logo elektroda
logo elektroda
X
logo elektroda
Advanced Search
logo elektroda
Build faster with M5Stack »
Build faster with M5Stack » Stick • Unit • Cardputer
ADVERTISEMENT
Dostępna jest polska wersja

Czy wolisz polską wersję strony elektroda?

Nie, dziękuję Przekieruj mnie tam

50W COB 30V-36V LED minitest from China - will it really be 50W?

p.kaczmarek2 26622 39

TL;DR

  • A cheap Chinese 50W COB LED module bought on eBay was tested to see whether its 30V-36V, 1500mA, 4700LM claims were real.
  • The module was powered from a TP-1603 laboratory supply as a constant-current source, with voltage checked by a UT210E clamp meter and a BM857s multimeter.
  • At 1.5A and 34.4V, the LED drew 51.6W; at 2A and 36.2V, it reached 72.4W.
  • The module really does hit about 50W, but most of that power becomes heat, and the small heat sink only worked well up to 500mA, under 15W.
ADVERTISEMENT
Treść została przetłumaczona polish » english Zobacz oryginalną wersję tematu
Report a violation of the law
Reply Cool? Ranking DIY | New topic
Notify about new articles
📢 Listen (AI):
  • #31 18942878
    metalMANiu
    Level 21  
    The "amount" of light is expressed in lumens and is determined in the lumen meter. A luxmeter is rather useful for measuring the illumination of work surfaces. And the Made in China lux meter for PLN 30, I don't know if it can be useful ... Although not - it can measure these Chinese lumens ?
  • ADVERTISEMENT
  • #32 18942948
    pawlik118
    Level 33  
    lumen determines the total amount of light generated by a given light source - the measurement is made inside the ulbricht sphere. Lux is the light intensity - its quantity in relation to a given unit area. In practice, it is rather impossible to determine the efficiency of the LED diode as lm / W at home without hardware. But as I wrote earlier, if these Chinese LEDs do not have lumens, the quality of this light is still poor due to the low CRI. It's such a thing that on the one hand it hurts the eyes and on the other hand it does not illuminate the area well.
  • ADVERTISEMENT
  • #33 18944038
    zgierzman
    Level 31  
    metalMANiu wrote:
    And the Made in China lux meter for PLN 30, I don't know if it can be useful ...


    Of course it might be useful.
    Not for an accurate absolute measurement, because you would have to calibrate it with something first, but it can be as a comparative measurement device. Many measurements are made not to know the absolute numerical value, but to compare the reading with something.

    For example: I have a table lit by a 4 x 18 W fluorescent lamp. The Chinese lux meter shows X. I change the fluorescent lamps to LED and the lux meter shows Y. Then you know how the light intensity has changed. A subjective impression is also important, if only because of the CRI mentioned here, but thanks to the measurements you can form an opinion whether such a change made sense.
  • ADVERTISEMENT
  • #35 18950813
    Ibuprom
    Level 26  
    I used these 10W LEDs to illuminate the sidewalk outside the house. They lasted 4-6 months at half the rated power. I replaced 4 of them in one lamp. 3W led bulb with a 12V lid, gouged out of the original housing and inserted into the "halogen", turned out to be many times more durable; is over 15 months old and still shines flawlessly. I did not observe the difference in brightness. All COBs are put on thermo paste and additionally screwed to the housing. Such a Chinese COB 10W cost 2-4 PLN, in lidl I paid 5.99 where the set includes a current source to power this LED. I suspect that in the case of the COB from this article, the durability will be similar. How many people have seen flashing "halogen" LEDs where the problem was the damaged COB LED.
  • #36 18950880
    keseszel
    Level 26  
    CRI - is it the color of light? This is the first time I hear about anything like this. Until I checked on google.
    Generally, I once bought 25W cheap for ridiculous money. For 220 V. I wanted to convert to 12 V for the car. For now, I gave up.
  • ADVERTISEMENT
  • #37 18951606
    CMS
    VIP Meritorious for electroda.pl  
    keseszel wrote:
    CRI - is it the color of light?
    Not so much the color of the light as the colors of the illuminated objects.
  • #38 18951660
    E8600
    Level 41  
    In general, with light, it is not always brighter is better. Cheap LEDs have low photometric efficiency, which explains why they are cheap (they emit a lot of heat and a lot of UV). There are tests on YT and, for example, expensive Power LED 1 W can give as much "light" as cheap Chinese 3 W.

    As for the measurement of light intensity, on most phones you can install the application, it is not a laboratory measurement, but by specifying the distance of the light source, it is always a reference point.
  • #39 18954166
    keseszel
    Level 26  
    Most cheap floodlights give white light that is unacceptable to me.
  • #40 18957432
    oldking
    Level 33  
    Without checking for at least 10 hours, the entire test adds nothing. We only know that it shines for a moment and that's it.
Reply Cool? Ranking DIY | New topic
Notify about new articles
📢 Listen (AI):
Report a violation of the law

Topic summary

✨ The discussion revolves around the performance and reliability of a 50W COB LED module purchased from a Chinese seller. Users share their experiences with similar modules, highlighting issues such as overheating, inadequate heat dissipation, and the importance of proper cooling solutions. Many participants express skepticism about the actual output power and longevity of these inexpensive modules, suggesting that they often underperform compared to their specifications. Concerns about the color rendering index (CRI) and luminous efficacy are also raised, with some users noting that cheaper LEDs emit more heat and less light. The consensus indicates that while the modules can function at 50W, they may not sustain this output effectively over time without significant cooling measures.
Today's popular

FAQ

TL;DR: Lab tests show the €2 “50 W” COB draws 51.6 W at 1.5 A [Elektroda, p.kaczmarek2, post #18937205]; “it heats up very much” [Elektroda, p.kaczmarek2, post #18937694] Proper cooling, not voltage, decides service life.

Why it matters: Overshooting temperature cuts LED life more than 50 %.

Quick Facts

• Forward voltage: 30–36 V DC at 1.5 A [Elektroda, p.kaczmarek2, post #18937205] • Claimed luminous flux: 4 700 lm (≈ 94 lm/W) [Elektroda, Joker., post #18939454] • Measured power: 0.25 W–72 W in test range [Elektroda, p.kaczmarek2, post #18937205] • Safe passive cooling: ≤ 15 W on 52 × 40 mm sink [Elektroda, p.kaczmarek2, post #18937205] • Typical lifetime in cheap floodlights: 5 475 h before failure [Elektroda, teskot, post #18939759]

How bright is it in lumens?

Seller quotes 4 700 lm, equal to ~94 lm/W efficacy [Elektroda, Joker., post #18939454] Independent lux or integrating-sphere data were not posted, so real light may vary ±30 %.

What current and voltage should I use?

Run it from a constant-current driver: 1.5 A nominal, 30–36 V forward range. Start at 500 mA for testing to cut heat by 70 % [Elektroda, p.kaczmarek2, post #18937205]

How big should the cooler be?

Field users needed sinks six times larger for fanless operation at full power [Elektroda, Tomek515, post #18937617] A PC CPU cooler plus fan handles 50 W with case temps below 60 °C.

What is CRI and why care?

CRI gauges colour fidelity. These budget chips sit around 60, causing washed-out colours and glare [Elektroda, pawlik118, post #18938359] High-CRI (>90) alternatives improve vision and comfort.

Can I build a handheld flashlight with it?

Possible, but energy and heat constraints bite. Two 18650 cells give ~60 min at 30 W while the head overheats quickly [Elektroda, PPK, post #18940963] Add thermal throttling or derate below 20 W.

Could it replace a projector lamp?

Unlikely. Projectors need 20 000–30 000 lm and near-perfect CRI; the COB lacks both and focusing a 52 mm light patch onto a 5 mm DLP window is impractical [Elektroda, CMS, post #18940663]

How do I drive it safely from mains?

Use an isolated 36 V constant-current supply, not a voltage source. Over-voltage spikes or thermal runaway can destroy the array in seconds [Elektroda, p.kaczmarek2, post #18937205]

Three-step home test for luminous output?

  1. Mount LED on final heatsink and power at target current.
  2. Place phone light-meter app or luxmeter 1 m away, perpendicular.
  3. Record lux; multiply by 12.57 to estimate lumens for a bare emitter pattern. “Comparative readings beat guessing.” [Elektroda, zgierzman, post #18944038]

What edge-case failures should I expect?

LED dies from single overheated bond wire; cheap silicone thermal paste worsens it. One user lost 2 of 3 modules despite 30 × 50 cm sink and light duty [Elektroda, zgierzman, post #18937984]
ADVERTISEMENT