SolarPVInvestor Forums

Back of the envelope... they could shut down module production and become and absolute giant in project development. They do know that business. If the worked on the BOS side they could use Chi modules and likely install a massive amount of GW's in the desert southwest alone in the coming years? Sounds like a plan.

Quoted from "explo"

Quoted from "larryvand"

All you have to know to figure out that FSLR is in trouble is the fact that they did not want to give guidance for 2013.

The fact that they can buy a module with 15% conversion efficiency, saving a lot of BOS cost compared to their own 13%, for 65 cents while having a 68 cent cost to make their own inferior modules, says that their module production is holding back their project profitability.

Modules too bad, sad but true?
Projects too good to be true, over time?

There are other intangibles within differing technologies that is not an apples to apples comparison of a performance of a watt in a real world scenario. Current cSi technology loses up to 25% efficiency depending on hot days. They break down much faster with elevated temps. You do not see companies bragging about this in their charts.

You also have different performances in low light absorption due to angles of lights(AM/PM). Companies that do better in these areas are starting to point these out. SOL it touting better performance numbers from low light angles in their specs now. If you can perform 10-15% better through the day on average, that $0.03 cost difference is a $0.03 advantage.

Quoted from "littleguyintucson"

So what you are stating is, they have room to grow margins.

Or did I suggest that their projects margin is too good to be true in the future?

Some of the c-si modules have better temperature coefficient and better NOCt performance than first solar. Some others have worse. I should know i compared 950 different module specs in q3.

For apples to apples you need to consider:

- approx. 5 cents in warranty and shipping costs that is included in FS COGS but not in Chi-Solar costs
- an additional 3 cents in end-of-life recycling provision, included in FS COGS but not in Chi-Solar costs - my guess is that even though this is not necessarily a must for c-Si modules most projects will incur a decommissioning provision.
- a yield advantage of 5-9% relative to c-Si under hot climate conditions (FS target markets)
- a BOS penalty resulting from lower efficiency of the CdTe panels

Do the math and you end up more or less at parity with the most competitive Chi-Solars. I'm eager to see the YGE figures to see at what cost per watt they exited 2013...

However imho the value of FS results not from their current position but from the technological potential going forward. It seems to me that the room for higher efficiencies is larger for FS than for c-Si. For one the record cell efficiencies for c-Si have been stagnant for a long while (mono record set in 1999, poly record in 2004), and c-Si modules have already pushed the envelop considerably relative to these records. On the CdTe side the record cell efficiency is being pushed up at a much higher frequency and there is more room to expand module efficiency relative to record cell. What is quite unique in the case of FS is that the very same company owning the upstream cell record is the same marketing the electricity.

Quoted from "Klothilde"

- approx. 5 cents in warranty and shipping costs that is included in FS COGS but not in Chi-Solar costs
- an additional 3 cents in end-of-life recycling provision, included in FS COGS but not in Chi-Solar costs - my guess is that even though this is not necessarily a must for c-Si modules most projects will incur a decommissioning provision.
- a yield advantage of 5-9% relative to c-Si under hot climate conditions (FS target markets)
- a BOS penalty resulting from lower efficiency of the CdTe panels

Klothilde I agree with you in many things so dont take this the wrong way. What I heard is that Canadian solar did not include warrenty and shipping costs in the cost estimate they made, (source of this is a photon international article where one of the stock analysists says the number from canadian solar does not include warrenty and freight.) I have not had this confirmed by other sources or if for example Trina solar dont include this. I know REC includes it in the costs they make and they had also lower cost than first solar in q4. (REC with yesterdays euro to dollar conversion had 0.65$/watt cost vs first solar including all costs 0.68, granted the real costs of first solar is lower as they point out in the presentation.) Things gets more interesting if you add the opex costs. Rec is at 0.08$/watt opex while fslr is around 0.19$/watt. (sg&a r&d) Trina solar including one off (rec includes one off in its opex so i assume this for both) is 0.17$/watt. They have lower processing cost than rec at 0.51 vs 0.56 for rec. Also trina has underutilization like fslr while REC is at allmost 100% capacity. Meaning that the real cost of both trina and first solar is lower if they produced at 100% (spreading deprecation costs over more volume.) Trina commented deprecation was at 9 cents and could down to 6 cents with 90%+ utilitization. (Currently at 70%.)

When it comes to the yield advantage of FSLR vs c-si modules I am more sceptical. This is research done by FSLR, but there are other modules out there that perform exellent in tropical climates and hot climates. Again you just have to look at the temperature coefficent number and nominal operation temperature.

Regarding your previous comment about sun at early and late night, this is because of the spectrum of radiance, mostly at the red wavelenght, REC is using backside passivation sensitive to this and yes this is an advantage some manufactures tout to try to achive higher ASP. The higher ASP is justified by higher effiency during the whole day of operation of the module.

I should add REC has high interest costs so this is another factor to consider. Odyd makes a good presentation of these facts in his quarterly results report. I belive first solar is in a very good interest rate shape, while trina is a bit worse and rec kinda bad because they made many loans in the boom period of the economy.

Quoted from "eysteinh"

Klothilde I agree with you in many things so dont take this the wrong way.

Don't worry, I'm not here to push or bash but simply to exchange thoughts on facts. I believe that FS is a good technology play but I also want to be challenged on this because I don't want to lose money. I'm looking to open a c-Si position and am currently looking for the highest long-term value.

Quoted from "eysteinh"

What I heard is that Canadian solar did not include warrenty and shipping costs in the cost estimate they made, (source of this is a photon international article where one of the stock analysists says the number from canadian solar does not include warrenty and freight.) I have not had this confirmed by other sources or if for example Trina solar dont include this. I know REC includes it in the costs they make and they had also lower cost than first solar in q4. (REC with yesterdays euro to dollar conversion had 0.65$/watt cost vs first solar including all costs 0.68, granted the real costs of first solar is lower as they point out in the presentation.)

TSL AR 2011 Page 49: "Selling expenses consist primarily of provisions for product warranties, outbound freight, employee salaries, pensions, share-based compensation expenses and benefits, travel and other sales and marketing expenses."

YGE AR 2011 Page 76: "Our operating expenses consist of: • Selling expenses, which consist primarily of advertising costs, salaries and employee benefits of sales personnel, salesrelated travel and entertainment expenses, sales related shipping costs, warranty costs, amortization of intangible assets (including backlog and customer relationships), share-based compensation expenses and other selling expenses including sales commissions paid to our sales agents.

I like REC's cost position. From the Q3 numbers I get COGS of 52 €-Cts *1.31 = 68 uscents. No point in arguing about a few cents, this is extremely competitive.

Quoted from "eysteinh"

Things gets more interesting if you add the opex costs. Rec is at 0.08$/watt opex while fslr is around 0.19$/watt. (sg&a r&d) Trina solar including one off (rec includes one off in its opex so i assume this for both) is 0.17$/watt. They have lower processing cost than rec at 0.51 vs 0.56 for rec. Also trina has underutilization like fslr while REC is at allmost 100% capacity. Meaning that the real cost of both trina and first solar is lower if they produced at 100% (spreading deprecation costs over more volume.) Trina commented deprecation was at 9 cents and could down to 6 cents with 90%+ utilitization. (Currently at 70%.)

FSLR gets most the revenue from systems at a much higher ASP per watt, so an OPEX comparison on watt basis becomes less of apples to apples. OPEX over Revenue is less of a difference. However I look at the high OPEX not as a burden but as an investment in proprietary technology. They plan on increasing efficiency by a full percentag point during 2013, which by itself will provide them with 10 cents of value at system level. 2013 will be a crucial year for FS, they will have to prove the value of the technology.

BTW but this off-topic: I'm puzzled by the 9 cents depreciation comment of Trina, doesn't make sense to me at all. I had them at 27.5 million depreciation per quarter and relative to 415 MW that is 6.6 cents.

Quoted from "eysteinh"

When it comes to the yield advantage of FSLR vs c-si modules I am more sceptical. This is research done by FSLR, but there are other modules out there that perform exellent in tropical climates and hot climates. Again you just have to look at the temperature coefficent number and nominal operation temperature.

Juwi did a comparison between CdTe and FS a few years back, and they came up with an average of 5,3% yield advantage for CdTe in their European project portfolio. I expect the advantage to be larger in sunbelt countries. FS speaks of 5-9 % yield advantage. The deciding factor will be how much of this yield advantage FS can monetize for the customer through higher performance guarantees. My guess is that they assume a conservative position on this, maybe 5%.

Quoted from "eysteinh"

Regarding your previous comment about sun at early and late night...

I think this comes from littleguy. I'm keener on the yield advantage under hot climates.

Quoted from "Klothilde"

BTW but this off-topic: I'm puzzled by the 9 cents depreciation comment of Trina, doesn't make sense to me at all. I had them at 27.5 million depreciation per quarter and relative to 415 MW that is 6.6 cents.

Interesting since the 9 cents is what they reported during conference call.

Quoted

I like REC's cost position. From the Q3 numbers I get COGS of 52 €-Cts *1.31 = 68 uscents. No point in arguing about a few cents, this is extremely competitive.

Yeah its just rounding difference and what usd / eur exchange rate you use. I agree it is around that place.

Quoted

TSL AR 2011 Page 49: "Selling expenses consist primarily of provisions for product warranties, outbound freight, employee salaries, pensions, share-based compensation expenses and benefits, travel and other sales and marketing expenses."

YGE AR 2011 Page 76: "Our operating expenses consist of: • Selling expenses, which consist primarily of advertising costs, salaries and employee benefits of sales personnel, salesrelated travel and entertainment expenses, sales related shipping costs, warranty costs, amortization of intangible assets (including backlog and customer relationships), share-based compensation expenses and other selling expenses including sales commissions paid to our sales agents.

Exellent so there we have it straight from the reports, freight and warrenty is put into opex. Thanks again klothilde. No more rumors then it is true This explains perhaps why REC have a lower opex because it puts the cost into cogs, while also explaning the higher number for chinese c-si manufatures on opex.

Quoted

FSLR gets most the revenue from systems at a much higher ASP per watt, so an OPEX comparison on watt basis becomes less of apples to apples. OPEX over Revenue is less of a difference. However I look at the high OPEX not as a burden but as an investment in proprietary technology. They plan on increasing efficiency by a full percentag point during 2013, which by itself will provide them with 10 cents of value at system level. 2013 will be a crucial year for FS, they will have to prove the value of the technology.

Yes all opex is not the same. I also like the opex people spend on r&d if it makes results. I notice a few companies put a lot more into r&d than others.

(REC has and still put a lot of money into R&D)

Quoted

Juwi did a comparison between CdTe and FS a few years back, and they came up with an average of 5,3% yield advantage for CdTe in their European project portfolio. I expect the advantage to be larger in sunbelt countries. FS speaks of 5-9 % yield advantage. The deciding factor will be how much of this yield advantage FS can monetize for the customer through higher performance guarantees. My guess is that they assume a conservative position on this, maybe 5%.

On average across all module producers I think this is still the trend. All I mean to say is there are modules out there that outperform FSLR also in hot countries.

Quoted

I think this comes from littleguy. I'm keener on the yield advantage under hot climates.

My appologies then. Hopefuly answering his question then

Quoted from "eysteinh"

On average across all module producers I think this is still the trend. All I mean to say is there are modules out there that outperform FSLR also in hot countries.

Can you point out which ones? I'd be interested to have a look at them. I know you did an NOCT / STC comparison but forgot where you put the data...

Data is from q3 2012:
22% noct efficency loss: (what is noct? http://pvcdrom.pveducation.org/MODULE/NOCT.htm)
SCHOTT PROTECT ASI 107
SCHOTT PROTECT ASI 100
SCHOTT PROTECT ASI 105
SCHOTT PROTECT ASI 103

23.2% effiency loss
SILIKEN
SLK48P6L2180WBLACK
23.3%
REC240PEPLUS
23.4%
REC235PEPLUS
23.7%
REC245PEBLK
REC245PEECO
REC245PEPLUS
23.8%
REC240PEBLK
REC240PEECO
REC235PEBLK
REC235PEECO
24.0%
REC250PEPLUS
24.2%
REC260PEBLK
24.3%
REC255PEBLK
24.3%
REC255PEECO
REC255PEPLUS
24.4%
REC250PEBLK
REC250PEECO
24.9%
FIRST SOLAR
FS-275
FS-385
25.0%
FS-272
FS-382
FS-270
FS-280
FS-380
FS-390
FS-387
FS-277
25.6%
RENESOLA
JC250M-24/Bbv
25.7%
JC300M-24/Abv
25.7%
SUNPOWER
SPR-327NE-WHT-D
SPR-245NE-WHT-D
25.8%
RENESOLA
JC260M-24/Bbv
JC260M-24/Bb
JC310M-24/Abv
JC310M-24/Ab