Four different LED GU10s - showing the variation in colours available

Nowadays, it’s nigh-on impossible to buy a luminaire from the DIY sheds or a high-street retailer which doesn’t have a GU10 cap. 12v lamps seem to have had their day as far as the mass retailers are concerned.

It isn’t difficult to match the light output of a 20W or 35W halogen GU10, so we have only tested LED lamps that claim equivalence to a 50W.  

The European DIM2 regulation says that to make this claim, the LED GU10 lamp must emit more than 345 lm in a 90-degree cone. The total lumen output, including light that falls outside this cone, may be considerably higher. Some of the lamps tested emit over 500 lm so they would clearly be seen as an improvement if you retrofitted them in an existing installation of halogen GU10 (with the same beam width). 

Our understanding is that all products currently on the market must conform to DIM2. But we regularly hear of lamps that don’t” 

We’ve asked several experts, and our understanding is that all products currently on the market must conform to DIM2. But we regularly hear of lamps that don’t. 

The problem for Joe Public is that some lamps don’t state on the box whether the lumen output quoted is the total emitted or within the 90-degree cone. Some quote both figures. Reputable manufacturers use phrases like ‘usable lumens’ or ’90-degree cone’ so you know what you are buying.  However, until all suppliers mark their packaging clearly, it is impossible for the purchaser to make fair comparisons between the different products. No wonder people buy the cheapest or the one quoting the biggest lumen value. 

The situation is made worse because there is no effective policing of the market. This, in effect, benefits the poor quality, low-cost suppliers to the detriment of the manufacturers of good-quality lamps.

One last point to make is that a 90-degree cone isn’t really what you would call a spotlight. For the non-trigonometrists amongst you, that’s a two-metre wide illuminated patch from a lamp one metre away. 

If you are as confused as I am by all the test standards that cover LEDs, there is a useful summary on the LIA website. Look for Technical Statement TS01. It’s nine pages long and covers existing and proposed European and US regulations and guidance.

None of the 11 lamps we’ve tested here had a particularly high power factor. The best was Osram at 0.88 and the poorest were the Aurora and Bell at 0.52. Why does this matter? Because it means that you are drawing more current than might be apparent from just looking at the wattage. An electrician complained to me about this because he had to resize all his fuses in a (large) domestic house purely because of equipment with a poor power factor.

Before you buy the lamps, you should check what the beam looks like. Some have coloured edges or don’t have a smooth gradation of light from the centre to the edge. The three on the left of the main photo for this article are all rated at 2700K and the one on the right is rated at 3000K.


Summary of results
We were pleased to see that, generally, the performance was as claimed. We measured the total light output. 

Note that we tested just one lamp from each supplier. There are always tolerances in manufacturing and so the lamp you buy may not perform exactly as the one we tested. However, these lamps are made by the million; it would be surprising if they varied that much.

Our testing was done at the independent LIA Laboratories in Telford.
Thanks to LIA Labs for their help!


Aurora Enlite 3000K, 5W

The Enlite is a brand new range of lamps. What really sets it apart from the other lamps we tested is the efficacy of 96 lm/W – almost 500 lm from 5W. It also has a 60-degree beam, wider than the others tested, which makes it much more useful for lighting areas where you want a general spread of light.

  • Output 493 lm
  • Power 5.1W
  • Efficacy 96 lm/W
  • CCT 2979K
  • CRI 82
  • Power factor 0.52

Contact – Aurora

British Electric Lamps 3000K, 6W

Bell slightly undersells itself inasmuch as the packaging indicates 6W consumption whereas we measured 4.65W. However, the power factor was 0.52 and the measured lumen output 358 lm, just above the minimum 345 lm allowed by DIM2.

Ra is 82 per cent.

  • Output 358 lm
  • Power 4.7W
  • Efficacy 76 lm/W
  • CCT 3073K
  • CRI 82
  • Power factor 0.52

Contact – British Electric Lamps

Integral LED 2700K, 7.5W

Integral’s 7.5W lamp performed almost exactly as claimed on the box. 7.4W versus 7.5W claimed and 499 lm instead of 500. That’s what you call tight manufacturing tolerances. The CRI was 80 and the measured power factor was 0.76, better than most. The packaging quotes both total lumens and those in a 90-degree beam. 

  • Output 499 lm
  • Power 7.4W
  • Efficacy 67 lm/W
  • CCT 2633K
  • CRI 80
  • Power factor 0.76

Contact – Integral LED


Kosnic LED

This is a good little lamp. It quotes both total and 90-degree lumens on the package, 480lm and 450lm respectively. It’s actually an understatement because we measured a total 505 lm. 7W consumption is claimed but our sample was 5.6W. Bearing in mind it is 2700K, the measured 90 lm/W is remarkable. It has a 38-degree beam with quite a soft edge. 

  • Output 505 lm
  • Power 5.6W
  • Efficacy 90 lm/W
  • CCT 2723K
  • CRI 83
  • Power factor 0.53
  • A nice little lamp


Contact – Kosnic 

Megaman 2800K, 7W

This particular 7W lamp consumed eight per cent more power and emitted nine per cent less light than the claims on the box indicated. The CRI was a whisker over 80. The CCT was pretty much as claimed at 2879K compared with a quoted 2800K. The power factor was a respectable 0.67.

  • Output 500 lm
  • Power 7.6W
  • Efficacy 66 lm/W
  • CCT 2879K
  • CRI 82
  • Power factor 0.67

Contact – Megaman 

Osram Parathom 2700K, 5.3W

As you would expect from Osram, this lamp performed almost exactly as stated on the box – 5.15W, 2688K, 355 lm against a stated 350 lm. The power factor at 0.88, was the highest we tested, so the electricians won’t have to worry about fuse sizes. In a way, Osram undersell themselves because they don’t quote total lumens on the box. A non-specialist – i.e. 99 per cent of purchasers – might think it doesn’t perform so well as the cheaper competition. 

  • Output 355 lm
  • Power 5.2W
  • Efficacy 68 lm/W
  • CCT 2688K
  • CRI 81
  • Power factor 0.88

Contact – Osram 

Philips MasterLED Spot 4000K, 5.5W

We tested the Master LEDspot MV, which is nominally a 5.5W unit. The measured wattage on the samples was 5.4W with a PF of 0.66. This lamp performed slightly better than the claims on the packaging, with a lumen output that was 8 per cent higher. It also had an Ra of 86; which is more than most products of this type.

  • Output 417 lm
  • Power 5.4W
  • Efficacy 77 lm/W
  • CCT 3890K
  • CRI 86
  • Power factor 0.66


Soraa Brilliant 3000K, 7.5W

Although this lamp had the highest wattage, it also had almost the highest lumen output and CRI. We tested the 10-degree narrow spot and this produces a good, clean beam with a clearly defined centre. This lamp is one of the Brilliant series, with a CRI of 80+ (Soraa is better known for its Vivid range, with CRIs above 95). There is also a useful range of clip-on lenses and accessories which can change the beam width or colour temperature.

  • Output 503 lm
  • Power 7.7W
  • Efficacy 66 lm/W
  • CCT 2834K
  • CRI 85
  • Power factor 0.78

Contact – Soraa 

Sylvania 5W RefLED ES50

This is a really good lamp. As you would expect, all the information is on the box. The power factor is much better than claimed and was the second highest we tested. The total measured output was 413 lm and this bumps its efficacy to third best. However, forget about the figures, what sets this lamp apart, both on and off, is its appearance. A lot of effort has gone to making it look as much as possible like a conventional halogen lamp. The front face really sparkles when you switch it on. If you care about lighting, this is the lamp for you. 

  • Output 413 lm
  • Power 5.2W
  • Efficacy 79 lm/W
  • CCT 2768K
  • CRI 82
  • Power factor 0.83
  • BEST OVERALL: Good quality and great looking

Contact – Sylvania

Verbatim 2700K, 6W

The packaging says it is equivalent to a 57W lamp, but certainly the 465 lm output (in a 90-degree cone) means it easily betters a 50W halogen GU10. You can see from the table that the efficacy is one of the best we tested. Cooler lamps would be even more efficient than the 2700K version we tested. 

  • Output 465 lm
  • Power 5.6W
  • Efficacy 83 lm/W
  • CCT 2697K
  • CRI 82
  • Power factor 0.66

Contact – Verbatim

V-Tac COB Spotlight

Although described as a spotlight, the beam is 110 degrees which could be misleading to the non-specialist – it would light from floor to ceiling if it was 1.5m away. The package claims equivalence to a 50W GU10 but the 90-degree beam lumen figure isn’t on the box, so there is no way of knowing. We measured 423 lm and 5.4W rather than the 450 lm and 6W claimed.

  • Output 423 lm
  • Power 5.4W
  • Efficacy 78 lm/W
  • CCT 2838K
  • CRI 81
  • Power factor 0.55
  • Data could be clearer

Contact – V-Tac



See the latest in lighting technology at LuxLive.

The show takes place on Wednesday 13 November and Thursday 14 November 2019 at ExCeL London. Be a part of Europe’s largest annual lighting event. Entry is free if you pre-register HERE.