      Right now, I do not have reliable data for cost breakups for these countries, but I am asking my team to figure these out and will provide a response here as soon as I can

      Many thanks once again, take care

  18. SP

    Hi Mr. Narasimhan ,

    Would you like to provide any update and feedback about below :

    As per JNNSM Phase-II, Batch-III and the current bid went to 4.63 from some developer. What is your feedback on
    1. tariff fix for 25 years @ 4.43/kWh.
    2. Loan at 10% for 8 years
    3 70:30 debt/equity ratio .
    4. VGF = 0

    Under this scheme Do you think people can get handsome return as per your calculation .

    What about taking loan from international bank I hear some banks are ready to give on 4-6 % as well .

    I wanted to setup in Rajasthan ( start with 5 MW and next year 50 MW) I have my own land. Could guide me it would be profitable and how ?

    Any contact information for initial discussion ?

    SP

    1. Narasimhan Santhanam Post author

      Hi SP,

      Many thanks for your detailed question

      I am sure you would have punched in all the data you have mentioned above into a financial model.

      I did the same with the financial model that Solar Mango uses for MW scale power plants, and the following are what I got:

      Firstly, I am assuming that you would be building a power plant on the scale of about 50 MW. Some of the capital costs I have put down are based on this scale.

      Second, I am assuming you would be using trackers, as that can significantly enhance your returns.

      Third, I am assuming that you would be paying for the land and not instal solar power plant, which means you need to add the cost of land too.

      Fourth, I am taking the interest at 10% (and the 4-6%)

      Finally, capital costs in my template do not include taxes, costs for approvals and costs for getting right of way.

      Whoa, how many conditions!

      Anyway, with the above, I am taking three different capital costs and CUFs and looking at how your returns are:

      Optimistic: Capital cost Rs 5.9 crores/MW, CUF: 23%
      Equity IRR: 16.3%
      Project IRR: 11.2%

      Likely: Capital cost: Rs 6.1 crores/MW, CUF: 21.5%
      Equity IRR: 12.7%
      Project IRR: 9.7%

      Pessimistic: Capital cost Rs 6.3 crores/MW, CUF: 20.5%
      Equity IRR: 10.2%
      Project IRR: 8.6%

      Now, all the above are assuming the use of trackers. If I were to estimate stuff without trackers, the following is what I get for the likely scenario

      Capital Cost: Rs 5.8/MW, CUF: 18%
      Equity IRR: 8.7%
      Project IRR: 7.8%

      The numbers in the no tracker scenario do not look very endearing, do they!

      So, the overall inference from my brief analysis would be:

      1. If you are able to get financing at 10% for years and keep the capital cost at Rs 6.1 crore/MW or less, you can expect about 9.5% Project IRR at Rs 4.43/kWh. not sure if this would be acceptable for you. (for completion sake, at tariff of 4.63. Project IRR goes to 10.3%, at Rs 5/kWh it is 11.5% and at Rs 5.25/kWh, it is 12.3%)

      2. If you do not use trackers, all the above returns are going to be significantly lower.

      Thanks once again for asking an interesting question and turning an otherwise unproductive Saturday afternoon into an intellectual one (as intellectual as it can get for me :-))

      Hope the above analyses are of some use

      And, all the very best in your project

      1. SP

        Thanks Sir, for taking time and replying over weekend. I hope this is compliment for me and seriousness for me to start something. I want to start with small, So I get all experience before jump to big one , e.g. start with 1-5MW first .
        Yes, I did prepared a financial model and counting below , Please guide me if I missing something , I tried to get your number although they are differ by 2-3 %
        Total Income = Unit x Tariff
        Total Expense =
        O& M Expense ( 1% of project Cost ) 5% escalation every years
        + Interest Payment
        + Insurance ( .15%)
        + Any Taxes ?? ( I have no idea I think its 10 yr tax holiday ?)

        Depreciation 80 % & Tax shield
        So Equity IRR = Total Income – Total Expense + Tax Shield for depreciation – Principal Repayment
        So Project IRR = Total Income – Total Expense + Tax Shield for depreciation + Interest Payment
        Added the equty and project portion and used IRR formula.

        Sir,
        – What would be the good tarif for VGF model and non -VGF model. I am confused what is the current 2016 good tariff and good IRRs
        – What is this collateral people discussing about – what sense I got is you have to give some security to bank like some property paper or gold to give loan. But the plant itself is the security – How come bank ask for other property – In USA you can not ask for some extra property paper . It means if you default you take the property /business on which bank agreed to give loan.

        Sir in JNNSM schme is something like this
        1. 4.43/Unit ( no escalation for 25 years )
        2.80 Lacs VGF fund from GOVT ( 50% on commission and 10% on rest of 5 years )

        I am getting below values, looks like something is wrong in my calculation , is these are good ?>
        Equity IRR 25.91%
        Project IRR 15.84%

        Regards, SP

      2. SP

        Sir, with AD these data will be affected ?

        is depreciation tax benefits advise to include while calculating Project / Equity IRRs ,

        I am not CA but was trying to get some resource in google but not clear answer

        Please help

  19. SP

    Hello Sir,

    For 5 MW what is the minimum Transmission line capacity needed ,? Could you provide some insight about GSS infrastructure and how the GOVT GRID gets electricity and where they are metered

    what is 11 kV, 33Kv, 132kv lines ?

    SP

  20. Dakshinamurthy

    Dear Narasimhan

    Excellent and detailed post, many thanks for the same

    A timely one too…

    Thanks once again, happy pongal and new year

    Daksh

    1. Narasimhan Santhanam Post author

      Dear Daksh

      Many thanks indeed for your kind appreciation

      I hope I can keep up the good work for a long time to come

      Thanks again and all the best!

  21. Meganathan

    Is there a revised CERC benchmark capital cost for MW scale solar PV farm? Something that has been updated in 2016?

    1. Narasimhan Santhanam Post author

      Yes, there is a recent Dec 2015 CERC benchmark costs.

      The link is here – http://www.cercind.gov.in/2015/orders/SO17N.pdf

      Data from the link (in Rs lacs /MW) for benchmark costs for 2016-17

      Modules 310
      Land 25
      Civil & gen works 35
      Mounting structures 35
      Inverters 30
      Electricals (including grid extension and bay extension) 40
      Preliminary costs 26

      Total 501

      The CERC has set the benchmark costs at Rs 501 lacs / MW for projects without trackers, over Rs 100 lacs less than the earlier benchmark!

      Solar Mango however thinks this is too ambitious and optimistic an estimate for small MW power plants (1-5 MW).

      Costs for panels and for electricals including grid extension and bay extension have been underestimated, and costs for civil works have been underestimated as well. Both these could be OK estimates for power plants of sizes larger than 10 MW, but for solar power plants less than 10 MW, especially sub 5MW, Solar Mango estimates the total project cost to be around Rs 575 lacs/MW, perhaps Rs 560 lacs/MW if you are able to negotiate really well.

      Solar Mango’s realistic estimates for 2016-17, for small scale power plants in the 1-5 MW range:

      (in Rs lacs/MW)

      Modules 325
      Land 25
      Civil & gen works 50
      Mounting structures 35
      Inverters 30
      Electricals (including grid extension and bay extension) 60
      Preliminary and other costs (includes preliminary expenses, approvals, interest during construction, expenses for right of way, taxes) 50
      Total 575

      I am conservatively estimating a realistic project cost to be Rs 575-590 lacs/MW for 2016-17, this should be an acceptable estimate starting April 2016.

      As you can see from the main blog post, as of Dec 2015, the equivalent estimates were in the range Rs 610-630 lacs/MW without trackers. You can thus see a 5-7% drop in price between Dec 2015 and Apr 2016, for overall costs.

      Hope this helps and thanks indeed for asking

  22. Williams

    Dear sir

    Would the project costs you have mentioned above be the same for a solar pv project that is about 10 MW in size?

    If no, what could be the project costs and optimal tariffs? We wish to bid for Karnataka sir

    1. Manasij

      Williams

      Project costs for solar power plants at 10 MW scales will be about 5% lower than those for 1-2 MW, mainly on account of some fixed costs spread over a larger number of MW and also small reductions in the prices of modules and inverters

      Manasij from MSK Solar

    1. Narasimhan Santhanam Post author

      It is usually at the substation for any grid connected solar power plant feeding into the grid

  23. Sateesh Kumar BS

    Are there few Indian BANKS that accept 0 zero collateral for grid connected solar power plants??

    If banks ask for high collateral how can small busines put up solar plants???

    1. Narasimhan Santhanam Post author

      Dear Sateesh

      I share your concern, but as of now, for small businesses, it will be difficult to get bank loans for solar without any collateral

      As I have explained elsewhere, some banks are willing to ease up on collateral and perhaps relax it to 60% instead of 100%, or if you are lucky, perhaps 50% even, but it is difficult for me to see banks lowering the collateral requirement below this

      IREDA asks for a much lower collateral, at 30% or perhaps you can try negotiating for even lower

      Hope this helps

    1. Karthikeyan BVS

      Dear Sir

      Single axis trackers can give upto 24% and this has already been witnessed in some solar power plants in S India. Some of these are in Telangana state and other S Indian states

      Dual axis trackers might even give more, but there are v few installations in India with these trackers to give you reliable estimates

      But pl note 24% CUF (21 lac units/MW/year) is the highest that has been achieved as far as I know, many single axis tracker plants get about 22%.

    2. Manoj Reddy

      Surjeeth Sir

      You can get upto 21 lac units per MW if you instal a good quality single axis tracker at a good DNI site (more than DNI of 5.5 kWh/m2/day). this will be about 24% CUF

    3. preetam

      In our solar farm in Rajasthan (near Jaisalmer), we have not got more than 19.5 lacs/MW using tracking systems…

      I think you should take 19-19.5 lacs/MW as what you will get with trackers

  24. Shyam

    Modi jee should focus and people indian and local people to be part od solar revolution and talk to banks to come forward and remove this collataral requirement and encourage general people to participent.

    Now it is looks like Modi jee is only interested to bring 100% FDI and sell whole indian land and the infrastructure to foreign companies.

    It is NOT make in india its is Giving indian business to foreign banks and big forigen companies they have access to 4% loan and without collatral so can bid for 4.43 and even lower . Yesterday Rajasthan got 4.35 bid w/o VGF funding.

    It is pushing very hard for indian businessmen to join party.

    The cost of Solar panel was came down exponentially … Now onwards it would go 10% more , but now needs govt intervention for aggresive bidding.

    1. Narasimhan Santhanam Post author

      Shyam

      Many thanks for your comment

      Some of your concerns are shared by a lot of Indians, I am sure

      On collateral, well, we can’t really blame the banks, they are just following rules. Unless as you mentioned, the govt intervenes and allows them to change the rules or provides them a support framework re investing in projects in new domains such as solar, they cannot do much about the collateral requirement.

      For sure, one of Modi govt’s main focus is to get large firms invest in solar as that is one of the ways to get to a huge target like 100 GW in the next 6 years. But balancing that with social equity is indeed a tough game – let’s just say I don’t envy Modi or Piyush Goyal!

      1. Shyam

        No doubt Modi Jee is doing good job across the world to call for investment , but feeling indian people not part of it. if any scheme or solar park develop who have specfic scheme for new people and provide incentive would be great .

        E.g.

        1. Only allow big companies to invest >100MW and on low tariff

        2. Encourge new people or small companies provide them intial funding or discount , good tariff >6 actually its kind of reservation:)
        Smal project give to small people it will boost the confidence of people.

        But anyway Lets india to get uninterrupted power 24X7 who produce it not matter , atleast people can get it . Farmer need not to wait for 4 hr of slot and start pump and small business man . Power cut is one of the big problem.

        Ok Lets mute this thread here

  25. Anamika Sampath

    Mr Narasimhan

    Many thanks for your detailed insights in this post, really terrific for business people like us who are new to solar

    I read today that a record new low tariff of Rs 4.35/unit was bid by an international co in Rajasthan ( http://economictimes.indiatimes.com/industry/energy/power/solar-power-tariff-at-record-low-drops-to-rs-4-34-a-unit/articleshow/50640342.cms ) whereas your numbers in the blog suggest an ideal tariff would be above Rs 6/unit

    Can you kindly explain why this conflict in data?

    Thank you once again for taking time to educate the market in such great detail; it is also refreshing to see an expert such as yourself taking time to answer each question posed…

    Your service to the Nation will indeed be appreciated

    1. Narasimhan Santhanam Post author

      Dear Anamika

      Many thanks for your kind query

      There are two main reasons why my estimate in the blog post above is quite different from the bid tariffs

      1. Scale – many of these projects bid at 4.35-4.63/kWh are happening at 50+ MW scales. At these scales, components such as modules and inverters could cost 5-7% lower, a significant reduction. Some of the fixed costs such as approvals and other liasioning works also get spread over a larger capacity. In addition, construction costs could be slightly lower too. Overall, you can expect the total project upfront cost to be about 7-8% lower for plants of these scales as against the 1-5 MW solar power plants.

      2. Some of the bidders are almost surely raising money from outside India at rates slightly less than 10%, whereas I have taken mostly 11.5-12% as interest rates.

      3. It is not entirely clear from the reading of the documentation for these NTPC bids whether cost of land has to be borne by the developer as all these developers are allotted land in an existing solar park.

      4. Finally, it is quite possible that NTPC has given a commitment to ensure 100% availability of grid for export of power from these power plants, whereas in our calculations for smaller MW plants, it is prudent to take a 2-3% grid non-availability

      The two big drivers for cost and tariff reduction however would however be the capital cost and the interest rates. Significant reductions in these two alone could result in a much lower LCOE, making a much lower tariff possible.

      Hope this helps

      On a related note, you might want to read this too – Softbank Too Bids at Rs 4.63/kWh for 350 MW Solar – An Analysis of the Magic Number 4.63 – http://www.solarmango.com/blog/2015/12/16/softbank-too-bids-at-rs-4-63kwh-for-350-mw-solar-an-analysis-of-the-magic-number-4-63/

  26. Kanagavel

    Dear sir

    Is it correct that in south India using of single axis trackers is best and in north India we should use dual axis trackers?

    please tell me

    Thank you

    1. Narasimhan Santhanam Post author

      Yes, for south India, single axis trackers are more optimal than dual axis

      The general rule of thumb is the higher the latitude, dual axis trackers more sense.

      Chennai, for instance is 13 degree North while Ajmer is 26 degree north

      So, we could say that dual axis trackers could be relatively more optimal for Ajmer than it is for Chennai

      As I mentioned, this is a general rule of thumb, but it will be a good idea to check with your EPC or other technical experts based on the exact location you plan to have your power plant at

  27. Michael Durai

    You have mentioned that the costs are for 1-5 MW scale solar plants

    but Mr Narsimhan, scale should not matter in solar plants as these are supposed to be modular…

    So why should unit costs for small scale MW plants be much lower than large scale plants?? though I can understand minor reductions in cost with volumes

    Do enlighten

    1. Narasimhan Santhanam Post author

      Dear Michael

      Many thanks for your question

      I request you to see my response to the commentor Anamika Sampath above…that might to a significant extent answer your uery

      Let me know if it doesn’t, and I will try once more 🙂

  28. Krish Vasu

    You have mentioned that power plants have reported upto 20 lac units/MW/year with trackers…

    But some EPCs tell me they can generate upto 22 lac units with trackers. about 24% CUF…is this possible?

    1. Narasimhan Santhanam Post author

      Dear Krish

      Many thanks for your kind comments

      Yes. we are aware of some solar power plants in S India reporting 24% CUF with single axis trackers, but it is too early to say whether this number can be replicated as some of these power plants have been operational for not much more than one year.

      It might be a good idea to wait for a year more to see if 22 lac units is something that can be achieved by these power plants consistently

      Which was why we have taken a more conservative 20 lac units / MW / annum

      Thank you once again for your question and observation

  29. Manav Shukla

    Just for your knowledge, costs for construction of civil structures will also depend on the type of soil – we shud first do soil testing and only then we will estimate the costs

    loose soil means higher costs for civil wheras firm soil will mean less cost

    also, the topography of the site and how level the land is will coming into play for overall construction costs

    1. Narasimhan Santhanam Post author

      Manav

      Thank you and completely agreed.

      I am glad that you pointed this out

      So, yes, for our estimations, we have assumed that the soil is firm soil and that the land is reasonably even so the costs for levelling are not very high.

      Thank you once again for pointing this out

  30. Shiv Shankar prasad

    Dear sir

    I have a question on owner’sw engineer.

    Will having owner’s engineer make a really large difference to the power plant quality or can we trust that the epc will do a good enough job?

    1. Narasimhan Santhanam Post author

      Dear Shiv

      Thanks for your interesting question

      In my opinion, it will be a good idea to employ owner’s engineers, mainly because they do not cost much (a few lacs /MW), so even if they are able to just do nothing more than double checking what the EPC has already done and are able to correct a few critical errors along the way, it is worth the money paid to them

      On the other hand, if the owner’s engineer is really smart and is able to do a lot more original stuff than the usual auditing done by conventional owner’s engineer, you might get significant benefits. I appreciate that most owner’s engineers do more of ticking off a checklist than anything original, but perhaps there are a select few who can offer higher value

      It hence might be worth your while to talk to a few of these and see if there are some who can provide more than commodity stuff.

      Overall however, employing an owner’s engineer makes sense whatever the scale of plant as long as they do not cost you a bomb

      1. SP

        Hello Sir,

        Could you please share some EPC list who are bests and cost effective as well and experienced in Rajasthan.

        SP

  31. MILIND

    On what basis are the O&M costs estimated by EpCs?

    Some EPCs I discussed with said it will cost Rs 8 lacs/MW with trackers and 6 lacs/MW without trackers/annum, but after negotiations i think they will come to the same numbers you have mentioned (7 and 5 lacs…)…

    But I really do not know based on what they quote these estimates for O&M as no one has actually showed me the costing for this

    Any insights on this will be highly useful sir

  32. Giriraj Verma

    Where and what company can provide me modules with the least amount of annual degradation? Is it possible to have less than 0.5% degradation at least for the 10 years which is the crucial period for us to pay back all the loans etc?

  33. Francis Xavier

    Dear Narasimhan

    At the outset, thanks for such a detailed and wonderful post

    By far the best post I have seen for a solar power plant anywhere on the Internet

    You are indeed doing a great service as some others have mentioned before

    There was an earlier comment on latest CERC benchmarks I to which you had mentioned that in your opinion the CERC benchmarks were too low. I wish to point out one reason why your estimates could be quite different from the estimates of CERC

    CERC actually quotes per MW DC, but what we want is MW AC. As you will agree, 1 MW AC could require upto 1.1 MW DC, owing to the losses at the cables and the inverters etc. Thus. the actual cost of 1 MW AC will be 1.1* cost of 1 MW DC…

    Does this explain the difference?

    Do let me know, and a million thanks once again for such a phenomenally useful post

    Francis, EFF Solar
    Now, if you take a few moments and compare 1.1*CERC benchmarks

    1. Narasimhan Santhanam Post author

      Dear Francis

      If I were asked to thank one and only one person today, that would be you!

      Yes, this possibly explains the difference

      To make sure, I went through the CERC document on the benchmark costs once more. http://www.cercind.gov.in/2015/orders/SO17N.pdf

      I do not see any place where they have mentioned that this is for 1 MW DC, but as you will agree, if they do not mention it, it should be for 1 MW DC; besides, the cost component breakups they have considered are all for DC.

      So, yes, this nails it

      Thanks a million once again. I hope CERC also makes a note of it; else, developers will be haggling with vendors and EPCs for a price that is not feasible

  34. Javed

    Also wish to know this –

    Will a 11 KV line be enough for a 2 MW solar power plant?

    What KV line is required for a 5 MW solar power plant?

    1. Narasimhan Santhanam Post author

      Javed

      Good qn, thanks

      You can take the following as benchmarks for evacuation line capacity for various scales of solar farms:

      1 to 5 MW – 11 KV
      5 to 20 MW – 33 KV
      50 to 100 MW – 110 KV

      Ho#e this is of help

  35. Arun Karunakaran

    Dear Narasimhan

    Thanks for your blog and details

    I am keen on a 5 MW solar power plant; I want to know if locating this in an existing solar park will be less costly than buying my own land and constructing the infrastructure separately

    I have been approached by two solar park owners and I don’t know whether I should go with them, as we also own some land banks ourselves though these are about 6 KMS from the substation.

    Your advice will be highly appreciated

    1. Narasimhan Santhanam Post author

      Arun

      Thanks for the question

      I think a decision on which is better – solar park or your own land – will be easy if you can share brief details of the terms from the solar park.

      How are they charging you – is it just for the land alone – and what are the other costs involved in the terms

      If you share some critical inputs on costs, then we can do a comparative analysis for you at Solar Mango

      Thanks again for your interest in us

      1. SP

        Sir some one asked for if lets say Solar Park is present and if you would like to setup there.

        I am not positive that Solar park will allow small MW1 1-10 MW as the other coat will oncrease for it.

        Regarding terms and condition like recent Rajasthan bid 4.35/- was below from Solar park
        http://www.rrecl.com/PDF/Solar%20Park%20Bhadla%20Presentation%203.pdf

        Do you think how much it help to getvtariff low such as 4.3/-

        One Time:
        o Non-refundable processing fee of Rs. 10 lakhs per project plus service tax & other charges (as per the Rajasthan Solar policy)
        o Land will be allotted to the developer on prevailing DLC rate at the time of executing of lease deed agreement, which is Rs. 32,873 per Bigha (Rs. 2,03,057 per Hectare )at present.
        o Development Charges of Rs. 10 lakhs per hectare for the year 2015-16 o Thereafter it shall escalate cumulatively @10% per annum
        o Metering System Cost at 220 kV side will be charged on actual basis.
        o Connectivity charges @ 2 Lakh/ MW (as per RERC) Annualy
        o Lease rent will be charged @5% of DLC rate for first 2 years and thereafter it will escalate for every year @5% per annum
        o Annual O&M charges shall be @5% of development charges for the first year and thereafter will escalate cumulatively @10% per annum.
        o Rs. 1 Lakh per MW every year for the entire life cycle of the project from the time of commissioning for the Rajasthan Renewable Energy Development Fund
        o Water charges will be payable on actual consumption basis @ Rs 16.5/kL (at present current rate)

        The charges above are payable with applicable taxes, wherever relevant

        1. Narasimhan Santhanam Post author

          Wow, thanks SP, that was a really good reference compilation

          Appears that there are indeed a good number of concessions and incentives – just look at the land cost, a fraction of what one would pay otherwise

          With these incentives factored, it indeed might JUST be possible to quote 4.34 and still take home something…

          What do you think??

          Thanks again

          1. SP

            Specifically in Rajasthan land is very cheap , e.g. 10Lakh ( buy for 99 year ) or some time free hold as well . But in this Solar park you are getting the incentive of your transmission line is not long. AFAIK the LOSS always happen due to GRID not because of xmission line. if SPD have 33kv max in 3.98 % and if 132kV transmission losses are almost zero ( this based on data from RRECL )/

            Now being in solar park you will get 99% GRID availability. But it is not clear to me that does O&M is responsibility of Solar Park ( as they take heavy fee ) or SPD .

            Solar Park is guarantee of land.

            So I think for big solar project Solar Park is good as you can cover the cost. But if it is small 1-20MW it Solar park ask for big money which is not reasonable .
            Best to buy land near to Solar Park

            SP

  36. Kunal Sood

    Should we use CUF or PR (Performance Ratio) to measure Solar Power Plants

    In your analyses you have used CUF but many industry veterans in India and also in Eu and USA use PR

    Can you tell me why you have not used PR in your evaluation?

    Kunal

  37. Makrand

    If I instal a grid connected a solar farm for captive use, what will be the losses that the discom will consider and what will be the other expenses if I wheel this power to my factory?

    I live in Maharashtra.

    Thank you

  38. Mahesh Khanna

    Thanks indeed for useful blog in Solarmango

    I will be putting a 10 MW power plant in Rajasthan with trackers

    Other than tracker maintenance for breakdown, what are the most frequent reasons for solar power plant failures and breakdowns that will need repairs and replacements

    1. SP

      Hello Mahesh,

      Where do you putting in Rajasthan, I am also planning for it , Let me know if we can share the information and would like to partnership .

      SP

  39. SHIV

    I am from Kerala and I wish to have a solar power plant in some other state as you know my state does not have a very good sunshine as other states and also land is very scarce and costly

    Most of India has good radiation (DNI) and many places I understand has DNI above 6. Also I know that high temperatures can decrease the performance of solar panels

    So how should I choose the location? Should it be medium-high DNI and reasonable temperatures because almost all the very high DNI locations will also have very hot temperatures, right?

    Your insights will be useful and thanks for creating this good resource

  40. MANAS DAS

    In your calculations you have not considered acceleration depreciation benefit and this can make a significant difference to the IRR…would be grateful if you could do some analyses using data from your post but with AD benefits included and show us the results

    Thanks – MANAS

    1. Narasimhan Santhanam Post author

      Dear Manas

      Thanks for your note

      Yes, I have not included AD benefits in my calculations, for two reasons

      1. Not all power plant developers can avail AD benefit, as if it an SPV formed for development of solar power plants, AD is not applicable as there is a 10 year tax holiday

      2. Even in cases where the developer happens to be a profit making company for whom AD is applicable, the extent of AD benefit actually depends on the amount of tax he has to pay, and one should also factor in the minimum alternate tax (MAT).

      As a result of the above two, only a select % of developers can avail the complete benefits of AD.

      All the same, I will do some work soon and see if I can show how AD benefits affect costs and returns

      Thanks once again for taking time to drop in here, take care

      1. SP

        Hello Sir,

        This was my question as well,

        Ibtalked to few plant owner in Rajasthan 1-10MW owner they avail AD and 10 year tax holiday both . It is allowed only you get lower tarrif if AD used . But since central Govt fixed at 4.43/- so this also gone . But due to MAT payable 20% . There income statement is ruined and having difficult now . With low tariff it is getting difficult day by day.

        Please share those data , I made a model if you can share your id can send for verification

        SP.

  41. SP

    With Tracker in Rajasthan usually you get 22-24 % Also if your land soil is not loose . Better to have concept of farming with Solar farm as it is done in USA , Use the water which is used to clean the panel.

    Earn extra money .

    SP

    1. Narasimhan Santhanam Post author

      Hi SP

      Thanks for the interesting insight on farming along with solar energy generation

      Do you have any details on live case studies of agriculture and solar in the same field? What crops do they grow?

      Any links on real life case studies of solar + agriculture will be most useful…

      The real concern is that only some crops can grow in the shade, and unless these crops have enough value in the market, the farmer might not really get much from it…

      Look fwd to more details on this, thanks

      1. Narasimhan Santhanam Post author

        In fact, in this context, I would request any other reader of this blog to share her/his insights on what crops could actually be grown in shade, and if they know of any real life instances where this is being done

        I was told by a senior director of the Intersolar event during last year’s event in Mumbai that such cropping under solar farms is being trialled in many places in China, but he was not able to give much more specifics. I plan to do more research on this, but would like to have insights from others as well.

        Interesting topic, I must say

        1. SP

          Hello Narsi,
          Hope you will not mind if I call you Narsi.

          Currently I am in US. Helping my father and brother to setup plant in Rajasthan. In US in my research and seen in California “Small project” do to vegetable farming as there height are small.

          Big project >50MW don’t consider this option as farming might create hazard and laws doesn’t allow.

          Regards, SP

  42. Chetan Shah

    I am a Gujarati and wish to put up a solar power plant near Surat

    I have contacts in business who are liking to purchase power from me at about Rs 7/unit for a 15 year period, will it be profitable for me to sell to them at existing costs?

    1. Narasimhan Santhanam Post author

      Dear Chetan,

      Thanks for your query.

      Please use the following template for calculations:

      On one side, you need to compute your total cost for power to be supplied to your buyer.

      Total cost of supply = LCOE + costs for wheeling/banking and other charges as specified by your discom that owns the grid

      LCOE is the levelised cost of energy; you compute it using capital and O&M costs, and by factoring in the time value of money

      costs for wheeling/banking, and cross subsidy etc., depend on your state policy; some states levy much smaller amounts, while some could be much higher; please check out what these are for Gujarat for solar power; please also add any taxes applicable.

      Once you have arrived at the total cost of solar power supply, add the minimum profit margin you would like to add to it.

      So, the lowest price you can sell (L) = Total cost of supply + your minimum margin

      Check what your prospective buyer is paying to the grid for electricity. Let us say it is G

      You have a strong business case if G>>L, that is, your prospective buyer is paying a much higher price to the grid today that a price you can charge him and still make a decent profit.

      To give an example, let us say your LCOE is Rs 5.5/kWh, and all other costs put together is Rs 2. the total cost of supply is Rs 7.5, and if you wish to have a minimum margin of Rs 50 paise/kWh, the minimum price you can sell is Rs 8/kWh.

      If in Gujarat, your prospective buyer is paying over Rs 10/kWh to the grid, you have a really good business case (Rs 10 vs Rs 8). On the other hand, if the price he is paying is only slightly above 8 (or less), you have little business case.

      Just make sure you add all the relevant taxes and peak demand charges etc while you compute the total price of electricity paid by your prospect to the grid.

      I know I could have done a better job of explaining it, all the same, hope it helps to some extent

      Take care, and all the best in your solar power efforts.

  43. Vipul SHARMA

    Dear Mr Narasimhan

    What INVERTERS ARE BEST for MW solar farms – central or string?

    My technical contacts says string inverters losses will be low and their prices are coming down to same as central inverters

    1. Narasimhan Santhanam Post author

      Dear Vipul

      Many thanks for the question

      This is a question on which long winded debates are happening worldwide, but here I would like to keep it short and actionable for you.

      Given the current cost structures of central and string inverters, I really doubt string inverters are anywhere as economical as central inverters for large scale solar farms, or MW solar farms as you call them. The costs differ from one brand to another, both for central and string inverters, but string inverters are still much costlier than central inverters. I am not aware of any MW scale Indian solar power plant using string inverters as of early 2016.

      Now, had your question been for rooftop solar, the race between central and string inverters would have been a more tight one, and in fact some prominent system integrators for rooftop solar almost solely use only string inverters, even for commercial installations over 200 kW. While string inverters could still be slightly costlier than central inverters, the difference might not be as high as it is for large scale power plants, as at smaller scales, the cost of central inverters/kW start moving towards those of string inverters. Besides, string inverters provide some unique benefits for rooftop solar power plants, and as a result, we see a neck to neck between string and central inverters for MW scale solar power plants.

      I hope my answer has been of some help

      All the best

  44. SV Rajshekar Reddy

    Can some solar power farm owners share their experiences here based on live info and data please on costs and yields?

    Thank you

    Rajshekar

    1. Narasimhan Santhanam Post author

      Dear Rajshekar

      Excellent request

      I have asked developers who we have assisted to provide us with the data, and will share those here once I receive some inputs from them

      In the meantime, I am also placing my request here for insights on live data from operational solar power plants in India

  45. Alexander

    Only chinese panels are bought by all Indian developers except those who are developing under DCR…

    How are Indian panel manufacturers surviving or have they all closed their shops??

    1. Narasimhan Santhanam Post author

      Dear Alexander

      Indian module makers are indeed living on thin branches right now…

      Many of them, especially the smaller ones, are barely eking out an existence, some of the large ones are able to do strategic marketing deals and stuff of that sort to do relatively better

      Allocations under the DCR schemes of the National Solar Mission and rooftop subsidies that mandate domestic solar panels are giving them some hope, but unless the government comes up with a comprehensive plan to boost the Indian solar manufacturing ecosystem, many module makers in India will be on life support

  46. Thilak Arivalagan

    Dear Sir

    Tell me best way to have very good quality power plant with very good generation at economical prices.

    Is it just the right EPC will do this for me or should I also research extra to achieve this??

    1. Narasimhan Santhanam Post author

      Dear Thilak

      Many thanks for your question

      The following posts from Solar Mango could answer your question to a significant extent

      Choosing the Right Inverters & panels for MW solar farm – http://www.solarmango.com/blog/2015/12/17/choosing-the-right-panels-and-inveters-for-your-mw-scale-power-plant/

      Quality of Indian solar installations compromised, says industry leaders – http://www.solarmango.com/blog/2015/12/05/indias-solar-leaders-feel-quality-of-solar-installations-compromised-in-the-race-to-low-tariffs/

  47. Partha Basu

    Will my module costs decrease if I use THIN FILM modules?

    First Solar has a lot of installations in India, and they are a REPUTED firm from USA.

    Should we consider them also, or should we consider only crystalline modules for solar farms?

    1. Narasimhan Santhanam Post author

      Dear Partha

      The thin film vs crystalline modules debate has been a long and colorful one. Until two years back, at least in the Indian context, it was a tough fight

      But I can tell you with confidence that in the last 18 months, crystalline panels have convincingly won the battle. If I were to hazard a guess, over 90% maybe even 95% of all utility scale installations in India in the last one year would have been with crystalline panels.

      Thin films earlier were able to give a tough fight to crystalline they were cheaper per kWh of electricity, but now with the falling prices of polysilicon last 4 years, the price gap between thin film and crystalline has become neglgible. Besides, until 2013, thin film panels were exempted from the DCR rule so projects under DCR could import thin film panels while they could not do so for crystalline. Now this gap has also been plugged.

      To give a short answer, go for crystalline. I do not see any advantages to you by using thin film panels for solar farms in India, at least not as things stand on commercials

    1. Narasimhan Santhanam Post author

      Dear VSR

      Many thanks for your query

      It will be difficult to give precise estimates of costs for specific brands, as the only reliable quotes are the ones you get from the companies themselves!

      However, I can say that inverter prices for prominent per MW (DC) can vary between Rs 30 lacs to 40 lacs / MW depending on the specific brand. Now, I have seen quotes from some of these companies at slightly less than Rs 30 lacs/MW too, but that could be for selective cases…

      Hope I was able to provide some useful inputs

      NS

  48. A Solar EPC

    Very few talk about fencing and lighting, even though these are important parameters when you look at an infrastructure that should be maintained well for 25 years

    While some consider chain linked fencing to be good, many others are going for concrete wall fencing…Also, on lighting and security, only a few are going for security camera based surveillance, most just feel that these are not required

    These decisions also will impact costing but more important plant performance

    1. Narasimhan Santhanam Post author

      Hi

      At the outset, thanks for opening up a topic that is fairly important but hardly discussed

      Well, I agree with you – both chain linked fencing and concrete wall fencing are relevant – the former being more prevalent as it costs less

      Some developers feel investing in concrete wall fencing is worth the extra expense as it offers better protection, but opinion is divided on this – as some experts feel that chain linked fencing is better than concrete wall, as, unless the wall is tall enough, animals (cats and some other felines for instance) can actually jump over the wall while it will be difficult for them to do it through a chain linked fencing.

      I would be glad if you could provide any further insights you have in this regard, but with our work so far in solar power plants, we feel it is better to go for chain linked fencing. Look fwd to hearing from you on this

  49. Arvind

    I want to be my own EPC for my 2 mw power plant…but I need training on some aspects especially power plant design because we are good in electricals and construction…

    Where can I find good design experts who can train us?

    1. Narasimhan Santhanam Post author

      Dear Arvind

      Many thanks for the query, and an interesting one at that

      Finding high quality independent design consultants is fairly difficult – ask me! as I have been trying the same for the last couple of weeks for one of our clients with a similar profile as yours

      So far we have received only very few good profiles

      Keeping fingers crossed

    2. Narasimhan Santhanam Post author

      I thought I’d add one more here: Solar Mango offers something called the PMC+ solutions to companies such as yours who wish to be their own EPCs and learn the design component especially for their own solar power plants.

      Through this PMC+, we provide expert services that all EPCs provide, but without being an EPC…essentially, you get trained on all the design and other solar-specific expertise while building your solar PV farm.

      If you are interested in this service from Solar Mango, send in a note to Ramya – ramya@solarmango.com with subject: PMC+ services from Solar Mango

  50. Muralidhar Rao

    My bank wants to know how they can be sure that the Telangana discom will pay on time and how to reduce this payment risk, for state government solar power allocations?

    1. Narasimhan Santhanam Post author

      Dear Muralidhar

      Thanks for your comment

      Your bank is fully justified in its concern and for raising this question.

      At the same time, you and I can do little about this too.

      The only aspect you can consider is to check your PPA terms to see whether the discom has mentioned specific clauses for late payment or if it is providing an LC facility for timely payment. Short of these, the common man has little control over this situation and it really has to be your bank’s call.

    1. Narasimhan Santhanam Post author

      Thanks SP for the link…

      As the news report itself says, the muted response could be owing to the fact that it is SECI and not NTPC so perhaps the perceived offtaker risk is higher, and also: land needs to be identified and procured by the developed unlike in NTPC where land is allocated in an existing solar park.

      Overall I agree with you that things are very volatile right now in the context of tariffs and government allocations…

  51. SP

    Thanks Narasimhan for such a good forum and most important you keep the discussion going and educating people.

    1 Could you please add AD in your financial model , So people get final sense. people would like to use AD , Any pros /cons of not using AD benefits over normal Depreciation ?
    2. Also what are the normal
    – Wheeling charges
    – Transmission charges
    – SLDC charges .

    I am not able to find any information regarding these charges from Internet , Any idea ? Mostly for Rajasthan ?

    SP

    1. Narasimhan Santhanam Post author

      Thanks for the query SP.

      I will revert later on the wheeling, transmission charges etc.

      On AD as well, I will need to spend some time computing the returns with AD benefits. In this context, I also need to have detailed discussion with our financial expert because, even with AD, I understand a Minimum Alternative Tax (about 19%) needs to be paid on book profits. Now, I am not entirely sure about the applicability of MAT for solar power plants, and I am not able to get a reliable and clear explanation, but if it mandates a minimum tax to be paid on book profits, things can be quite different…

      Will revert soon, and am hoping to add clarity to this discussion

        1. Narasimhan Santhanam Post author

          Hi

          This data should be available at the Rajasthan Electricity Regulatory Commission site ( http://rerc.rajasthan.gov.in/ ), but I was not able to find them on the site – presumably you have done the same too? I will try with other sources and revert to you on the wheeling charges and cross subsidy surcharge for solar

          On MAT, our finance expert tells me that MAT is applicable for solar power plants, but of course the actual impact of MAT can be different depending on whether your solar project is under an SPV or whether it is under an existing business with a balance sheet.

          For SPV, my financial guide says we should take a MAT of 19% for the first 10 years of the solar project, and this can be offset against the taxes after the 10th year

          Hope these are of help, and take care

  52. ROHIN BHATIA

    Dear Narsi – We have one of the longest performing ( Dec 2011 ) – (record every May month last 4 yrs – over 900 + MW total installed farms in Gujarat )- output solar farms with Single Axis Tracker.
    As per ALL official published data- and our info -NO solar Farm has given Generation more than 19 L per MWp( this too is achieved by one farm in high radiation area of Rajasthan – and with Tracker )
    Per MWp – safe to take gen of around 18.5 L – with Single Axis Tracker .
    If u have different Generation data- pls share
    SAT will be 16 to 20 % higher over Fixed Tilt

    Regds
    ROHIN

    1. Narasimhan Santhanam Post author

      Dear Rohin

      Firstly, many thanks for taking time to give us such critical data

      Next, I agree with you on the 16-20% increase for single axis tracker over fixed tilt as the estimate to go with. Until about 6 months back, this was the estimate we used for all our assignments too

      But in the last 6 months, we are seeing some farms from south India starting to report better numbers – between 20-21 lacs units on an annualized basis.

      Sure, it is too early, and we have results only from a few months, so I surely am guarded about its extensibility, but it does appear that the tracker technology is improving all the time and as a result EPCs are confident enough to give a performance guarantee upto 19 lac units / MW for a single axis tracker.

      It will be indeed instructive if you could provide data from any recently implemented solar power plants with single axis trackers. I say recently because the use of trackers in Indian solar power plants has been rather limited and only picking up recently, and hence you might have different insights from data sets acquired from recently implemented projects

      Thank you once again for providing us with real life insights and look forward to continuing the discussion

      All the best

  53. SP

    APPC and REC : Narsi Sir ,

    In APPC SPD gets REC certificate and its traded at PXE , after sell how does SPD supply power or DISCOM supply when the energyvisbeing sold to local DISCOM .

    Once REC sold does SPD not paid from DISCOM – how this system works

    SP

  54. Avinash Mudaliar

    Sir –

    Is it a good idea for small time developers who wish to put up 1-5 MW SPPs to get involved in solar parks projects where installations are typically 50 MW and above?

    1. Narasimhan Santhanam Post author

      Dear Avinash

      The size of power plants allowed at solar parks depends on the solar park in question

      The solar parks we hear about these days where large scale (50 MW and above) power plants are put up are the NTPC constructed solar parks, specially for large scale power plants

      At the same time, there are private solar power parks as well, for some of whom the total capacity itself is only about 1000 MW; so you can easily see that these parks can accommodate smaller sized solar power plants, as small as 1 MW. You should try contacting one of these solar parks

      All the best!

  55. Mithilesh Reddy

    Sir, I also want to know about solar parks and why they are different from other solar power plants

    What are solar parks and how are they different from conventional utility scale solar farms? How are the costs reduced and what are the economies of scale in such project?

  56. Dinesh Venky

    Sir

    Solar parks are nothing but levellised land and readily available evacuation infrastructure. Such a set up will help companies wishing to set up solar power plants the trouble of identifying land and putting up their own evacuation to the nearest grid point

    Solar farms on the other hand represent solar power plants

    You can say thus that solar farms are located within solar parks

    Hope I am clear

    1. SP

      At what time solar plant generating energy and stop .

      Plant up time and off time ?

      What is the usual time for peak hours.

      Usually sun is out for 8-10 hr so what is the graph of production ?

  57. Krishna

    Dear Sir,
    The bankers are actually going by the becnmark price determined by CERC which does not include trackers, vat, additional costs. Do solar mango offer financial assistance

    Thanks & Regards
    krishnaveni

  58. Chinmay

    Hello sir,

    Great info…….

    Sir I want to invest money on small scale solar power plant..Could u pls guide me and assist me the approximate cost of the set up on BOT model…

  59. Pingback: Super Big Resource on Solar Power Costs & Returns - Solar Mango – #1 guide for solar

  60. MANTESH MAHAJAN

    We are Yarn manufacturer in Jalgaon Dist in Maharashtra. Due to 2nd largest component in our yarn costing power cost of MSEDCL comes to Rs 7.0 /Kwh which is about 10-12% of sales price. To reduce down we want to go for SOLAR PLANT OF 2.5 MW ON BOT basis with minimal interest rate for duration of 15-20 years. We have taken loan from NCDC by way of land,plant & machinery. So there is nothing to be kept with company & also no bank guarantee can be provided. Request to guide for companies doing business in this field of solar power generation on BOT.

  61. Dinesh Borade

    Sir,Please tell me how many cost is required of substation for 50 MW Solar Power Plant in India as well as in Abroad.Also tell me how many cost required Transmission Line for 50 KM Distance.

  62. Prasad

    Dear Narasimhan:

    This is a great article. But with current prices of the panel and trackers these values and assumptions are probably not true anymore. I wonder if you can update this article or write another article on

    1. Are trackers worth it if the price of panel is Rs xxx/Watt
    I believe we have already reached this point or going to reach soon.

    2. What are the LCOE and Tariffs to consider with 2018 market in mind ?

    Thanks,

  63. Indika

    Hi,
    We are planning to start a 0.5MV solar power farm in Sri Lanka. Could you please tell me how much cost will be supply and installation of panels and etc
    Land area requirement
    Construction duration
    Thanks,
    Indika

  64. Triston

    Worcester MA Solar Power is dedicated to giving our clients professional advice on how to make the most of their Tesla Powerwall during a power outage. Additionally, we desire to assist customers in comprehending their energy usage.

    Contact

Leave a Reply to A Solar EPC Cancel reply

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

India Renewable Energy Expert