(HELP!) CXB 3590 36V @ 2.8A CB/CD - Need some data to calculate...

Vescovi · Jul 27, 2015

Hi guys, can you help me?!

Here is the deal, I have two situations I would like to calculate PPFD, but i don't know how to calculate the exact efficiency of this COB @ 2.8A. I know how to calculate PPFD, you guys thought me, but I can't seem to find correct efficiency @2.8 for these COBs (CB/CD).
I know from @SupraSPL spreadsheet it will dissipate something like 103W per COB.

One lamp is 3X cxb3590 36V CD 3500K, each COB uses one driver 36V 2.8A 100W+- 86% efficiency, dimmable. Area (3x2 feet).

The other is 6x cxb3590 36V CB 3500K, each COB uses one driver 36V 2.8A 100W+- 86% efficiency, dimmable. Area (3.25x3.25 feet).

Do you think it will get too hot? The heatsink I will use is 50cm SLAB of this:

Dimensions in mm. With 120mm 12v fan. The flat base has 6mm solid aluminium.

The driver is:

ModelWSCD-100-2A80W
OutputDC output30-36V
Current range2.8A
Rated power100W
Ripple&Noise±25% (sure this is a typing error maybe is 2.5%)
Current regulation±3%
Starting time3000mS/230Vac, 5000mS/115Vac
InputVoltage range90Vac-264Vac
Frequency range47Hz-63Hz
Power factorPF>0.95/230Vac,PF>0.98/115Vac
Efficiency85%typ @100Vac; 87%typ @220Vac
Inrush currentAt cold start and 25 deg C,
No damage should occurred and input fuse should not blow-up
Leakage current0.25mAmax @ 230Vac
ProtectionShort circuitOutput short circuit protection
Protection:Auto recovery after the fault removed
EnvironmentWorking temperature&humidity-20°C to +50°C;20%-90%RH,Without condensation
Storage temperature&humidity-30°C to +80°C
Vibration resist10-500Hz,2G ,X/Y/ZAXIS 60 mintues
Safty&EMCSafetyUL8750,GB4943
HI-POTI/P-O/P: 1K5Vac
Insulation resistanceI/P-O/P: 100M Ohms/500Vdc
EMIEN55015, GB17743 CLASSB
Harmonic currentIEC61000-3-2 Class C compliant
Connection terminalsAC input terminalsSilicon rubber cable-Φ7.0mm / Brown L,Blue N, Green G/1mm2
DC outputV:Brown V:Blue
OthersDimension170*80*40mm(L*W*H)
Net weight1.0kg

Thanks guys!

Vescovi · Jul 27, 2015

Is this 43.1% or 2.8A a dumb way to run this?
The dollar is hitting the ceiling in my country and import taxes here are around 60%, so I am trying to make the most with fewer cobs. Try to understand it...

UKpeanuts · Jul 27, 2015

It might be worth considering HPS if they are produced in your country etc.. avoid import tax

salmonetin · Jul 27, 2015

...yes 43,14 % @ 2080 mA...

3 cxb3590 @ 2080 mA ...In a space with 3 x 2 feet...

3 x 2 feet = 6 sqft
Vf @ 50ºC @ 2080 mA = 37,12 V x If @ 2080 = 103,94 W x 3 (cobs) = 311,82 Dissipation W
312 Dissipation W * 43,14 % efficiency (for 3500ºK at 2080 mA) = 134,6 PAR W
135 PAR W * .9 reflector / lens / wall losses = 121,5 PAR W
121,5 PAR W / 6 (3 x 2 feet) = 20,25 PAR W / ft²
20.3 PAR W / ft² * 4.65 umol/s * 10.7 (10.764 sqft = ~1 m^2) = 1.007,85 .... 1.008 umols / m2 * sec ...or PPFD

...In a 3 x 2 Feet.... yes...

...312 dissipation w - 135 par w = 177 Heat W...

....but today check my math... im too :joint:

....

...for the rest i prefer listen other opinions ... wilson ...back to chozo...

:peace:

Saludos

Vescovi · Jul 27, 2015

salmonetin said:
...yes 43,14 % @ 2080 mA...

3 cxb3590 @ 2080 mA ...In a space with 3 x 2 feet...

3 x 2 feet = 6 sqft
Vf @ 50ºC @ 2080 mA = 37,12 V x If @ 2080 = 103,94 W x 3 (cobs) = 311,82 Dissipation W
312 Dissipation W * 43,14 % efficiency (for 3500ºK at 2080 mA) = 134,6 PAR W
135 PAR W * .9 reflector / lens / wall losses = 121,5 PAR W
121,5 PAR W / 6 (3 x 2 feet) = 20,25 PAR W / ft²
20.3 PAR W / ft² * 4.65 umol/s * 10.7 (10.764 sqft = ~1 m^2) = 1.007,85 .... 1.008 umols / m2 * sec ...or PPFD

...In a 3 x 2 Feet.... yes...

...312 dissipation w - 135 par w = 177 Heat W...

....but today check my math... im too ....

...for the rest i prefer other listen other opinions ... wilson ...back to chozo...

Saludos

Thanks ALOT man. But I don't understand one thing... @SupraSPL spreadsheets say the 72V CXB3590 is 45.59 efficient.
You and Supra calculated my last PPFD using 49.7 eff. Thats where i got lost brow.

4x cxb 3590 @1050mA
3X2 feet 6sqft
75.5W X 4 = 302 Dissipation W
302 * 49.7% efficiency = 150 PAR W
150 * .9 reflector/lens/wall losses = 135 PAR W
135/6 = 22.5 PAR W/ft²
22.5 * 4.65 umol/s * 10.7 = 1119 PPFD In a 3x2 Feet.

salmonetin · Jul 27, 2015

...maybe i and supra missunderstud your 3590 elected (comes on two version)...
...but for 72 v cxb 3590 @ 1050 mA is 45,59 % ...yes...

...sorry... sometimes people dont said v version or kelvin... and errors ocurrs too...

...but now you catch the idea ok... ...sorry ...back to chozo...

:peace:

Saludos

Vescovi · Jul 27, 2015

salmonetin said:
...maybe i and supra missunderstud your 3590 elected (comes on two version)... but for 72 v cxb 3590 is 45,59 % ...yes...

...sorry... sometimes people dont said v version... and errors ocurrs too...

...but now you catch the idea ok... ...sorry ...back to chozo...

Saludos

Yes, now I get it... I thought you guys had some freakin' secret formula to get to that eff. number.

Thanks man!!

salmonetin · Jul 27, 2015

...no secret formula... its hard way... ...data extrapolated from graphic on datasheet...

...or maybe the cree PCT tool...

Saludos

Vescovi · Jul 27, 2015

Now I have my doubts about this driver... Mean Well in this config only without dimmer.
In the summer I am fu**ed without dimmer.

Vescovi · Jul 27, 2015

UKpeanuts said:
It might be worth considering HPS if they are produced in your country etc.. avoid import tax

Dude...

salmonetin · Jul 27, 2015

...top bin leds needs good driver/s with high efficiency too... 90 % or better with pf corrected... for my pov

...i know Wilson... i Know... back to chozo...

:peace:

Saludos

SupraSPL · Jul 27, 2015

Vescovi said:
Thanks ALOT man. But I don't understand one thing... @SupraSPL spreadsheets say the 72V CXB3590 is 45.59 efficient.
You and Supra calculated my last PPFD using 49.7 eff. Thats where i got lost brow.

Yep 49.7% is for the 3500K CD bin.

alesh · Jul 28, 2015

SupraSPL said:
Yep 49.7% is for the 3500K CD bin.

That's not at 2.8A, is it?

I'd guess that CD bin will be <40% at 2.8A as it seems impossible to keep temperature at 50°C.

SupraSPL · Jul 28, 2015

alesh said:
That's not at 2.8A, is it?
I'd guess that CD bin will be <40% at 2.8A as it seems impossible to keep temperature at 50°C.

The CXB3590 3500K CD 72V @ 1.05A (75.54W) I have it at 49.7%. Temp droop measures 3.5-4.5% with minimal cooling. I think 2.5-3% would be possible on a CPU cooler, maybe will get around to testing that soon.

The CXB3590 3500K CD 36V @ 2.8A (104W) I have at 43.14% (Tj 60C) and I agree Tj 50C is probably not possible with ambient temps of 25C. I will try to test that as well, should be able to get data up to 3.2A.

3X CXB3590 36V CD 3500K 36V @ 2.8A in a 3'x2'
104W X 3 = 312 Dissipation W
312 * 43.14 efficiency = 134.6 PAR W
134.6 * .9 reflector/lens/wall losses = 121.14 PAR W
121.14/6 = 20.19 PAR W/ft²
20.19 * 4.65 umol/s * 10.7 = 1004.5 PPFD

alesh · Jul 28, 2015

SupraSPL said:
The CXB3590 3500K CD 72V @ 1.05A (75.54W) I have it at 49.7%. Temp droop measures 3.5-4.5% with minimal cooling. I think 2.5-3% would be possible on a CPU cooler, maybe will get around to testing that soon.

The CXB3590 3500K CD 36V @ 2.8A (104W) I have at 43.14% (Tj 60C) and I agree Tj 50C is probably not possible with ambient temps of 25C. I will try to test that as well, should be able to get data up to 3.2A.

3X CXB3590 36V CD 3500K 36V @ 2.8A in a 3'x2'
104W X 3 = 312 Dissipation W
312 * 43.14 efficiency = 134.6 PAR W
134.6 * .9 reflector/lens/wall losses = 121.14 PAR W
121.14/6 = 20.19 PAR W/ft²
20.19 * 4.65 umol/s * 10.7 = 1004.5 PPFD

I still think we underestimate Tj. I feel a mistake to be there but I don't where. I know we've already discussed this and I haven't been able to think of something new since.
...

From the CXB3590 spec sheet:
- they're binned at Tj = 85°C

- from the relative flux chart, we can check that Tc 55°C = Tj 85°C at binning current

- at binning current (2400mA ~ 87W), CXB3590 are about 45% efficient, producing ~48W of heat
- from that we know that 48W of heat makes 30°C difference between Tc and Tj
- simple math: Rth = ΔT/P = 30°C/48W = 0.625°C/W
- knowing the thermal resistance between Tc and Tj, we can conclude that Tj of CXB3590 at 2.8A will never be lesser than 60-65°C in 25°C ambient, even with a perfect heat sink. Since perfect heat sinks don't exist I'd say that 75°C is a realistic figure.

SupraSPL · Jul 28, 2015

I agree there would be a minimum Tj that corresponds with a given dissipation power because it depends on thermal resistance of the package. The trick is how to test this in real world usage. We can monitor the instantaneous Vf drop output drop when applying that much power to a COB mounted on a large 25C heatsink, but Cree does not provide that voltage change coefficient, possibly because the rate is variable depending on current.

In this case Cree is suggesting that CXB3590 36V @ 2.4A, warming up from Tj25C to Tc55/Tj85C results in a temp droop of 9%. So if Tj is 65C (Tc35C) we could expect temp droop of 6%. If Tj is 75C (Tc45C) temp droop should be 7.5%.

My CXB3590 spreadsheet assumes Tj 60C so it is applying a temp droop of ~5.25%. Maybe that is a bit generous but not by much. You know me I am normally focused on how close to the 25C lumen figure we can get by running at very low currents

Will be very interesting to see the test results on a good CPU cooler though.

Vescovi · Jul 28, 2015

Got a slightly better driver for U$ 27,5:

 Self-adapting Output Voltage, and range is large enough to match different LED load

 High power factor, Low harmonic current

 High Efficiency (up to 90%)

 Protections: Short circuit, Over Current, Over Voltage, Over Temperature

 Compliance to the testing requirement of double 85

 4 in 1 dimming function (0~10Vdc , PWM signal, resistance, timing controlled dimming

function, Infrared remote control)

 Max 5 stages programmable in timing controlled dimming function

 Cooling by free air convection

 IP67 design for indoor or outdoor installations

 Suitable for LED lighting and street lighting application

 5 years warranty

WSCA-105-020VN-AXXXX-YY 105 is the maximum output power; 020 is the minimum output operating voltage; XXXX range from 0580 to 2910, it defines the initial output current; YY is

the internal identification code

WSCA-105-060VN-AXXXX-YY 105 is the maximum output power; 060 is the minimum output operating voltage; XXXX range from 0280 to 1400, it defines the initial output current; YY is

the internal identification code

WSCA-105-120VN-AXXXX-YY 105 is the maximum output power; 120 is the minimum output operating voltage; XXXX range from 0140 to 0700, it defines the initial output current; YY is

the internal identification code

SPECIFICATION

Model WSCA-105-020VN WSCA-105-060VN WSCA-105-120VN

Output Current Range 0.58A~2.91A 0.28A~1.4A 0.14A~0.7A

Output Voltage 20~54Vdc, 60~120Vdc 120V~270V

Rated Power 105W 105W 105W

Ripple Current (max.) 1100mA 1100mA 1100mA

Output current regulation ±3% ±3% ±3%

Turn-on Delay time 3Smax.@220Vac input & Full load, 5Smax.@100Vac input & Full load.

OUTPUT

Turn-on Rise time 300mSmax.@ Full load

Voltage ramge 90V~305VAC 90V~305VAC 90V~305VAC

Frequency Range 47Hz~63Hz 47Hz~63Hz 47Hz~63Hz

Power Factor(Typ.) PF>0.99 at 115VAC PF>0.99 at 115VAC PF>0.99 at 115VAC

Efficiency (Typ.) 90% 90% 90%

AC Current 1.5Amax@100Vac-277Vac & Full Load

Inrush Current 50A max @ 230Vac input 50A max @ 230Vac input 50A max @ 230Vac input

INPUT

Standby input power (max.) 3Wmax.@ Nominal input 3Wmax.@ Nominal input 3Wmax.@ Nominal input

Over Voltage <60Vdc <132Vdc <300Vdc

When the output voltage is over the limitation, the product will shut down output, it can recovery when the fault condition is removed.

Short Circuit The input power shall decrease when the output rail is shorted, the power supply shall have no damage, and shall recovery when the fault

PROTECTION condition is removed.

Over Temperature When Tc > 85℃, the output current will be decreased to protect the LED driver. When the temperature of the case go down below 85℃, the

product will self-recovery. The minimum output current will be limited to 30% (typ.) of the rated output current in OTP function. The LED

driver could survive in 125℃ for 2hrs.

Operating Temp. -40~60℃ -40~60℃ -40~60℃

Operating Humidity 0～95% RH 0～95% RH 0～95% RH

Storage Temp -40~85C -40~85C -40~85C

Water proof IP67 IP67 IP67

ENVIORNMENT

Vibration The LED power supply can survive vibration towards three mutually perpendicular direction (X, Y, Z), each direction for 72 minutes. The

vibration is in accordance with the sine wave with 2mm amplitude, and its frequency range from 10Hz to 500Hz with 5G acceleration

MTBF >200Khours @ 25℃

RELIABILITY

E-CAP LIFETIME The life time shall be at least 5 years at 40℃, Full load and nominal input condition

SAFETY STANDARDS UL8750, EN61347-1/A2:2013, EN61347-2-13：2006, IEC61347-1, IEC61347-2-13, EN62493：2010， GB19510.1-2009, GB19510.14-2009,

WITHSTAND VOLTAGE I/P-O/P: 3750VAC I/P-FG: 1650VAC O/P-FG: 500VAC

Leakage Current 0.75mA @ 240Vac 50Hz input

ISOLATION RESISTANCE 50MΩ min. at primary to secondary with 500Vdc test voltage

EMC EMISSION Compliance to EN55015 (CISPR15), EN61000-3-2; EN61000-3-3, GB17743, GB17625.1

SAFETY &EMC

EMC IMMUNITY Compliance to EN61547; EN55042, EN61000-4-2,3,4,5,6,8,11, GB/T18595, GB17626

DIMENSION 170*80*40mm

WIGHT 920±50g

COLOR BLACK BLACK BLACK

MECHANICAL

PACKING 10pcs/10Kg 10pcs/10Kg 10pcs/10Kg

OTHERS / / / /

NOTE 1. Ripple current are measured at full bandwidth of the oscillator

2. The power supply will be operated in combination with final equipment. Since EMC performance

(HELP!) CXB 3590 36V @ 2.8A CB/CD - Need some data to calculate...

Vescovi

Active Member

Vescovi

Active Member

UKpeanuts

Well-Known Member

salmonetin

Well-Known Member

Vescovi

Active Member

salmonetin

Well-Known Member

Vescovi

Active Member

salmonetin

Well-Known Member

Vescovi

Active Member

Vescovi

Active Member

salmonetin

Well-Known Member

SupraSPL

Well-Known Member

alesh

Well-Known Member

SupraSPL

Well-Known Member

alesh

Well-Known Member

SupraSPL

Well-Known Member

Vescovi

Active Member