The warm white light of excellent, reportedly only 2900 Kelvin does not seem exaggerated to me. I can actually see no difference in color to halogen light. This LED lamp is not conceived with their very high for this size power as cheap Far East goods: The 12 SMD LED's are due to the diffusion voltage (approx 3,1-3,6V) also a 3 LED grouped in 4 strands. All strands are supplied together but with only a single common series resistance (0.130 ohms 125 mW) analog. This very small and weak fixed resistor is not directly to the current limit, but only indirectly: Its voltage drop is directly proportional to the total load current is measured. One diode produces an almost constant second reference voltage. The IC amplifies the difference, thus controls the total load current to an inflated, but temperature-compensated stable value and overloaded so easy the LED's. On the back there is a bulky Elko with coil and a small SMD IC. With these components, a clock with a constant pulse duration and frame rate (about 8US = 125kHz = long wave) is obviously produced, the pulse-width modulates the excessive stabilized power (PWM) passes during the pulses by the LED's. During the subsequent periods they can cool off again. The relationship between pulse and pause (measured: approx Impuls2us + Pause6us = Rahmen8us) is automatically controlled so that, according to my measurements at each copy a constant voltage and temperature compensated RMS current of 0.22 ~ 0,23A arises. With this concept you move the operating point of the loss-making center in the two analog low-loss digital edges of the voltage spectrum, thus achieving minimal adjustment losses as well as equal load for a maximum service life. On the FM band, I could not identify any interference from the PWM. The impact on long wave I could not examine and should in particular be taken into consideration by radio amateurs. Actual Secondary rms power is therefore 12.0V x (0.22 ~ 0,23A) = 2.64 ~ 2,76W. The size indicated on the package with 2.5W rated power is converted only when a transformer undervoltage of 10.87 ~ 11,36V. On the energy label a power consumption of 2,88kWh / 1000h is specified, which corresponds to a converted power from 2,88W. I assume that this 2.5W LED power + 15% (= 0,375W) transformer losses = 2,875W total power meant. As actual primary energy consumption are according to my measurement 2.64 ~ 2,76kWh / 1000h + 15% = 3,036 ~ 3,174kWh / expected 1000h. The maximum power of a halogen or LED secondary transformers is limited not only by its cooling ability, but also by the iron core cross-section as well as the response threshold of the electronic short-circuit protection and therefore can not be covered for a short time. Even more secondary transformer minimum power than the corrected actual effective power thus require pulse width modulated light sources. If one lamp per transformer of the plus-factor is the ratio of pulse to pulse frame duration (2,76W / 2us * 8US = 11,04W). After Shannon, the spikes are distributed from a lamp number of 2 * (frame duration / pulse duration) within the shortest possible delay virtually uniformly over the entire pulse frame, so that no PWM power surcharge is no longer required then (here 2 * 8US / 2us = 8Lampen / transformer) , With 2 lamps per transformer and at least 50% duty cycle twice the minimum power is needed, otherwise (as here) not. In lamp numbers between these limiting cases the transformer minimum benefits evenly after the exponential function. Too tight dimensioning of the transformer secondary output of these controls back so inaccurate reviews as LED bulb is just as bright as 10W Halogen "or LED transformer does not really 50W" arise. The laying of the secondary line at the AC output of an electronic halogen transformers in the ring "leading to this result, is there better not be done and is recommended only for DC voltage. The part upgrade bulbs to halogen transformer also results in a distorted brightness comparison. This LED lamp is on Correct Port actually as indicated at least as bright as a 20W halogen lamp. The calculated minimum power transformer is then rounded up to the next size of the rod. Although even more dimensioning increases the power redundancy and thus prolongs the life, but also reduces slightly the efficiency. At a furniture transformer continues to consider that this fits into the intended niche. Transformers in cavity ceilings are fit to maintain the revision capability at a given installation depth through the next lights cut. This is sometimes very scarce, especially in small spots to D60mm. On introduction by a wooden ceiling silhouette should be avoided, since a replacement transformer quickly will no longer fit there and then sometimes extensive rewiring would be announced. The cooling plate of the light source ranks with their enormous diameter of 30mm hard at the limit of the fit. Your cut is not as usual perforated and broken, but all around clean punched circular to waste any further valuable space. The assembly quality, however, is as bumpy as you know from other Far Eastern goods. As long as the product works, I do not, however, rather than a higher price. The terminal pins do not come off the shelf (5.08mm), but actually have a pin spacing of only 4 mm and must not be bent conical. You are from outside edge plastic web 9.0mm long and had me with a calliper and electronics side cutters on 7,0mm be shortened. Due to their still erbärmlicheren assembly quality (lateral offset, lateral and vertical skew, and also extreme Hochwasseraxialversatz 0.0 ~ 2.0mm) I had all desoldering, strip and aligned flush and re-equip, since the lights would not otherwise fit into the luminaires. The latter is absolutely no longer be tolerated and will be punished by me with a deduction of 2 stars. Obviously Almighty large LED leader sebson is unable to bring its suppliers in a subordinate clause suggests that this mess has to be another. At the end of all the lamps fitted straight times so narrowly eccentrically in the cabinet luminaires my kitchen wall cabinets manufacturer Hera ". With closure of the glass pane everything was compressed to press fit (additionally promotes cooling by heat dissipation of the glass pane). Alternatively, you could here remove the luminaire reflector. The then in However, the lamps shared view not everyone's true flavor. In the mirror cabinet lighting manufacturer L + S CALDEN "LED lamps did not fit. Here, the reflectors can not be easily removed, since they also include the engaging holes of the glass cover support ring. After removing the plastic ridge on the lamp base, the lamps were used. The glass lid supporting rings stood at the mirror end from approximately 0.5mm normal. If you do not jump in the glass wants, one should resist especially when brushing too much pressure is applied. A modified layout in small Platinenabflachung for the base and a tiny platinum recess for the plastic web of the connector pins would alleviate the problem. The lamp would also move toward the center of the light, so that despite still larger outer diameter to fit would be assured. The seller sebson one of the few Amazon who hereby proved relatively consistent specify all characteristics specified by the manufacturer in the product description and main product in the product name. The latter is required for a display already in the list view and thus a prerequisite for a time-saving customers looking for the optimal product. Too bad that the actual lamp power and the power consumption are slightly higher and the minimum transformer services for this PWM bulbs are also not given. The product is both cost-saving 10-pack, and available separately. The latter is important to avoid being forced to excess orders and to be able with its attractive price-performance ratio in the amateurish run ordering portal Amazon ever recognize this product as such (no sorting function by price per unit without shipping "). The (Compliance) Delivery time 2-7 days seems pleasantly short, is for my expectations, however, the lowest rank. In summary, this product is an outstanding engineering feat that deserves my sincere respect. Unfortunately, this power is clearly destroyed the inexpedient layout and especially on the production incompetent execution. Since I do not know any better G4 lamp, I can still recommend any electronics, which is equipped and willing besides price includes an additional hour per 10 lamps for reequipping, reduction of the base and, if necessary, to invest distance of the base web. Paradoxical as it may sound: For all other customers of this product can unfortunately represent only nervenzermürbender electronic waste. However, I know of no alternative product without others even greater hook. This indictment of our economy is reminiscent of the situation in the GDR. A reasonably acceptable, but even worse alternative product is B00DINK9MK ASIN. This has a much smaller, non-specified power with enormous manufacturing tolerances, only lossy, not temperature-compensated analog current limit and an even worse production quality, but it fits with its much smaller design problems in almost any G4 Light, is only half as bright and costs only the half.
