112 January 2014

When combing through the PHOTON da-tabase for all inverters currently avail-

able on the market worldwide, only isolated devices can be found whose manufacturers promise a conversion efficiency of 99 percent or higher. Spanish manufacturer Jema Energy SA, for example, specifies 99 percent for its IF-225 TL plus, a 275 kW central inverter. In the category comprised of lower capacities, this issue’s test candidate, the 20000TL3-HE made by Growatt, is the only device whose manu-facturer promises 99 percent. This device was given to the PHOTON Laboratory in October 2013, and the obligatory test agreement was signed.

The 99-percent mark was broken for the first time a few years ago by the Sunny Trip-ower STP 20000TLHE-10, for which its manu-facturer, SMA Solar Technology AG, specified 98.5 percent (see PI 12/2011, p. 140). However, the SMA inverter made it to 99.15 percent only after the reverse current diodes had been de-activated – when the protective diodes were operating, its efficiency was slightly below 99 percent. Furthermore, SMA used silicon car-

Pushed to new heights

bide transistors, while Growatt manages with-out utilizing this silver bullet.

Of course, very good conversion efficiency is, on its own, not an adequate criterion for judging what is a very good inverter overall. Being able to break the 99-percent mark, how-ever, shows that the manufacturer is playing in the technological major league.

Construction

The 20000TL3-HE is one of two three-phase transformerless devices in the HE fam-ily (HE stands for »High Efficiency«). Along with the inverter tested for this report, with its nominal AC capacity of 20,000 W, there is a slightly smaller 18,000 W device in the series.

The construction of the 20000TL3-HE is highly compact and production-friendly. The power element is housed on three circuit boards, two on the lower level directly on the cooling element, and one on the second level. This level is also where the control and com-munication circuit boards are installed, with

A 20 kW device made by Growatt, the 20000TL3-HE, is only the second inverter tested by PHOTON Lab to achieve a conversion efficiency higher than 99 percent. The device scores third place overall, even without silicon carbide transistors

Highlights

The 20000TL3-HE is a 20 kW trans-•formerless inverter featuring three-phase feed-in and one MPP trackerThe device reaches a conversion ef-•ficiency of 99.02 percent – the sec-ond-highest value ever measured by PHOTON Lab. Its PHOTON efficiency for medium irradiation totals 98.0 per-cent, which scores it an »A+« and third place in the PHOTON ranking There are still a few weaknesses •when it comes to MPP tracking at low voltages and high powers. New firmware should be able to rectify this and could push the 20000TL3-HE even higher up the rankings

science & technology | inverters | test

A+1/2014

www.photon.info/ laboratory

98.4 % at high irradiation

Growatt New Energy Growatt 20000TL3-HE

A+1/2014

www.photon.info/ laboratory

98.0 % at medium irradiation

Growatt New Energy Growatt 20000TL3-HE

Text: Heinz Neuenstein, Anne Kreutzmann

January 2014 113

Small, low in weight, compact and highly efficient: The 20000TL3-HE, a 20 kW inverter made by Growatt, achieves high maximum efficiencies despite its rather Spartan equipment features. The thermographic image shows component temperatures of up to 80 °C on the circuit boards (on the surface of the grid disconnect relay). The two electrolytic capacitors that are visible are, in contrast, in the green or, in this case, the blue part of the temperature scale.

80.078.9

45.342.0

20

28

36

44

52

60

68

76

83

90

80.078.9

45.342.0

20

28

36

44

52

60

68

76

83

90

phot

on-p

ictu

res.

com

(2)

the display connected to the latter. The power semi-conductors all feature a discrete design and are fitted from the solder side. They have been screwed onto a cooling element below them, which protrudes through a cutout in the base of the housing. The control and com-munication circuit boards are interconnected by flat ribbon cables. Their plug-in connec-tors have been affixed with a silicon-like ad-hesive that has not been cleanly applied in this case. The DC input and the AC output with the interference filter components, the gird disconnect relay, the fault current moni-tor and the automatic circuit breaker are on the lower power element circuit board.

The three chokes for the AC-side sine-wave filter are installed in a separate compart-ment in the upper part of the housing. This compartment is a half-shell made of cast alu-minum with small cooling fins in which the chokes are positioned. The half-shell is fully cast, ensuring the thermal coupling of the chokes to the surface of the housing is opti-mal. Unfortunately, this also couples it acous-tically, so the device emits an appreciable

amount of noise. This upside-down half-shell is fitted onto the upper surface of the hous-ing. The device features passive cooling only, with no fans being used, a consequence of the high efficiency. The cooling element is a com-plex high-performance cooling element with hollow fins molded into an extruded alumi-num baseplate, with the fins also made of alu-minum. The housing is formed by a welded frame made of steel sheet and is protected by a cover made of thick aluminum plate. The housing is therefore comprised of four parts, and it features a IP 65 protection type.

The display is found behind a cutout in the housing cover and is covered by transparent film. The device status is indicated by the dis-play and an LED.

The electrolytic capacitors used in the pow-er element belong to temperature class 105 °C, a good choice for the prevailing ambient temperatures. Varistors have been utilized in the AC output to limit overvoltage, but not in the DC input.

The solar generator is connected using six pairs of PV connectors made by Amphenol.

There is only one block terminal covered by a small, separate cap for connecting to the grid.

The internal DC disconnect is installed on the bottom of the device. There are no string fuses. All in all, the device is rather Spartan in design. When it comes to communication op-tions, there is an RS232 interface for commu-nicating with a PC and an RS485 interface for communicating with several other inverters along with a monitoring unit. Wi-Fi is avail-able as an optional extra using a plug-in card on the RS232 interface, as is a data logger.

Operation

The device is delivered to the user well packaged and protected by a cardboard box with foam half-shells. The inverter is in-stalled on a wall with the aid of a bracket. The Growatt 20000TL3-HE, which weighs 52 kg, is lightweight in terms of its nomi-nal DC power. Once the solar generator has been correctly dimensioned and the internal DC disconnect has been switched on, the inverter requires approximately another 42

114 January 2014

Conversion efficiency

A rare sight: The conversion efficiency exceeds 98 percent over almost the entire operating range. The maximum of 99.02 percent is found at 30 percent of nominal power and an MPP voltage of 603 V. A conver-sion efficiency of more than 99 percent has only been measured at the PHOTON laboratory once before, when the Sunny Tripower STP 20000TLHE-10 made by SMA was tested.

TM TR UM UR VM VR NMMRUMRVOSMPSNRSOSSPUSQVSSNSTPSUQSVSTMTTNVTPNTQOTRQTSRTTTTUUUMM

M R NM NR OM ORM

OMM

QMM

SMM

UMM

NIMMM

NM OM PM QM RM SM TM UM VM NMM NNM NOMTMTRUMURVMVR

NMM

VU

VT

VV

VVNM OM PM QM RM SM TM UM VM NMM NNM NOM

RUMRVOSMPSNRSOSSPUSQVSSNSTPSUQSVSTMTTNVTPNTQOTRQTSRTTTTUUUMM

η áå=B s

ât B

η áå=BB=kçãáå~ä=éçïÉê=EmjmmF

s jmm=áå=s

VV

VVVU

VU

VV

NM OM PM QM RM SM TM UM VM NMM NNM NOMRUMRVOSMPSNRSOSSPUSQVSSNSTPSUQSVSTMTTNVTPNTQOTRQTSRTTTTUUUMM

B

B=kçãáå~ä=éçïÉê=EmjmmF

sjmm=áå=s

TM TR UM UR VM VR NMMRUMRVOSMPSNRSOSSPUSQVSSNSTPSUQSVSTMTTNVTPNTQOTRQTSRTTTTUUUMM

M R NM NR OM ORM

OMM

QMM

SMM

UMM

NIMMM

NM OM PM QM RM SM TM UM VM NMM NNM NOMTMTRUMURVMVR

NMM

NM OM PM QM RM SM TM UM VM NMM NNM NOMRUMRVOSMPSNRSOSSPUSQVSSNSTPSUQSVSTMTTNVTPNTQOTRQTSRTTTTUUUMM

ηpìã=áå=B s

ât B

ηpìã=áå=BB=kçãáå~ä=éçïÉê=EmjmmF

s jmm=áå=s

VT

VU

× MPPT adjustment efficiency

A minor goof-up: The MPPT adjustment efficiency is highly uniform, and very high, over almost the entire operating range. At low powers there are problems with the MPP tracking, and the MPPT adjustment ef-ficiency falls below 99 percent and, at times, to less than 98 percent. When MPP voltages are low and DC powers are high, the inverter reacts by reducing power.

= Overall efficiency

The maximum overall efficiency range is found at low voltages and medium powers. The overall efficiency maximum of 98.9 percent is found at 35 percent of nominal power and an MPP voltage of 580 V.

science & technology | inverters | test

January 2014 115

seconds for a variety of tests. It then con-nects to the grid and starts work.

The graphics-capable display is below the front cover and features white backlighting. It is difficult to read. It is activated by tapping the cover of the housing near the display. The menu is also operated this way. English, Ger-man, French and Spanish are available as op-erating languages.

Along with diverse status and error messag-es, the following actual values are displayed: DC power, grid frequency, DC voltage, DC current and a bar diagram showing the energy feed-in curve in kilowatt-hours over the course of the day, the week, the year, or as a total. The most important values are therefore provided.

Operating manual

An English-language operating manual is provided with the device. Along with general explanations, it covers installation, informa-tion on operating characteristics, mainte-nance, the display, error messages and in-structions for the communication options and a number of block circuit diagrams of a system configuration.

Circuit design

The inverter features a single-stage circuit design. The energy from the PV generator flows past an EMI filter and directly into the intermediate voltage circuit, to which a ca-pacitor half bridge and a three-phase capaci-tor half bridge are connected. Growatt calls the technology used for this output bridge, which is comprised of an optimized three-voltage-level bridge, »Smart Topology«. According to statements made by Growatt, conventional silicon FETs (field effect transistors) and IG-BTs (insulated gate bipolar transistors) are used in the branches of the output bridge, and no silicon carbide transistors have been used. There are mounting points for reverse current diodes, but these have not been installed.

The output-side filter smoothens the modulated voltage blocks into sine-wave voltage with the grid frequency of 50 Hz. An automatic circuit breaker disconnects the inverter from the grid if the grid voltage or grid frequency deviates from the defined thresholds, doing the same if a fault current or a DC component occurs on the grid side. Furthermore, the ground resistance and the discharge current are checked on the DC side. Any radio interference occurring is eliminated by an output filter, which is found directly in front of the grid terminals.

Measurements

All of the following measurements are based on a grid voltage of 230 V. The maximum DC voltage the Growatt 20000TL3-HE can process totals 1,000 V, with the nominal DC power be-

ing 20,000 W. The manufacturer recommends connecting a maximum power of 20,800 W.

Locating the MPP: When measuring be-gan, both the DC side and the AC side were switched off. At a predefined IV curve with nominal power and an MPP voltage of 684 V, the inverter requires around 42 seconds before it connects to the grid. Around 88 more sec-onds pass before the inverter reaches the MPP. It needs 11 seconds to change from 684 V to 673 V, and changing to the next-highest MPP range of 696 V also took 11 seconds.

MPP range: The MPP range extends from 580 to 800 V and is equivalent to that of a nor-mal range inverter. The maximum MPP volt-age of 800 V is at a good distance away from the maximum input voltage of 1,000 V.

Conversion efficiency: The inverter can operate at 105 percent of nominal power in the MPP voltage range spanning 580 to 800 V. The efficiency could therefore be determined for this part of the diagram.

At a maximum DC voltage of 1,000 V, there is a hatched area that indicates limitations when thin-film modules are used due to the insufficient distance to the maximum MPP voltage and the maximum DC voltage.

The efficiency reaches more than 98 percent over the entire voltage range and at powers of more than 10 percent. The maximum of 99.02 percent is found at 30 percent of nomi-nal power and an MPP voltage of 603 V. This means the manufacturer’s specification of 99.0 percent was reached. Toward high MPP volt-ages, the conversion efficiency falls by around 0.4 percentage points, falling by around 0.05 percentage points toward low powers.

At low powers of less than 15 percent of nom-inal power, the efficiency of this device falls by 1.6 to 2.1 percentage points. At nominal power, the power factor cos φ was around 1.

Weighted conversion efficiency: Euro-pean efficiency reaches its maximum at the lowest MPP voltage and is, at 98.8 percent, another 0.2 percentage points higher than the manufacturer’s specification of 98.6 percent. The difference between maximum conversion efficiency and maximum European efficiency only totals 0.2 percentage points. Californian efficiency takes a virtually identical course to European efficiency.

MPPT adjustment efficiency: The MPPT adjustment efficiency is highly uniform and very high over almost the entire operating range – but there are two limitations. At low powers, there are problems with the MPP tracking, and the MPPT adjustment efficiency falls below 99 percent and, at times, to less than 98 percent. When the MPP voltage is low and DC power is high, the inverter reacts by reducing output. This effect does, however, depend on the grid voltage level, meaning the ratio of DC voltage to grid voltage. Because this inverter features a single-stage concept, the DC voltage may not, in some cases, be suf-ficient to feed in at a 230 V or 400 V level. The

adjustment problems in the DC voltage range spanning 580 to 626 V could be eliminated with a slightly lower grid voltage of 225 V, sup-porting this theory.

Overall efficiency: Overall efficiency, which is the product calculated from conver-sion and MPPT adjustment efficiency, reaches its highest level at low voltages and medium powers. The maximum of 98.9 percent is found at 35 percent of nominal power and an MPP voltage of 580 V.

Overall efficiency curves, average over-all efficiency and PHOTON efficiency: The PHOTON efficiency for medium irradiation totals 98.0 percent, and the PHOTON efficien-cy for high irradiation is 98.4 percent.

Feed in at nominal power: The inverter feeds in 100 percent of nominal power over almost the entire input voltage range spanning 580 to 800 V and at an ambient temperature of 25 °C.

Displayed output power: At a constant MPP voltage of 684 V, which is the medium range, the output power displayed by the in-verter at different powers between 5 and 105 percent of nominal power was compared with the value identified by a power analyzer. At low powers, the display deviates by up to -9.3 percent from the value measured by the power analyzer. At nominal powers of 20 percent or more, the error is still within a range of -3 to -5 percent. The level of accuracy exhibited by the display is therefore insufficient.

Operation at high ambient temperatures: The inverter keeps feeding 100 percent of its nominal power output into the grid without any reductions up to an ambient temperature of around 61.6 °C. The operating point select-ed was at 20,000 W and an MPP voltage of 684 V. The efficiency fell by around 0.3 percentage points over this temperature range.

Overload behavior: If the Growatt 20000TL3-HE is fed an overload of 1.3 times the nominal input power, making it 26,000 W, the inverter limits output to a DC power of 20,917 W. This is equivalent to an overload of 4.6 percent. The device therefore features a small overload range. When power is limited in this way, the device pushes the operating point on the IV curve toward higher input voltages. The DC voltage value was around 762 V. The operating point chosen was found at an MPP voltage of 684 V and an ambient temperature of 24 °C.

Own consumption and night consump-tion: The energy consumed by the device in its basic tested state totals around 0.4 W on the AC side and 16.2 W on the DC side, with the manufacturer not specifying a figure for this. At night, the inverter also draws around 0.4 W of effective power from the grid. The manufac-turer specifies less than 0.5 W.

Thermography: The thermographic cam-era shows the inverter from above while it is operating at an ambient temperature of 24 °C at nominal power. It reveals component

116 January 2014

Weighted conversion efficiency

Exemplary: European efficiency reaches a maximum at the lowest MPP voltages and is, at 98.8 percent, 0.2 percentage points higher than the manufacturer’s specification of 98.6 percent. The difference between maximum conversion efficiency and maximum Euro-pean efficiency only amounts to 0.2 percentage points. Californian efficiency follows a curve that is almost identical to European efficiency.

RUM RVO SMP SNR SOS SPU SQV SSN STP SUQ SVS TMT TNV TPN TQO TRQ TSR TTT TUU UMMTM

TO

TQ

TS

TU

UM

UO

UQ

US

UU

VM

VO

VQ

VS

VU

NMM

=bìêçéÉ~å=ïÉáÖÜíÉÇ==== ηbìêçj~ñ=Z=VUKU=B

=`~äáÑçêåá~å=ïÉáÖÜíÉÇ=== η`b`j~ñ==Z=VUKU=B

sjmm=áå=s

tÉáÖÜíÉÇ=ÅçåîÉêëáçå=ÉÑÑáÅáÉåÅó=ηbìêçI=η`b`=áå=B

ηbìêç=ã~åìÑ~ÅíìêÉê=ëéÉÅáÑáÉÇ=Z=VUKS=B

R NM NR OM OR PM PR QM QR RM RR SM SR TM TR UM UR VM VR NMM NMR NNM NNR NOMTM

TO

TQ

TS

TU

UM

UO

UQ

US

UU

VM

VO

VQ

VS

VU

NMMlîÉê~ää=ÉÑÑáÅáÉåÅó=ηpìã=áå=B

B=kçãáå~ä=éçïÉê=EmjmmF

==sjmm=Z=RUMKM=s=EsjmmãáåF ηpìãj~ñ=Z=VUKVQ=B

==sjmm=Z=UMMKM=s=Esjmmã~ñF ηpìãj~ñ=Z=VUKSN=B

==^îÉê~ÖÉ=çîÉê~ää=ÉÑÑáÅáÉåÅó= η^îÖpìãj~ñ=Z=VUKTU=B

ηmãÉÇ=Z=VUKM=BI=ηmÜáÖÜ=Z=VUKQ=B

ηpìãj~ñ

R NM NR OM OR PM PR QM QR RM RR SM SR TM TR UM UR VM VR NMM NMR NNM NNR NOMJPMJOUJOSJOQJOOJOMJNUJNSJNQJNOJNMJUJSJQJOMOQSU

NMNONQNSNUOMOOOQOSOUPM

=

B=kçãáå~ä=éçïÉê=EmjmmF

aÉîá~íáçå=áå=B

Overall efficiency at different voltages

Third place in the overall list of rankings: The PHOTON efficiency for medium irradiation is 98.0 percent, a level that only two test candidates have exceeded to date. If there were no decline at low MPP voltages (blue curve) and high powers, the result would have been even better.

Accuracy of inverter display

No longer up-to-date: At low powers, the display dif-fers by up to -9.3 percent from the value measured by the power analyzer. At 20 percent of nominal power of more, the error range totals -3 to -5 percent. This is something other inverters are considerably better at.

science & technology | inverters | test

January 2014 117

temperatures on the circuit boards of up to 80.0 °C, measured on the surface of the grid disconnect relay. Two additional temperature markers can also be seen in the upper area, which is the part of the housing where the cast chokes are located. Only a small strip of this die-cast aluminum cap was blackened in the thermographic images made by PHOTON Laboratory. A surface temperature of 45.3 °C was measured there. The other, non-black-ened area has an emission factor that is too low for thermographic imaging, so the value of 42.0 °C determined there cannot be accu-rate. The two visible electrolytic capacitors in the power element are in the non-critical part of the temperature scale. The multi-lay-er construction of the device only allows the components that are visible to be shown.

Summary

Chinese company Growatt New Energy Technology Co. Ltd. has developed a top device with only a few small weaknesses in the form of its 20000TL3-HE. The inverter features a very compact construction and is production-friendly, and it is very light and small for a three-phase device. The num-ber of additional functions has been kept to a minimum, producing a low-cost device, which can be supplemented by a number of additional external components. The loud operating noises are unpleasant, and the rather imprecise inverter power display is a drawback.

The conversion efficiency is very good and uniformly high over the entire operat-ing range and remains almost consistently above 98 percent. A peak value of 99.02 percent was measured, which confirms the manufacturer’s specification of 99.0 percent. Its European and Californian efficiencies are just 0.2 percentage points lower than the maximum conversion efficiency. The very consistent, and high, MPPT adjustment ef-ficiency also results in a high and uniform overall efficiency. A number of weak points in the MPP tracking emerge at low powers and low voltages, as well as at high powers, which the manufacturer intends to tackle by overhauling the software. However, even with these problem areas, the PHOTON efficiency for medium irradiation totals a very high 98.0 percent, with the PHOTON efficiency for high irradiation being another 0.4 per-centage points higher, at 98.4 percent, as the problem encountered at low powers are not weighted as highly.

Efficiencies as high as these can only be achieved by devices that do not feature a (loss-prone) boost converter in the DC input. The smaller input voltage range featured by the 20000TL3-HE, which spans 580 to 800 V, also helps. The maximum MPP voltage of 800 V is at a good distance away from the maximum input voltage of 1,000 V; however, there are

Manufacturer’s response

We have verified the test results relating to the MPPT efficiency.

limitations with the use of thin-film modules. The choice of a single-stage topology is one

reason for the high efficiencies, along with an optimized output bridge, which Growatt has christened »Smart Topology.« Silicon carbide transistors have not been employed. The high efficiency also allows active cooling by fans to be dispensed with. Due to the very wide tem-perature range spanning -25 to +60 °C, and protection type IP 65, the inverter can be in-stalled under the roof or outdoors. No power reduction was observed at high temperatures.

The 20000TL3-HE finds itself in third place in the PHOTON ranking for the »medium ir-radiation« category (see table, p. 52 - 54), shar-ing a place with the Sun2000-20KTL made by Huawei Technologies Co. Ltd. and the Plati-num 16000 R3 made by Diehl AKO Stiftung & Co. KG. Weighted for high irradiation, the 20000TL3-HE is able to claim second place, allowing it to overtake the Refusol 020k SCI made by Refusol GmbH, which is equipped with silicon carbide transistors.

52 January 2014

Inverter test resultsInverter Observed

voltage range*1

Medium irradiation High irradiation PI issue

etaPmed Grade as of 2011

Grade before 2011

Position etaPhigh Grade as of 2011

Grade before 2011

Position

science & technology | inverters | test

SMA's STP 20000TLHE-10*3 580 - 800 V 98,5 % A+ – 1 98,6 % A+ – 1 12/2011

Refusol's 020k SCI 490 - 800 V 98,2 % A+ – 2 98,3 % A+ – 3 7/2012

Growatt's 20000TL3-HE 580 - 800 V 98,0 % A+ – 3 98,4 % A+ – 2 1/2014

Huawei Technologies Co. Ltd.'s Sun2000-20KTL 480 - 800 V 98,0 % A+ – 3 98,1 % A+ – 4 6/2013

Diehl AKO's Platinum 16000 R3 350 - 720 V 98,0 % A+ – 3 98,0 % A+ – 5 3/2013

Donauer Solartechnik's High Efficiency 3.6 350 - 650 V 97,8 % A – 6 97,9 % A – 6 12/2012

Steca's StecaGrid 3600 350 - 600 V 97,7 % A – 7 97,8 % A – 7 12/2011

Goodwe Power Supply Technology's GW17K-DT 500 - 800 V 97,6 % A – 8 97,8 % A – 7 10/2013

Steca's Stecagrid 3000 350 - 700 V 97,5 % A – 9 97,8 % A – 7 9/2012

Siemens’ Sinvert PVM20 480 - 850 V 97,5 % A – 9 97,7 % A – 10 4/2011

Sungrow's SG30KTL 480 - 800 V 97,5 % A – 9 97,7 % A – 10 2/2013

Siemens’ Sinvert PVM17 460 - 850 V 97,4 % A – 12 97,7 % A – 10 4/2011

Refusol's 017K 460 - 850 V 97,4 % A A+ 12 97,6 % A A+ 13 12/2010

Refusol's 020K (808) 490 - 850 V 97,4 % A – 12 97,6 % A – 13 11/2013

Global Mainstream Dynamic Energy Technology's Soldate 318KTLE

490 - 800 V 97,3 % A – 15 97,6 % A – 13 7/2013

Refusol's 013K 420 - 850 V 97,3 % A A+ 15 97,6 % A A+ 13 12/2010

Siemens' Sinvert PVM13 420 - 850 V 97,3 % A – 15 97,6 % A – 13 4/2011

Refusol's 020K 480 - 850 V 97,3 % A – 15 97,5 % A – 18 3/2012

SMA's STP 17000TL 400 - 800 V 97,3 % A A+ 15 97,5 % A A+ 18 12/2010

Chint Power's CPS SCA12KTL-DOHE 430 - 800 V 97,2 % A – 20 97,4 % A – 21 11/2013

SMA's STP 10000TL-10 320 - 800 V 97,1 % A – 21 97,5 % A – 18 10/2011

Chint Power's CPS SC20KTL-O 500 - 800 V 97,1 % A – 21 97,4 % A – 21 11/2011

Siemens' Sinvert PVM10 380 - 850 V 97,0 % A – 23 97,4 % A – 21 1/2011

Delta Energy Systems' Solivia 20 EU G3 TL 350 - 800 V 97,0 % A – 23 97,2 % A – 26 3/2012

Hosola's Bright 4200TL 300 - 500 V 97,0 % A – 23 97,2 % A – 26 12/2013

Zeversolar New Energy's Eversol-TLC 17k*2 550 - 720 V 96,9 % A – 26 97,3 % A – 24 4/2011

Mastervolt's Sunmaster CS20TL 350 - 800 V 96,9 % A – 26 97,2 % A – 26 5/2011

Power-One's Trio-27.6-TL-OUTD-S2-400 500 - 800 V 96,9 % A – 26 97,2 % A – 26 2/2013

Refusol's 011K*3 380 - 800 V 96,9 % A A+ 26 97,2 % A A+ 26 9/2008

Goodwe Power Supply Technology's GW4000-SS 280 - 500 V 96,9 % A – 26 97,1 % A – 32 12/2012

Kaco's Powador 60.0 TL3 (firmware V2.02) 480 - 850 V 96,9 % A – 26 97,0 % A – 37 9/2013

SMA's SMC 8000 TL*3 335 - 487 V 96,9 % A A+ 26 97,0 % A A+ 37 10/2007

SMA's SMC 11000TL*3 333 - 500 V 96,9 % A A+ 26 96,8 % A A+ 53 7/2010

B&B Power's SF 4600TL 250 - 500 V 96,8 % A – 34 97,3 % A – 24 7/2013

Growatt's 5000MTL (version 2) 250 - 540 V 96,8 % A – 34 97,1 % A – 32 12/2012

Sputnik's Solarmax 13MT*4 250 - 750 V 96,8 % A – 34 97,1 % A – 32 9/2011

Diehl AKO's Platinum 6300 TL*3 350 - 710 V 96,8 % A A+ 34 96,9 % A A+ 49 2/2009

Kaco's Powador 60.0 TL3 (firmware V2.10) 480 - 850 V 96,8 % A – 34 96,9 % A – 49 9/2013

Power-One's TRIO-20.0-TL-OUTD S2-400 410 - 800 V 96,7 % A – 39 97,1 % A – 32 9/2012

Danfoss's TLX 15 k 430 - 800 V 96,7 % A A+ 39 97,0 % A A+ 37 6/2010

Samil Power's Solarlake 15000TL 380 - 800 V 96,7 % A – 39 97,0 % A – 37 6/2012

Zeversolar New Energy's Eversol-TL 4600 290 - 500 V 96,7 % A – 39 97,0 % A – 37 9/2011

Sunways' NT 4200 340 - 750 V 96,7 % A A+ 39 96,8 % A A+ 53 3/2010

Sunways' PT33k 460 - 800 V 96,7 % A – 39 96,8 % A – 53 6/2012

Primevolt's PV-10000T-U 550 - 850 V 96,6 % A – 45 97,2 % A – 26 11/2013

Conergy's IPG 15T 450 - 800 V 96,6 % A A+ 45 97,0 % A A+ 37 8/2010

Hangzhou Zhejiang University Sunny Energy Sci-ence and Technology's ZDNY-TL 17000

400 - 800 V 96,6 % A – 45 97,0 % A – 37 9/2013

Kinglong's KLNE Solartec D 15000 480 - 750 V 96,6 % A – 45 97,0 % A – 37 3/2013

Kinglong's KLNE Sunteams 5000 280 - 440 V 96,6 % A – 45 97,0 % A – 37 5/2012

Sungrow's SG15KTL 380 - 800 V 96,6 % A – 45 97,0 % A – 37 2/2012

SMA's SMC 7000TL*3 333 - 500 V 96,6 % A A+ 45 96,8 % A A+ 53 5/2010

Sunways' NT 11000 340 - 750 V 96,6 % A – 45 96,7 % A – 62 11/2012

Danfoss's TLX 10 k 430 - 800 V 96,5 % A A+ 53 97,0 % A A+ 37 8/2010

Eaton Phoenixtec MMPL's SV 20000s 450 - 850 V 96,5 % A – 53 96,8 % A – 53 5/2013

Samil Power's Solarriver SR4K4TLA1 200 - 500 V 96,5 % A – 53 96,8 % A – 53 8/2011

January 2014 53

Eltek Valere's Theia 4.4HE-t*5 230 - 480 V 96,5 % A – 53 96,7 % A – 62 11/2011

Schneider Electric's Conext TL 20000 E 350 - 800 V 96,4 % B – 57 97,1 % A – 32 8/2013

Power-One's Aurora PVI-12.5-OUTD-FS*3 360 - 750 V 96,4 % B A 57 96,9 % A A+ 49 4/2010

SLD Power Technology's SLS5KH65 225 - 500 V 96,4 % B – 57 96,7 % A – 62 5/2013

Sputnik's SolarMax 5000P 260 - 480 V 96,4 % B – 57 96,7 % A – 62 1/2014

B&B Power's SF 3000TL 250 - 450 V 96,3 % B – 61 96,9 % A – 49 5/2013

Helios's HSI20 350 - 800 V 96,2 % B – 62 97,0 % A – 37 3/2012

Growatt's 5000 MTL 250 - 550 V 96,2 % B – 62 96,8 % A – 53 7/2012

Kaco's Powador 4000 supreme DCS (9 kHz)*3 350 - 510 V 96,2 % B A 62 96,7 % A A+ 62 1/2010

Kstar's New Energy KSG-5K (version 2) 280 - 480 V 96,2 % B – 62 96,6 % A – 67 12/2012

Kstar's New Energy KSG-3K 190 - 440 V 96,1 % B – 66 96,6 % A – 67 8/2012

Trannergy's PVI 4600TL 300 - 500 V 96,1 % B – 66 96,6 % A – 67 8/2012

Growatt's 5000 TL 280 - 500 V 96,0 % B – 68 96,8 % A – 53 2/2011

Fronius's IG TL 5.0 350 - 700 V 95,9 % B A 69 96,2 % B A 72 9/2010

Kstar's New Energy KSG-17K 400 - 720 V 95,7 % B – 70 96,3 % B – 70 8/2013

Kaco's Powador 4000 supreme DCS (18 kHz)*3 350 - 510 V 95,7 % B A 70 96,1 % B A 74 1/2010

SMA's SB 5000TL-20*3 175 - 440 V 95,7 % B A 70 96,0 % B A 76 5/2009

Sungrow's SG4KTL 210 - 420 V 95,6 % B – 73 96,3 % B – 70 1/2011

Eaton Phoenixtec MMPL's SV 10000s 350 - 850 V 95,5 % B – 74 96,2 % B – 72 10/2013

Omron's KP100L (OD-EU) 320 - 825 V 95,5 % B – 74 96,1 % B – 74 1/2013

Sanjing Electric's SAJ Sununo TL5K 200 - 440 V 95,5 % B – 74 96,0 % B – 76 5/2012

Power-One's Aurora PVI-6000-OUTD-S*3 180 - 530 V 95,4 % B A 77 95,9 % B A 78 3/2009

Jingfuyuan Tech.'s Suntree 10000TL 500 - 720 V 95,2 % B – 78 96,8 % A – 53 12/2013

Omnik New Energy’s Omniksol-2k-TL 120 - 450 V 95,2 % B – 78 95,9 % B – 78 1/2012

Aros' Sirio 4000*3,8 250 - 450 V 95,1 % B A 80 95,7 % B A 81 12/2008

Dasstech's DSP-123K2 200 - 450 V 95,1 % B – 80 95,7 % B – 81 3/2011

Kstar's New Energy KSG-5K (version 1) 280 - 480 V 95,1 % B – 80 95,1 % B – 88 12/2012

Conergy's IPG 5 S*3 275 - 750 V 95,0 % B A 83 95,8 % B A 80 9/2009

Fronius's IG Plus 100*3 230 - 500 V 94,8 % C B 84 95,0 % B A 92 11/2010

SMA's SB 3000HF-30 210 - 560 V 94,7 % C – 85 95,2 % B – 86 2/2012

Power-One's Uno-2.5-I-OUTD-S 200 - 470 V 94,6 % C – 86 95,4 % B – 84 5/2013

Yisun New Energy Tech.'s Yisun-2K-TL 120 - 450 V 94,6 % C – 86 95,4 % B – 84 12/2012

Fronius's IG Plus 150 V-3 230 - 500 V 94,6 % C – 86 95,1 % B – 88 10/2012

Sunways' AT 4500 250 - 600 V 94,6 % C B 86 94,8 % C B 97 7/2008

Sungrow's SG3KTL (version 2) 180 - 420 V 94,5 % C – 90 95,7 % B – 81 8/2011

Fronius's IG Plus 50 230 - 500 V 94,5 % C B 90 94,8 % C B 97 8/2008

Phoenixtec's PVG 2800 (updated model) 250 - 450 V 94,4 % C B 92 95,1 % B A 88 5/2008

Kaco's Powador 8000xi (new software; since Jan. 2010)*3

350 - 600 V 94,4 % C B 92 94,7 % C B 101 3/2010

Kaco's Powador 2500xi DCS*3 350 - 600 V 94,3 % C B 94 95,0 % B A 92 1/2010

Motech Industries' PVMate 5000E 200 - 550 V 94,3 % C – 94 94,9 % C – 96 6/2013

Sunways' AT 2700 181 - 600 V 94,3 % C B 94 94,8 % C B 97 8/2009

Sputnik's SolarMax 6000S 220 - 550 V 94,3 % C B 94 94,7 % C B 101 11/2009

Effekta's ES5000 (new software, PV00113L) 150 - 450 V 94,2 % C – 98 94,8 % C – 97 2/2012

Carlo Gavazzi's ISMG150DE 200 - 450 V 94,1 % C B 99 95,0 % B A 92 5/2010

Xantrex's GT5.0SP*6,3 240 - 550 V 94,1 % C B 99 94,7 % C B 101 1/2009

Conergy's IPG 5000 vision*3 301 - 706 V 94,0 % C B 101 94,7 % C B 101 7/2007

Kaco's Powador 8000xi (old firmware; till Jan. 2010)*3

350 - 600 V 94,0 % C B 101 94,7 % C B 101 3/2010

Kostal's Piko 10.1 400 - 850 V 94,0 % C B 101 94,4 % C B 113 7/2009

Delta Energy Systems' SI 3300*3 150 - 435 V 93,9 % C B 104 94,7 % C B 101 5/2008

Mitsubishi's PV-PNS06ATL-GER 260 - 650 V 93,9 % C B 104 94,6 % C B 107 6/2008

SMA's SMC 7000HV*3 335 - 560V 93,9 % C B 104 94,2 % C B 115 9/2009

Sunways' NT 2600 (lower range)*3 350 - 623 V 93,8 % C B 107 95,1 % B A 88 11/2007

Steca’s Stecagrid 9000 3ph*3 350 - 680 V 93,8 % C B 107 95,0 % B A 92 7/2010

Sputnik's SolarMax 2000C*3 165 - 515 V 93,8 % C B 107 93,1 % D C 128 4/2007

Sungrow's SG3KTL (version 1) 180 - 420 V 93,7 % C – 110 95,2 % B – 86 8/2011

Inverter test resultsInverter Observed

voltage range*1

Medium irradiation High irradiation PI issue

etaPmed Grade as of 2011

Grade before 2011

Position etaPhigh Grade as of 2011

Grade before 2011

Position

54 January 2014

Inverter test resultsInverter Observed

voltage range*1

Medium irradiation High irradiation PI issue

etaPmed Grade as of 2011

Grade before 2011

Position etaPhigh Grade as of 2011

Grade before 2011

Position

Kaco's Powador 4202 200 - 510 V 93,7 % C B 110 94,6 % C B 107 10/2010

SMA's SB 2100TL 200 - 480 V 93,7 % C B 110 94,6 % C B 107 6/2009

Oelmaier's PAC 4 330 - 600 V 93,6 % C B 113 94,6 % C B 107 12/2009

Mastervolt's Sunmaster XS6500 180 - 480 V 93,6 % C B 113 94,1 % C B 116 2/2010

Ingeteam's Ingecon Sun 3.3 TL 159 - 414 V 93,4 % D C 115 94,3 % C B 114 8/2007

SMA's SB 3800*3 208 - 395 V 93,2 % D C 116 93,6 % C B 120 2/2007

Dasstech's DSP-123KH*3 350 - 600 V 93,0 % D C 117 94,6 % C B 107 10/2010

Diehl AKO's Platinum 4600S 320 - 628 V 92,9 % D C 118 93,3 % D C 125 4/2008

Power-One's Aurora PVI-2000-OUTD-DE*3 210 - 530 V 92,8 % D C 119 94,0 % C B 117 2/2010

Diehl AKO's Platinum 2100S 206 - 390 V 92,8 % D C 119 93,3 % D C 125 10/2009

Kaco's Powador 3501xi*3 125 - 391 V 92,6 % D C 121 92,9 % D C 129 6/2007

Kaco's Powador 2500xi*3 350 - 597 V 92,5 % D C 122 93,4 % D C 122 12/2007

Sunways' NT 2600 (upper range)*3 476 - 749 V 92,3 % D C 123 93,9 % C B 118 11/2007

Solon's Satis 40/750 IT*7 375 - 575 V 92,3 % D C 123 93,5 % C B 121 11/2008

Mastervolt's QS 2000*3 212 - 366 V 92,3 % D C 123 92,7 % D C 130 1/2008

Opti-Solar’s GT 4000 (new software, V2.07) 250 - 400 V 92,1 % D – 126 94,6 % C – 107 6/2011

Powercom’s SLK-4000 205 - 408 V 92,0 % D C 127 93,4 % D C 122 11/2010

Phoenixtec's PVG 10000 320 - 720 V 91,8 % F D 128 93,3 % D C 125 6/2010

Riello's HP 4065REL-D*3,8 255 - 435 V 91,7 % F D 129 93,9 % C B 118 9/2007

Effekta's ES5000 (old software) 150 - 450 V 91,7 % F – 129 92,2 % D – 131 2/2012

Fronius's IG 30 150 - 397 V 91,4 % F D 131 92,2 % D C 131 1/2007

Powercom’s SLK-4000 (new software, V2.07) 250 - 450 V 91,1 % F – 132 93,4 % D – 122 6/2011

Siemens' Sitop solar 1100 Master*3 200 - 552 V 90,2 % F D 133 91,7 % F D 134 5/2007

Danfoss's ULX 1800 HV IN*3 260 - 500 V 89,2 % F F 134 91,3 % F D 136 4/2010

SMA's SB1100*3 139 - 320 V 89,1 % F F 135 90,5 % F D 138 10/2009

Opti-Solar’s GT 4000 (old software, V1.09) 200 - 450 V 87,8 % F – 136 92,1 % D – 133 6/2011

Ehe New Energy's EHE-N2K5 200 - 400 V 87,4 % F – 137 91,4 % F – 135 7/2011

SunnySwiss' SSP-6000 250 - 480 V 86,8 % F – 138 91,2 % F – 137 2/2011

Ehe New Energy's EHE-N5K 300 - 650 V 80,3 % F – 139 86,3 % F – 139 7/2011

Phoenixtec's PVG 2800 (original model)*3 255 - 435 V 78,4 % F F 140 85,8 % F F 140 2/2008

*1 range at which the model was tested and to which the grade applies

*2 Eversolar New Energy Co. Ltd. and Zof New Energy Co. Ltd. merged at the end of 2011 and altered their name to Zeversolar New Energy Co. Ltd.; Zeversolar now calls the device the Eversol

TL 17k; however, the power data differs from the tested Eversol-T

*3 device no longer being produced

*4 renamed Solarmax 13MT3 since April 2012

*5 name changed from Eltek Valere to Eltek

*6 now Schneider Electric Industries SA

*7 prototype; device no longer being produced

*8 the identical solar inverter brands Helios Power (Riello UPS) and Sirio (Aros) are now marketed under a single brand, Aros Solar Technology GmbH, and distributed by Aros Neufahrn

*9 inverters that have already been tested by PHOTON Lab but for which results have not yet been published in the magazine

New grades in PHOTON Lab’s inverter test as of 2011The table showing the results achieved by

the inverters tested by PHOTON Lab looks

slightly different due to a new grading sys-

tem as of 2011. All of the inverters tested

before that point have two grades: one

based on the old system and one related to

the new method.

The grades are based on the PHOTON

efficiency at medium and high irradiation.

More detailed information about the invert-

ers can be found in the corresponding test

reports (the issue in which each report was

published is noted in the last column of the

table). The rankings are also based on the

PHOTON efficiency.

The changes to the grading system were

made to reflect the current status in the sector;

and the system will be updated again in the fu-

ture to reflect technical advancements. Now, in-

verters have to get a higher PHOTON efficiency

to secure a better grade: what would have

gotten an A in 2010 with 96.4 percent, would

now get a B. Should manufacturers further im-

prove their devices, these inverters could even

get downgraded to a C as our grading system

changes to reflect the current times.

Moreover, to align grades with our US

sister publication, we have changed the

letter »E« to »F,« based on the American

school grading system.

Grading system for inverter tests as of 2011A++ A+ A B C D F*1

PHOTON efficiency ≥ 99 ≥ 98 - < 99 ≥ 96.5 - < 98 ≥ 95 - < 96.5 ≥ 93.5 - < 95 ≥ 92 - < 93.5 < 92

Deviation from next grade 0.5 1 1.5 1.5 1.5 1.5 –

*1 to align grades with our US sister publication, we have changed the letter »E« to »F«

science & technology | inverters | test